3380_PSG_R05_Aug1989 3380 PSG R05 Aug1989
User Manual: Pdf 3380_PSG_R05_Aug1989
Open the PDF directly: View PDF 
.
Page Count: 404
| Download | |
| Open PDF In Browser | View PDF |
•
If.'. '
C
F
L
-:==;.-- - -----c -- --,---
1'.
.../
~-
c
(
{
(
(~
(
c
-
PSG
-®
('
PCM*
MIM
INTRO
MLX
START
ISOLATE
REPAIR
MD
LOG
PROG
SC-FRU
INDEX
(MASTER)
TRANS
LGND/GLOS
LOC
PWR
CARR
MAP
SAFETY
INDEX
VOL R05
VOL. R10
VISUAL
INDEX
CATALOG
NUMERIC
INDEX
VOL. R20
MOM
MICROFICHE
CARD/CABLE/
VOLTAGE
CHARTS
A-BOARD LRM
B-BOARD LRM
C-BOARD LRM
POWER
DIAGRAMS
RD
INST
VOL. R30
ECM*
ECDs
IND·EX
VOL. R40
MSM*
HELP
SENSE
OPER
DIAG
INDEX
VOL. R60
* These volumes are shipped in microfiche and are located in volume R30.
3380
(
(
Maintenance Library
PIN 4519888 2 Sides
EC 476581
01Aug89
© Copyright IBM Corporation 1987
Direct Access Storage
Models J and K
Maintenance Information
FRONT-'
The drawings and specifications contained in this manual shall not be reproduced in. whole or in part
without written permission from the IBM Corporation.
".
"
I
)
IBM has prepared this maintenance manual for the use of IBM service representatives in the installation,
maintenance, and repair of the specific machines indicated. IBM makes no representation that it is suitable
for any other purpose.
Information contained in this manual is subject to change. Any such change will be included in following
revisions.
This material may contain reference to, or information about, IBM products (machines and programs),
programming, or services .that are not announced in your coun~.I. Such references or information does not
necessarily mean that IBM intends to announce such IBM products; programming, or services in your
country.
Requests for copies of IBM publications should be made to your IBM representative or to the IBM branch
office servicing your locality.
A form for reader's comments is supplied at the back of this manual. If the form has been removed, send
comments to:
International Business Machines Corporation
Department G26/61C
5600 Cottle Road
San Jose, California 95193
You can use this form to communicate your comments about this manual, its organization, or its subject
matter with the understanding that IBM can. use and distribute the information you supply in any way it
believes correct without any obligation to you.
PIN 4519888 02 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987. 1988
FRONT-2
,.,.
i . )
'\
/
./
Preface
Maintenance Manual Ordering Procedure
The Product Service Guide (PSG). Volume R05 of
the maintenance library. is designed to help the
IBM service representative maintain the IBM 3380
Direct Access Storage. Models AJ4. BJ4. AK4. and
BK4.
A Documentation Request form (order number
972-8390) is located in the microfiche pocket (see
the MDM. Volume R30). Use this request form to
order microfiche or printed volumes of the
maintenance manuals. Only the most current
version of any manual will be sent. No automatic
updates to any man'Jal ordered by this card will be
sent.
When maintaining any of these models. use the
IBM Maintenance Device diskette and the
maintenance library manuals shipped with the
machine. and/or the manuals shipped with the AJ4
or AK4 unit to which the machine and the IBM
Maintenance Device connect.
Always start any maintenance action on a 3380
Model AJ4. BJ4. AK4. or BK4 by following the
instructions in the PSG. Volume R05.
c-
When a 3380 model AJ4 or AK4 is connected to the
same storage control as a 3380 Model AA4. AD4. or
AE4. instructions in the MD or notes in this manual
indicate when to use the information in the other
3380 maintenance library manuals.
In this manual. the following terms are used to
simplify the distinction among the various 3380
models:
3380
3380-J
3380-K
3380-JK
A unit
B unit
c-
For USA
Parts of this manual can be ordered from the IBM
San Jose plant by using the Wiring Diagram/Logic
Page Request. Order No. Z150-0130 (U/M 015). In
the logic-page columns, enter the part number [1]
and engineering change [2] number shown at the
bottom of each page. For example.
PN 1234567
EC 123456
= all 3380 models
= 3380 AJ4. BJ4
= 3380 AK4. BK4
= 3380 AJ4. BJ4. AK4.
= 3380 AJ4, AK4
= 3380 BJ4, BK4
It is recommended that one hard copy set of all
microfiche manuals be ordered for each account.
Because a machine serial number is not required
for ordering, ECs affecting the contents of these
manuals will update the microfiche copies only.
The customer engineer will be instructed by the EC
to reorder new printed support documentation.
BK4
The complete model name will be used to specify a
single model. (For example. 3380 Model BJ4.)
[1]
[2]
PRE-1
Whole volumes can be ordered by including the
volume number and the machine serial number.
For Europe, Middle East, and Africa
(EMEA)
Wiring diagram and logic page requests are
handled through the MLC plant in Mainz, Germany.
Printed manuals can be ordered by sending the
Documentation Request form (found in the
microfiche pocket) to the MLC plant in Mainz,
Germany.
('
('
((
PIN 4519890 08 Sides EC 475248 25Apr88
~ Copyright IBM Corp. 1987. 1988
PRE-1
Subsystem Microcode Ordering
Information
Diskette Ordering Procedure
Note: When ordering a diskette to replace a
defective diskette, the EC number of the
defective diskette must be specified.
Otherwise, the latest level diskette is
shipped.
Subsystem microcode diskettes must always be at
compatible levels with each other and with the
supporting maintenance package documentation.
Updates to these diskettes will be through EC
control procedures.
3380-JK/3880 Subsystem: This subsystem
configuration requires three microcode diskettes as
follows:
To order a diskette for the 3380-JK, supply the
following information indicated for that diskette:
•
Storage Control functional diskette
•
3380-JK/3880 subsystem functional microcode
diskette.
Machine Type =
Serial# = xxxxx
•
3880 MD diskette.
GID# = xxxxxx
EC# = xxxxxx
•
3380-JK MD diskette.
PIN
3380-JK/3990 Subsystem: This subsystem
configuration requires two microcode diskettes as
follows:
3380-JK MD diskette.
•
3380-JK/3990 subsystem functional microcode
diskette.
xxxx
= xxxxxxx
-,/
•
•
\
j
3880 MD Diskette (if the 3880 Model AJ4 or AK4
unit is attached to a 3880)
PIN = 4518573 (No other data required)
•
3380-JK MD Diskette
PIN
EC#
= 4516289
= xxxxxx
j
PRE-2
PIN 4519890 08 Sides EC 475248 25Apr88
ro Copyright IBM Corp.
1987, 1988
~
(i-.
(:
('(~
(-
How To Update This Manual
This manual is under engineering change control.
Put pages in by part number and page number.
The EC level and date are given in the EC history
block on the last page of the part-numbered group
of pages. For example,
EC History of Front Matter
EC HISTORY OF P/N 4519890
EC Number
Date Of EC
EC Number
475245
14Nov86
475248
475246
21Ju187
475247
llSep87
('
(
(
f
Related Publications
•
IBM 3990 Storage Control Introduction,
GA32-0098
The following is a list of documents that may help
to understand and/or repair the 3380:
•
IBM 3990 Storage Control Reference Manual,
GA32-0099
•
Environmental Record Editing and Printing
Program: User's Guide and Reference,
GC28-1378
•
Device Support Facilities User's Guide and
Reference Manual, GC35-0033-12 or later
•
Maintaining IBM Storage Subsystem Media,
GC26-4495
•
Device Support Facilities Primer for the User of
IBM 3380 Direct Access Storage, GC26-4498.
•
Device Support Facilities Primer for the User of
IBM 3380 Direct Access Storage, GC26-4498
•
Storage Subsystem Library Master Index,
GC26-4496
•
IBM 3380 Direct Access Storage General
Information, GA26-1670.
•
IBM 3380 Direct Access Storage Introduction,
GC26-4491
•
IBM 3380 Direct Access Storage User's Guide,
GA26-1671.
{
(
•
(
•
(
(("
Date Of EC
25Apr88
•
•
Using the IBM 3380 Direct Access Storage in an
MVS Environment, GC26-4492
Using the IBM 3380 Direct Access Storage in a
VM Environment, GC26-4493
Using the IBM 3380 Direct Access Storage in a
VSE Environment, GC26-4494
IBM 3880 Storage Control, Models 1,2,3, and 4
Reference Manual, GA26-1661
(,
(PIN 4519890 08 Sides EC 475248 25Apr88
to Copyright
IBM Corp. 1987. 1988
PRE-3
About The Maintenance Library
PWR
The Power section contains information
concerning power distribution and operation,
voltage checks, power failures, and power
maintenance analysis procedures.
The Maintenance Library for the 3380 Model AJ4
and AK4, and for attached 3380 Model BJ4 and BK4
units, is comprised of the following manuals:
•
Maintenance Information Manual, Volume R 10
•
Parts Catalog Manual, Volume R20
CARR
The Checks, Adjustments, Removals, and
Replacements section supplies procedures to
check, adjust, remove and replace
assemblies in a 3380.
•
Maintenance Diagrams Manual, Volume R30
MAP
•
Error Condition Manual, Volume R40
•
Maintenance Support Manual, Volume R60
•
Product Service Guide, Volume R05
Pro~uct
Service Guide, Volume R05
The Product Service Guide, Volume R05 (PSG) is
designed to teach you how to service the 3380
Models AJ4, BJ4, AK4, and BK4. For a description
of the Product Service Guide, see "About This
Manual" on page PRE-6.
"J1
\
/
/
The Maintenance Analysis Procedure section
contains maintenance procedures for detailed
failure analysis. The MAP section is entered
from the IBM Maintenance Device (MD).
SAFETY
The Safety section is a safety inspection
guide: An aid in identifying unsafe conditions
on 3380 machines. The SAFETY section
ensures that all safety equipment, including
safety labels, are installed.
Parts Catalog Manual, Volume R20
Maintenance Information Manual, Volume R10
The Maintenance Information Manual, Volume R10
(MIM) contains information about the failure
symptoms for the 3380.
The Parts Catalog Manual, Volume R20 (PCM)
contains diagrams, descriptions, and part numbers
of each field replaceable part and assembly in the
3380-JK subsystem.
TRANS
The Translation section contains the Danger
and Caution statements used in the 3380-JK
maintenance library, translated into the
necessary World Trade languages.
VISUAL INDEX
The Visual Index section contains summary
diagrams of the 3380-JK, defining the
assembly name and the catalog figure
number where the assembly can be found.
LGND/GLOS
The Legend/Glossary section defines the
graphic symbols, abbreviations, acronyms,
and terms used in the 3380 Maintenance
Library.
CATALOG
The Catalog section contains detailed
diagrams of each field replaceable assembly
in the 3380-JK, with lists of part numbers and
descriptions of each part.
LOC
NUMERIC INDEX
The Numeric Index section is an index of the
part numbers of field replaceable assemblies
and parts in the 3380-JK.
The Location section shows the location of
the assemblies in a 3380.
PRE-4
PIN 4519890 08 Sides EC 475248 25Apr88
@ Copyright IBM Corp. 1987, 1988
---
----------
/
""
...•..
~.
c
(
('
('
Maintenance Diagrams Manual, Volume R30
Error Condition Manual, Volume R40
The Maintenance Diagrams Manual, Volume R30
(MDM) contains the schematics and logic
diagrams. It also contains the installation
instructions for the 3380-JK subsystem.
The Error Condition Manual, Volume R40 (ECM) is
used with the PSG to service the 3380.
CARD/CABLE/VOLTAGE CHARTS
This section contains charts for the cards,
cables, and voltages.
B-BOARD LRM
The B-Board Logic Reference section
contains the logic index and logic charts for
the B-board.
(
(
C-BOARD LRM
The C-Board Logic Reference section
contains the logic index and logic charts for
the C-board.
POWER DIAGRAMS
The Power Diagrams section contains the
power logic diagrams.
INST
('
The Error Condition Descriptions section
describes each circuit or function whose
operation is checked or verified. Topics (for
each circuit or function) include "How the
check is indicated" and "How and when it
works." Additional reference material,
diagnostics, and service aids for each
symptom code are also supplied.
MICROFICHE
The Microfiche section contains the PCM,
ECM, and MSM manuals in microfiche form.
It also contains the order card for ordering a
printed copy of the PCM, ECM, and MSM
manuals.
A-BOARD LRM
The A-Board Logic Reference section
contains the logic reference master index, the
A-board logic index. and logic charts.
(
ECDs
The Installation section describes the
procedures to install, relocate, and remove a
3380. It also includes an installation checklist.
Maintenance Support Manual, Volume R60
The Maintenance Support Manual, Volume R60
(MSM) is used with the PSG to service the 3380.
HELP
The purpose of the Help section is to supply
the customer engineer with DASD
maintenance information not supplied in the
PSG or the MIM.
SENSE
The Sense section contains the 3380 sense
data descriptions which identify the contents
of each format by messages, bytes, and bits.
It supplies page references to additional
information on each error symptom along
with the priority of service actions and the
primary symptom code.
OPER
The Operation section contains the theory of
operation for the 3380.
DIAG
The Diagnosis section contains the run
procedures for each diagnostic routine, along
with isolation codes associated with the
routine.
(~
(
("
PIN 4519890 08 Sides EC 475248 25Apr88
Cl Copyright IBM Corp, 1987, 1988
PRE-S
About This Manual
ISOLATE
The Product Service Guide, Volume R05 (PSG) is
used with the Maintenance Information Manual,
Volume RIO (MIM) when servicing the 3380-JK
subsystem:
The Isolate section contains information about:
The PSG is divided into ten sections, each with
specific purposes. A brief description of each
section follows:
INTRO
The Introduction section describes the correct
service procedures for the IBM 3380 Direct Access
Storage, Models AJ4, BJ4, AK4, and BK4. In
addition, the INTRO section describes the 3380
device attributes, Miscellaneous Equipment
Spe,eifications, storage controller interface
information, emergency procedures, microcode and
documentation ordering procedures, and, tools and
test equipment information.
•
•
•
•
•
Problems found during physical inspection.
Controller and machine checkout procedures.
Problem analysis with Error Log data.
Problem analysis using Console messages.
Problem analysis using ICKDSF messages.
\
The ISOLATE section helps isolate a problem to a
Field Replaceable Unit (FRU) or to the smallest
group of FRUs possible.
REPAIR
The Repair section contains information about:
•
•
•
•
•
FRU (card) replacement procedures.
Power offlon procedures.
Post-repair checkout procedures.
End of guided maintenance procedures.
End of call procedures.
MLX
The Maintenance Library Cross-Reference (MLX)
section is the connection between this maintenance
library and other machine maintenance libraries.
The REPAIR section includes charts of FRUs,
including a Repair Affects column to help you
determine the portion of the machine or string that
will be affected by replacement of the logic card.
START
MD
The Start section contains information about:
The Maintenance Device (MD) section describes
how to connect and use the MD. It also contains
explanations of MD messages.
•
•
•
•
•
•
How to plan a repair action with minimum
impact on the customer.
How to determine if corrective action is
required.
How to prioritize the failure.
How to locate the failing string.
How to determine the failing FRUs.
How to schedule the hardware for
maintenance.
The START section contains Maintenance Analysis
Procedures (SMAPs), organized to guide you in
diagnosis and repair. Complete the procedures in
the START section before going to the ISOLATE
section.
PRE·6
j
LOG
The LOG section describes the organization and
use of the error log, which is maintained in the
storage control. Data from the log can be retrieved
through the MD. The LOG section contains
operating instructions and options. The LOG
section contains information concerning the error
logs, which are equipment check logs, whose
functions are transfer, merge, search, erase, and
change log mask.
PIN 4519890 08 Sides EC 475248 25Apr88
~ Copyright IBM Corp. 1987, 1988
((
c'
(
PROG
Master Index
The Program section contains information about
EREP System Exception Reports, which are needed
to maintain the 3380.
The Master Index section in this volume (R05)
contains an alphabetical list of items contained in
the 3380-JK DASD maintenance library manuals
(volumes R05, R10, R30 (INST section)! R40, and
R60). Use the Master Index section to find items in
this manual and in the other 3380-JK DASD
maintenance manuals.
...
.
,
,'.
(
(
SC-FRU
The Symptom Code to Field Replaceable Unit
(SC-FRU) section contains tables that relate
symptom codes to FRUs. The SC-FRU tables in the
PSG are for reference only, and should not be used
to determine field replaceable units (FRUs) when
the IBM Maintenance Device is available.
('
(
(
('
(
('
(
(
(
('
c'
PIN 4519890 08 Sides EC 475248 25Apr88
10 Copyright
IBM Corp. 1987, 1988
PRE-7
----~-
...
---.~--
.. - - - - .
EC History of Front Matter
EC HISTORY OF P/N 4519890
Date Of EC
EC Number
EC Number
475245
14Nov86
475246
21Ju187
llSep87
475247
475248
"\
Date Of EC
25Apr88
)
Notes:
/
/
PRE-8
PIN 4519890 08 S,des EC 475248 25Apr88
~ Copyright IBM Corp_ 1987, 19S8
---.-------~-
(:
(1'
,;1,:
"
Contents
(
(:.-
(
(-
(
(
(
(
', ..
"I:
(
Introduction ......................
About the 3380-JK .................
Miscellaneous Equipment Specifications
(MES) ........................ .
Selective Feature Codes ......... .
MES Summary ................. .
3380-JK Connected to a 3880 Storage
Control ....................... .
3880 Storage Control Requirements ..
Subsystem Configurations ........ .
Example of 3380-JK Strings Attached
To a 3880 Storage Control ....... .
3380-JK String and Port Configurations
(3880 Subsystem) .............. .
Dynamic Path Selection (DPS) ..... .
Device Level Selection (DLS) ...... .
Control Interface Fencing ......... .
Power Sequencing .............. .
Front Operator Panel (3880 Subsystem)
Front Operator Panel Labels ...... .
Rear Unit Emergency Power Off Panel
(3880 Subsystem) ............... .
Unit Emergency Switch ......... .
Controller Ready LEDs ......... .
Maintenance Device (MD) Connector
Rear Unit Emergency Power Off Panel
Labels ..................... .
3380-JK Connected to a 3990 Storage
Control ...................... .
3990 Storage Control Requirements .
Subsystem Configurations ....... .
Example of 3380-JK Strings Attached
To a 3880 Storage Control ...... .
3380-JK String and Port
Configurations (3990 Subsystem) ..
3380-JK 2-Path String Configuration
3380-JK 4-Path String Configuration .
3380-JK Port Configurations (3990
Subsystem) ................. .
Dynamic Path Selection (DPS) .... .
Device Level Selection (DLS) ..... .
Four Path .................... .
Control Interface Fencing ........ .
Subsystem Fencing ............ .
Quiesce / Resume Storage Path .. .
Quiesce and Resume Controller .,.
Power Sequencing ............. .
PIN 4519891 06 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
INTRO-1
INTRO-1
INTRO-2
INTRO-2
INTRO-2
INTRO-3
INTRO-3
INTRO-3
INTRO-4
INTRO-5
INTRO-6
INTRO-6
INTRO-6
INTRO-6
INTRO-?
INTRO-9
INTRO-10
INTRO-10
INTRO-11
INTRO-11
INTRO-11
INTRO-12
INTRO-12
INTRO-12
INTRO-12
INTRO-14
INTRO-14
INTRO-15
INTRO-17
INTRO-18
INTRO-18
INTRO-18
INTRO-18
INTRO-18
INTRO-19
INTRO-19
INTRO-19
Front Operator Panel (3990
Subsystem) ................. . INTRO-20
Front Operator Panel Labels ..... . INTRO-23
Rear Unit Emergency Power orf Panel
(3990 Subsystem) ............... . INTRO-24
Unit Emergency Switch ......... . INTRO-24
Controller Ready LEDs ......... . INTRO-25
Maintenance Device (MD) Connector
INTRO-25
Rear Unit Emergency Power Off Panel
Labels ..................... . INTRO-25
Device Characteristics Summary .... . INTRO-26
Track Layout - Models AJ4 and BJ4 . INTRO-2?
INTRO-28
Track Layout - Models AK4 and BK4
Disk, Arm, and Head Layout ..... . INTRO-29
Servo System ................ . INTRO-30
Data Check Detection and Correction INTRO-30
Switches and Indicators ........... . INTRO-30
A 1 Board Voltage and Clock LEDs .. INTRO-30
B1 Board Power Card LEDs ...... . INTRO-31
B1 Board - PES/PROM Card LEDs .. INTRO-31
Phase Rotation LEDs ........... . INTRO-31
Controller Voltage Regulator Card
LEDs ....................... INTRO-32
Controller Power Supply Circuit
Pru£ectors ................... INTRO-32
24 Vdc Panel - A and B units ...... INTRO-32
Single Actuator Maintenance (SAM)
Control Panel - A and B Units .... INTRO-34
Undervoltage Detection ............ INTRO-36
Service Call Strategy .............. INTRO-36
EREP and Console Messages ..... INTRO-3?
Physical Identifiers ............. INTRO-37
Maintenance Device (MD) ........ INTRO-3?
Error Log ..................... INTRO-38
Diagnostics ................... INTRO-38
Analysis Procedures ............ INTRO-38
FRU Replacement .............. INTRO-38
Subsystem Microcode Ordering
Information .................. INTRO-38
Documentation Ordering Procedure ... INTRO-39
Tools and Test Equipment .......... INTRO-40
EC History of Introduction Section .... INTRO-41
MLX .............................. MLX-1
Entries from Other Maintenance Libraries . MLX-2
Exits to Other Maintenance Libraries ..... MLX-3
EC History of MLX Section ............. MLX-4
Start Procedures .................. START-1
SMAP 0010: Start ................ START-1
SMAP 0015: Isolate a Problem ...... START-2
SMAP 0016: Using EREP to Determine
IfThere Are 3380 Failures ........ START-4
Contents-1
SMAP 0017: Guidelines for Periodic
Review of EREP Reports ......... START-6
SMAP 0020: Corrective Action
Determination With EREP ......... START-8
SMAP 0025: Prioritizing Symptom
Codes ...................... START-12
SMAP 0030: Locating the Failing
String and Devices ............ START-14
SMAP 0033: Oetermine the Probable
Failing Unit .................. START-16
SMAP 0035: Corrective Action
Determination With Sense Data ... START-17
SMAP 0040: Prioritizing Symptom
Codes ...................... START-19
SMAP 0045: Locating the Failing
String and Devices ............ START-21
SMAP 0050: Planning the Repair
Action for Data Checks ......... START-23
SMAP 0051: Determining Tracks
Reporting Data Checks ......... START-25
SMAP 0052: Corrective Action
Determination from Device Support
Facilities Messages ............ START-27
SMAP 0053: Locating the Failing Unit
from Device Support Facilities
Messages ................... START-30
SMAP 0054: Prioritizing Data Check
Symptom Codes using Device
Support Facilities Messages ..... START-31
SMAP 0056: Prioritizing Data Check
START-32
Symptom Codes using Sense Data
SMAP 0057: Determining the Failing
Ports (EBxx) .................. START-33
SMAP 0058: Determining the Failing
FRUs ....................... START-34
SMAP 0060: Scheduling the Hardware START-35
SMAP 0070: Locate the Failing
Machine with Console Messages .. START-36
EC History of Start Section .......... START-38
Isolate Procedures ............... ISOLATE-1
IMAP 0200: Isolate the Problem ... ISOLATE-1
IMAP 0205: Physical Inspection ... ISOLATE-2
IMAP 0210: Miscellaneous Physical
Symptoms .................. ISOLATE-6
IMAP 0212: Checking For
Undervoltage ................ ISOLATE-7
IMAP 0215: Running Diagnostics .. ISOLATE-8
IMAP 0218: Controller Diagnostics . ISOLATE-9
IMAP 0220: Normal Sequence
Diagnostics ................ ISOLATE-12
PIN 4519891 06 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
IMAP 0225: Trouble Setting CE
Mode .................... -. ISOLATE-14
IMAP 0228: Undervoltage With Data
Che.cks .................. . ISOLATE-15
IMAP 0230: Diagnostics for Data
Checks .................. . ISOLATE-16
IMAP 0232: Determine the Failing
Devices and Controllers with the
Device Status Test .......... . ISOLATE-18
IMAP 0233: Determine the Failing
Devices and Controllers with the
Error Log ................. . ISOLATE-20
IMAP 0235: Symptom Code Analysis ISOLATE-23
IMAP 0240: Problem Analysis With
Error Log Data ............. . ISOLATE-24
IMAP 0245: Device Not Ready .. . ISOLATE-27
IMAP 0247: Device Switch Enabled
LED Off ....... .' .......... . ISOLATE-30
IMAP 0248: Device Sequencer
Status LED Off ............. . ISOLATE-31
IMAP 0250: Controller Not Ready . ISOLATE-32
IMAP 0252: Status Pending LED Is
On Continuously ........... . ISOLATE-33
IMAP 0255: Console Messages .. ISOLATE-34
IMAP 0260: Console Message Intervention Required ....... . ISOLATE-35
IMAP 0265: Console Message Condition Code 1 ........... . ISOLATE-38
IMAP 0270: Console Message Condition Code 3 ........... . ISOLATE-42
IMAP 0275: Console Message Missing Channel End and Device
End (Missing Interrupt) ....... ISOLATE-46
IMAP 0280: Console Message Write Inhibited .............. ISOLATE-49
IMAP 0285: Console Message Volume Fenced from Storage Path ISOLATE-51
IMAP 0290: Miscellaneous Console
Messa,ges ................. ISOLATE-53
IMAP 0295: Controller Not
Connected Correctly ......... ISOLATE-55
EC History of Isolate Section ...... ISOLATE-56
Repair Procedures ................ REPAIR-1
RMAP 0300: FRU Replacement
Procedures .................. REPAIR-2
RMAP 0310: Replacing FRUs ...... REPAIR-5
RMAP 0320: Post-Repair Checkout .. REPAIR-8
RMAP 0330: Post-Repair Controller
Checkout ................... REPAIR-11
RMAP 0340: End of Guided
Maintenance ................ REPAIR-13
. Contents-2
"'~_~
-"
j
"
./
(,
(.":~
f
(
"
{
-'
--.-~/
j
C~
(,
('\
(~:
(
('
(~
(
(~:
(~
('
RMAP 0350: End of Call ......... REPAIR-15
DDC/DTB Card (A1W2 and A1B2)
Switch Settings (2-Path) .......... REPAIR-18
String Address ................ REPAIR-18
DDC/DTB Card (A1W2 and A1B2)
Switch Settings (4-Path) .......... REPAIR-19
String Address ................ REPAIR-19
1/0 Control Card (A1V2 and A1C2)
Switch Settings (String ID) ........ REPAIR-20
String ID .................... REPAIR-20
1/0 Control Card (A1V2 and A1C2)
Switch Settings (2-Path String
Configuration) ................. REPAIR-21
String Configuration: Switches 1 and
2 ......................... REPAIR-21
1/0 Control Card (A1V2 and A1C2)
Switch Settings (4-Path String
Configuration) ................. REPAIR-22
String Configuration: Switches 1 and
2 ......................... REPAIR-22
String Configuration: Switches 3 and
4 ......................... REPAIR-22
Power OnlOff Procedures ......... REPAIR-23
24 Vdc Panel for A and B units ... REPAIR-23
Single Actuator Maintenance Control
Panel for A and B Units ........ REPAIR-24
String Power Off Procedure ...... REPAIR-25
String Power On Procedure ...... REPAIR-25
Unit Power Off Procedure ....... REPAIR-26
Unit Power On Procedure ....... REPAIR-26
Controller Power Off Procedure ... REPAIR-27
Controller Power On Procedure ... REPAIR-27
Drive Power Off Procedure ...... REPAIR-28
Drive Power On Procedure ...... REPAIR-28
Device Power Off Procedure ..... REPAIR-29
Device Power On Procedure ..... REPAIR-29
EC History of Repair Section ....... REPAIR-30
Maintenance Device (MD) Operation ..... MD-1
MD Menus ....................... MD-2
Maintenance Device (MD) Display Reference Number Table ........... MD-3
Prepare the MD ................... MD-5
Set/Reset CE Mode (MD Main Menu
Option 0) ....................... MD-7
Run Diagnostic Tests (MD Main Menu
Option 1) ....................... MD-10
Diagnostic Menu Option 0 - Normal
Sequence ...................... MD-11
Diagnostic Menu Option 1 - Controller
Test ........................... MD-13
PIN 4519891 06 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
Diagnostic Menu Option 2 - Extended
DPS Test .......................
Diagnostic Menu Option 3 - Device Test
Diagnostic Menu Option 4 - Connection
Test ............. ,..............
Diagnostic Menu Option 5 - HA Scan ...
Display Modes ....................
Diagnostic Menu Option 6 - Motor Test .
Diagnostic Menu Option 7 - Device
Status .........................
Diagnostic Menu Option 8 - CE Track
Repair .........................
Diagnostic Menu Option 9 - Storage
Control Status ...................
Diagnostic Menu Option A Non-Repeatable Runout (NRRO) Test
Isolation Code Error Display ........ .
Analyze Symptom Code (Main Menu
Option 3) ...................... .
Diagnostic Aids (Main Menu Option D) .
Error Display .................... .
Review Option List (Main Menu Option
R) ........................... .
010 SELECT AN OPTION ........... .
020 SELECT CONTROLLER ......... .
030 SET I RESET CE MODE ......... .
035 DEVICES IN CE MODE/DEFAULT
DEVICES ...................... .
050 DEFINE THE FAILURE BOUNDARY .
MD 0050: ....................... .
051 DEFINE THE FAILURE BOUNDARY
(EBXX) ........................ .
MD 0051: ....................... .
052 DEFINE THE FAILURE BOUNDARY
(DATA CHECKS) ................ .
MD 0052: ....................... .
090 WRITE INHIBIT ............... .
092 STORAGE PATH FENCED ....... .
093 FENCED DEVICES ............. .
100 DIAGNOSTIC MENU ........... .
200 SELECT CONTROLLER ......... .
205 INVALID CONTROLLER SELECTED .
210 SELECT DEVICE(S) ............ .
215 ~NVALID DEVICE SELECTED ..... .
310 DO NOT FORGET TO RESET CE
MODE ........................ .
320 DO YOU WANT TO RUN
CONNECTION TESTS? ............ .
330 WHICH CONTROLLER CONNECTION
DO YOU WANT TO TEST? ......... .
MD-14
MD-15
MD-17
MD-19
MD-23
MD-24
MD-26
MD-32
MD-34
MD-37
MD-41
MD-42
MD-44
MD-48
MD-49
MD-52
MD-52
MD-53
MD-54
MD-55
MD-55
MD-58
MD-58
MD-61
MD-61
MD-63
MD-63
MD-64
MD-64
MD-65
MD-65
MD-65
MD-65
MD-65
MD-66
MD-66
Contents-3
340 CONTROLLER AX CONNECTION
TESTS READY. . ................. . MD-67
350 DPS TESTS CANNOT BE RUN .... . MD-67
360 HA SCAN ..........•......... MD-67
395 SCAN COMPLETE / SCAN REVIEW
MD-70
400 SELECT ONE OF. THE FOLLOWING
ERROR LOG FUNCTIONS .......... . MD-71
405 SELECT CONTROLLER (ERROR
LOG) ......................... . MD-71
410 ENTER SEARCH ARGUMENTS ... . MD-72
415 ENTER LOG NUMBER .......... . MD-72
420 ENTER THE DATE ............. . MD-72
425 ENTER LOGS TO BE MERGED ... . MD-72
430 ENTER THE SENSE FORMATS TO BE
LOGGED ...................... . MD-73
440 ERROR LOG MASK/TRANSFER
STOPPED BY KEYBOARD ENTRY ... . MD-73
460 CONFIGURATION ............. . MD-73
462 CONFIGURATION DfoTA/UNKNOWN·
STORAGE CTRL ................ . MD-75
465 DEVICE STATUS .............. . MD-76
468 DEVICE TYPE ................. . MD-77
480 STORAGE CONTROL STATUS .... . MD-78
MD-78
500 ISOLATION CODE ERROR DISPLAY
550 ENTER THE SYMPTOM CODE ..... MD-79
605 WARNING: HDA CABLES SWAPPED MD-79
615 MD DOES NOT PROVIDE ANALYSIS
FOR DISPLAYED IC .............. . MD-81
700 SELECT REVIEW OPTION FROM
FOLLOWING LIST ............... . MD-81
742 PRESS ENTER TO ANALYZE, OR 'N'
TO DISPLAY NEXT ERROR ......... . MD-81
800 DIAGNOSTIC AIDS ............. . MD-82
900 CONTROL UNIT AND MD DISKS
ARE NOT COMPATIBLE ........... . MD-83
EC History of Maintenance Device Section
MD-84
Error Log Function (MD Main menu Option
2) ............................... LOG-1
Introduction ...................... LOG-1
Select Error Log Function ........... LOG-2
Transfer Log (Option 1) ............. LOG-4
Search Log (Option 2) .............. LOG-5
Merge Log (Option 3) ............... LOG-8
Change Log Mask (Option 4) ......... LOG-9
Erase Log (Option 5) .............. lOG-10
EC History of Log Section ............ LOG-11
EREP Reports .................... PROG-1
EREP System Exception Reports .... PROG-1
System Error Summary (Part 2) Report PROG-2
DASD Subsystem Exception Report .. PROG-5
DASD String Summary ........... PROG-9
PIN 4519891 06 Sides
EC 476581 O1Aug89
@ Copyright IBM Corp. 1987. 1988
DASD Informational Messages ... .
DASD Data Transfer Summary .... .
DASD Symptom Code Summary .. .
Event History Report ........... .
EC History of Program Section ...... .
Symptom Code to FRU Charts ...... .
How to Use the SC-FRU Charts ... .
SC=4XYY (4X= Format 4 Byte 07,
YY = Format 4 Byte 23) ........ .
SC = 90XX (XX = Format 1 Byte 10)
SC = 91 XX (XX = Format 1 or 9 Byte
11) ........................ .
SC = 92XX (XX = Format 9 Byte 12)
SC=93XX (XX= Format 9 Byte 13)
SC = 94XX (XX = Format 9 Byte 14)
SC=95XX (XX= Format 1 Byte 15)
SC=97XX (XX= Format 1 Byte 17)
SC=98XX (XX= Format 9 Byte 18)
SC = 9CXX (XX = Format 1 Bytes
11-15) .................... .
SC=9DXX (XX= Format 1 Byte 15)
SC=9E10 (Format 1) .......... .
SC=9FXX (XX= Format 1 or 9 Byte
07) ....................... .
SC=9FXX (Format 1 and 9) ..... .
SC = D1XX (XX = Format 7 Byte 11)
SC=D2XX (XX= Format 7 Byte 12)
SC=D3XX (XX= Format 7 Byte 13)
SC = D4XX (XX = Format 7 Byte 14)
SC = D5XX (XX = Format 7 Byte 15)
SC = D6XX (XX = Format 7 Byte 16)
SC=D8XX (XX= Format 7 Byte 18)
SC = DEXX (XX = Format 7 Byte 11)
SC = DFXY (X = Format 7 Byte 10
Bits 0-3, Y= Byte 7 Bits 4-7) ... .
SC=DFXX (Format 7) ......... .
SC=EOXX (XX= Format 8 Byte 10)
SC=E1XX (XX= Format 8 Byte 11)
SC=E2XX (XX= Format 8 Byte 12)
SC=E3XX (XX= Format 8 Byte 13)
SC=E5XX (XX= Format 8 Byte 15)
SC=E6XX (XX= Format 8 Byte 16)
SC=E7XX (XX= Format 8 Byte 17)
SC=E8XX (XX= Format 8 Byte 18)
SC=EBXX .................. .
SC = ECXX (XX = Format 8 Bytes 11,
12,13,18) ................. .
SC = EDXX (XX = Format 8 Byte 09)
SC=EEXX (XX= Format 8 Byte 14)
SC = EFXX (XX = Format 8 Byte 07)
SC = EFXX (Format 8) .......... .
PROG-11
PROG-12
PROG-17
PROG-21
PROG-24
SC-FRU-1
SC-FRU-1
SC-FRU-2
SC-FRU-3
SC-FRU-3
SC-FRU-4
SC-FRU-4
SC-FRU-5
SC-FRU-6
SC-FRU-19
SC-FRU-22
,'-. .
SC-FRU-22
SC-FRU-23
SC-FRU-23
SC~FRU-24
SC-FRU-25
SC-FRU-26
SC-FRU-27
SC-FRU-28
SC-FRU-29
SC-FRU-30
...
SC-FRU-31
SC-FRU-33
SC-FRU-34
SC-FRU-35
SC-FRU-37
SC-FRU-37
SC-FRU-38
SC-FRU-39
SC-FRU-39
SC-FRU-40
SC-FRU-41
SC-FRU-41
SC-FRU-42
SC-FRU-45
SC-FRU-46
SC-FRU-46
SC-FRU-48
SC-FRU-49
SC-FRU-49
Contents-4
J
'.
/
c
( '"
-~
(
EC History of SC-FRU Section ...... SC-FRU-50
How to Use the Master Index .. MASTER INDEX-1
EC History of Master Index
..
MASTER INDEX-39
'\
./
(~,
(':
(:
(/
(~
•
c
(~~\
(~
(.':
./
PIN 4519891 06 Sides
EC 476581 01Aug89
~ Copyright IBM Corp. 1987, 1988
Contents-5
EC History of Table of Contents
EC HISTORY OF P/N 4519891
EC Number
Date Of EC
EC Number
475245
14Nov86
475248
475246
21Ju187
476573
llSep87
475247
476581
Date Of EC
25Apr88
30Nov88
01Aug89
Notes:
PIN 4519891 06 Sides
© Copyright
EC 476581 01Aug89
Contents-6
IBM Corp. 1987, 1988
\ .....
c
()
(j
Introduction
Before starting service procedures on page
START-1, read and become familiar with the
information contained in this section. The Product
Service Guide. Volume R05 is designed to teach
you how to service the 3380 Models AJ4. BJ4, AK4.
and BK4. Collectively. these models are referred
to in this manual as 3380-JK.
The first time you use this guide, carefully read and
follow the step-by-step procedures. As you gain
experience, you will not need to follow the PSG as
carefully.
(
()
()
C
Go to page
A
About this manual
PRE-6
B
About the product
INTRO-1
C
Front Operator Panel (3880 Subsystem)
Front Operator Panel (3990 Subsystem)
INTRO-7
INTRO-20
0
Service call strategy
INTRO-36
E
Ordering subsystem microcode
PRE-2
F
Tools and test equipment
INTRO-40
G
Subsystem Configurations (3880)
Subsystem Configurations (3990)
INTRO-5
INTRO-14
INTRO Section Entries
3380-JK consists of the following four basic models:
Each model of the 3380-JK contains two drives with
one head and disk assembly (HDA) per drive. Each
drive has two access mechanisms. Each model
therefore contains four access mechanisms.
Two A-unit models
AJ4
AK4
•
Failure to follow these instructions can cause
damage to the machine.
Select one of the following
About the 3380-J K
(/
When removing. replacing, or repairing any part of
this machine, follow directions. This ensures that
you use the correct remove. replace, or repair
procedure (including the correct power onloff
procedure) for this machine.
Entry
Figure 1.
•
It is important that you use the correct service
procedures, especially where personal safety
and/or possible machine damage is involved.
= A unit with two 1.26 gigabyte HDAs
= A unit with two 3.78 gigabyte HDAs
Two B-unit models
BJ4
BK4
= B unit with two 1.26 gigabyte HDAs
= B unit with two 3.78 gigabyte HDAs
Each access mechanism is associated with a
unique device address and operates independently
of other access mechanisms in the unit.
Models AJ4 and AK4 contain controllers as well as
access mechanisms. A controller. regardless of
3380 model. contains the logic needed to control
and transfer data to and from access mechanisms.
(C.~.
/
C
C
PIN 4519893 41 Sides EC 476573 3ONov88
@ Copyright IBM Corp. 1987, 1988
INTRO-1
{)
Miscellaneous Equipment Specifications (MES)
Selective Feature Codes
The A,.unit models may be ordered for attachment
in anyone of three Subsystem Configurations. The
Subsystem Configuration is determined by the
customer and indicated by an associated Selective
Feature Code.
The three Selective Feature codes are defined
below:
•
•
•
SF 9431: 3880 Model 3 or 23 Storage Control
attachment.
SF 9432: 3990 Storage Control attached in a
Two Path configuration.
SF 9433: 3990 Storage Control attached in a
Four Path configuration.
The B-unit models may be attached to an A-unit
model no matter what Selective Feature Code is
installed. There are no Selective Features required
for B units.
CONVERSION
SF 9431 to SF 9432
SF 9432 to SF 9433
SF 9431 to SF 9433
SF 9433 to SF 9432
SF 9432 to SF 9431
SF 9433 to SF 9431
Model J HDA to Model K HDA
Model K HDA to Model J HDA
Figure 2.
c
MES Summary
A units ship from the factory with one of the three
Selective Features installed. An MES is required to
convert from one Selective Feature to another.
The HDA type shipped from the factory is
determined by the model ordered by the customer.
A units and/or B units may have HDAs changed to
increase or decrease storage capacity.
The MESs available for these changes are defined
in the following table:
3880
3990
3880
3990
3990
3990
1.26
3.78
DESCRIPTION
Attach to 3990 2-Path
2-Path Attach to 3990
Attach to 3990 4-Path
4-Path Attach to 3990
2-Path Attach to 3880
4-Path Attach to 3880
Gb/HDA to 3.78 Gb/HDA
Gb/HDA to 1.26 Gb/HDA
Attach
4-Path Attach
Attach
2-Path Attach
Attach
Attach
o
o
c
;~
"~
c
Miscellaneous Equipment Specifications (MES) Summary Table
o
o
INTRO-2
PIN 4519893 41 Sides EC 476573 3ONov88
@Copyrlght IBM Corp. 1987. 1988
o
-----------.
(j
3380-JK Connected to a 3880 Storage Control
3880 Storage Control Requirements
Note: For 3990 Storage Control
requirements, see page INTRO-12.
The 3380-JK connects to 3880 Models 3 and 23.
The 3380-JK cannot be connected to other models
of the 3880. The 3380-JK cannot be connected to a
3880 with the speed-matching buffer feature.
(
(
('/
3880 Model 3
To connect a 3380-JK to a 3880 Model 3, the
3880 must have the following two features
installed:
Subsystem Configurations
The two controllers in a 3380 Model AJ4 or AK4
string attach to two 3880 Storage Directors. Some
possible configurations are:
•
To both Storage Directors in the same 3880
Model 3.
•
To either Storage Director in two different 3880
Model3s.
•
To both Storage Directors in the same 3880
Model 23:
To either of the paired Storage Directors in two
different 3880 Model 23s in dual frame
configuration.
•
The 3380 AD4/ AE4 Support Feature. This
feature upgrades the 3880 to include a
total of 12K of high speed storage.
•
•
The 3380-JK Support Feature. This feature
upgrades the 3880 to include a full track
buffer required to support improved data
correction. This feature can only be
installed on 3880's with 22 card position
boards.
The following figures illustrate some of the ways
that the 3380-JK may be attached to the 3880.
There are many more configurations that may be
used. See the MDM, Volume R30 INST section for
more details.
(~'
3880 Model 23
To connect a 3380-JK to a 3880 Model 23, the
3880 must have the 3380-JK Support Feature
installed. This feature upgrades the 3880 to
include a full track buffer required to support
improved data correction. This feature can
only be installed on 3880s with 22 card
position boards.
(~
(~
c
PIN 4519893 41 Sides EC 476573 3ONov88
to Copyright
IBM Corp. 1987, 1988
INTRO-3
o
Example of 3380-JK Strings Attached To a 3880 Storage Control
Figure 3 is an example of two 3380-JK 2-path
strings sequentially connected to both storage
directors of a 3880. In this example, one string
contains a 3380 Model AJ4 controller and the other
a 3380 Model AK4 controller with a mixture of BJ4
and BK4 units on both strings.
(C?:::.~,
~j
3880 Model 3 or 23
Storage
Director
Al
Figure 3.
AJ4
AK4
BK4
BK4
BK4
BJ4
BJ4
BJ4
3380 Model AJ4 and AK4 2-path strings attached to a 3880.
Figure 4 is an example of 3380 Models AD4, AE4,
AJ4, and AK4 strings connected to two storage
directors in different 3880 model 3s or in different
Model 23s in dual frame configuration.
3880 Model 3 or 23
I
I
Storage
Director
Al
1 A2
I
Storage
Director
'I
Al
c -I
1
1 A2
Al
1 A2
[~,
.'~
I
J
Al
1 A2
AD4
B
AK4
B
B
B
B
B
B
B
I
'I
"-
-I
Storage
Director
AJ4
B
B
AE4
Figure 4.
I
Storage
Director
,I
"-
,I
3880 Model 3 or 23
c
B
B
o
Intermixed 2-path strings, attached to a 3880.
o
o
INTRO-4
PIN 4519893 41 Sides EC 476573 3ONov88
Il:) Copyright IBM Corp, 1987, 1988
o
o
3380-JK String and Port Configurations (3880 Subsystem)
Figure 5 represents th~ 3380-JK maximum string
configuration for 2-path attached to a 3880.
To the 3880 storage Control
I
2-pa th configurations have a
maxi mum of 16 devices per string.
I
CUr. Al
Port
1 2 3
o
Fra[
-
DPS
-
CUr. A2
Port
1 2 3
o
I
A-Unit
- -
I
Dev.
00
Dev.
01
Dev.
02
Dev.
03
Dev.
04
Dev.
05
Dev.
06
Dev.
07
I
I
Dev.
08
Dev.
09
Dev.
OA
Dev.
OB
I
I
I
I
Dev.
OC
Dev.
OD
Dev.
OE
Dev.
OF
(
f/
Figure 5.
2-Path SIring Configuration
c
c
PIN 4519893 41 Sides EC 476573 3ONov88
© Copyright
19M Corp. 1987, 1989
INTRO-5
Dynamic Path Selection (DPS)
Power Sequencing
When the 3380-JK is attached to a 3880 Storage
Control, the controllers in each model have
dynamic path selection (DPS). DPS controls tie
breaking between storage controllers. DPS also
records information about devices that are shared
between controllers.
Warning: Do not use the Unit Emergency switches
to normally power the string off and on. Possible
data loss can occur with the heads landing on the
data surface and not on the park areas of the disk
surface. Use the Controller OnlOff switches to
power the string off and on (see Power OnlOff
procedures on page REPAIR-25).
Each controller has a DPS storage matrix which
contains an exact copy of the information in the
DPS storage matrix in the other (companion)
controller. When a controller updates its matrix,
the matrix in the other (companion) controller is
also updated. DPS has checking circuits to ensure
that both storage matrixes always contain the
same information. When a string is powered on,
both DPS storage matrixes are initialized by one of
the ,storage directors.
Power on control is provided by the 3880 when the
power control cable is connected from the 3880 to
the 3380-JK, and the 3380-JK Local/Remote switch
is in the Remote position.
o
c
When either Storage Director in the 3880 is
powered on, the controllers attached to positions
D1, D3, D5, and D7 in the 3880 will power on. Ten
seconds later the controllers attached to positions
D2, D4, D6, and D8 in the 3880 will power on.
Device Level Selection (DLS)
Each model 3380-JK has Device Level Selection
(DLS). DLS provides an independent path from
each controller to each device in a string. While
one controller is controlling a device, another
controller can control any other device in the same
string. This capability permits the transfer of data
from more than one device at the same time.
A Storage Director may be powered off without
dropping power on the controllers if any Storage
Director is left powered on. Power will drop on the
controllers when the last Storage Director is
powered off.
Control Interface Fencing
The 3880/3380-JK subsystem supports the same
control interface fenCing provided on earlier
models of the 3380. Control interface fencing is an
operation that results in a failing controller being
partitioned off a controller interface (CTL-I) so that
it does not interfere with the operation of a second
controller on the same CTL-1.
A controller is fenced off when an operation to it
causes a permanent controller check-1 condition.
()
o
INTRO-6
PIN 4519893 41 Sides EC 476573 3ONov88
«:l Copyright
IBM Corp. 1987, 1988
'.
C
/
!
Front Operator Panel (3880 Subsystem)
(
STRING
DEVICE
ID
ADDRESS
[ill]
OJ
STORAGE DIRECTOR 10
@D]
~
PATH
(
At
A2
@]
OJ
=00 =Ot =02
Ej~~
.g
=04- ~ =06
tj~~
~ READY
READY
~ENABLE
DISABLE
~ENABLE
DISABLE
I
=AC POWER ON
CONTROLLER
At
=
0
(/
(
A2
=
08
=
READY
OA
=
~Ej~
D~~ON
POWER
09
=
OFF
~ READY
~ENABLE
DISABLE
UNIT
READY
POWER
ENABLE
IDEVICES NOT I NSTALLED I
OC
=
OD
=
OE
=
~~~
g
~ENABLE
EMERGENCY
POWER OFF
DISABLE
(IB7624PI3
Figure 6.
((
3380-JK Operator Panel for 3880 Attachment
The front operator panel has the following controls
and LEOs:
•
Controller Power On/Off switches for each
controller
•
Unit Emergency switch
•
Controller Ready LEOs
•
AC Power-On LED
•
Device Enable/Disable switches for each device
•
Device Ready LEOs
-~.
(/
(~;
C:
c
PIN 4519893 41 Sides EC 476573 30Nov88
@ Copyright IBM Corp. 19B7, 19BB
INTRO-7
,r"
(~
.
"'-....;PI
Unit Emergency Switch
Controller Ready LEOs
The Unit Emergency switch on the operator panel
is only for emergency power off of the string. It
switches off all power, except 24 Vdc in the A Unit
only, bypassing the normal power-off sequence.
Each controller has one Ready LED. The Controller
Ready LED has four states. The Controller Ready
LED will:
The Unit Emergency switches are normally in the
Power Enable (on) position. In this position,
3380-JK power can be controlled by either the host
system (remote) or the 3380-JK (local).
Warning: Do not use the Unit Emergency switches
to normally power the string off and on. Possible
data loss can occur with the heads landing on the
data surtace and not on the park areas of the disk
surtace. Use the Controller OnlOff switches to
power the string off and on (see Power OnlOff
pro~edures on page REPAIR.25).
AC Power-On LED
The AC Power-On LED is on when line voltage is
present and the AC phase rotation is correct.
Controller Power OnlOff Switches
Each controller in an A unit has one Power On/Off
switch. These switches control powering the
controller and the drives On and Off.
Setting either controller power On/Off switch to On
will power on the controller and all the drives in
the string.
Providing one controller power On/Off switch is set
to On, setting the other controller power On/Off
switch to Off will power off that controller but not
the drives in the string.
Setting both controller power On/Off switches to Off
will power off both controllers and all the drives in
the string.
•
Be Off when the controller is powered off, a
controller check 1 is active, or a controller
fenced condition exists.
•
Blink during a controller or a string power on
sequence. It is normal for the controller ready
LEOs for both controllers to blink at different
rates during a string power on sequence.
•
Be dim (approximately half brightness) when
the controller is ready and there is no activity.
•
Get brighter (approach full brightness) as
activity increases. Flashing or flickering is
normally apparent when there is activity.
----
I~
U
If the switch is in the Enable position device
interrupts are permitted across the control
interface (CTL-I).
If the switch is in the Disable position, it disables
interrupts from the device. If the system attempts
to access the disabled device the storage control
responds with INTERVENTION REQUIRED.
If the switch is changed from Enable to Disable, the
device will not be disabled until the device finishes
the current sequence. This permits the device to
complete the current operation before it is
disabled.
PIN 4519893 41 Sides EC 476573 3ONov88
----
o
Each device has an Enable/Disable switch. The
switch permits device interrupts from the device to
the storage path. The switch enables or disables
the device from access by the customer program.
@ Copyright IBM Corp. 1987. 1988
------~
,
Device Enable/Disable Switch
If the switch is changed from Disable to Enable, an
HDA attention interrupt is generated. The interrupt
is sent by all storage directors to all the attached
systems.
INTRO-8
0'--
C)
r
0
(
Device Ready LEOs
Front Operator Panel Labels
Each of the sixteen possible devices has one
Ready LED. The LEOs are active only for the
installed devices.
The following labels are placed on the Front
Operator panel at installation time.
The Ready LED is on when the device is ready for
use but is not being used. When the device is
being used the LED flashes off during seek
operations. Because the seek times of a 3380-JK
are in the millisecond range, the flashing of the
ready LED may not be visible. During periods of
inactivity the Device Ready LEOs will flash off
periodically. The flashing indicates that a SAFE
operation is being performed.
(
(
The Ready LED is off when the device is not
available to the customer (disabled) or when an
error condition is present on the device.
String 10
The physical ID set in the controller. In the
3380-JK all controllers in the string must be
set to the same 10.
String Address
The address of the string (0 or 1).
Storage Director 10 (SOlD)
The physical IDs of the 3880 storage directors.
A 3880 contains two storage directors and
each storage director has a unique 2
hexadecimal digit SOlD. The SOlD labels
identify the storage directors connected to
controllers A 1 and A2.
(
(-
c;
Ci
PIN 4519893 41 Sides EC 476573 3ONov88
© Copyright
IBM Corp. 1987, 1988
INTRO-9
c
C
0
I,"
Rear Unit Emergency Power Off Panel (3880 Subsystem)
,
r
.,:'
0
~I
(-,
~)
UNIT EMERGENCY
SWITCH
~
1
.
______
UNIT
EMERGENCY
(~~
PATH
D~ CONTROLLER
REAOY LEOS
POWER
~
CONTROLLER
ENABLE
D
POWER OFF
~)
0
r~
MD
STRING ADDRESS
STRING 10
SUBSYSTEM 10
o
D
D
D
D
IS000127NI
L
Figure 7.
.J
3380-JK Rear Unit Emergency Power Off Panel for 3880 Attachment
The rear unit emergency power off panel has the
following controls and LEDs:
•
Unit Emergency switch
•
Controller Ready LEDs
•
Maintenance Device Connector
Unit Emergency Switch
The Unit Emergency switch on the rear Unit
Emergency Power Off panel is only for emergency
power off of the string. It switches off all power,
except 24 Vdc in the A unit only, bypassing the
normal power-off sequence.
The Unit Emergency switches are normally in the
Power Enable (on) position. In this position,
3380-JK power can be controlled by either the host
system (remote) or the 3380-JK (local).
Warning: Do not use the Unit Emergency switches
to normally power the string off and on. Possible
'data loss can occur with the heads landing on the
data surface and not on the park areas of the disk
surface. Use the Controller On/Off switches to
power the string off and on (see Power On/Off
procedures on page REPAIR.25).
o
INTRO-10
PIN 4519893 41 Sides EC 476573 3ONov88
© Copyright
IBM Corp. 1987, 1988
(
('"
(~
Controller Ready LEDs
Each controller has one Ready LED. The Controller
Ready LED has four states. The Controller Ready
LED will:
•
Be Off when the controller is powered off, a
controller check 1 is active, or a controller
fenced condition exists.
•
Blink during a controller or a string power on
sequence. It is normal for the controller ready
LEDs for both controllers to blink at dirferent
rates during a string power on sequence.
(
•
Be dim (approximately half brightness) when
the controller is ready and there is no activity.
(
•
Get brighter (approach full brightness) as
activity increases. Flashing or flickering is
normally apparent when there is activity.
(
Rear Unit Emergency Power Off Panel
Labels
These labels are put on the rear unit emergency
power off panel during installation.
String Address
Label identifying the String Address (0 or 1).
String 10
Label identifying the physical 10 set in the
controller. In the 3380-JK all controllers in
the string must be set to the same 10.
Subsystem 10
Not used.
Path
Maintenance Device (MD) Connector
The MO connector provides the means to
physically connect the maintenance device (MO) to
the 3380-JK.
(
Label identifying the path connected to
controllers A 1 and A2.
Controller
Label identifying the controllers, A 1 or A2.
Storage Director 10 (SOlD)
The SOlD label locations are not shown on
the panel. The SOlOs identify the storage
directors that are attached to controllers A 1
and A2. The SOlD labels are placed on the
panel below the path for each controller.
((~
f:
(
"
,-,/
C~
C
(
'
"
. ./
('
PIN 4519893 41 Sides EC 476573 3ONov88
@ Copyright IBM Corp. 1997, 1999
INTRO-11
o
o
3380-JK Connected to a 3990 Storage Control
3990 Storage Control Requirements
Note: For 3880 Storage Control
requirements, see page INTRO~3.
The 3380-JK connects to all 3990 models. No
special 3990 Features are required.
Subsystem Configurations
The two controllers in a 3380AJ4 and AK4 attach to
two 3990 storage paths. Some possible
configurations are:
•
Storage path 0 and storage path 1 in the same
3990 Model 1.
•
Storage path 0 and 2, or storage path 1 and 3
in the same 3990 Model 2 or Model 3 for a
2-path string.
•
Storage path 0, 1, 2, and 3 in the same 3990
Model 2 or Model 3, for a 4-path string.
Example of 3380-JK Strings Attached To a 3880 Storage Control
The following figures illustrate some of the ways
that the 3380-JK may be attached to the 3990.
There are many more configurations that may be
used. See the MDM, Volume R30 INST section for
more details.
Note: One storage path from each cluster is
attached to each string. This is a
configuration requirement, to ensure that
there will be an operational path to each
string if one of the clusters is powered off.
Figure 8 is an example of four 2-path strings. Two
string pairs are sequentially connected to each of
the paired storage paths in the same 3990 Model 2
or Model 3.
3990 storage Control
Cluster a
spa
t
Cluster I
SPI
-;=J-'\
r-
SP2
1 SP3
-- 1--,
II
,1--,
~1.-lA~
~1~A~
~l~A~
~1.-lA~
AK4
AJ4
AA4
AE4
B
B
B
B
B
B
B
B
B
B
B
B
Figure 8.
c
Intermixed 2-path Strings Attached to a 3990
INTRO~12
PiN 4519893 41 Sides EC 476573 3ONov88
© Copyright IBM Corp.
1987. 1988
~~,
,{,.j
(.
('
Figure 9 is an example of two 3380-JK 4-path
strings sequentially connected to the paired
storage paths in the same 3990 Model 2 or Model
3.
(
(
_.,
3990 Storage Control
Cluster 0
SPO
SPI
'--
f
Cluster 1
SP2
I SP3
1- -
I
II
I
B B B ~1!A~ ~3!A~ B B B
AK4
AK4
I I I I
B B B ~1!A~ ~3,lA~ B B B
AJ4
f·'
Figure 9.
(.
(:~
AJ4
3380-JK 4-palh Strings Attached to a 3990
Figure 10 is an example of a 3380-JK 4-path string
and two 2-path strings sequentially connected to
two paired storage paths in a 3990 Model 2 or
Model 3.
3990 Storage Control
Cluster 0
SPI
SPO
-
(
(
.'
B B B
(-
(
I-
~1!A~ ~3!A~
B B B
AK4
Lil
I
.•.
I-
I I
AJ4
~'
Cluster I
SP2
SP3
I
-'I
....
Note: One storage path from each cluster is
attached to each 2-path string. This is a
configuration requirement. It is required to
ensure that there wi/l be an operational path
to each 2-path string if one of the clusters is
powered off.
~l-.LA~
~l-.LA~
AE4
AD4
B
B
B
B
B
B
-.~/
C'
Figure 10.
Intermixed 4-path and 2-path Strings Attached to a 3990.
PIN 4519893 41 Sides EC 476573 3ONoV88
@ Copyright IBM Corp. 1987. 1988
(
INTRO-13
o
3380-JK String and Port Configurations (3990 Subsystem)
The 3380-JK has two maximum string
configurations, one for 2-path and one for 4-path
when attached to a 3990. Figure 11 represents the
3380~JK
3380-JK maximum string configuration for 2 path.
Figure 12 on page INTRO-15 represents the
3380-JK maximum string configuration for 4 path.
o
o
o
2-Path String Configuration
To the 3990 Storage Control
2-pa th configurations have a
maxi mum of 16 devices per string.
Fra[
J
I
CUr. Al
Port
o1 2 3
CUr. A2
Port
0 123
I
A-Unit
Dev.
00
I
I
Dev.
02
Dev.
03
J
1
Dev.
05
Dev.
06
Dev.
07
J
I
l
Dev.
09
Dev.
OA
Dev.
OB
Dev.
01
I
Dev.
04
Dev.
08
-
-
/
(
-
()
o
I
Dev.
OC
Dev.
OD
Dev.
OE
Dev.
OF
Note: DPS is not installed in
the 3380-JK when attached
to a 3990 Storage Control.
c
Figure 11. 2·Palh String Configuration
o
INTRO-14
--~
PIN 4519893 41 Sides EC 476573 3ONov88
© Copyright IBM Corp. 1997, 19S9
- - ..
-.~---.
_.
('
(~"
3380-JK 4-Path String Configuration
SUBSTRING 0
('
To the 3990 Storage Control
4-path configurations have a maximum
of 32 devices per string.
(
Fra[
A-Unit
(/
I
Ctlr. Al
Port
136
o2 7 S
Ctlr. A2
Port
I
III
III
Dev.
00
Dev.
01
11
E]
02
i
I
136 4
9i 2 T11 s j
II I II
Al-4
A2-4
- -A3-0
A4-0
I
L..-.
III
--
Dev.
03
Al-S
A2-S
- -A3-1
A4-1
L--
I
111
hi
III
II.
Dev.
Dev.
OS
Dev.
Dev.
04
(',
I
06
--
07
I
(
"""'"
,/
("
III
III
III
III
Dev.
08
Dev.
09
Dev.
OA
Dev.
OB
III
III
III
hi
Dev.
OC
Dev.
OD
Dev.
OE
Dev.
OF
I
I
Al-6
A2-6
A3-2
A4-2
Al-7
A2-7
A3-3
A4-3
(~
Figure 12 (Part 1 of 2).
Full 4-Path String Configuration
()
(
"!
J
C
c
PIN 4519893 41 Sides EC 476573 30Nov88
@.Copyright IBM Corp, 1987, 1988
INTRO-1S
0'
,
To the 3990 Storage Control
'.
o
SUBSTRING I
o
AI-4----~~~~----~~~~------~------~~-------r------~
A2-4-------------------------------r
A3-0----'
-------7 -------7
-------~
A4-0--
Dey.
10
Dey.
11
Dey.
12
Dey.
13
Jell
A-Unit
Al-S----~~~------~~~~------,---------~------_r------~
A2-S-------------------------------r
A3-1-----'
A4-1---
Dey.
14
-------~
Dey.
15
-------7 -------,
Dey.
16
Dey.
17
AI-6----~~~--------~~------_.--------._------~------__,
A2-6-------------------------------r
I
A3-2
I
I
A4-2
,I'I
A3-3
A4-3
I
Figure 12 (Part 2 of 2).
I
.TI
Dey.
Dey.
19
,r!
rI
Dey.
IC
Dey.
1D
18
AI-7
A2-7
-------~
-------T -------,
'1'1
II
Dey.
1A
Dey.
1B
d
Dey.
IE
II
Dey.
IF
c
Full 4-Path String Configuration
q'{ ""'....
c'
•
~j)
()
INTRO-16
PIN 4519893 41 Sides EC 476573 3ONov88
© Copyright
IBM Corp. 1987. 1988
o
(
3380-JK Port Configurations (3990 Subsystem)
(
A string of 3380-JK devices consists of one or two
substrings.
•
•
•
Al Controller
CDP Interface
Only 4-path subsystems have a second
substring (CDPSS1).
Each substring contains four ports. .
Each port connects to four devices in one unit.
1st A-Unit Devices
· .................
c
Port 0
,
CDPSSO
CARD
(-
2 Path
or
4 Path
Port I
Port 2
(
·
·Port
'
-"
AI€]
A2
A3
A4
DEV
03
2nd A-Unit Devices
..................
4
CDPSSI
CARD
Port 5
4 Path
Only
Port 6
Figure 13.
01
DEV
02
07)
To 1st B Unit CD EV 04
OB)
To 2nd B Unit CD EV 08
To 3rd B Unit CD EV OC - OF)
··
·
............·
(~:
AI€]
A2
A3
A4
L.-
·· .................
Port 7
f
~A10
A2
DEV
A3
A4
··
f
r-
·
Port 3
·
(
··
~Al€]
DEV
· A2
00
· A3
·· A4
·· ~Al€]
A2
DEV
··
··
·
···
A3
A4
10
r-
A2 l i
DEV
A
!]
A3
12
A4
0 A10
A1
A2
- A3
DEV
- A2
A3
·· A4 11
A4
·..................
DEV
13
To 4th B Unit CD EV 14 - 17)
To 5th B Unit CD EV 18 - 1B)
To 6th B Unit CD EV lC - IF)
Port Configuration
(~\
,j
PIN 4519893 41 Sides EC 476573 3ONov88
© Copyright
(
,
""
'
IBM Corp. 1987, 1988
INTRO-17
o
Dynamic Path Selection (DPS)
Control Interface Fencing
When the 3380-JK is attached to the 3990 storage
control, the DPS function is moved from the
controllers to the Shared Control Array (SCA) in
the 3990. The SCA controls tie breaking between
storage paths and records information about
devices that are shared between controllers. The
information contained in the Shared Control Array
is duplicated in each storage cluster in that
subsystem.
The 3990/3380-JK subsystem supports the same
control interface fencing provided on earlier
models of the 3380. Control interface fencing is an
operation that results in a failing controller being
partitioned off a Director Device Connection (DOC)
so that it does not interfere with the operation of a
second controller on the same DOC.
A controller is fenced off when an operation to it
causes a permanent controller check':1 condition.
Device Level Selection (DLS)
Subsystem Fencing
Each model 3380-JK has Device Level Selection
(DLS). DLS provides an independent path from
each controller to each device in a string. While
one controller is controlling a device, another
controller can control any other device in the same
string. This capability permits the transfer of data
from more than one device at the same time.
Four Path
The 4-path feature provides four controller paths to
each DASD in the string. The Shared Control Array
(SCA) in the 3990 storage control, along with a Path
Interrupt Table in each device, coordinate the
necessary information for the four paths to each
device.
Four path combines two 2-path 3380-.JK A units and
adds another CDPSS card (ports 4-7) to each
controller. Each A unit is built to attach B units
from either side. This allows both A units to be
bolted together and have the operator's panel from
each A unit facing the same direction. The eight
ports can control 32 devices. Each B-boardand
R/W channel board receives cables from all 4
controllers.
The maximum 4-path configuration is 32 devices in
string 0 and 32 devices in string 1. This requires a
64 address range on a 64 address boundary. A
substring is an A1 and A2 controller and its
associated B units for substring 0 (OO-OF or 20-2F)
or an A3 and A4 controller and its associated B
units for substring 1 (10-1F or 30-3F).
INTRO-18
In addition to control interface fencing there is
another level of fencing performed when the
3380-JK is attached to a 3990 in 4-Path mode.
Subsystem initiated fencing automatically removes
failing components from a 4-path 3990/3380-JK
subsystem, by altering the selection path to a
device.
The fencing action results in a path to a device
being made unavailable to the host processor.
This prevents recurring path related error
conditions from being posted in the system
Environmental Recording Data Set (EROS) and
degrading system performance. In addition,
subsystem fencing permits better identification of
the failing hardware or media in console messages
and EREP reports.
Subsystem fencing can be initiated by anyone of
the four storage paths in the 3990 4-path
subsystem. Three levels of fencing are possible:
Channel, Storage Path, and Device fencing. Each
level of fencing has its own set of error and usage
counters. When an error counter overflows, a
process is started that can result in a permanent
subsystem fence by the storage control to a path or
subsystem component.
A permanent device fence is reported as a Format
o message 3 with message to operator bit on. The
system operator is notified of the fence condition
by a console message generated from the Format 0
sense data.
PIN 4519893 41 Sides EC 476573 3ONov88
© Copyright
IBM Corp. 1987, 1989
c
c
c
o
c
o
(
C/
(
(
Reset of a fenced condition should only be
attempted upon completion of a repair action as
directed by the 3380-JK maintenance package. The
following list defines the methods for resetting a
fenced condition:
•
Using the MD Main Menu option 2: End of Call.
•
Pressing the storage path restart switch.
•
Issuing a Resume Command at the 3990
maintenance panel.
•
Using the ICKDSF control command with the
CLEARFENCE parameter.
•
Performing an IML of the fenced storage path.
Quiesce I Resume Storage Path
(
(
(
c
("
Quiesce storage path and Resume storage path
are functions provided by a 4-path 3990 subsystem.
These functions, which can be invoked at the 3990
maintenance panel, provide the ability to service a
CTL-I without impacting the other storage paths in
a 4-path 3990 subsystem. (This includes the
non-disruptive install/remove of strings).
Quiesce and Resume Controller
Quiesce Controller and Resume Controller are
functions provided by a 3990 DASD subsystem.
Quiesce controller can be used on active 2-path,
4-path, and mixed 4-path/2-path subsystems. This
function is invoked at the 3990 maintenance panel.
It provides the ability to service a controller, but
not the CTL-I, without impacting the subsystem.
The 3990 support facility (SF) upon processing the
Quiesce Controller command, fences the selected
controller. No host activity will be routed to that
controller. All errors developed by the selected
controller are not recognized or reported by the
functional code. With the controller quiesced,
maintenance can be performed without impacting
the other controllers operations on the CTL-1.
Diagnostic routines can continue to detect all of the
errors on a controller that is quiesced.
Following a repair action, a Resume Controller
command removes the internal controller fence
condition.
For further information, see the MNT PNL section of
the 3990 maintenance documentation.
The Quiesce storage path command is entered at
the 3990 maintenance panel. Under control of the
3990 support facility (SF), the storage path is fenced
and all host activity is internally routed through
another storage path. The Quiesce storage path is
not successful if an alternate storage path to the
devices is not available. A message is sent to the
operator identifying the part of the subsystem that
has been quiesced. This is an informational
message and no action is required by the system
operator. With the path quiesced, maintenance can
be performed on the CTL-1.
Power Sequencing
Following a repair action the Resume storage path
command is entered at the 3990 maintenance panel
to remove the internal storage path fence
condition.
The clusters may be powered off without dropping
power on the controllers if any cluster is left
powered on. Power will drop on the controllers
when the last cluster is powered off.
PIN 4519893 41 Sides EC 476573 3ONoV88
~ Copyright IBM Corp. 1987, 1988
Power on control is provided by the 3990 when the
power control cable is connected from the 3990 to
the 3380-JK, and the 3380-JK Local/Remote switch
is in the Remote position.
When either cluster in the 3990 is powered on, the
controllers attached to positions 02 to 08 in the
3990 will power on. Ten seconds later the
controllers attached to positions 010 to 016 in the
3990 will power on.
INTRO-19
(5"
V
Front Operator Panel (3990 Subsystem)
Figure 14 shows the A unit operator panel for a
2-path 3990 attachment.
The first A unit operator panel for a 4-path 3990
attachment is not shown. It is the same as
Figure 14 except the label under PATH indicates
FOUR PATH instead of TWO PATH.
Figure 15 on page INTRO-20 shows the second A
unit operator panel for a 4-path 3990 attachment.
STRING
DEVICE
ADDRESS
10
o
CJ
SUBSYSTEM 1.0.
I
I
I
=00 =Ot =02
tjt:j~
I
~ READY
EjENABLE
DISABLE
I
PATH
Al
I
= ~ =06
~~Ej
O~
A2
o
o
TWO PATH
~ READY
tjENABLE
DISABLE
CONTROLLER
A2
READY
O
o
POWER ON
POWER OFF
=AC POWER DN
I
I
At
o
=08 =09 =OA
~~~
~ READY
~ENABLE
DISABLE
IDEVICES NOT I NSTALLEDI
OC
00
OE
= =
~~~
=
I
g
READY
~ENABLE
UNIT
EMERGENCY
g
I~
POWER
ENABLE
POWER OFF
DISABLE
"'"
"
IA7625PI.
Figure 14.
c
3380-JK Operator Panel for 2-Path 3990 Attachment
The front operator panel for 2-path 3990 attachment
has the following controls and LEOs:
•
Unit Emergency switch
•
AC Power-On LED
•
Controller Power On/Off switches for
controllers A 1 and A2
INTRO-20
•
Controller Ready LEOs
•
Device Enable/Disable switches for devices 00
through OF
•
Device Ready LEOs
PIN 4519893 41 Sides EC 476573 3ONov88
@ Copyright IBM Corp. 1987. 1988
C'
o
o
,
(
DEVICE
I
(
I
=10 =11 =12
Ej ~ ~
c
g
~ENABLE
DISABLE
I
I
=ILl-
(
tj
~
~
=16
g
READY
tjENABLE
tj
DISABLE
I
(
READY
=AC POWER ON
I
=18 =19 =lA
tj tj tj
~ READY
~ENABLE
DISABLE
IDEVICS5 NOT I NSTALLED I
Ie
=
10
IE
= =
~ ~ ~
~ READY
~ENABLE
UNIT
EMERGENCY
POWER
ENABLE
POWER OFF
DISABLE
I7626P I
L
Figure 15.
.J
3380-JK Second A unil Operator Panel for 4-Path 3990 Attachment
The second A-unit front operator panel for 4-path
3990 attachment has the following controls and
LEOs:
(-
I
•
Unit Emergency switch
•
AC Power-On LED
•
Controller Power On/Off switches for
controllers A3 and A4
•
Controller Ready LEOs
•
Device Enable/Disable switches for devices 10
through 1F
•
Device Ready LEOs
c
c
PIN 4519893 41 Sides EC 476573 3ONov88
@ Copyright IBM Corp. 1987, 1988
INTRO-21
n \
\l.)
Unit Emergency Switch
Controller Ready LEOs
The Unit Emergency switch on the operator panel
is only for emergency power off of the string. It
switches off all power, except 24 Vdc in the A Unit
only, bypassing the normal power-off sequence.
Each controller has one Ready LED. The Controller
Ready LED has four states. The Controller Ready
LED will:
The Unit Emergency switches are normally in the
Power Enable (on) position. In this position,
3380-JK power can be controlled by either the host
system (remote) or the 3380-JK (local).
•
Be Off when the controller is powered off, a
controller check 1 is active, or a controller
fenced condition exists.
•
Blink during a controller or a string power on
sequence. It is normal for the controller ready
LEOs for both controllers to blink at different
rates during a string power on sequence.
•
Be dim (approximately half brightness) when
the controller is ready and there is no activity.
•
Get brighter (approach full brightness) as
activity increases. Flashing or flickering is
normally apparent when there is activity.
In 4-path configurations the Unit Emergency
Switches in each A unit are Interlocked. That is,
either switch will power off both substrings.
Warning: Do not use the Unit Emergency switches
to normally power the string off and on. Possible
data loss can occur with the heads landing on the
data surface and not on the park areas of the disk
surface. Use the Controller OnlOff switches to
power the string off and on (see Power OnlOff
procedures on page REPAIR-25).
AC Power-On LED
The AC Power-On LED is on when line voltage is
present and the ac phase rotation is correct.
Controller Power OnlOff Switches
Each controller in an A unit has one Power OnlOff
switch. These switches control powering the
controller and the drives On and Off.
Setting either controller power OnlOff switch to On
will power on the controller and ali the drives in
the string. In 4-path configurations, powering on
anyone of the controllers will power up all the
drives in the string.
Providing there is at least one controller power
OnlOff switch set to On, setting another controller
power OnlOff switch to Off will power off that
controller but not the drives in the string.
Setting all controller power On/Off switches in the
string to Off will power off all controllers and all the
drives in the string.
INTRO-22
Device EnablelDisable Switch
Each device has an Enable/Disable switch. The
switch permits device interrupts from the device to
the storage path. The switch enables or disables
the device from access by the customer program.
If the switch is in the Enable position device
interrupts are permitted across the control
interface (DOC).
If the switch is in the Disable position, it disables
interrupts from the device. If the device is
accessed by the customer program, Intervention
Required is generated.
If the switch is changed from Enable to Disable, the
device will not be disabled until the device finishes
the current sequence. This permits the device to
complete the current operation before it is
disabled.
If the switch is changed from Disable to Enable. an
HDA attention interrupt is generated. The interrupt
is sent by all storage paths to all the attached
systems.
c
PIN 4519893 41 Sides EC 476573 3ONov8B
@ Copyright IBM Corp. 1987, 1988
o
(
(
Oevice Ready LEOs
Front Operator Panel Labels
Each of the thirty-two possible devices has one
Ready LED. The LEOs are active only for the
installed devices.
The, following labels are placed on the front
operator panel at installation time.
The Ready LED is on when the device is ready for
use but is not being used. When the device is
being used the LED flashes off during seek
operations. Because the seek times of a 3380 are
in the millisecond range, the flashing of the ready
LED may not be visible.
(
(
During periods of inactivity the Device Ready LEOs
will flash off periodically. The flashing indicates
that a SAFE operation is being performed.
String 10
The physical 10 set in the controller. In the
3380-JK all controllers in the string must be
set to the same 10.
Address
The address of the string (0 or 1).
Subsystem ID (5510)
The Subsystem 10 is an assigned number that
identifies the physical components of a logical
DASD Subsystem.
The Ready LED is off when the device is not
available to the customer (disabled) or when an
error condition is present on the device.
Sense byte limitations for 3380-JK require that
all subsystem IDs must have "00" as the first
two characters of the 10.
When looking at sense data for devices
attached to 3990, 32 bytes will be displayed.
Byte 21, the SSID, is the same for all paths
attached to a particular device. Byte 21 is the
low order byte of the SSID.
The operator panel has an area for only one
SSID number on either a 2-path or 4-path
panel when attached to 3990. When 2-path
string(s) are attached to a four path string all
strings have the same SSID number and are
within one contiguous 64 address range. The
2-path string attached to 3990 paths 0 and 2
will have the lower of the 16 address range
and if another 2-path string is attached to
3990 paths 1 and 3 they will occupy the higher
16 address range.
(
(
Configurations without 4-path strings will be
in 2-path mode. Paths 0 and 2 will have one
SSID number and paths 1 and 3 will have
another SSID number.
('
Path
The storage path (O, 1, 2, or 3) in the 3990 that
is attached to controller A 1, A2, A3, or A4.
(,
PIN 4519893 41 Sides EC 476573 30Nov88
© Copyright IBM Corp. 1987. 1988
INTRO-23
u
(['''I.,
1
~i
Rear Unit Emergency Power Off Panel (3990 Subsystem)
~
"J
,
r
~\
0
1
('\
~J
(-\
UNIT E10ERGENCY
SWITOi
~
1
________
UNIT
EMERGENCY
'0
PATH
CJ~ CONTROLLER
READY LEOS
POWER
CONTROLLER
ENABLE
CJ
POWER OFF
0
MD
STRING ADDRESS
STRING 10
SUBSYSTEM IO
o
CJ
CJ
D
D
IS000127NI
L
Figure 16.
.A
3380·JK Rear Unit Emergency Power Off Panel for 3990 Attachment
The rear unit emergency power off panel has the
following controls and LEOs:
•
Unit Emergency switch
•
Controller Ready LEOs
•
Maintenance Device Connector
Unit Emergency Switch
The Unit Emergency switch on the rear unit
emergency power off panel is only for emergency
power off of the string. It switches off all power.
except 24 Vdc in the A unit only. bypassing the
normal power-off sequence.
\
,
"'-/
The Unit Emergency switches are normally in the
Power Enable (on) position. In this position.
3380-JK power can be controlled by either the host
system (remote) or the 3380-JK (local).
In 4-path configurations the Unit Emergency
Switches in each A unit are interlocked. That is.
either switch will power off both substrings.
Warning: Do not use the Unit Emergency switches
to normally power the string off and on. Possible
data loss can occur with the heads landing on the
data surface and not on the park areas of the disk
surface. Use the Controller On/Off switches to
INTRO-24
PIN 4519893 41 Sides EC 476573 3ONov88
@Copyrlght IBM Corp. 1987, 1988
C:
C
power the string off and on (see Power On/Off
procedures on page REPAIR-25).
Rear Unit Emergency Power Off Panel
Labels
Controller Ready LEDs
The following labels are placed on the rear unit
emergency power off panel during installation.
These labels have the same meaning as the
corresponding Front Operator Panel labels
described on page INTRO-23.
Each controller has one Ready LED. The Controller
Ready LED has four states. Tile Controller Ready
LED will:
(-
(
•
Be Off when the controller is powered off, a
controller check 1 is active, or a controller
fenced condition exists.
•
Blink during a controller or a string power on
sequence. It is normal for the controller ready
LEOs for both controllers to blink at different
rates during a string power on sequence.
•
Be dim (approximately half brightness) when
the controller is ready and there is no activity.
•
Get brighter (approach full brightness) as
activity increases. Flashing or flickering is
normally apparent when there is activity.
Maintenance Device (MDl Connector
f
The MO connector provides the means to
physically connect the maintenance device (MO) to
the 3380-JK.
String Address
Label identifying the String Address (0 or 1).
String 10
Label identifying the physical 10 set in the
controller. In the 3380-JK all controllers in
the string must be set to the same 10.
Subsystem 10 (5510)
Label identifying the subsystem identifier
when attached to 3990.
Path
Label identifying the storage path connected
to controller A 1, A2, A3, or A4.
Controller
Label identifying the controllers connected to
3990 storage paths 0, 1, 2, or 3.
('\
(
(
C
PIN 4519893 41 Sides EC 476573 3ONoV88
© Copyright
IBM Corp. 1987, 1988
INTRO-2S
Device Characteristics Summary
Access Mechanisms
per unit
Models
A.J4 and B.J4
Models
AK4 and BK4
4
4
Cylinders
per access mechanism
Data
Al ternate data
CE
Diagnostic
885
2,655
4
1
4
Data Tracks
per cylinder
per access mechanism
15
13,275
15
39,825
47,476
712,140
47,476
712,140
Maximum Data Capacity
in bytes:
per track
per cylinders
in Mbytes:
per access mechanism
per unit
per 4-unit string
4-path configuration:
per 8-unit string
Figure 17.
1
1
630 Mb
2,520 Mb
10,083 Mb
20,166 Mb
1
1,890 Mb
7,526 Mb
30,210 Mb
60,500 Mb
Characteristics of the 3380·JK Models
1\I
"
~/
INTRO-26
PIN 451989341 SIdes EC 476573 3ONov88
~ Copyright IBM Corp. 1987. 1988
,r'j
~j
Track Layout - Models AJ4 and BJ4
Outer Diameter (00)
Physical:
Data
Track
Decimal
logical
Data
Track
Decimal
logical
Data
Track
Hex
I
Head
landing
Zone
-11
I
I
Reserved
I
-10
Guard
Band Outer
Diameter
(GBOD)
(
-4
-3
-3
I
-2
I
I
Reserved
-1
(
(
(,
0
0
0
884
884
374
885
885
375
I
888*
I
Guard
BandInner
Diameter
(G8ID)
I
*
( '-'
.. /"
(
",'
• • • r'
3380-J has 3 buffer cylinders between the
alternate cylinder and the CE cylinder. These
cylinders are not accessible from the system. If the
system attempts to access the CE cylinder (886) the
access mechanism will be positioned at physical
cylinder 889 (logical cylinder 886). The home
addresses on the buffer cylinders are flagged as
defective with no alternate assigned.
I
Alternate
886
376
890
887
377**
891
888
892
889
378** D379**
893
890
37AH
CE
I 1C
,.grS
t-
Req~ired
895
Inner Diameter (10)
Data
889
894
(
customer
BUf~er
I
886*
887*
(-"
HA Map
FFFD
for
servo
I
896
**
Diagnostic cylinders are accessible only by the
diagnostics. They cannot be accessed from the
system.
c
C"
'.
-,'
PIN 4519893 41 Sides EC 476573 30Nov88
«:J Copyright
IBM Corp, 1987, 1988
INTRO-27
(L)
(~)
Track Layout - Models AK4 and BK4
Outer Diameter (00)
0,-y
Physical
Data
Track
Decimal
logical
Data
Track
Decimal
logical
Data
Track
Hex
r'"
0
I
Head
landing
Zone
-11
-10
Guard
Band Outer
Diameter
(GBOD)
I
Guard
BandInner
Diameter
(GBID)
I
-4
-3
2669
2668
A6D
A6C
-2
2667
A6B
Reserved
(~'
I
I
0\
HA Map
~,.;i
~j
Reserved
0
0
000
2654
2654
A5E
2655
2655
Alternate
2656
2656
A5F
A603E
2657
2657
A613E
I
2658
2658
A62
A633E3E
2659
2660
2659
2660
A64**
2661
2661
A65**
2662
2662
A66**
customer
Data
I
("
~/
Buf+er
CE
r-~
(
2664
I ."
1a.rS 1C
0"
3380-K has 2 buffer cylinders between the
*alternate
cylinder and the CE cylinder. These
cylinders are not accessible from the system.
Req~ired
for
servo
2665
Inner Diameter (rD)
.
~7'
I
2663
.
\---./
I
-1
I
~\
I
I
/
\,_
7
("'
\
\c.,/
**
Diagnostic cylinders are accessible only by the
diagnostics. They cannot be accessed from the
system.
c
INTRO-28
PIN 4519893 41 Sides EC 476573 3ONov88
IC Copyright
IBM Corp. 1987, 1988
o
()
Disk, Arm, and Head Layout
Arm
1
Left Access Mechanism
Arm
Arm
Arm
2
3
4
('"
rO
r-
r3
'--
'--
F
f-
f-
-
(
2
-
(
Drive
Pulley
8
-
-1
-
rE
r-
r-
-4
'--
:0-
B
7
f-
-
I-
r- rD A
'-- r-
c-
r- r9
6
'-- r-
r5
'--
-
0
rE
r- r-
r-
r--
2
(
r-"
r-
-
r-
D A
r- -
1
r-
(
!-
C
rC
r- r-
9
r-
r-
'--
r-
3
F
8
B
-
!-
-
r-
5
!-
I-
l-
r--
0
6
-
!-
4
-
r-
7
"-
(
Arm
1
{~i
Figure 18.
Arm
2
3
Right Access Mechanism
Arm
4
Not e : Head F is t h e servo h ead.
Arms, Heads, and Disks in the HDA
PIN 4519893 41 Sides EC 476573 3ONov88
© Copyright IBM Corp. 1987, 19B8
f'
Arm
INTRO-29
Servo System
Switches and Indicators
The tracks on the 3380-K are spaced closer
together than other 3380 models. The servo control
loop now contains a digital control processor. The
addition of the micro-processor provides improved
control of the servo system.
A1 Board Voltage and Clock LEOs
Data Check Detection and Correction
A record is partitioned into subblocks. For each
subblock there is a set of check bytes. These
check bytes are used for the first-level ECC
detection and correction. As in other 3380 models
there are check bytes at the end of each record. In
3380-JK these bytes are used for second-level ECC
detection and correction.
The additional check bytes used for first-level
detection and correction do not decrease the
maximum data capacity of the track.
This scheme provides for detecting and correcting
multiple defects within a single record. ECC
correction including data area ECC correction is
always performed in the storage control buffer by
the storage control. In previous 3380 models data
area ECC correction was performed by the host.
Controller
Controller
A2 /
Al /
A4
0
-
+5 V
0
0
-
Clock-
0
0
-
-5 V
0
0
- -4.25 V
0
OIA-A1l4
A3
OIA-AIM4
Each controller has Voltage LEOs that show + SV.
-SV. and -4.2SV. The Voltage LEOs come on when
the Controller Power On/Off switch is on.
Each controller also has a Clock LED. When the
controller clocks are running. the clock LEOs are
slightly dimmer than the voltage LEOs. If a clock
LED is off or is at the same brightness level as the
voltage LED. it means the clock is stopped.
Single data area correction is done in the storage
control. The method of reporting data checks is
different from previous 3380 models. Both first-level
and second-level correctable data checks are
reported in the format 4 sense data. The format 4
symptom codes (4xxx) in sense bytes 22 and 23
have been modified to provide additional
information about the data check.
(,
o
INTRO-30
PIN 4519893 41 Sides EC 476573 3ONov88
© Copyright
IBM Corp. 1987, 1988
(
(
B1 Board Power Card LEOs
Front
Drive
(
f
B1 Board - PES/PROM Card LEOs
Rear
Drive
o
left Sequencer Status
o - - +5/24 V Common
o - - +5 V Common
o
o -
0
Right Sequencer Status --
BlG2
0
0
BlR2
(
0
--- left Device Ready - left Switch Enabled - -
0
0
-- Right Switch Enabled ---
0
0
--- Right Device Ready - -
0
0
(
BlG3
(
(
0
BlR3
Each B1 Board has LEOs showing Left and Right
Sequencer status, +5124V Common, +5 V
Common, Left and Right Device Ready, and Left
and Right Switch Enabled for each drive.
After initial power on, the Sequencer Status LEOs
for all 3380-JK models should come on for a few
seconds then flash approximately 3 minutes while
the heads are sweeping across the disk surface.
The Sequencer Status LED is on continuously and
is not flashing after t·he device becomes ready.
(
Right
Front
Device
left
Front
Device
o
+15 V
o
o
+5 V
o
o -o --
-5 V
o
o
-15 V
BID3
BlK3
Left
Rear
Device
Right
Rear
Device
o
o
+15 V
o --
-5 V
+5 V
o -- -15 V
BlN3
o
o
o
0
BlU3
Each B1 board has LEOs showing +15V, +5V, -5V,
and -15V. There is one row of LEOs on each
PES/PROM card for each actuator (device).
Phase Rotation LEOs
The phase LEOs are located on the primary power
box in the A unit (see the LOC section in the MIM,
Volume R10). If the Green Phase LED is on, all ac
input line voltages are present and phasing is
correct.
If the Red Phase LED is on, all ac input line
voltages are present and the phasing is wrong, or
phase C is missing.
(
If both Phase LEOs are off, at least one of the input
voltages is missing or CB200 is tripped.
("
(
(/
PIN 4519893 41 Sides EC 476573 3ONov88
@ Copyright IBM Corp. 1987, 1988
INTRO-31
Controller Voltage Regulator Card LEDs
Controller Power Supply Circuit Protectors
Each controller power supply contains a Controller
Voltage Regulator Card with a + 1.7 UV LED and a
+ SV Special UV LED.
Each controller power supply (there are two
controller power supplies in an AJ4 or AK4 unit)
has three circuit protectors:
The controller voltage regulator cards are located
in the controller power supplies at the front of each
A unit.
CP305
Protects the +SVdc to the controller
power supply regulator card and to the
A 11 A3 or A21 A4 controller.
The Controller Voltage Regulator Card LEOs are
normally off. When the corresponding controller is
turned off or when an error prevents the controller
from completing the normal power-on sequence
the Controller Voltage Regulator Card LEDs will
turn on.
CP306
Protects the +S.SVdc to the controller
power supply regulator card and to the
A 11 A3 or A21 A4 controller.
CP307
Protects the -S.SVdc to the A 1/A3 or A21 A4
controller.
-",
24 Vdc Panel - A and B units
DO NOT
~
~~~~~d1,~L~, ~
C:PLlOLJ.
C:PLlOS
• C:PLl06
BLOWER
2LJ. V
2L1 V
C:PLl01
C:PLl08
FRT
RR
• SWLl86
REMOTE
CP405
(NTL 2
• SWLl87
REMOTE
24 Vdc Panel for A and B Units
The 24 Vdc Panel is located in the rear of the unit
to the left of the Single Actuator Maintenance
panel. The 24 Vdc panel contains the following
circuit protectors and switches:
CP404
(NTL 1
• C:PLll1
IA5084-L I.
• CONTAINED IN THE A UNIT ONLY
Figure 19.
• C:PLll 0
CP404 will trip when a blower motor
thermal circuit problem exists.
Protects the + 24 Vdc circuit to CP404,
CP407, and CP40S. The +24 Vdc from
CP40S provides voltage to CP404, CP407,
and CP40S.
INTRO-32
CP406* Protects the +24 Vdc to the Unit
Emergency Power OnlOff switch circuit.
CP406 also protects the surge complete
relay (K415) and the Phase LED Detection
circuits.
CP407
Protects the + 24 Vdc to the thermal in the
front motor start assembly (MSA), the front
motor thermal, and the front belt guard
switch. If an open condition exists in this
circuit, CP407 will trip, shutting off + 24
Vdc to the front motor run relay (K418) and
front motor brake relay (K416).
PIN 4519893 41 Sides EC 476573 3DNov88
@ Copyright IBM Corp. 1987, 1988
rr----......·\
(
CP408
f
(
Protects the + 24 Vdc to the thermal in the
rear motor start assembly (MSA), the rear
motor thermal, and the rear belt guard
switch. If an open condition exists in this
circuit, CP40B will trip, shutling off + 24
Vdc to the rear motor run relay (K419) and
rear motor brake relay (K417).
SW486* Local/Remote switch for controller A 1. In
the Local position, + 24 Vdc Source for
power sequencing is provided through the
Local/Remote switch. In the Remote
position, + 24 Vdc source for power
sequencing is provided through the power
control cable and relay contacts in the
attached storage control.
(
(
(
.
CP410* Protects the + 24 Vdc to the A 1 Controller
Power On/Off switch. The + 24 Vdc from
the A 1 controller On/Off switch is
distributed to the A 1 controller regulator
(/'
card in the controller A 1 power supply and
picks the A1 controller power relay (K217).
CP411* Protects the + 24 Vdc to the A2 Controller
Power On/Off switch. The + 24 Vdc from
the A2 controller On/Off switch is
distributed to the A2 controller regulator
card in the controller A2 power supply and
picks the A2 controller power relay (K218).
SW487* Local/Remote switch for controller A2. In
the Local position, + 24 Vdc Source for
power sequencing is provided through the
Local/Remote switch. In the Remote
position, + 24 Vdc source for power
sequencing is provided through the power
control cable and relay contacts in the
attached storage control.
* The switch or circuit protector is on the 24 Vdc
panel of 3380 Model AJ4 and AK4 units, but not on
BJ4 or BK4 units .
'
,~'
(
(
PIN 4519893 41 Sides EC 476573 30Nov88
@ Copyright IBM Corp. 1987, 1998
INTRO-33
0
'
\
tt"\
l,~J'
c\
Single Actuator Maintenance (SAM) Control Panel - A and B Units
CJl
FRONT DRIVE
.J675
..1676
..1677
..1678
FRONT
• •• ..167'3 MOTOR
::: [J
•••
•••
•
··• ....•• ··•• ....•••• ··• ....•• ··•• ....••••
·.. ·..
J685
J686
.J687
,
REAR DRIVE
.J688
REAR
..1681
MOTOR.J680 •••
101= ON -=fB1
0
~)
ca>
0
..1683
(f-,\
(.
··• ....•• ··• ....•• ··• ....••
··.... ·.. ·..
J6'30
J6'31
,)6'32
~
~
CP61q
CPSIS
CPS16
~
~
~
REAR SWr.95 CP623
LEFT
ACTUATOR
CP62q
CP625
CP626
~
~
REAR SW696 CP633
RIGHT
ACTUATOR
ENAaLE
bl ~
DJ5A1!1L£
CP63Q
CP635
CP636
~
~
bl
qp
LOGIC'
:~~ LOGIC
[J
cPqp
LOGIC
cp
~.,,)
• ••
CP606
EMAaLE
DJSAI!ILE
i"\
• ••
• ••
CP605
CP60~
FRONT SW69q CPS1)
RIGHT
ACTUATOR
i
"L.J
•••
J689
0
FRONT REAR
rr~
~)
J68~
[J :::
•••
l.§J OFF l§J •
5W6'31
5W692
"'5
"'5
CP602
CP622
o
..1682
,""- .. /
'
IS00012SQL
Figure 20.
Single Actuator Maintenance Control Panel • A and B Units
The single actuator maintenance panel (CE power
sequence panel) has the following controls:
•
Logic Enable/Disable .switches (SW693, SW694,
SW695, SW696)
•
Motor On/Off switches (SW691, SW692)
•
•
+ 5V Common Circuit Protectors (CP602,
CP622)
Other Circuit Protectors (CP603 to CP606,
CP613 to CP616, CP623 to CP626, and CP633 to
CP636)
INTRO-34
PIN 4519893 41 Sides EC 476573 3ONov1!8
@Copyright IBM Corp. 1987, 1988
c
(
(
Motor ONIOFF Switch
Actuator Circuit Protectors
The Motor On/Off switch controls the front or rear
HDA drive motor. With the switch on, the motor is
powered on and off through normal power on/off
sequence. If the switch is off, the motor is
prevented from a normal power on.
Each actuator has a set of circuit protectors that
protect the associated logic circuits from excessive
current levels. The Circuit Protectors are used for
single actuator maintenance.
If the switch is set to Off when the motor is running,
the motor will power off.
(
+ 5 Volt Common Circuit Protectors
(
The + 5V Common Circuit Protectors (CP602,
CP622) protect the power cards.
('
CP602 controls the + 5V to B1 G2.
CP622 controls the +5V to B1R2.
Logic EnablelDisable Switches
Each actuator has one Logic Enable/Disable switch
to permit single actuator maintenance.
Voltage
+/-36v
+/-15v
-5v
+5v
Device
Front HDA
Rear HDA
Left
Right
Right
Left
CP603
CP604
CP605
CP606
CP613
CP614
CP615
CP616
CP623
CP624
CP625
CP626
CP633
CP634
CP635
CP636
Warning: Never power off the CPs for both devices
on the same HDA. without powering off the drive
first. Machine damage will result (see Power
OnlOff procedures on page REPAIR-29).
Warning: Never power off the CPs (CP603 through
CP636) for more than two actuators (devices)
without powering off the entire unit. Machine
damage will result (see Power OnlOff procedures
on pageREPAIR-29).
Warning: When powering off a drive, NEVER power
off the CPs for either of the associated actuators
before turning off the Motor Switch. Machine
damage will result (see Power OnlOff procedures
on page REPAIR-28).
(
c
(
o
PIN 4519893 41 Sides EC 476573 3ONov88
~ Copyright IBM Corp. 1987. 1988
,INTRO-3,S
r: ' .
I.
.'--/
Undervoltage Detection
Service Call Strategy
The 3380-JK provides undervoltage detection for
voltages available on the A1 and B1 boards as
follows:
The service call strategy for the 3380-JK models is:
A 1 Board, + 5 volts and + 8.5 volts
B1 Board, +/-5 volts and +/-15 volts
Undervoltage detection provides a means of
identifying marginal voltages. An undervoltage
condition by itself will not cause an equipment
check condition.
An undervoltage condition will only be reported
when it occurs in conjunction with a controller
check-1, controller check-2, device check-1, device
check-2, or data check.
The maintenance package assigns a higher priority
to undervoltage check conditions and it analyzes
them before attempting to analyze other failures.
INTRO-36
•
Collect all available information from EREP and
console messages.
•
Determine if corrective action is required.
•
Prioritize Symptom Codes.
•
Locate the failing string and devices.
•
Determine the probable failing unit.
•
Perform a physical inspection of the machine.
•
Vary offline (if possible) all required devices on
all required paths.
•
Attach the maintenance device (MD) to the MD
connector in the 3380-JK A unit.
•
Set the devices to be serviced in CE mode.
•
Run the diagnostics.
•
Analyze the SENSE data from the system or
error log if the diagnostics do not fail.
•
Begin problem analysis and follow instructions
displayed by the maintenance device (MD).
See the PSG, beginning with page MD-1).
•
Follow correct power off/on, and repair
procedures.
•
Reset CE mode for all devices at the end of the
call.
•
Reset Write Inhibit and Fenced conditions in the
Storage Control if they exist.
PIN 4519893 41 Sides EC 476573 3ONov88
@Copyright iBM Corp. 1987, 1988
o
G
()
(-,
(
EREP and Console Messages
Maintenance Device (MD)
The primary method of defining the failure type and
boundary is by reviewing EREP reports and console
messages. EREP is an important tool that must be
in place and running when it is needed for the
analysis of DASD problems.
The maintenance device (MD) is a maintenance
tool to be used under the direction of this manual.
There are eight basic functions provided.
(
The maintenance package for the 3380-JK requires
the use of the Environmental Recording, Editing,
and Printing (EREP) program at release 3.3.2 or
later. If your customer has not installed EREP at
release level 3.3.2 or later, provide the assistance
necessary to help get it installed.
(
The EREP System Exception Reports (SYSEXN) are
used by this maintenance package for analysis of
3380-JK problems. The customer should normally
run the System Exception reports on a daily basis.
For information on running the EREP System
Exception reports, see Environmental Record
Editing and Printing Program: User's Guide and
Reference, GC28-1378.
•
Permits you to set or reset CE mode in all
storage paths or storage directors for the
specified devices. CE mode causes the storage
path or storage director to present 'Intervention
Required' to any system that tries to access the
specified devices.
•
•
c/
cc
=
p=
DD =
(-
Controller String 10
With 3380-JK, all controllers in the same
string must have the same 10.
Path
Device
In EREP, the CC.P-DD is defined as a physical
address and is used to locate the failing string,
path, and device. At installation, strict guidelines
are given to properly label the 3380-JK A-unit
operator panel with the necessary information.
The DASD Symptom Code Summary report in EREP
also has a new heading "SSIO-STRING" where:
SSID = Subsystem 10 from sense byte 21.
STRING = String address (0 or 1) from sense byte
4, bit 2.
PIN 4519893 41 Sides EC 476573 3ONov88
© Copyright
IBM Corp. 1987. 1988
Error Log Function
Permits you to perform the following:
Transfer the contents of the error logs to
the MD.
Merge the transferred logs.
Search the contents of each transferred log
or the merged log.
Analyze specific log entries.
Change the error log mask.
Erase (reset) the contents of the error log
in each storage director or storage path.
Physical Identifiers
f
Run Diagnostics
Permits you to run a specified diagnostic
sequence. Nine options are available.
For information about EREP reports, see "EREP
System Exception Reports" on page PROG-1.
The physical identifier reporting format has
changed from previous DASD models (SS-CC-DD).
The format for 3380-JK is "CC.P-DD" where:
Set/Reset CE Mode
•
Analyze Symptom Code
Permits you to analyze a symptom code that
you entered into the MD.
•
Review
Permits you to review the following:
Devices set in CE mode.
The last four diagnostic errors previously
displayed.
The contents of the error logs.
The last repair action previously displayed.
HA scan results
INTRO-37
('(, I\
o
•
Diagnostic Aids
Permits you to run any diagnostic and to enter
parameters.
•
End of Call
Displays the status of the storage control and
permits you to reset CE mode, reset write
inhibit conditions, and (for 3990 storage control
only) reset fenced conditions. Refer to "RMAP
0350: End of Call" on page REPAIR-15.
Error Log
The MD permits you to review error log entries that
are contained in the two storage paths or storage
directors attached to the 3380-JK. The error logs
can be used to help isolate a failure if EREP or
console messages are not available.
Failure data is displayed for review before going
into analysis procedures in the MD. Devices do not
need to be varied offline or placed in CE mode to
use this option. Refer to "Error Log Function (MD
Main menu Option 2)" on page LOG-1.
Diagnostics
Diagnostic options are selected from the MD
Diagnostic menu. Errors are displayed for review
before going into the analysis procedures in the
MD. Before device diagnostics are run, the
devices should be varied offline from all attached
systems and placed in CE mode. Devices may be
placed in or out of CE mode using MD Main menu
option a (Set/Reset CE Mode). Most controller
diagnostics run in-line with customer operation.
Most device diagnostics run concurrent with
customer operation on other devices in the string.
Analysis Procedures
The analysis procedures in the MD can be entered
following the review of error log entries, following a
diagnostic failure, or by entering a symptom code.
The analysis procedures provide a list of suspected
field-replaceable units (FRUs) and references to
documentation. The MD need not be attached to
the 3380-JK when entering the analysis procedures
by entering a symptom code or after reviewing an
INTRO-38
error log entry that was previously transferred to
the MD from the storage control.
o
o
FRU Replacement
The maintenance package provides a primary and
secondary FRU list for most failure symptoms. The
primary list contains all of the high probability
FRUs. The FRU replacement strategy for solid
failures is different from that for intermittent
failures.
c
o
Solid Failures: A solid failure is indicated by a
consistent physical symptom or a consistent
diagnostic program detected error. Solid failures
should be isolated to the failing FRU by replacing
the FRUs indicated in the maintenance package
one at a time until the failure symptom disappears.
The FRUs are listed in the order of probability.
Intermittent Failures: An intermittent failure is a
failure that is not readily reproducible by
diagnostics or is only occasionally indicated by
physical symptoms. On the first call, all FRUs in
the primary FRU list should be replaced as a block
of FRUs. On a repeat call, the secondary FRUs
should be replaced as a block of FRUs and all
additional action items should be reviewed to
determine what type of action will be required if
there is another repeat call. On a second repeat
call, all additional action items should be
performed. If the HDA is suspect, an HDA cable
swap should be performed to help isolate the
problem if there is a repeat call.
Subsystem Microcode Ordering
Information
Subsystem microcode diskettes must always be at
compatible levels with each other and with the
supporting maintenance package documentation.
Updates to these diskettes will be through EC
control procedures.
3380-JK/3880 Subsystem: The 3880 subsystem
configuration requires three microcode diskettes as
follows:
•
3380-JK/3880 subsystem functional microcode
diskette.
PIN 451989341 Sides EC 476573 3ONoV88
~ Copyright IBM Corp. 1987. 1988
c
o
.
,
···\·
0
·
o
("
(
•
3880 MD diskette.
•
3380-JK MD diskette.
3380-JK/3990 Subsystem: The 3990 subsystem
configuration requires two microcode diskettes as
follows:
(-
•
3380-JK MD diskette.
•
3380-JK/3990 subsystem functional microcode
diskette.
Microcode and diskette ordering information can
be found in the Preface, on page PRE-2.
Note: When ordering a diskette to replace a
defective diskette, the EC number of the
defective diskette must be specified.
Otherwise, the latest level diskette is
shipped.
Documentation Ordering Procedure
A Documentation Request form (order number
972-8390) is located in the microfiche holder (see
the MOM, Volume R30). Use this request form to
order microfiche or printed volumes of the
maintenance manuals. Refer to "Maintenance
Manual Ordering Procedure" on page PRE-1 and
"Related Publications" on page PRE-3.
(
f
.
,.
'
... _/J
(~:'
(
f
{--
C
PIN 4519893 41 Sides EC 476573 3ONov88
@ Copyright IBM Corp. 1987, 1988
INTRO-39
(i
. . .,
0'
o
Tools and Test Equipment
The following lists the special tools, test equipment,
and maintenance supplies needed to provide
adequate support for installation and maintenance
actions. When possible, the 3380-JK uses the tools
and test equipment already provided for the
3380-DE.
Description.
asci Iloscope
Maintenance Device
3380 on site tool kit
Air pressure gauge
Bracket, air system tool
Socket assembly ground
Manifold plug
Fitting adapter
Tubing
Bmm Allen extension socket
Spring height gauge
Duct covers
Coil covers
Socket Adapter, 3/8" to 1/4"
3380-DE on site tool kit
ESD cover
16mm/17mm wrench
ESD handling kit
Small grounding wrist strap
Large grounding wrist strap
Digital voltmeter
Vacuum cleaner
Torque wrench
Ratchet drive, 3/8"
High-voltage test probes
Red
Black
Electrical safety analyzer
Electrical safety analyzer
(Not necessary, but useful.)
Pulley-locking clamp
Pulley shipping lock
DOC (CTU) terminators
PIN
2759150
2760742
2135182
2759155
2759954
2760473
2760746
27t>0793
2760883
2759075
2759076
2108746
2179657
2317384
2135145
6428316
6428167
6428169
2515283
6428140
o
Qty
1
1
1
1
1
location
On site (may be located in branch office)
On site (may be located in branch office)
On site (may be located in branch office)
16
2
1
2
1
1
4
4
/.
I
'
"- ~'
1
1
4
On site (may be located in branch office)
1
1
1
1
1
1
1
1
1749249
1749250
1650792
1650792
1
2759423
OOF6760
2315764
1
2
1
1
On site
On site
On site
On site
Branch
Branch
Branch
(may be
(may be
(may be
(may be
office
office
office
Branch
Branch
Branch
Branch
office
office
office
office
located
located
located
located
in
in
in
in
branch
branch
branch
branch
office)
office)
office)
office)
Shipped with each 3380-JK unit
Shipped with each 3380-JK unit
Shipped with each Storage Control
c
INTRO-40
PIN 4519893 41 Sides EC 476573 3ONov88
@ Copyright IBM Corp. 1987, 1988
o
o
(
EC History of Introduction Section
EC HISTORY OF P/N (,519893
(
EC Number
475245
Date Of EC
14Nov86
475246
21Ju187
llSep87
475247
c
EC Number
Date Of EC
475248
25Apr88
476573
30Nov88
-
--
Notes:
(
(
f
(
( -,/
(~
PIN 4519893 41 Sides EC 476573 3ONovBB
© Copyright
(
IBM Corp. 1987, 1988
INTRO-41
3.
MLX
(
This section makes going between this
maintenance library and other machine
maintenance libraries easier.
(
(-
When coming to this library from some other
library:
(
1.
(
2.
Note the exit number on the MIM page or MD
display you are leaving in the other
maintenance package.
Find the same number in the column for that
machine in the Entries from Other Maintenance
Libraries chart on the MLX-2 page for this
machine (3380).
Go to the page listed (in the same row) in the
column labeled Go to page (in this maintenance
package).
When you need to go from this maintenance
package to the maintenance package for some
other machine, you will be given an entry number
for the other package by the page or MAP doing
the sending. Find that number in the 3380 column
in the entry chart in the MLX pages for the other
machine.
To make tracing backwards easier, the chart on
MLX-3 in this manual contains a list of the entry
numbers specified in this maintenance package (for
3380) and the reason for going to the other
maintenance package.
(
(
(
(-
(/
(
PIN 4519895 04 Sides
© Copyright
r-
EC 476581 01Aug89
IBM Corp. 1987, 1988
MLX-1
r··
'''-./
Entries from Other Maintenance Libraries
Use this page to get to the right page or procedure for a given entry to this maintenance package.
Reason for coming to this maintenance package
Go to page
(in this maintenance
package)
N/A
DDC Interface Check FRUs. SC 2004 (sync in check). No
diagnostics failed.
MAP-1, Entry AG (See
MIM, Volume R10)
2
N/A
DDC Interface Check FRUs. SC 2020 (DDC interface
check). No diagnostics failed.
MAP-1, Entry AG (See
MIM. Volume R10)
3
N/A
DDC Interface Check FRUs. SC 2080 (data overrun). No
diagnostics failed.
MAP-1, Entry AG (See
MIM. Volume RIO)
4
4
DDC Interface Check FRUs. The symptom code is a
Format 2. No diagnostics failed.
MAP-1, Entry AG (Se~
MIM. Volume R10)
5
5
3380 symptom code indicated. Device symptom codes
were entered by the CEo The control interface was
tested. No diagnostics failed.
SMAP 0010 Step 001 on
page ST ART-1
6
N/A
No power sequence complete signal. The storage control
did not sense a power sequence complete signal from
the device.
PWR-1, Entry AS (See
MDM, Volume R30)
7
N/A
Routine D3 failed. Routine 77 correct at the storage
control tailgate. The 3380 diagnostic routine D3 found an
error. The CTL-I cable wrap (routine 77) ran without error
when the wrap was performed at the storage control
tailgate.
MAP-1, Entry AK (See
MIM. Volume RIO)
8
N/A
Routine CO failed. Routine 03 correct. The 3380
diagnostic routine CO found an error. The CTL-I cable
wrap (routine 77) ran without error when the wrap was
performed at the storage control tailgate.
MAP-1, Entry AC (See
MIM. Volume R10)
9
N/A
Routine D3 failed (CTRL error IC == D31 B). The 3380
diagnostic routine D3 found an error. The error indicates
that the 3380 controller is failing.
MAP-1, Entry AK (See
MIM, Volume RIO)
10
N/A
DDC Interface Check. The storage control directs the CE
to perform a CTL-I check.
MAP-1, Entry AA (See
MIM. Volume RIO)
N/A
11
CTL-I test failed. Routine 77 correct at the storage
control tailgate.
MAP-1, Entry AB (See
MIM, Volume R 10)
N/A
12
Controller not connected correctly.
IMAP 0295 Step 001 on
page ISOLATE-55
From
a
3880
From
a
3990
1
",---"
-',
...,../
f'\,
\t.j
PIN 451989504 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1997, 1988
MLX-2
(-
Exits to Other Maintenance Libraries
Use this page when tracing back from another maintenance package to understand the condition that
caused entry to that package.
(
(
Exit
from
3380
Reason for going to other package
1
Exit to storage control interface analysis procedure.
(
(
(
(~
•
If attached to a 3880, exit to storage control interface
analysis procedure. Use MD option 1 at the storage
control and answer yes when asked if control interface is
to be tested.
•
If attached to a 3990, select and execute the external DDC
wrap procedure when the manual intervention menu is
displayed.
Exit to 3880
Exit to 3990
MLX Entry 1
MLX Entry 1
2
Exit to the storage control to run diagnostics. The 3380
diagnostics received an error code. The error was found by
the storage control, probably while trying to read from a
diskette.
MLX Entry 2
MLX Entry 2
3
Exit to the storage control to run diagnostics. The 3380
diagnostics found an error in the storage control. The MD
display or the printed MAP contains error information.
MLX Entry 3
MLX Entry 3
4
Not used.
5
Not used.
N/A
N/A
N/A
N/A
6
Exit to the storage control to run diagnostics. The symptom
code does not indi~ate a 3380 error.
MLX Entry 6
MLX Entry 6
7
The status pending light for the storage director is on. Perform
analysis from the storage director.
MLX Entry 7
N/A
8
Exit to the storage control to quiesce a storage path.
MLX Entry 7
9
Exit to the storage control to quiesce a controller.
MLX Entry 8
10
Exit to the storage control for string removal procedure.
N/A
N/A
N/A
11
Exit to the storage control for a 8-unit removal procedure.
N/A
MLX Entry 10
12
Exit to the storage control for an HDA replacement procedure.
N/A
MLX Entry 11
13
Exit to the storage control for a procedure to deactivate the
Maintenance Panel.
N/A
MLX Entry 12
14
Exit to the storage control for an installation procedure.
N/A
MLX Entry 13
15
Exit to the storage control to display 3990 Vital Product Data
(VPD).
N/A
MLX Entry 14
MLX Entry 9
(.
(
..' \
-'
(/
PIN 4519895 04 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987. 1988
MLX-3
EC History of MLX Section
EC HISTORY OF P/N 4519895
EC Number
Date Of EC
EC Number
475245
14Nov86
476573
475246
21Ju187
476581
475248
25Apr88
Date Of EC
30Nov88
01Aug89
Notes:
\"'--_7
,~
PIN 4519895 04 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1987, 1988
MLX-4
J
3380,,",K
(Start
(
SMAP 0010: Start
This section contains analysis procedures
(SMAPs) that help to:
• Determine if a repair action is necessary.
• Prioritize errors.
• Locate the failing unit.
• Determine the probable failing FRUs (field
replaceable units).
• Plan the repair action.
001
(From Page MlX-2)
Questions you must answer and actions you
must perform are printed in bold print. All other
text is explanatory and is designed to help the
service person with less experience with the
3380-JK. As experience is gained. the explanatory text may be skipped. If a question seems
difficult. return to the explanation for that step.
Select an entry from the following table and go
to the MAP or entry point Indicated.
The machine does not have to be available to
perform the following procedures:
Entry
Select One of the Following
Go To
A
Isolate a problem (START REPAIR)
SMAP 0015 Step 001 on page START-2
B
Guidelines for Periodic Review of EREP
Reports (Not problem related)
SMAP 0017 Step 001 on page START-6
C
Install, remove, or relocate a machine
MDM, Volume R30. INST section
D
Perform a safety check
MIM. Volume RfO, SAFETY section
E
Power Off and On
Page REPAIR-1, Entry B through F
F
Perform preventive maintenance
Page PM-1
Figure 1. Start Section Entry Table
(
(
~
Copyright IBM Corp. 1992
START-1
3380-.11<
SMAP 0015: Isolate a Problem
001
(From SMAP 0010 Step 001 on page START~1)
(fromIMA'P 0200 Step 002 on page ISOlATE-1)
(From 'MAP 0290 Step 009 on page ISOLATE-54)
(From .MAP 0290 Step 010 on page ISOLATE-54)
The following SMAPs contain analysis procedures used to determine if a repair action is necessary. If a repair action is necessary; you can
determine (if not already known) the probable
failing FRUs and the impact to the customer of
replacing those FRUs. Good planning allows
you to schedule the correct subsystem
resources. The SMAPs will analyze error data
that was collected at the time of the failure. The
machine does not have to be available to
perform the following procedures.
Collect Information about the failure
Discuss the problem with the customer and
collect all available error information (for
example, EREP reports and console messages).
Sources of error information are physical symptoms, EREP, Device Support Facilities messages,
and console messages. The preferred information for isolating most problems is contained in
the EREP System Exception reports (SYSEXN).
However, if there are known physical symptoms,
EREP reports or sense data will not be necessary.
Physical Symptoms: If you can locate the failing
string of devices due to a physical symptom
such as a Ready LED being off or a tripped CPo
EREP or Sense data is probably not required
because you can enter the maintenance package
with the physical symptom.
EREP Reports: EREP should be installed at
release level 3.3.2 or later to support the
3380-JK. If your customer has not installed
release level 3.3.2, provide the assistance necessary to get it installed.
This maintenance package uses the EREP
System Exception DASD Report. the DASD
and
Symptom Code Summary r&pOrt,
the DASO
Data Transfer Summary report to isolate problems. These reports are part of the EREP
System Exception Reports.
t'''\.
I .
V
Refer to the Environmental Recording and
Editing Printing Program User's Guide,
GC28-1378.
The customer should normally run the System
Exception reports on a daily basis. If you are
isolating temporary I/O errors, it is normally sufficient to use the daify reports provided by the
customer. If you are isolating a permanent I/O
error, you should ask the customer to run the
EREP System Exception reports for 33XX
devices.
r~'
\ ....../
Warning:
Do not attempt to off-load device statistics when
running EREP (SYSEXN) if devices or paths are
failing. A device or path problem can prevent
EREP from successfully collecting statistics and
the EREP job will not complete successfully. To
prevent off-loading statistics, a working data set
must be created from EROS and then EREP run
against the working data set.
To run EREP for a specific problem, two steps
are required:
1. Create a working data set. Have the customer use the following parameters:
PRINT=NO, ACC=Y. ZERO=N, TYPE =0.
and TABSIZE = 999K.
2. Run EREP against the working data set and
print. Have the customer use the following
parameters: SYSEXN=Y, HIST. ACC=N.
TABSIZE=999K and DEV=(3380).
The System Exception reports are normally all
that is required to isolate a problem. If addilional information is required. an Event History
report (EVENT) may be run against the working
data set. If this is done. have the customer use
the following parameters: EVENT=Y. HIST.
ACC-N. TABSIZE=999K. DEV=(3380). and
CUA = (xxx-xxx). where xxx-xxx is the device
address (CUA) range of the string.
(~
"
.1
"L./
(
(
Device Support Facilities (ICKDSF) Messages: If a
customer reports a problem while trying to perform
media maintenance using Device Support Facilities,
collect all Device Support Facilities reports
including the results of ANALYZE DRIVE TEST and
ANALYZE SCAN if available. If possible, collect all
EREP reports for the problem that caused the
customer to initiate a media maintenance action.
Console Messages: Console messages are also a
good source of information, particularly if they
contain sense information that pertains to the
failure. However, EREP reports will provide more
detailed information and should be obtained if at al\
possible.
f
Select your best source of information about the
problem from Figure 2 and go to the MAP or entry
point indicated.
Information Source
Go to
Physical symptoms (sense data not required)
IMAP 0205 Step 001 on page ISOLATE-2.
EREP System Exception reports
SMAP 0017 Step 001 on page START-6.
Device Support Facilities (ICKDSF) messages
SMAP 0052 Step 001 on page START-27.
Sense data from other sources (such as other EREP
reports, console messages, etc.)
SMAP 0035 Step 001 on page START-17.
Console error messages' without sense data
None of the above. The customer cannot provide console
messages or sense data but does provide enough
information to locate the failing string.
Figure 2.
. dMAP 0070 Step 001 on page START-36.
IMAP 0205 Step 001 on page ISOLATE-2.
Information Sources
(
('/
{~:
C
PIN 4519897 38 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1987. 1966
START-3
C'
:,,
SMAP 0016: Using EREP to Determine If There Are 3380 Failures
001
Use the Subsystem Exception DASD report and
refer to Figure 3 on page START-5 to determine if
Storage Control errors have been reported. If
Storage Control errors are reported, the
PROBABLE FAILING UNIT column will contain SCU,
a Storage Control physical identifier, and the
Storage Control type.
See "DASD Subsystem Exception Report" on
page PROG-5 for additional information.
Are there any Storage Control errors reported?
Yes No
I
2.
Review the TOTALS PERM and TEMP columns
to determine permanent and temporary error
counts.
3.
Review the IMPACT OF TEMPORARY ERRORS
columns to determine the type of temporary
errors. To further analyze the problem, the
following procedures will use the DASD
Symptom Code Summary report. In addition, if
data checks are reported (determined by
counts in the RD column), the following
procedures will use the DASD Data Transfer
Summary report. These reports should be
included in the same output as the Subsystem
Exception DASD report.
o
Go to SMAP 0020 Step 001 on page START-S.
002
Go to Step 004 in this SMAP
003
Go to the Start procedure in the Storage Control
maintenance package for the indicated Storage
Control type.
004
(From Step 002 in this SMAP)
Determine if there are 3380 errors reported in the
PROBABLE FAILING UNIT column. If there are
3380 errors reported, the PROBABLE FAILING UNIT
column will contain 3380, 3380-DE, 3380-JK, or
3380-CJ.
Are there 3380 errors reported for the failing
subsystem?
Yes No
I
005
Continue to review EREP periodically for
changes in error conditions.
006
Perform the following procedure:
1.
Review the PHYSICAL ADDRESS column to
determine which physical addresses are
reporting errors. This information will be used
later to locate the failing string and devices.
PIN 4519897 38 Sides
EC 476581 01Aug89
~ Copyright IBM Corp. 1987. 1988
(f ",
'tj
START-4
\,
(
SUBSYSTEM EXCEPTION
DASD
('
(
(
REPORT DATE 156 87
PERIOD FROM 133 87
TO 134 87
B-BUS OUT PARITY CHK C-CHECK
DA~A
PROBABLE
FAILING
UNIT
IMPACT OF TEMPORARY ERRORS
PHYSICAL -TOTALS- EQU
ADDRESS
PERM TEMP
CHK SKS RD OVRN OTHER
l
FAILURE
AFFECT
CPU
CHK
D-DISKETTE CHK
I-INVOKED OFFSETS S-SER
************************************************************~********************
Use this column to ~
determine the physical
address of the machine
reporting errors.
Use this column to determine
the probable failing unit and
machine type.
SCU
CTLR
MULT
DEV
VOL
=
=
=
=
=
Storage Control Unit
Controller
Multiple Devices
Device
Volume
f
r-
I
l Use these columns to
I
determine the count of
each temporary error type.
EQU CHK
SKS
RD
OVRN
OTHER
= Equipment check
= Seek errors
= Data checks
= Overruns
= Other errors including
System Information
Messages (SIM)
Note: The count in the SKS and
RD columns represent the number
of times the temporary threshold
was exceeded (not the total
number of errors).
Use these columns to determine
the total permanent and temporary
error counts.
Figure 3.
Subsystem Exception DASD Report
(
(-
c
PIN 451989738 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1987, 1988
START-5
SMAP 0017: Guidelines for Periodic Review of EREP Reports
001
•
(From SMAP 0010 Step 001 on page START-1)
(From SMAP 0015 Step 001 on page START-3)
(From SMAP 0051 Step 007 on page START-26)
This SMAP contains guidelines for reviewing EREP
on a regular basis to determine if there is a
potential 3380 problem. EREP should be installed
at release level 3.3.2 or later. If your customer has
not installed EREP at release level 3.3.2, provide
the assistance necessary to help get it installed.
The EREP report used in the following procedure is
the Subsystem Exception DASD Report, which is
contained in the EREP System Exception Reports
(SYSEXN). System Exception Reports should be
run on a daily basis. Limits should not be selected
when running the reports because error thresholds
are maintained in the storage control. For
information on running the EREP System Exception
Reports, see Environmental Record and Editing
Printing User's Guide and Reference, GC28-1378.
A recommended schedule for reviewing the System
Exception Reports is based on the number of 3380
units (all models) in the account. The schedule can
vary based on account performance and customer
expectations. The minimum recommended
schedule is as follows:
NUMBER OF 3380 UNITS
IN ACCOUNT
FREQUENCY OF
REVIEW
The number of each type of error from the
Impact of Temporary Errors columns
EQU CHK - Equipment Checks
SKS - Seek Checks
RD - Read Errors (Data Checks)
OVRN - Overruns
OTHER
Determine if corrective action is required.
Use the following guidelines:
•
If permanent errors are indicated in the report,
corrective action is required.
•
If temporary seek checks or data checks are
indicated, a threshold has been exceeded, and
corrective action is required.
•
If 3 or more temporary errors (other than seek
checks, data checks or overruns) are indicated,
corrective action is required.
According to the guidelines, is corrective action
required?
Yes No
I
002
Continue to review EREP periodically for
error conditions.
003
Go to SMAP 0020 Step 001 on page START-8.
32 or more units
Daily
8 to 31 units
Weekly
Less than 8
Monthly
When reviewing the reports, use the System
Exception DASD Report and Figure 4 on
page START-7 to determine the following:
•
The device type is 3380-JK
•
The number of permanent and temporary
errors in the Totals Perm and Temp columns
PIN 4519897 38 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
START-6
:"-
'"
"' .... j '
SUBSYSTEM EXCEPTION
DASD
REPORT DATE 156 87
PERIOD FROM 133 87
TO 134 87
B-BUS OUT PARITY CHK C-CHECK DATA CHK
PROBABLE
FAILING
UNIT
FAILURE
AFFECT
CPU
D-DISKETTE CHK I-INVOKED OFFSETS S-SER
IMPACT OF TEMPORARY ERRORS
PHYSICAL -TOTALS-- EQU
ADDRESS
PERM TEMP CHK SKS RD OVRN OTHER
*********************************************************************************
<1> SCU 00101Z3.1 SCU
TOTAL
1
(
3990.0Z
A
0010123.1
1
<2> CTLR OZ.o-XX SCU/CTLR
338o-JK
A
TOTAL
OZ.O-XX
4
4
A
(
<3>
(
<4> DEV 41.X-02
3380-JK
('
XX.83-XX SCU/CTLR
3380-DE
DEV
6
6
6
6
TOTAL
3l-83-XX
1
1
1
1
A
TOTAL
41.0-02
1
1
A
A
TOTAL
62.0-05
6Z.1-05
10
5
5
<5> VOL
ABC123
DATAXFR
33So-JK
33So-JK
Figure 4.
Subsystem Exception DASD Report
Based on the guidelines for this report, the
following actions are required:
<1> A permanent error was logged for the 3990
Storage Control 0010123, Storage Path 1. A service
action is required. The probable failing unit is the
storage control. More information about the error
can be obtained from the DASD Informational
Messages report. The service action should begin
in the 3990 maintenance package.
(
/
(,
('
<2> Four permanent errors and six temporary
equipment checks were logged for 3380-JK string
02, path O. A service action is required. The
probable failing unit is the controller attached to
path O. More information about the errors can be
determined from the DASD Symptom Code
Summary Report.
<3> One temporary equipment check was logged
for 3380-DE controller 83. No service action is
required. If a service action were required, the
3380-DE maintenance package would be used
because the failing string is a 3380-DE string.
c
c
<4> The temporary seek error threshold was
exceeded once for 3380-JK string 41, path 0, device
02. The probable failing unit is device 02. Because
PIN 4519897 38 Sides
10
5
5
the threshold was exceeded, the next 8 errors
occurring for device 02 will be logged. A service
action is required. More information about these
seek errors can be determined from the DASD
Symptom Code Summary report.
<5> The temporary error threshold for data
checks has been exceeded ten times for 3380-JK
string 62, device 05, volume ABC123. Two paths
connected to device 05 are reporting errors. The
probable failing unit is volume ABC123 (or device
05). Each time the threshold is exceeded, the next
24 errors are recorded. 240 temporary data checks
have been logged for device 5 (10 x 24 = 240).
More information about these data checks can be
determined from the DASD Data Transfer Summary
report.
The customer should run ICKDSF ANALYZE DRIVE
TEST on device 05 (volume ABC123) to determine if
there is a suspected hardware problem. If there is
a suspected hardware problem, a service action is
required; otherwise, the customer should perform
media maintenance in accordance with instructions
in Device Support Facilities Primer for the User of
IBM 3380 Direct Access Storage, GC26-4498 and
Maintaining IBM Storage Subsystem Media,
GC26-4495.
START-7
EC 476581 01Aug89
(') Copyright IBM Corp. 1987, 1988
-----
1
1
"'--'-----------
SMAP 0020: Corrective Action Determination With EREP
Use the EREP Subsystem Exception DASD
Symptom Code Summary report and the EREP
System Exception DASD Data Transfer Summary
report to determine if corrective action is required.
See the PROG section of this manual for sample
EREP reports and descriptions.
Exception DASD Symptom Code Summary
report, Figure 5 on page START-10.)
•
001
(From SMAP 0016 Step 006 on page START-4)
(From SMAP 0017 Step 003 on page START-6)
(From IMAP 0233 Step 006 on page ISOLATE-21)
4xCx, 9E12, 9FFF, 9F9F
DFFF, EFFF, EFFO
(See the EREP System Exception DASD
Symptom Code Summary report, Figure 5
on page START-10.)
Determine if corrective action is required. Do you
need help?
Yes No
•
I
002
Go to SMAP 0025 Step 001 on page
START-12.
003
Use the DASD Symptom Code Summary report and
Figure 5 on page START-10 to determine the
following:
1.
Determine that the failure is a 33SQ-JK failure.
Verify that the Device Type Column contains
3380-JK or the symptom code is Dxxx.
=
Determine the kind of errors (Permanent or
Temporary) from the OCCURRENCES
PERMITEMP columns.
3.
Determine if corrective action is required.
Use these guidelines:
•
•
~.
I
If the threshold for temporary seek errors
(symptom codes 9Dxx and 9F1E) was
exceeded on the DASD Symptom Code
Summary report, corrective action is
required. (See the EREP System Exception.
DASD Symptom Code Summary report,
Figure 5 on page START-10.)
o
START-8
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
-----
--_._._--
--~
.------~
-~-
:
~I
Symptom codes 9Dxx and 9F1E will not be
presented in the Symptom Code Summary
report unless the threshold is exceeded.
When the threshold for seek checks is
exceeded, a format 0 sense record
(symptom code 0001) is logged and a
If permanent errors are indicated in the
DASD Symptom Code report, corrective
action is required. (See the EREP System
PIN 4519897 38 Sides
If the threshold for temporary data checks
was exceeded (indicated by a nonzero
value in either the TEMPORARY OFFSET
INVK YES or NO columns of the Data
Transfer Summary report), corrective action
is required. (See the TEMPORARY OFFSET
INVK YES and NO columns of the EREP
System Exception DASD Data Transfer
Summary report, Figure 6 on
page START-11, for additional information
about data checks (Symptom Code 4xxx».
When the threshold for data checks without
. offset is exceeded, the storage control goes
into logging mode. After logging mode is
completed, a format 0 sense record
(symptom code 0002) is logged and a
message is presented in the DASD
Informational Messages report (see
Figure 5 on page PROG-11).
There are cases where 3380-JK errors
(symptom code = Dxxx) are reported with the
wrong device type. For these cases a symptom
code of Dxxx is a definite indication of an error
on a 3380-JK device type. There are also
cases where 3380 and 3380-DE format 7 errors
(symptom code
7xxx or Bxxx) are reported
for 3380-JK device type. For these errors, refer
to the 3380 and 3380-DE maintenance package.
2.
If temporary errors with one of the following
symptom codes is reported in the DASD
Symptom Code report, corrective action is
required:
-- --------
(('
message is presented in the DASD
Informational Messages report (see
Figure 5 on page PROG-11).
(-
(-
•
If three or more temporary errors (other
than above) are indicated on the DASD
Symptom Code Summary report, corrective
action is required, (See the EREP System
Exception DASD Symptom Code Summary
report, Figure 5 on page START-10.)
•
If fewer than three temporary errors (other
than above) are indicated on the DASD
Symptom Code Summary report,
maintenance action is not generally
required.
Temporary errors are usually transparent
to the customer's normal operation unless
the errors are occurring so often that
performance is degraded.
c:
Continue to review EREP each day for additional
errors. If the error condition persists, corrective
action is required.
According to the guidelines, is corrective action
required?
Yes No
I
(/
(/
c
004
Continue to review EREP daily for changes in
the error conditions.
005
Go to SMAP 0025 Step 001 on page START-12.
PIN 4519897 38 Sides
© Copyright
EC 476581 01Aug89
IBM Corp, 1987, 1988
START-9
DASD SYMPTOM
COD~
SUMMARY
o
REPORT DATE 034 87
PERIOD FROM 032 87
TO 03487
;:r~
:,,-~j
SEQUENCE BY PROBABLE FAILING UNIT
SYMPTOM PHYSICAL OCCURRENCES FAILURE
CODE
ID
PERMITEMP AFFECT
DEVICE
TYPE
0 0 0 0
DATE AND TIME OF
FIRST OCCURRENCE
LAST OCCURRENCE
SENSE FROM FIRST OCCURRENCE
0 0 0 0
0 0 1 1
1 1 1 1
1 1 1 1
2 2 2 2
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 78 9 0 1 2 3
PHYSICAL
ERROR
ADDRESS SSID-STRING PATH CPU{S)
*******************************************************************************
PROBABLE FAILING UNIT:
SEQUENCE BY _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
L~
338G-JK
Look in these columns for counts of
permanent and temporary errors.
(
Use this column to identify the device type.
\,
'",-/
Use this column to identify the symptom code.
Valid codes for 3380-JK are:
Figure 5.
4xxx
9xxx
Dxxx
Exxx
Headings for the DASD Symptom Code Summary Report.
additional information about this report.
See "DASD Symptom Code Summary" on page PROG-17 for
rf"'
\l)
PIN 4519897 38 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1987. 1988
START-10
(
(
DASD DATA TRANSFER SUMMARY
PROBABLE FAILING UNIT - VOLUME
REPORT DATE 034 87
PERIOD FROM 032 87
TO 034 87
SENSE COUNTS
TEMPORARY
OFFSET INVK THRESHOLD
PERM
NO
YES LOGGING
'J
**********************************************************************************
Use this column to find the number of permanent
data checks that occurred in the report period.
Use this column to find the number of times the temporary
data check threshold was exceeded in the report period.
The error records associated with numbers in this column
contain the volume ID and the device address of the
device that exceeded the threshold.
(
Use this column to find the number of temporary data checks - - - - -....
recovered with offset.
(
Use this column to find the number of temporary data checks
that were logged during threshold logging.
(
Figure 6.
----------------~
Headings for DASD Data Transfer Summary Report
('
(
(
..
-_/
(:
C;
--,'
PIN 4519897 38 Sides
EC 476581 01 Aug89
© Copyright IBM Corp. 1987, 198B
START-11
()
SMAP 0025: Prioritizing Symptom Codes
001
(From
(From
(From
(From
SMAP 0020 Step 002 on page START-8)
SMAP 0020 Step 005 on page START-9)
SMAP 0030 Step 003 on page START-14)
MD 0050 Step 003 on page MD-55)
Determine the highest priority symptom code.
Check for symptom codes that indicate
undervoltage.
All undervoltage or non-repeatable runout
problems should be resolved before attempting to
resolve other problems. Review the symptom
codes for the following:
DFFF
EFFF
EFFO
9FFF
9F9F
Controller undervoltage
Controller undervoltage
Device undervoltage
Device undervoltage
Non-repeatable runout
Do you have one of these symptom codes?
Yes No
I
002
Go to Step 004 in this SMAP.
003
This is the highest priority symptom code. Record
the symptom code and go to SMAP 0030 Step 001
on page START-14.
004
(From Step 002 in this SMAP)
Use the DASD Symptom Code Summary report to
answer the following questions and to find the
highest priority symptom code. (A description of
the DASD Symptom Code Summary report starts
on PROG-17.)
Is there more than one symptom code listed?
Yes No
I
005
(Step 005 continues)
PIN 4519897 38 Sides
© Copyright
EC 476581 01Aug89
005 (continued)
The symptom code listed is the highest
priority symptom code. Record the symptom
code and go to SMAP 0030 Step 001 on page
START-14.
o
o
006
(From Steps 010 and 012 in this SMAP)
The following symptom codes are low priority
symptom codes and should not be used to isolate a
problem when other symptom codes with a value
of Dxxx or Exxx are reported:
Low Priority Symptom Codes
DFx3
DFx5
DFx8
DFx9
DFxA
DFxB
DFxC
DFxD
Is there a low priority symptom code?
Yes No
I
007
Go to Step 010 in this SMAP.
008
Are there other Dxxx or Exxx symptom codes
(temporary or permanent)?
Yes No
I
009
The highest priority symptom code is the
first low priority symptom code listed.
Record the symptom code and go to SMAP
0030 Step 001 on page START-14.
Note: Symptom code DF05 may be
caused by a control interface fenced
condition or a control/er power off
condition. There will usually be a
preceding symptom code.
010
(From Step 007 in this SMAP)
(Step 010 continues)
START-12
IBM Corp. 1987, 1988
)
(-
(
010 (continued)
Are there any symptom codes (other than the low
priority symptom codes listed in Step 006 in this
SMAP) with permanent errors?
Yes No
I
011
Determine the highest priority symptom code
group. Use the following guidelines:
1.
Symptom codes are grouped by the first
character of the code.
2.
If there are symptom codes in more than
one of the following groups, select the
group closest to the top of the list.
(
(
Symptom
Code
Group
Meaning
3xxx
2xxx
Dxxx
Exxx
9xxx
4xxx
Oxxx
3.
(
(\
(-~
Storage Control Check 1
Storage Control Equipment
Check
Controller Check 1
Controller Check 2 or
Device Check 1
Device Check 2
Data Check
Program or System Check
After the highest priority group is
determined, go to Step 015 in this SMAP.
012
Is there more than one symptom code (other than
the low priority symptom codes listed in Step 006
in this SMAP) with permanent errors?
Yes No
I
013
The symptom code with permanent errors is
the highest priority symptom code. Record
the symptom code and go to SMAP 0030
Step 001 on page START-14.
Symptom codes are grouped by the first character
of the symptom code. Refer to the following table
to group the symptom codes with permanent
errors. If there are symptom codes with permanent
errors in more than one of the groups listed in the
table, select the group closest to the top of the
table.
Symptom
Code
Group
Meaning
3xxx
2xxx
Dxxx
Exxx
Storage Control Check 1
Storage Control Equipment Check
Controller Check 1
Controller Check 2 or
Device Check 1
Device Check 2
Data Check
Program or System Check
9xxx
4xxx
OXXX
After the highest priority group is determined, go to
Step 015 in this SMAP.
015
(From Steps 011 and 014 in this SMAP)
Is there more than one symptom code in the
selected group?
Yes No
I
016
The symptom code in the group selected is
the highest priority symptom code. Record
the symptom code and go to SMAP 0030
Step 001 on page ST ART-14.
017
The symptom code in the selected group that
occurs most often (permanent and temporary) Is
the highest priority symptom code. Record the
symptom code and go to SMAP 0030 Step 001 on
page START-14.
014
Determine the highest priority symptom code
group.
PIN 4519897 38 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1997, 1999
START-13
SMAP 0030: Locating the Failing String and Devices
001
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
SMAP 0025 Step 003 on page START-12)
SMAP 0025 Step 005 on page START-12)
SMAP 0025 Step 009 on page START-12)
SMAP 0025 Step 013 on page START-13)
SMAP 0025 Step 016 on page START-13)
SMAP 0025 Step 017 on page START-13)
IMAP 0205 Step 003 on page ISOLATE-2)
RMAP 0310 Step 001 on page REPAIR-5)
MO 0050 Step 003 on page MO-55)
MO 0052 Step 005 on page MO-62)
Warning: Use the data in the Physical Address
column of the report to identify the physical
controller, path, and device. Do not use the Error
Path column or the Physical 10 column to Identify
the failing paths or devices. The Error Path column
identifies the path the error was reported on and
the device that was selected when the error was
reported, not necessarily the failing physical
address.
•
To identify the string. match the controller 10
(CC) in the Physical Address column with the
string 10 on the A unit operator panel.
•
To identify the Storage Control. match the SOlD
or SSIO from the SSID-STRING column with the
SOlD or SSIO on the storage control.
•
To identify the controller. match the path bit (P)
in the Physical Address column with the path
character on the A unit operator panel.
•
To identify devices, match the device address
(DO) in the Physical Address column with the
device addresses on the A unit operator panel.
Locate the failing machine. Do you need help?
Yes No
I
002
Go to SMAP 0033 Step 001 on page
START-16.
003
Use the entry with the highest priority symptom
.
code.
Using the OASO Symptom Code Summary report
(see Figure 7 on page START-15). perform the
following steps.
Go to SMAP 0033 Step 001 on page START-16.
1.
Verify that the Device Type column contains
3380·JK or the symptom code is Dxxx.
Note: If two strings are attached to the control
interface, there are some cases where format 7
errors (sense byte 7= 7x) are reported for both
strings. For these cases, it is necessary to
locate both strings. To locate 3380 or 3380-DE
strings, use the appropriate 3380 maintenance
package.
2.
Determine the physical string, controllers,
devices, and storage control related to the
failure.
c
If there are multiple symptom codes in the
symptom code group determined in SMAP 0025
Step 001 on page START-12. use all physical
addresses listed within the selected group (both
permanent and temporary errors) to make this
determination (not just those physical addresses
listed under the highest priority symptom code).
PIN 4519897 38 Sides
EC 476581 01Aug89
@Copyright IBM Corp. ISS7, 19a5
START-14
(
REPORT DATE 034 87
PERIOD FROM 032 87
TO 034 87
DASD SYMPTOM CODE SUMMARY
(
SEQUENCE BY PROBABLE FAILING UNIT
SYMPTOM PHYSICAL OCCURRENCES FAILURE
CODE
ID
PERM/TEMP AFFECT
DEVICE
TYPE
(
«
(
DATE
FIRST OCCURRENCE
SENSE FROM
0 0 0 0 0 0 0 0 0 0 1 1 1
0 1 2 3 4 5 6 7 8 9 0 1 2
PHYSICAL
ERROR
ADDRESS SSID-STRING PATH
AND TIME OF
LAST OCCURRENCE
FIRST OCCURRENCE
1 1 1 1 1 1 1 2 2 2 2
3 4 5 6 7 8 9 0 1 2 3
CPU(S)
*******************************************************************************
PROBABLE BY
FAILING
UNIT:
SEQUENCE
________________________________________________________
__
- DATE TIME - - - -
- - - -
CC.P-DD
Use this column to identify
I
the failing controllers.
~
paths, and devices:
CC = Controller String ID
P = Physical Path
DD = Device Physical Address
- DATE TIME -
- SENSE - DATA
SC-S
L
- - - -
- - - -
Use this column to identify
the Storage Control and
string address (0 or 1):
SC = storage Control
S = String (0 or 1)
SC-CC-DD
For certain Format 7 errors,
I
the physical address is
~
reported in the same form
as previous 3880 models:
SC = Storage Control
CC = Controller String ID
DD = Device Physical Address
Figure 7.
DASD Symptom Code Summary Report, Physical Address field.
for additional Information about this report.
See "DASD Symptom Code Summary" on page PROG-17
(
(
(~\
PIN 4519897 38 Sides
«:> Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
START-15
o
SMAP 0033: Determine the Probable Failing Unit
001
(From SMAP 0030 Step 002 on page START-14)
(From SMAP 0030 Step 003 on page START-14)
1. Use the DASD Symptom Code Summary report to
determine the probable Falling Unit.
Determine the probable failing unit from the first
line following the report heading. See Figure 9.
2. Find the highest priority symptom code group in
Figure 8 and go to the SMAP, MLX entry, or
maintenance package indicated.
Symptom Code Groups
Go To
2xxx, 3xxx, or Fxxx Storage Control
errors
The storage control maintenance package, MLX entry 6.
Dxxx
SMAP 0058 Step 001 on page START-34.
EBxx (Controller/Device Port checks)
SMAP 0057 Step 001 on page START-33.
Exxx (except EBxx)
SMAP 0058 Step 001 on page START-34.
9xxx
SMAP 0058 Step 001 on page START-34.
1xxx, 6xxx, 7xxx, or8xxx
The 3380 Stage 2 maintenance package.
Axxx, Bxxx, or Cxxx
The 3380-DE maintenance package.
4xxx
Data Check
SMAP 0050 Step 001 on page START-23.
Oxxx
Figure 8.
The storage control maintenance package, MLX entry 6.
Symptom Code Groups.
DASD SYMPTOM CODE SUMMARY
REPORT DATE 034 87
PERIOD FROM 032 87
TO
034 87
SEQUENCE BY PROBABLE FAILING UNIT
SYMPTOM PHYSICAL OCCURRENCES FAILURE
CODE
ID
PERM/TEMP AFFECT
DATE AND TIME OF
FIRST OCCURRENCE
LAST OCCURRENCE
SENSE FROM FIRST OCCURRENCE
0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 2 2 2 2
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3
PHYSICAL
ERROR
ADDRESS SSID-STRING
PATH CPU(S)
DEVICE
TYPE
*******************************************************************************
J
PROBABLE FAILING UNIT: XXXXX~~
SEQUENCE BY
Use this field to determine
the PFU (probable failing
unit)
Figure 9.
DASD Symptom Code Summary Report, Probable Failing Unit.
PIN 4519897 38 Sides
© Copyright
Probable failing units are:
Channel
Multiple (more than
Storage Control
one device)
Device
Unknown
Volume
EC 476581 01Aug89
IBM Corp. 1987. 1988
START-16
o
(
SMAP 0035: Corrective Action Determination With Sense Data
(
001
006
(From SMAP 0015 Step 001 on page START-3)
(From Step 004 in this SMAP)
Determine if corrective action is required.
Go to the appropriate maintenance package as
indicated in the following list:
Do you need help?
Yes No
Symptom Meaning and
Code
Maintenance Package
I
(
002
(
(
6xxx
003
1.
2.
f
3.
(
8xxx
Determine the symptom code from
sense bytes 22 and 23.
Axxx
Bxxx
Cxxx
Determine that the failure is a 3380·JK failure
from the symptom code.
Is this a valid 3380-JK failure?
Yes No
I
004
Go to Step 006 in this SMAP.
(
7xxx
Determine if the error is permanent or
temporary from console messages.
Valid 3380-JK symptom codes are 4xxx (except
49xx), 9xxx, Dxxx, and Exxx. 49xx symptom
codes are not used for the 3380-JK.
(
1xxx
5xxx
Go to SMAP 0040 Step 001 on page
START-19.
005
Go to Step 007 in this SMAP.
Device Check 2 - 3380 Stage 2
Correctable data check
- 3380 Stage 2 or 3380-DE
Controller Check 2 or
Device Check 1 - 3380 Stage 2
Controller Check 1
• 3380 Stage 2
Controller Check 2 or
Device Check 1 - 3380 Stage 2
Device Check 2 - 3380-DE
Controller Chec.k 1 - 3380-DE
Controller Check 2 or
Device Check 1 • 3380-DE
007
(From Step 005 in this SMAP)
Determine if corrective action is required. Use
these guidelines:
•
If permanent errors are indicated, corrective
action is required.
•
If temporary errors with one of the following
symptom codes is reported, corrective action is
required:
(
4xCx,
DFFF,
•
9E12,
EFFF,
9FFF, 9F9F
EFFO
If the threshold for temporary data checks
(symptom codes 4xxx) was exceeded,
corrective action is required.
A threshold exceeded condition can be
determined from console messages or from a
format 0 sense record with a symptom code of
0002.
(C/
C'
PIN 4519897 38 Sides
EC 476581 01Aug89
«:) Copyright IBM Corp. 1987, 1988
START-17
•
If the threshold for temporary seek errors
(symptom codes 9Dxx and 9F1E) was exceeded,
corrective action is required.
A threshold exceeded condition can be
determined from console messages or from a
format 0 sense record with a symptom code of
0001.
Note: Symptom codes 9Dxx and 9Ft E
are not presented to the system unless
the threshold is exceeded.
•
If three or more temporary errors (other than
above) are indicated, corrective action is
required.
•
If fewer than three temporary errors (other than
above) are indicated, maintenance action is not
generally required.
Temporary errors usually do not affect the
customer's normal operation unless the errors
are occurring $0 often that performance is
degraded.
Review EREP each day for additional errors. If the
error condition persists, corrective action is
required.
According to the guidelines, is corrective action
required?
Yes No
I
008
Review EREP daily for changes in the error
conditions.
009
Go to SMAP 0040 Step 001 on page START-19.
PIN 4519897 38 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
START-iS
(
C
f
(
SMAP 0040: Prioritizing Symptom Codes
001
(From
(From
(From
(From
(From
SMAP 0035 Step 002 on page START-17)
SMAP 0035 Step 009 on page START-18)
SMAP 0045 Step 004 on page START-22)
SMAP 0053 Step 005 on page START-30)
MD 0050 Step 004 on page MD-56)
006
(From Steps 010 and 012 in this SMAP)
The following symptom codes are low priority
symptom codes and should not be used to isolate a
problem when other symptom codes with a value
of Dxxx or Exxx are reported:
Determine the highest priority symptom code.
Low Priority Symptom Codes
(
Check for symptom codes that indicate
undervoltage.
All undervoltage or non-repeatable runout
problems should be resolved before attempting to
resolve other problems. Review the symptom
codes for the following:
DFFF
EFFF
EFFO
9FFF
9F9F
Controller undervoltage
Controller undervottage
Device undervoltage
Device undervoltage
Non-repeatable runout
Do you have one of these symptom codes?
Yes No
DFx3
DFx5
DFx8
I
007
Go to Step 010 in this SMAP.
008
Are there other Dxxx or Exxx symptom codes
(temporary or permanent)?
Yes No
I
009
The highest priority symptom code is the
first low priority symptom code listed.
Record the symptom code and go to SMAP
0045 Step 001 on page START-21.
002
Go to Step 004 in this SMAP.
003
This is the highest priority symptom code. Record
the symptom code and go to SMAP 0045 Step 001
on page START-21.
Note: Symptom code DF05 may be
caused by a control interface fenced
condition or a controller power off
condition. There will usually be a
preceding symptom code.
(
('
(-'
(0,
004
(From Step 002 in this SMAP)
Perform the following MAP steps to find the highest
priority symptom code.
Is there more than one symptom code listed?
Yes No
I
005
The symptom code listed is the highest
priority symptom code. Record the symptom
code and go to SMAP 0045 Step 001 on page
START-21.
PIN 4519897 38 Sides
EC 476581 01 Aug89
@ Copyright IBM Corp, 1987. 1988
(
DFxC
DFxD
Is there a low priority symptom code?
Yes No
I
(
DFx9
DFxA
DFxB
010
(From Step 007 in this SMAP)
Are there any symptom codes (other than the low
priority symptom codes listed in Step 006 in this
SMAP) with permanent errors?
Yes No
I
(Istep 011
continues)
START-19
011
Determine the highest priority symptom code
group. Use the following guidelines:
table.
Symptom
Code
Meaning
Group
1.
Symptom codes are grouped by the first
character of the code.
2.
If there are symptom codes in more than
one of the following groups, select the
group closest to the top of the list.
3xxx
2xxx
Dxxx
Exxx
Symptom
Code
Group
Meaning
9xxx
4xxx
OXXX
3xxx
2xxx
Dxxx
Exxx
9xxx
4xxx
Oxxx
3.
Storage Control Check 1
Storage Control Equipment
Check
Controller Check 1
Controller Check 2 or
Device Check 1
Device Check 2
Data Check
Program or System Check
After the highest priority group is
determined, go to Step 015 in this SMAP.
012
Is there more than one symptom code (other than
the low priority symptom codes listed in Step 006
in this SMAP) with permanent errors?
Yes No
I
013
The symptom code with permanent errors is
the highest priority symptom code. Record
the symptom code and go to SMAP 0045
Step 001 on page START-21.
Storage Control Check 1
Storage Control Equipment Check
Controller Check 1
Controller Check 2 or
Device Check 1
Device Check 2
Data Check
Program or System Check
After the highest priority group is determined, go to
Step 015 in this SMAP.
015
(From Steps 011 and 014 in this SMAP)
Is there more than one symptom code in the
selected group?
Yes No
I
016
The symptom code in the group selected is
the highest priority symptom code. Record
the symptom code and go to SMAP 0045
Step 001 on ·page START-21.
017
The symptom code in the selected group that
occurs most often (permanent and temporary) is
the highest priority symptom code. Record the
symptom code and go to SMAP 0045 Step 001 on
page START-21.
014
Determine the highest priority symptom code
group.
Symptom codes are grouped by the first character
of the symptom code. Refer to the following table
to group the symptom codes with permanent
errors. If there are symptom codes with permanent
errors in more than one of the groups listed in the
table, select the group closest to the top of the
PIN 4519897 38 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
START-20
(~.\
\. >
(
SMAP 0045: Locating the Failing String and Devices
(
001
(From
(From
(From
(From
(From
(From
(From
(From
(From
(
("
3.
SMAP 0040 Step 003 on page START-19)
SMAP 0040 Step 005 on page START-19)
SMAP 0040 Step 009 on page START-19)
SMAP 0040 Step 013 on page START-20)
SMAP 0040 Step 016 on page START-20)
SMAP 0040 Step 017 on page START-20)
IMAP 0205 Step 003 on page ISOLATE-2)
RMAP 0310 Step 001 on page REPAIR-5)
MD 0050 Step 004 on page MD-56)
Sense Byte 4
Path
OOxx xxxx
01xx xxxx
10xx XXXX
11xx xxxx
Path
Path
Path
Path
Sense Byte 4
String Address
I
xxOx xxxx
xx1x xxxx
String 0
String 1
002
Go to Step 004 in this SMAP.
Sense Byte 4
Device Address
xxxO
xxxO
xxxO
xxxO
Device
Device
Device
Device
Locate the failing machine. Do you need help?
Yes No
(
(
003
Determine the physical strings, controllers, and
devices (CC.P.DD).
If there are multiple symptom codes in the
symptom code group determined on page
ST ART -19, use all sense records in the selected
group to make the following determination (not just
those sense records associated with the highest
priority symptom code).
('
Note: If there are two strings attached to
the control interface, there are some cases
where format 7 errors (sense byte 7= 7x) are
reported for both strings. For these cases it
is necessary to locate both strings. To locate
3380 or 3380-DE strings, use the appropriate
3380 maintenance package.
1.
(~.
2.
(
"',
_/
Use sense byte 3 or 14 to determine the string
10. See Figure 11 on page START-22.
For example, if the symptom code is 9xxx,
Dxxx, or Exxx, use sense byte 3 to determine
the String ID. If the symptom code is 4xxx, use
sense byte 14.
('
0000
0001
0010
0011
xxx11110
xxx11111
0
1
2
3
00
01
02
03
Device 1E
Device 1F
4.
To identify the controller, use sense byte 4 to
determine the path and device address. Next,
match the path and device address from sense
byte 4 with the Path and Device on the A-Unit
operator panel. For 3880 subsystems, see
page INTRO-7; for 3990 subsystems, see page
INTRO-20 or INTRO-22.
5.
To identify a device, match the device address
from sense byte 4 with the Device on the A unit
operator panel. For 3880 subsystems, see
page INTRO-7; for 3990 subsystems, see page
INTRO'-20 or INTRO-22.
Continue with Step 004 in this SMAP.
To identify the string, match the string ID
obtained from sense byte 3 or 14 with the
String 10 on the A-unit operator panel. For
3880 subsystems, see page INTRO-7; for 3990
subsystems, see page INTRO-20 or INTRO-22.
PIN 4519897 38 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
('
Use sense byte 4 to determine the path, the
string address, and the device address:
START-21
004
(From Steps 002 and 003 in this SMAP)
Find your highest priority symptom code group in
Figure 10 (determined in SMAP 0040 Step 001 on
page START-19) and go to the SMAP, MLX entry, or
maintenance package indicated.
Symptom Code Groups
Go To
2xxx, 3xxx, or Fxxx Storage Control
errors
The storage control maintenance package, MLXentry 6.
Dxxx
SMAP 0058 Step 001 on page START-34.
EBxx (Controller/Device Port checks)
SMAP 0057 Step 001 on page START-33.
Exxx (except EBxx)
SMAP 0058 Step 001 on page START-34.
9xxx
SMAP 0058 Step 001 on page START-34.
1xxx, 6xxx, 7xxx, or8xxx
Axxx, Bxxx, or
4xxx
Oxxx
Figur~
/~
The 3380 Stage 2 maintenance package.
Cxxx
The 3380-DE maintenance package.
Data Check
SMAP 0050 Step 001 on page START-23.
The storage control maintenance package, MLX entry 6.
10.
Symptom
Cod~
Groups
Sense Symptom Code
Error Type
Byte 7 (bytes 22 and 23)
Storage
Director
ID Byte
Controller
or String
ID Byte
Device ID
and
Path Byte
3x
Oxxx
9xxx
2xxx
3xxx
4x
4xxx
Data checks
Byte 21
Byte 14
Byte 4
7x
Dxxx
Controller check-1
Byte 21
Byte 3
Byte 4
8x
Exxx
Controller check-2 or device check-1
Byte 21
Byte 3
Byte 4
9x
9xxx
Device check-2 (R/W checks)
Byte 21
Byte 3
Byte 4
Fx
Fxxx
Storage Control check or message
Byte 21
*
*
Ox
1x
2x
Figur~
\
11.
Program or system check
Byte 21
Device check-2 (except R/W checks)
Byte 21
Byte 3
Storage Control equipment checks
Byte 21
Storage Control check-1
Byte 21
*
*
*
"ir
Byte 4
~'r
*
Physical 10 Bytes by Sense Format
* See the Storage Control maintenance package for sense byte definitions.
\
,
'''1:...../
r-~
PIN 4519897 38 Sides
© Copyright
EC 476581 01AugB9
IBM Corp. 1987, 198a
START-22
(
"'--
,/
(
f"
f.
SMAP 0050: Planning the Repair Action for Data Checks
001
3.
(From SMAP 0033 Step 001 on page START-16)
(From SMAP 0045 Step 004 on page ST ART-22)
The highest priority symptom found was data
checks (symptom code group 4xxx).
(
('
(
(
Use the EREP System Exception report, DASD Data
Transfer Summary If available. See Figure 12 on
page START-24. An analysis of this report will
assist you in making the decisions that are
required to isolate and fix data check problems.
Defects in the media will be fixed by the Device
Support Facilities program or other programs.
Hardware errors will require a repair action.
For information about media maintenance, see
Maintaining IBM Storage Subsystem Media,
GC26-4495 and Device Support Facilities Primer for
the User of IBM 3380 Direct Access Storage,
GC26-4498.
1.
Determine if there are permanent data checks.
Examine the PERM column in the DASD Data
Transfer Summary or examine console
messages for permanent data checks.
Warning: Do not use sense byte 1, bit 0 to
determine if a data check is permanent.
2.
('
('
Determine if there are data checks that required
offset for recovery.
Determine if there are data check threshold
exceeded conditions.
Examine the OFFSET INVK NO column in the
DASD Data Transfer Summary report or sense
byte 2, bit 2 of each sense record to find entries
for devices that were put in Logging mode.
Sense byte 2, bit 2 (first logging mode)
identifies the first record logged for a device in
logging mode.
A data check with a symptom code of 4xCx
may be reported in this column. 4xCx symptom
codes are all logged. (The threshold does not
apply.)
4xCx indicates that the data check was not
recoverable with one or more control unit retry
sequences. A data check of this type should be
treated as permanent. For additional
information on data check reporting, see
"Understanding Temporary Data Check
Reporting" on page PROG-1S.
Did more than one device in the failing string report
data checks.
Yes No
I
002
Go to SMAP 0051 Step 001 on page
START-25.
003
Go to SMAP 0056 Step 001 on page START-32.
Examine the OFFSET INVK YES column in the
DASD Data Transfer Summary report or
examine sense byte 23 to determine if data
checks with offset were reported. If byte 23
contains hex 2x, 6x, or Ax, offset was active.
('"
(~~
C,\
PIN 4519897 38 Sides
CO Copyright
('
EC 476581 01Aug89
IBM Corp. 1987. 1988
START-23
,ff
l, ~
REPORT DATE 034 85
PERIOD FROM 032 85
TO 032 85
DASD DATA TRANSFER SUMMARY
PROBABLE FAILING UNIT - VOLUME
Vi
SENSE COUNTS
TEMPORARY
OFFSET INVK THRESHOLD
PERM
NO
YES LOGGING
********************************************************************MKMMMKMKK*****
f~--"\
SEQUENCE BY VOLUME LABEL, PHYSICAL ADDRESS, HEAD, CYLINDER
UNITADDRESS xxxx DEVTYPE 338o-JK
VOLUME
CPU A PHYSICAL ADDRESS 52.1-0A
",,/
HSM800
/,c--....\
FAILURE AT ADDRESS: CYLINDER 0087 HEAD 01
0
00001000 4A570143 00570001 00035200 00010000 00804340
LAST SENSE AT: 032/85 15:29:50:71
1
o
15
FAILURE AT ADDRESS: CYLINDER 0095 HEAD 02
00001000 4A5F0243 005F0002 00035200 00010000 00804340
LAST SENSE AT: 032/85 18:33:51:07
0
o
o
1
FAILURE AT ADDRESS: CYLINDER 0105 HEAD 04
00001000 4A690443 00690004 00035200 00040000 00804340
LAST SENSE AT: 032/85 18:33:51:07
0
1
o
30
Use this column to find the number of permanent
data checks' that occurred in the report period.
~
J
\"",,)
Use this column to find the n~ber of times the temporary
data check threshold was exceeded in the report period.
Use this column to find the number of temporary data checks _ _ _ _..I
recovered with offset.
Use this column to find the number of temporary data checks - - - - - - - -.....
that were logged during threshold logging.
Above example shows:
I
Device OA on string 52, path 1, has exceeded the temporary
data check (no offset) threshold two times.
cylinder 87 head 1 -
a temporary error on this track caused the
cylinder 95 head 2 -
this track had a total of 1 temporary
error recorded while in threshold logging
mode.
cylinder 105 head 4 -
a temporary error on this track caused the
threshold to be exceeded. The track had a
total of 16 temporary errors.
threshold to be exceeded. The track had a
total of 31 temporary errors.
Recommended repair action for the above example:
Ask the customer to run ICKDSF INSPECT with PRESERVE and
CHECK(l) on cylinder 87 head 1 and on cylinder 105 head 4.
Figure 12.
Sample of DASD Data Transfer Summary Report
PIN 4519897 38 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1987, 1988
START-24
(
(
SMAP 0051: Determining Tracks Reporting Data Checks
(-
001
004
(From SMAP OOSO Step 002 on page START-23)
(From Steps 002 and 003 in this SMAP)
Use this SMAP to determine the failing track
addresses.
Warning: Have the customer dump the data from
the volume if one of the following conditions exist:
Do you need help?
Yes No
•
Permanent data checks on eleven or more
tracks.
•
Permanent data checks on any head at three or
more consecutive track addresses.
(
I
002
Go to Step 004 in this SMAP.
(
003
Determine the failing cylinder and head address.
Use the EREP System Exception report, DASD Data
Transfer Summary if available. Refer to Figure 12
on page START-24 for an example of this report.
Record the CYLINDER and HEAD values for each
head reporting errors from the "FAILURE AT
ADDRESS" lines. Cylinder and head values are in
decimal.
If a DASD Data Transfer Summary report is not
available, the cylinder and head address
(hexadecimal) may be determined from sense
bytes Sand 6 using the following diagram:
Sense
byte 5
Sense
byte 6
02
31
II
c
(~
I
cytHer
03 02
Have the customer perform media maintenance in
accordance with the procedures described in
Maintaining IBM Storage Subsystem Media,
GC26--: ms.
Note: Before using the Device Support
Facilities INSPECT command to inspect
tracks, the customer should always run the
Device Support Facilities ANAL YZE
DRIVETEST to ensure there are no hardware
problems.
For additional information about media
maintenance, see" Part 3: Using Device Support
Facilities" in Device Support Facilities Primer for
the User of IBM 3380 Direct Access Storage,
GC26-4498.
Did Device Support Facilities indicate a suspected
drive problem or present error messages?
Yes No
I
Head
01
005
Go to Step 007 in this SMAP.
In this example, byte S contains 02 and byte 6
contains 31. The address is cylinder X"0302," head.
X "01."
006
Go to SMAP 0052 Step 001 on page START-27.
Bytes 8 to 11 mayor may not contain the CCCC
HHHH in hexadecimal for the track having the data
check, depending on the nature of the data check.
Go to Step 004 in this SMAP.
PIN 4519897 38 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
START-25
-,
{~
r
"'-?
007
(From Step 005 in this SMAP)
Device Support Facilities does not Indicate a
suspected drive problem and does not present
error messages.
(,-.;---"'\
Perform the following actions:
1.
Determine if the problem was corrected.
2.
Continue to review EREP reports as required to
verify that the media maintenance action was
successful. Refer to SMAP 0017 Step 001 on
page START·6 for the guidelines to be used for
the review procedure.
3.
\... j
/'.
~
..•
If the problem was not corrected, perform the
following:
•
If possible, collect additional information by
having the customer run Device Support
Facilities ANALYZE SCAN.
•
Obtain all error data from Device Support
Facilities.
/-".
',-.. /
Go to SMAP 0056 Step 001 on page ST ART-32.
PIN 4519897 38 Sides
EC 476581 01Aug89
@ Copyright iBM Corp. 1987, 1988
START-26
(
SMAP 0052: Corrective Action Determination from Device Support Facilities Messages
001
(From SMAP 0015 Step 001 on page START-3)
(From SMAP 0051 Step 006 on page START-25)
Device Support Facilities reports can be used to
supplement the EREP System Exception reports for
isolating data check problems. This MAP deals
with error messages that can be presented in the
Device Support Facilities reports for the following
Device Support Facilities commands:
('
•
•
•
•
•
•
•
•
•
02 = 3380 Stage 1 or 2
06 = 3380-D
OA = 3380-E
16 = 3380-J
1E = 3380-K
Determine if corrective action is required.
Do you need help?
Yes No
I
002
Go to Step 005 in this SMAP.
ANALYZE DRIVETEST
ANALYZE SCAN
INSPECT
INIT
003
Corrective action is required if one of the following
messages is presented:
Collect information about the problem.
(/
("
1.
Collect Device Support Facilities reports.
2.
If possible collect EREP System Exception
Reports prior to running Device Support
Facilities programs. These reports will assist
with problem determination.
For information about Device Support Facilities,
and error messages, refer to Device Support
Facilities User's Guide and Reference, GC35-0033,
Level 12 or later.
SUSPECTED DRIVE PROBLEM
This message is an indication of a hardware
problem. For most hardware problems, one
or more of the following messages may also
be reported.
EQUIPMENT CHECK (permanent or temporary)
Equipment checks are indicated by the term
"EQUIPMENT CHECK" or by Sense byte 07 as
follows: OX, 1X, 2X, 3X, 7X, 8X, 9X, or FX.
For information on when and how to perform media
maintenance, refer to:
UNCORRECTABLE DATA CHECK OCCURRED ON
CE CYLINDER
This message can indicate either correctable
or uncorrectable data checks.
•
CE CYLINDER DATA CHECK THRESHOLD
•
Maintaining IBM Storage Subsystem Media,
GC26-4495.
Device Support Facilities: Primer for the User of
IBM 3380 Direct Access Storage, GC26-4498.
These manuals contain guidelines the customer
uses to determine when and how to perform media
maintenance and when to call IBM for service.
Determine that the failure is a 3380-JK failure.
(~
Review the Device Support Facilities reports to
ensure that PHYSICAL DEVICE
3380 and that
DEVICE DESCRIPTOR = 16 or 1E.
c
3380 device descriptor codes are as follows:
=
PIN 4519897 38 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1987, 1988
(~~CEEDED
This message can indicate either correctable
or uncorrectable data checks.
CORRECTABLE DATA CHECK LIMIT EXCEEDED ON
CE CYLINDER
This message is used to show that the
indicated head has exceeded the error
criterion. A previous message has indicated
the detailed sense information associated
with the failing head.
START-27
DATA CHECK THRESHOLD EXCEEDED ON ONE OR
MORE HEADS
This message is produced if one or more
heads exceed a threshold for data checks.
These data checks may be either correctable
or uncorrectable. The head or heads that
exceed the threshold are indicated in the
MOVEABLE HEAD ERROR TABLE that follows
the message.
DATAVER DATA CHECK EXCEEDED THRESHOLD
The number of data checks detected while
running ANALYZE SCAN exceeded a
threshold established for data checks on this
volume.
(non-repeatable) data check during surface
analysis of the indicated track. It is normal
and expected that some data checks could be
detected when INSPECT is run. This message
by itself is not necessarily cause for concern.
CORRECTABLE DATA CHECK OCCURRED ON CE
CYLINDER
It is expected that there will be some data
checks during a run of ANALYZE DRIVETEST.
According to the guidelines, is corrective action
required?
Yes No
I
004
DATAVER UNCORRECTABLE ERROR
This message is used by ANALYZE SCAN to
indicate that it has found a track that is not
correctable using ECC correction algorithms.
It does not mean that it would not be read
correctly using rereads, read with offset or be
ECC correctable as a result of those rereads.
DATAVER CORRECTABLE ERROR
This message is produced whenever
ANALYZE detects a correctable data check
which persists through multiple attempts to
read the data. The multiple read is intended
to eliminate data checks caused by random
noise.
Corrective action is not required for the following
messages unless one or more of the above
messages also occur.
SKIP DISPLACEMENTS CHANGED FOR TRACK
X"cccc hhhh."
This message is generated each time
INSPECT determines a media defect and
bypasses it by assigning a skip displacement.
It is normal and expected that some skip
displacements will be assigned when
INSPECT is run. It should be noted that skip
displacements may be changed even when
the NOSKIP option is selected.
TRACK X"cccc hhhh" EXPERIENCED
NON-RECURRING DATA CHECKS
This message occurs when the INSPECT
function detects a non-recurring
PIN 4519897 38 Sides
EC 476581 01Aug89
IBM Corp. 1987. 1989
© Copyright
Continue to review EREP periodically for
error conditions.
005
(From Step 002 in this SMAP)
Go to the message in Figure 13 on page START-29
and perform the indicated corrective action.
006
(From Step 011 in this SMAP)
Do the Device Support Facilities messages contain
sense data?
Yes No
I
007
A media maintenance action is required.
Have the customer perform media
maintenance in accordance with the
procedures in Maintaining IBM Storage
Subsystem Media, GC26-4495.
For 'additional information about media
maintenance, see Device Support Facilities:
Primer for the User of IBM 3380 Direct
Access Storage, GC26-4498.
If the media maintenance procedures
indicate that a maintenance action is
. aquired (for example, "Call your Service
Representative"), go to SMAP 0053 Step 001
on page START-30.
(Step 008 continues)
START-28
f
(
(
008
Determine if there is an undervoltage or loss of
track following condition:
•
Sense byte 23, bit 4 = 1, indicates that a
device undervoltage condition was detected.
•
Sense byte 23, bit 5 = 1, indicates that a
controller undervoltage condition was detected.
•
Sense byte 23, bit 7 = 1, indicates that a loss
of track following was detected.
Is there an undervoltage condition or loss of track
following condition?
Yes No
(
I
011
(From Step 009 in this SMAP)
A media maintenance action is required. Have the
customer perform media maintenance in
accordance with the procedures in Maintaining IBM
Storage Subsystem Media, GC26-4495.
For additional information about media
maintenance, see Device Support Facilities: Primer
for the User of IBM 3380 Direct Access Storage,
GC26-4498.
If the media maintenance procedures indicate that
a maintenance action is required (for example,
"Call your Service Representative"), go to SMAP
0053 Step 001 on page START-30.
009
Go to Step 011 in this SMAP.
010
A maintenance action is reCjuired. Go to SMAP
0053 Step 001 on page START-30.
C\
Message
Action
SUSPECTED DRIVE PROBLEM
Go to SMAP 0053 Step 001 on page START-30
EQUIPMENT CHECK
Go to SMAP 0053 Step 001 on page START-30
UNCORRECTABLE DATA CHECK OCCURRED ON CE
CYLINDER
Go to SMAP 0053 Step 001 on page START-30
CE CYLINDER DATA CHECK THRESHOLD EXCEEDED
Go to SMAP 0053 Step 001 on page START-30
CORRECT ABLE OAT A CHECK LIMIT EXCEEDED ON
CE CYLINDER
Go to SMAP 0053 Step 001 on page START-30
DATA CHECK THRESHOLD EXCEEDED ON ONE OR
MORE HEADS
Go to SMAP 0053 Step 001 on page START-30
DATAVER DATA CHECK EXCEEDED THRESHOLD
Go to SMAP 0053 Step 001 on page START-30
DATAVER UNCORRECTABLE ERROR
Go to Step 006 in this SMAP
DATAVER CORRECTABLE ERROR
Go to Step 006 in this SMAP
SKIP DISPLACEMENTS CHANGED FOR TRACK X"cccc
hhhh"
No action is required.
TRACK X"cccc hhhh" EXPERIENCED
NON-RECURRING DATA CHECKS
No action is required.
CORRECTABLE DATA CHECK OCCURRED ON CE
CYLINDER
No action is required.
,,/
C
C
Figure 13.
Device Support Facilities messages
PIN 4519897 38 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987. 1988
START-29
(.'
~"'.?,
SMAP 0053: Locating the Failing Unit from Device Support Facilities Messages
001
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
006
SMAP 0052
SMAP 0052
SMAP 0052
SMAP 0052
SMAP 0052
SMAP 0052
SMAP 0052
SMAP 0052
SMAP 0052
SMAP 0052
Step
Step
Step
Step
Step
Step
Step
Step
Step
Step
007
010
011
011
011
011
011
011
011
011
on
on
on
on
on
on
on
on
on
on
page
page
page
page
page
page
page
page
page
page
START-28)
START-29)
ST ART-29)
START-29)
START-29)
START-29)
START-29)
START-29)
START-29)
START-29)
(From Step 004 In this SMAP)
The sense data is from a Data Check. Locate the
failing unit with the physical ID information
contained in sense bytes 4 and 14 as follows:
1.
To identify the string, match the String ID
contained in sense byte 14 with the string 10 on
the A-unit operator panel. For 3880
subsystems, see Figure 6 on page INTRO-7;for
3990 subsystems, see Figure 14 on
page INTRO-20 and Figure 15 on
page INTRO-20.
2.
To identify the controller, use sense byte 4 to
determine the path and device address. Next,
match the path and device address from sense
byte 4 with the Path and Device on the A-unit
operator panel.
The best way to locate the failing unit is with the
physical address information contained in a sense
record.
Does the report contain error messages with sense
data?
Yes No
I
002
Use the Channel Unit Address (CUU) or the
volume identifier (DNAME, DDNAME or
SYSNAME) from the input statement with
existing CE or customer procedures to locate
the failing unit.
Schedule machine with customer and go to
IMAP 0230 Step 001 on page ISOLATE-16.
003
All Equipment Check problems should be resolved
before attempting to resolve a Data Check
problem.
Is there an Equipment Check message?
Yes No
I
004
Go to Step 006 in this SMAP.
005
Go to SMAP 0040 Step 001 on page START-19.
Sense Byte 4
Path
OOxx xxxx
01xx xxxx
10xx xxxx
11xx xxxx
Path
Path
Path
Path
Sense Byte 4
String Address
xxOx xxxx
xx1x xxxx
String 0
String 1
Sense Byte 4
Device Address
xxxO
xxxO
xxxO
xxxO
Device
Device
Device
Device
0000
0001
0010
0011
xxx11110
xxx11111
0
1
2
3
00
01
02
03
Device 1E
Device 1F
Go to SMAP 0054 Step 001 on page START-31.
\,--
"
)
PIN 4519897 38 Sides
© Copyright
EC 476581 01AugS9
IBM Corp. 1987. 1988
START-30
(
('
SMAP 0054: Prioritizing Data Check Symptom Codes using Device Support Facilities
Messages
('
001
(From SMAP 0053 Step 006 on page START-30)
('.'
(
closest to the top of the following table of sense
byte 22 values:
Use the following guidelines to determine the
highest priority symptom code.
1
Sense
Byte 22
Value
44
2
45
3
46
4
47
003
5
40
Determine the highest priority group of symptom
codes from the following list. Select the applicable
group that is closest to the top of the list.
6
41
7
42
8
43
Is there more than 1 symptom code listed?
Yes No
Group
I
002
The symptom code listed is the highest
priority symptom code. Record the symptom
code and go to SMAP 0058 Step 001 on page
START-34.
Priority 1
Priority 2
Priority 3
Priority 4
Sense Byte 23. bits 4 or 5 = 1
(undervoltage)
Sense Byte 23. bit 7 = 1 (lost track
following)
Message indicates uncorrectable
Other
Is there more than one symptom code in the
selected group?
Yes No
,("
The symptom code listed is the highest
priority symptom code. Record the symptom
code and go to SMAP 0058 Step 001 on page
START-34.
005
Reduce the size of the selected group by selecting
only those symptom codes that are in the group
\
/
I
006
The symptom code listed is the highest
priority symptom code. Record the symptom
code and go to SMAP 0058 Step 001 on page
START-34.
004
(~
(
Is there more than one symptom code in the
selected group?
Yes No
I
-"
PIN 4519897 38 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1987. 1998
Description
No Sync Byte Found,
HA Area
No Sync Byte Found,
Count Area
No Sync Byte Found,
Key Area
No Sync Byte Found,
•
Data Area
ECC Data Check,
HA Area
ECC Data Check,
Count Area
ECC Data Check,
Key Area
ECC Data Check,
Data Area
007
The symptom code that occurs most often is the
highest priority symptom code.
Record the symptom code and go to SMAP 0058
Step 001 on page START-34.
START-31
SMAP 0056: Prioritizing Data Check Symptom Codes using Sense Data
001
(From SMAP 0050 Step 003 on page START-23)
(From SMAP 0051 Step 007 on page START-26)
closest to the top of the following list of sense byte
22 values:
Use the following guidelines to determine the
highest priority symptom code.
Is there more than 1 symptom code listed?
Yes No
Group
Sense
Byte 22
Value
Description
1
44
No Sync Byte Found,
HA Area
2
45
No Sync Byte Found,
Count Area
3
46
No Sync Byte Found,
Key Area
4
47
No Sync Byte Found,
Data Area
5
40
fCC Data Check,
HA Area
Determine the highest priority group of symptom
codes from the following list. Select the applicable
group that is closest to the top of the list.
6
41
fCC Data Check,
Count Area
7
42
fCC Data Check,
Key Area
Sense Byte 23, bits 4 or 5 = 1
(undervoltage)
Sense Byte 23, bit 2 = 1 (offset active)
Sense Byte 23, bit 7 = 1 (lost track
following)
Permanent Data Check (Indicated in
EREP Data Transfer Summary or
console message)
Other
8
43
fCC Data Check,
Data Area
I
-
002
The symptom code listed is the highest
priority symptom code. Record the symptom
code and go to SMAP 0058 Step 001 on page
START-34.
003
Priority 1
Priority 2
Priority 3
Priority 4
Priority 5
Is there more than one symptom code in the
selected group?
Yes No
I
006
The symptom code listed is the highest
priority symptom code. Record the symptom
code and go to SMAP 0058 Step 001 on page
START-34.
Is there more than one symptom code in the
selected group?
Yes No
I
004
The symptom code listed is the highest
priority symptom code. Record the symptom
code and go to SMAP 0058 Step 001 on page
START-34.
005
',,-_./'
007
The symptom code that occurs most often is the
highest priority symptom code.
Record the symptom code and go to SMAP 0058
Step 001 on page START-34.
Reduce the size of the selected group by selecting
only those symptom codes that are in the group
'~
-.
\",
PIN 4519897 38 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987. 1988
START-32
/\
-"",
'----.--"
I
(
('
(
3380-JK
SMAP 0057: Determining the Failing Poris (EBxx)
1 001 1
f
f
(
(
(From SMAP 0033 Step 001 on Page START-16)
(From SMAP 0045 Step 004 on Page START-22)
(From IMAP 0220 Step 001 on Page ISOlATE-12)
Determine the failing ports and the devices and
controllers that must be removed from customer
operation to repair the problem.
The EBX)( Symptom Code represents a failure of
the Controller/Device Port (COP) interface.
Failing Ports and Devices cannot be determined
from the physical address column in the EREP .
DASD Symptom Code Summary report or from
sense byte 4.
Multiple Port Failures
Failing ports are determined from the EBxx
symptom code where bits 4-7 of "xx" identify the
failing port. See Figure 14. Multiple EBxx
symptom codes with different values of "xx" indicate that multiple ports are failing and that the
failure is probably in the controller.
Repairing the problem requires that the controller and one failing device be removed from
customer operation.
Single Port Failures
If all EBxx symptom codes have the same value
of "xx," a single port is failing. The failure may
be in the controller, or in the COP interface, or in
one of the four devices on the port.
If devices on more than one drive are reporting
errors, failing devices cannot be determined
from sense data. Isolating the problem requires
that the controller and all four devices on the
failing port be removed from customer operation. See Figure 14 to determine which port and
devices are associated with the failure.
Note:' Failure boundary definition for EBXX
symptom codes requires the use of MD
procedure 0051 (see MD 0051 Step 001 on
page MD-58).
Go to SMAP 0058 Step 001 on page START-34.
Table 1. PortlDevice Service Groups
EBxx
Bits 4-7
if
Port
Device
Address
EBxx
Bits 4-7
Port
Device
Address
x4
4
10-13
x8
0
00-03
x5
5
14-17
x9
1
04-07
x6
6
18-1B
xA
2
08-0B
x7
7
1C-1F
xB
3
OC-OF
(
('
f
(
(
~
Copyright IBM Corp. 1992
START-33
\
(
\.~
3380.JK
SMAP0058: Determining the Failing FRUs
I I
001
006 (continued)
The fafllng machine is not available for running
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
SMAP 0033
SMAP 0033
SMAP 0033
SMAP 0045
SMAP 0045
SMAP 0045
SMAP 0054
SMAP 0054
SMAP 0054
SMAP 0054
SMAP 0056
SMAP 0056
SMAP 0056
SMAP 0056
SMAP 0057
Step 001
Step 001
Step 001
Step 004
Step 004
Step 004
Step 002
Step 004
Step 006
Step 007
Step 002
Step 004
Step 006
Step 007
Step 001
on
on
on
on
on
on
on
on
on
on
on
on
on
on
on
page
page
page
page
page
page
page
page
page
page
page
page
page
page
page
START~16)
START-t6)
START-16)
START-22)
START-22)
START-22)
START-31)
START-31)
START-31)
START-31)
START-32)
START-32)
START -32)
START-32)
START-33)
diagnostics.
Follow this path to determine the probable failing
FRUs. based on the sense information that has
been collected. You can also determine the part
of the string that will be affected by the repair
action.
Use MO main menu option 3 to determine the
probable falling FRUs. For additional information
on the use of MD option 3. see page MD-42.
1. Power-on and IMl the MD. The MD need
not be attached to the 3380.
2. From the MD Main menu. select option 3
(ANALYZE SYMPTOM CODE).
If the symptom code is 7xxx or Bxxx, ensure that
the string is a 3380-JK string before continuing.
3. In response to the prompt. enter the
symptom code to be analyzed.
If the symptom code is 9F9F, go to START-39,
Entry 001 (do not continue with this SMAP).
4. If possible. answer all questions and enter
all data requested by the MD. If asked to
enter sense data when the data is not available. the step can be bypassed by pressing
the enter key. If a step is bypassed. a larger
FRU group will be given or the FRUs will not
be listed in order of highest probable failure.
Is the failing machine available for running diagnost,cs?
Yes No
~
Go to Step 006.
003
Are you working on a data check problem
(symptom code 4xxx)?
Yes No
~
Go to IMAP 0205 Step 001 on page
ISOLATE-2.
5. When a FRU group is displayed. record the
list of FRUs from the primary FRU list and
the secondary FRU list if there is one. Do
not replace any of these FRUs at this time.
6. If there is a MAP referenced under additional
action. review the map to determine the
additional action that may be needed to
isolate the problem. Do not perform any of
these actions at this time.
7. If you have time and want to learn more
about the problem, look up the error condition in the ECD seelion of the ECM. Volume
R40.
Go 10 SMAP 0060 Step 001 on page START-3S.
Go to IMAP 0228 Step 001 on page ISOLATE-1S.
(Step 006 continues)
.....
/
-/
f'
(
(
SMAP 0060: Scheduling the Hardware
001
(From SMAP 0058 Step 006 on page ST ART-34)
the affected devices varied offline. When the 3380
unit is available for service, continue with this step.
Ensure that the FRUs are a"ailable before
scheduling the 3380 unit for maintenance. Look up
the probable failing FRUs and determine what part
of the string will be affected by the repair action:
Are you working on a data check problem
(symptom code 4xxx)?
Yes No
•
•
I
002
Equipment checks: Go to IMAP 0205 Step
001 on page ISOLATE-2.
For FRUs on the A1 board, see Figure 15.
For FRUs on the 81 or ex boards, see
Figure 16.
003
Data checks: Go to IMAP 0228 Step 001 on page
ISOLATE-15.
Schedule maintenance time for the affected devices
and controllers with the customer. Arrange to have
Al Board Cards
Controller
Logic Card Name
DDC/DTB
I/O Control
CDPSSO
CDPSS1 (note 1)
DPS (note 2)
Controller Sequencer
DHPLO
MD adapter
Clock/SERDES/ECC 1
Clock/SERDES/ECC 2
Figure 15.
CUr
Al/A3
AlW2
AlV2
AlU2
AH2
A1S2
A1R2
A1Q2
AIM4
A1P2
AIN2
CUr
AUA4
AlB2
AlC2
A1D2
A1E2
A1F2
Al62
A1H2
AIL4
A1J2
AlK2
Repair Affects
Controller
Notes:
Controller
1.
2.
String
Controller
Controller
AJ4 and AK4 units do
not have CDPSSl cards
unless the 4-path
feature is installed.
AJ4 and AK4 units
attached to a 3990
storage control do
not have DPS cards.
Controller
Probable Failing FRU on the A1 Board
Bl and CX Board Cards
('
(='
(-~,
Device
logic Card Name
PES/PROM
SEQ/Servo/RPS
Po rt.lR-W
R/W channel board
Servo power amp
Power
Al Port Term
A2 Port Term
A3 Port Term
A4 Port Term
Figure 16.
left
Front
BlD2
BIE2
B1F2
COB2
C4Pl
B162
B1W2
81W3
81B2
8183
Probable Falling FRU on the B1 or
Right
Front
BIK2
B1J2
B1H2
ClB2
C4P2
B162
BIW2
81W3
81B2
8183
left
Rear
BlN2
B1P2
B1Q2
C2B2
C4P3
8lR2
81W2
81W3
81B2
B1B3
Right
Rear
BIU2
BH2
B1S2
C3B2
C4P4
BlR2
B1W2
81W3
8182
8lB3
Repair Affects
Device
Device
Drive
All devices
in the unit
ex Board
c
PIN 4519897 38 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 19S8
START-3S
o
SMAP 0070: Locate the Failing Machine with Console Messages
001
008
(From SMAP 0015 Step 001 on page START-3)
(From Step 007 in this SMAP)
Do you need aid in locating the falling machine?
Yes No
The physicallD form is "SSID.P.CC·DD," where:
I
002
Go to IMAP 0205 Step 001 on page
ISOLATE-2.
003
See the console message manual. Look for the
physical ID in the console messages. If present, it
will be in one of the following forms:
SSID = Storage Control subsystem ID or
Storage Director ID (SDID)
P
== Path
CC
= String ID
DD
Device address
=
Determine the failing string, paths, and devices
using the following procedure:
1.
To identify the string, match the String ID (CC),
obtained from the console messages, with the
String ID on the A unit operator panel. For
3880, see page INTRO-7; for 3990, see page
INTRO-20 or INTRO-22.
2.
To identify the controllers, match the path (P)
obtained from the console message, with the
path character on the A unit operator panel. If
P
X in the console message, the path must
be determined from the Channel Unit Address
(CUA) or the Channel Path ID (CHPID) in the
console messages. Match the path determined
with the path character on the A-unit operator
panel.
SSID.P-CC-DD
SS-CC-DD
Is one of these physical 10 forms In the console
messages?
Yes No
I
004
=
Use one of the following to locate the failing
machine:
•
•
The Channel Unit Address (CUA).
The Channel Path Identifier (CHPID) and
device number.
Locate the failing string of devices by using
the CUA (or the CHPID and device number)
and existing customer or CE procedures.
Go to IMAP 0205 Step 001 on page
ISOLATE-2.
005
Is the physical 10 form "SSIO.P·CC·OO"?
Yes No
3.
To identify devices, match the device address
(DD), obtained from the console messages,
with the device addresses on the A unit
operator panel. For 3880, see page INTRO-7;
for 3990, see page INTRO-20 or INTRO-22.
Schedule the machine with the customer and go to
IMAP 0205 Step 001 on page ISOLATE-2.
I
006
Go to Step 009 in this SMAP.
007
Go to Step 008 in this SMAP.
(,f
"
\L)
PIN 4519897 38 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1987, 1988
START-36
('
009
(From Step 006 in this SMAP)
The physicallD form is "55-CC-DD," where:
(
55
= Storage director 10 or
CC
DD
= String ID
= Device address
Storage Control subsystem 10
(
Determine the failing string, paths, and devices
using the following procedure:
(
1.
To identify the string, match the string 10 (CC)
obtained from the console messages with the
string 10 on the A unit operator panel. For
3880, see page INTRO-7; for 3990, see page
INTRO-20 or INTRO-22.
2.
This form of physical 10 does not contain path
information. The path must be determined from
the Channel Unit Address (CUA) or the Channel
Path 10 (CHPIO) in the console messages.
Match the path determined with the path
character on the A unit operator panel.
3.
To identify devices, match the device address
(~O), obtained from the console messages,
with the device addresses on the A unit
operator panel. For 3880, see page INTRO-7;
for 3990, see page INTRO-20 or INTRO-22.
(
(
(
('
(
Schedule the machine with the customer and go to
IMAP 0205 Step 001 on page ISOLATE-2.
(~
(
/
c
c
PIN 4519897 38 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
START-37
EC History of Start Section
EC HISTORY OF P/N 4519897
EC Number
Date Of EC
EC Number
475245
14Nov86
475248
475246
21Ju187
476573
llSep87
476581
475247
Date Of EC
25Apr88
30Nov88
01Aug89
Notes:
;",
tt
\'l..J
PIN 4519897 38 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
START-38
1
(
3380-JK
(
SMAP 0059: Determining the Failing FRUs (9F9F)
(
(From SMAP 0058 Step 001 on Start-34)
005 (continued)
Is the problem fixed?
Yes No
f
~
Symptom code 9F9F is designed to give an early
indication of 3380-JK bearing instabilities and is
not an error condition.
A cold operating environment andlor a wide fluctuation in operating temperatures might cause
symptom code 9F9F and result in unnecessary
HOA replacement. Replacing the HOA might
NOT fix the 9F9F symptom code if the environmental issue is not first addressed.
Call the next level of support.
007
No further action is required.
Balancing the room temperature to the recommended operating environment of 72°F. ±2°F
has proven to eliminate 9F9F symptom codes in
many cases.
{
Go to "3380-JK Input Air Temperature Check" on
page PM-2. Return here when finished.
·f
Was the air intake temperature within specifications?
Yes No
§J
After modifications are made to correct the
temperature. monitor for symptom code
9F9F for a minimum of 24 hours. If
symptom code 9F9F reoccurs. go to Step
005.
f
(
f
(
Is byte 9 of any symptom code 9F9F greater than
or equal to 10?
Yes No
~
Monitor EREP for the next occurrence of
symptom code 9F9F. No further action is
required.
(
Go to page MO-1. Entry C. and run Option A
(Non-repeatable Runout (NRRO) Test).
(Step 005 continues)
(
II:> Copyright IBM Corp. 1992
START-39
3380.JK
(
(
Preventive Maintenance
3380 Models J and K Preventive
Maintenance Procedures
Perform the preventive maintenance procedures
shown in Table 2 at the intervals shown.
Table 2. Preventive Maintenance for 3380 Models AJ4, AK4, BJ4. and BK4
(
(
f
Task
Frequency
In
Months
Comments
Check the box intake
air temperature
6
Refer to "3380-JK Input Air Temperature Check" on
page PM-2.
Check the blower
motor
6
Refer to "Blower Vibration Measurement Procedure" on
page PM-4.
Check the gate fans
6
Check the A and B gates for excessively noisy or failed
gate fans. Replace the fans as necessary. Refer to "Gate
Fan" on page CARR-43, Vol. R10.
Check the primary air
filter
6
Check the primary air filler and replace as necessary. Do
not attempt to clean the filter. Refer to page lOC-6. Vol.
R10.
Check the air pressure
6
Go to "Air Pressure Checks" on page CARR-20. Vol. R10.
Check the HDA motor
brakes
6
Check both motor brakes for excessive red dust around the
brake housing per ECA 149/150. Order the EC if required.
Check for cable
chafing
12
Order the cable chafe inspection kit. PIN 2317345. This kit
supplies instructions for the recommended inspection
points. cable ties. and cable wrap tape.
(
(
f
(
(
It")
(
Copyright IBM Corp. 1992
PM-1
3380..JK
338~K
Input Air Temperature
Check
8ackg.round
A cold operating environment might cause
symptom code 9F9F and result in unnecessary
HDA replacement. Replacing the HDA might
NOT fix the 9F9F symptom codes if the environmental issue is not addressed.
Note. Generally. out-of-specification environmental temperatures can be observed by a clustering of 9F9F symptom codes (multiple devices
in the same physical area within a 24 hour
period).
Balancing the room temperature to the recommended ideal operating environment of 72°F.
±2°F. as specified in the IBM Generallnformation Instal/ation Physical Planning Manual. has
proven to eliminate 9F9F symptom codes in
many cases.
Note: Symptom code 9F9F is NOT an error condition. but is an environmental symptom code
that might require further action to be taken.
Ensure that your customer understands that
symptom code 9F9F is NOT an error. but an indication that further diagnostics might be required.
Measuring the Input Air
Temperature
Note: Use SC 08 for all temperature checking
activities.
To measure the input air temperature, perform
the following:
1. Check for the presence of temperature strips
(PIN 2179638) on the pre-filter of aU A and 8
boxes. These were installed by ECA 163. and
measure the temperature of the input air
entering the pre-filter. See Figure 2.
If the temperature strips are not installed.
order one temperature strip (PIN 2179368)
for each J or K box in your account. Install
the strips as shown in Figure 2.
,. . - _. (fill:::.. @]::: -....
?'iF,-atl--(Jum·::
l®........ ...... .."....
....... cp.::
......,
....
...
::: um·:: lim·:: ......
.. III '"
"'..
" ' . '"
....
..
o
.. ......
.. .. '"
"'..
.,
.,
urn·::
.. "
......
.. '" '"
•••
",
:0
1~!rID
[lll1illJ .,.V ":"
I~ IilllIillHillll
ULYll
VYYY?
,
.....
00
,.... . - CN4D
G"",
.......... ,
CI'Wf
:::. lblccP cP
..-r .....
CJl'5I,J
O'&.~
"Ufll'
IC"","'"
=l~CCPCP
Tools Required
M-
CNJ
CN(oI
CPIOS
V
....
a..
~
0'&"
0'&11
~
~
CPNS
mK
AClUITat
The following tools are required to perlorm this
procedure:
:::'0 cPcp
lOGIC
___
I7U)
tpUI
• Temperature strips. PIN 2179638
Ollfll'
• NFPA approved pillows. PIN 2317361
:Ocpcp
These pillows are approved by the National
Fire Protection Agency (NFPA). If IBM takes
the responsibility for plugging the cable
cutouts. then this product (PIN 2317361) must
be used. If the customer chooses to do this
job, this is the recommended material.
'"
• • '"
~
CP»S
V
0'&»
ItNl'.
lOGIC
TEMPERATURE STRIP
~
V
15003565013
Figure 2.
2. Read the temperature on each strip and
record the value mad for each box.
3. Determine if the air entering the pre-filler is
within the limits of 18° to 32°C (65 0 to gO°F).
It is desirable to have the temperature
measure a few degrees above the 18C
minimum. The recommended ideal temper,r("""'
I
\ ... j
'
(
3380.JK
ature range of air entering the 3380 pre-filter
is 21"C to 23°C (70"F to 74°F).
If the temperature is below the minimum of
18~C, do steps 4 through 6 on page PM-3,
then recheck the temperature again in 24
hours.
(
4. Warning:
Do not decrease the number of vented noor
tiles in the computer room. II is important to
ensure that there is an adequate number of
vented noor tiles to maintain the proper circulation of air to and from the air handlers.
Without the proper circulation, the under
noor temperature can be too cold and the
room temperature too warm.
(
Seal or replace all missing and perforated
noor panels under, or partially under, the
3380 string.
Reference Documentation
1. IBM General Information Installation PhYSical
Planning Manual, GC22-7072-2 dated August.
1990.
Note: Chapter 4 - Air Conditioning. defines
the latest IBM temperature and humidity
design criteria. The design criteria is 72°F.
±2°F and 45% RH. ±5% relative humidity.
2. IBM Systeml360, Systeml370, 4300 and 9370
Processors Input/Output Equipment Installation Manual-Physical Planning, GC22-7064-12
dated February, 1990.
Note: The 370 I/O IMPP dated February 1991
renects that the minimum temperature
entering the 3380-JK pre-filter is 18°C (65°F).
5. Ensure that all vented noor tiles are 3 feet (1
meter) away from the louvered door.
6. Pack the noor panel cable openings under
the A boxes with a fire rated pillow. This is
very important if the air under the noor is
below 18°C (65°F). See "Tools Required" on
page PM-2 for the recommended pillow.
7. If the recommended input temperature
cannot be reached using the above steps.
contact your Branch Office Installation Planning Representative (IPR) for a thermal
balance review of your customer's DASD
area.
(
(
(
PM-3
(
(
o
"
3:J8&.JK .
Blower Vibration Measurement
Procedure
VIBAAT
NONITOA
This p.rocedure requires the use of a vibration
PIN 2317360. This is a branch office
monitor
tool. If your branch does not have it, order the
tool and place it in the B/O tool inventory.
D,
1. Connect the cable end of the magnetic
mounting base fJ to the 'XDCR' jack of the
vibration monitor.
2. Switch the vibration monitor power on by
pushing the Power On switch to On. The red
LED lights if the power is on.
3. Test the batteries using the Battery Test
switch and meter battery indicator. If the
batteries are low, replace both batteries with
9-volt batteries.
4. Test the bias using the Bias Test switch and
meter indicator. If the bias reading is not
within the indicated range, return the meter
to the vendor for repair.
S. Remove the cover on the vibration monitor
magnetic base by sliding it sideways.
6. Attach the vibration monitor magnetic base
to the end cap of the blower motor Q in the
12 o'clock position fJ. See Figure 3.
AI500356tfcl2
.-,/
Figure 3. Blower Motor Vibration Monitor Placement
7. With the blower motor running, read and
record the vibration monitor reading.
8. Move the vibration monitor magnetic base to
of the blower motor
the 3 o'clock position
end cap. See Figure 3.
II
9. With the blower motor running, read and
record the vibration monitor reading.
10. The highest of the two readings is the measured vibration level for that particular blower
motor.
11. If either reading is .25 or above, replace the
blower motor. See page CARR-22 in the
MIM for the blower motor replacement procedure.
\<;.", -",,-
~.
\.'" .)
("
Isolate Procedures
IMAP 0200: Isolate the Problem
001
(
(
The procedures in this section are designed to help
isolate the problem to a Field Replaceable Unit
(FRU) or to the smallest group of FRUs possible.
FRU isolation is based upon the results of
. diagnostic programs and the data collected at the
time of the original failure. The machine must be
available for you to perform these procedures.
003
Before starting the procedures in this section, you
should have completed the analysis procedures in
the START section.
Locate the entry point (A, B, C, etc. determined
from the START section procedures) In the
following table and go to the map and step
indicated.
(
I
002
Go to SMAP 0015 Step 001 on page START-2
and collect information about the failure.
If an entry point is not available, select a symptom
from the Description column and go to the map and
step indicated.
(
(
Have you completed the analysis procedures in the
START section?
Yes No
Entry
Description
Go to
A
Physical Inspection
IMAP 0205 Step 001 on page ISOLATE-2.
B
Equipment Checks (Sense data is available)
IMAP 0205 Step 001 on page ISOLATE-2.
C
Data Checks (Sense data is available)
IMAP 0228 Step 001 on page ISOLATE-15.
D
Console Message (Sense data is NOT available)
IMAP 0205 Step 001 on page ISOLA TE-2.
E
Status Pending LED is lit continuously
on the storage control (3880 only).
IMAP 0252 Step 001 on page ISOLATE-33.
F
Run Diagnostics
IMAP 0215 Step 001 on page ISOLATE-8.
G
Problem Analysis with Error Log Data
IMAP 0240 Step 001 on page ISOLATE-24.
H
Trouble Setting CE Mode
IMAP 0225 Step 001 on page ISOLATE-14.
J
Symptom Code Analysis
IMAP 0235 Step 001 on page ISOLATE-23.
Figure 1.
ISOLATE Section Entry Table
(
c·
(
PIN 4519899 56 Sides EC 475248 25Apr88
© Copyright
IBM Corp. 1987, 1988
ISOLATE-1
o
IMAP 0205: Physical Inspection
001
005
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
SMAP 0015 Step 001 on page START~3)
SMAP 0015 Step 001 on page START-3)
SMAP 0058 Step 004 on page START-33)
SMAP 0060 Step 002 on page START-34)
SMAP 0070 Step 002 on page ST ART-35)
SMAP 0070 Step 004 on page START-35)
SMAP 0070 Step 008 on page START-35)
SMAP 0070 Step 009 on page START-36)
IMAP 0200 Step 003 on page ISOLATE-1)
IMAP 0200 Step 003 on page ISOLATE-1)
IMAP 0200 Step 003 on page ISOLATE-1)
Will the customer allow service at this time?
Yes No
I
002
Schedule the machine for service, then
continue with Step 003 in this IMAP when
the machine is available.
003
(From Step 002 in this IMAP)
1.
Locate the rear of the failing string.
Locate the rear of the string you are going to
work on by referencing the STRING 10 on the
rear Unit Emergency Power Off panel of each A
unit in the string. See SMAP 0030 Step 001 on
page START-13 or SMAP 0045 Step 001 on
page START-20.
2.
Check rear circuit breakers and protectors.
Check that all circuit breakers and protectors at
the rear of each A unit and each B unit affected
by the failure are ON. If you are not sure which
B units are affec;ted, check the circuit
protectors on all B units in the string.
Are all circuit breakers and protectors on?
Yes No
I
004
Go to page PWR-1, Entry C, in the MIM,
Volume R10.
Check MOA (Maintenance Device Adapter) card
LEOs, A1-Board.
For each A unit in the string, remove the A1-board
logic card cover and check that both controller
clock LEOs and all voltage LEOs are on. When the
controller clocks are running, the clock LEOs are
slightly dimmer than the voltage LEOs. If a clock
LED is off or is at the same brightness level as the
voltage LEOs, it means that the clock is stopped.
Controller
A2 / A4
0 +5 V
0 Clock
0 -5 V
0 -4.25 V
A1l4
Controller
Al / A3
0 +5 V
0 Clock
0 -s V
0 -4.25 V
AIM4
Are all voltage LEOs on (disregard the clock LED)?
Yes No
I
j
!
006
Go to the front of the string and use the
following list to check each item in each A
unit in the string:
AC POWER-ON LED
Check that the AC Power-On LED
located on the front Operator panel is
on.
PHASE DETECTION LEOs
Check the green and red phase
detection LEOs to ensure that the green
LED is on and the red LED is off. The
green and red phase detection LEOs
are located on the primary power box.
at the front of the A unit.
FRONT CIRCUIT BREAKERS AND
PROTECTORS
Check that all circuit breakers and
protectors at the front of the machine
are on.
(Step 005 continues)
ISOLATE-2
PIN 4519899 58 Sides EC 475248 25Apr88
© Copyright IBM Corp. 1987. 1988
rf".'
~~
CONTROLLER POWER SUPPLY VOLTAGE
REGULATOR CARD LEOs
Check that the two LEOs for + 1.7 Vdc
and the + 5 Vsp on both controller
power supply regulator cards are off.
The + 1.7 Vdc LED is yellow and the
+ 5 Vsp is red.
(
The controller power supplies are
located at the front lower right of the A
unit (see the LaC section in the MIM.
Volume RIO).
(
(
After checking each item, refer to the
following list and find the first item, from the
top, that matches your findings. Go to the
specified reference.
(
Physical Symptom
Tripped CB or CP
Go to page PWR·1
(MIM, Volume RIO)
Entry J
Red phase LED is on
Entry AM
AC Power On LED
is off or dim
Entry AM
Controller power supply Entry AI
voltage regulator card
LEO(s) are on
Entry AI
007
Are the clock LEOs on and dimmer than the voltage
LEOs?
Yes No
(~'
-,j
I
008
C
£'
.-"
,
Are all A 1·board fans running?
Yes No
I
010
Refer to page PWR-1, Entry N, in the MIM.
Volume RIO to correct the problem. After
the problem if; corrected, return to Step 011
in this IMAP.
011
(From Step 010 in this IMAP)
Check Controller Ready LEOs. For each A unit in
the string, check that the Controller Ready LEOs
(located on the rear operator panel beside the MD
receptacle) are ON.
Are the Controller Ready LEOs on?
Yes No
Entry C
Green and red phase
LEOs are both off
None of the above
A1-board are running.
Go to IMAP 0215 Step 001 on page
ISOLATE-B.
009
Check A1·board gate fans. For each A unit in the
string, check to ensure that the gate fans for the
I
012
Go to IMAP 0250 Step 001 on page
ISOLATE-32.
013
Check the PES/PROM card LEOs.
Remove the B1-board logic card cover of the
machines that are affected by this failure and check
that the four, LEOs on each PES! PROM card are
on.
Note: LEOs may appear to be dim when they are
not on at all. This is due to light from the adjacent
LEOs.
left
Front
Device
0
0
0
0
BID3
+15V
+5V
-5V
-15V
Right
Front
Device
0
0
0
0
0
0
0
0
BIK3
Right
Rear
Device
left
Device
Rear
BIN3
+15 V
+5 V
-5 V
-15 V
0
0
0
0
BIU3
(Step 013 continues)
'
'""~~/
PIN 4519899 56 Sides EC 475248 25Apr88
@Copyright IBM Corp. 1987, 1989
ISOLATE-3
o
013 (continued)
Are all PES/PROM card LEDs on?
Yes No
I
014
Go to page PWR-1, Entry AJ, in the M/M,
Volume RIO.
015
Check the B1-Board gate fans. Check to ensure
that the gate fans for the B1-boards of the affected
units are running.
Are all B1-board fans running?
Yes No
I
016
Refer to page PWR-1, Entry N, in the MIM,
Volume R10 to correct the problem. After
the problem is corrected, return to Step 017
in this IMAP.
017
(From Step 015 in this IMAP)
Are any of the drive motors in the string running?
Yes No
I
018
Perform one of the following.
1.
If you are attempting to isolate a
problem where none of the drive motors
start during a normal string power up
sequence, go to page PWR-1, Entry L, in
the MIM, Volume RIO.
2.
If you are trying to isolate a problem
other than in item 1 above, go to
Step 019 in this IMAP.
•
If the HDA cables are not swapped, refer to the
left side of Figure 2 on page ISOLATE-5 for the
meaning of the LEDs on cards B1G2 and B1R2.
•
If the HDA cables are swapped, refer to the
right side of Figure 2 on page ISOLATE-5 for
the meaning of the LEDs on cards B1G2 and
B1R2.
Are all B1-Board Power Card LEDs on?
Yes No
I
020
Determine if the HDA cables are swapped. If
the cables are swapped, there should be a
cable swap label at the top of the B1-board
gate. If so, use Figure 2 on page ISOLATE-5
to determine the meaning of the LEDs.
In the following list, find the first item (from
the top) that matches your findings. Then go
to the specified reference.
Physical Symptom
(
Reference
+ 5/24 Vdc or + 5 Vdc Page PWR-1,
Common LED off
Entry AJ
(MIM, Volume RIO)
One or more Device
IMAP 0245 Step 001
Ready LEDs are off
on page ISOLATE-27
SW ENABLED LED
IMAP 0247 Step 001
is off
on page ISOLATE-30
One or more Device
Sequencer Status LEDs
are off
IMAP 0248 Step 001
on page ISOLATE-31
,('
,
j,
021
Go to IMAP 0210 Step 001 on page ISOLATE-5.
019
(From Step 018 in this IMAP)
Check B1-Board Power Card LEOs. While the
B1-board logic card covers are removed, check
that the eight LEDs on each Power card are on
(see Figure 2 on page ISOLATE-5).
ISOLATE-4
c
PIN 4519899 56 Sides EC 475248 25Apr88
© Copyright
IBM Corp. 1987. 1988
(
(
(
(
B1 Board LEOs - Cable Swap
FRONT DRIVE
NOTE.
FRONT DRIVE DOES
NOT EXIST IN MODEL
CJ2.
c
(
I
YA
REAR DRIVE
II
YB
I
YJ
I
I
YK
I
0
YO
YC
I
YM
I
I
YL
I
E
F
YE
G
H
YF
I
YN
I
J
K
L
M
I
YP
I
N
YG
YH
I
YQ
I
5
I
YR
I
T
U
BI-BOARD (REF)
(
BI-BOARD (FRONT DRIVE - G2 COMMON POWER CARD
REAR DRIVE
- R2 COMMON POWER CARD] - - - - - '
f
(
C
,
I
NORMAL(HDA CABLES NOT SWAPPED)
HDA CABLES SWAPPED
BEFORE THE HDA CABLE SWAP. TI-E LEOS MAKE
REFERENCE TO THE FOLLOWING.
AFTER THE HDA CABLE SWAP. THE SAME LEOS MAKE
REFERENCE TO TI-E FOLLOWING.
.
~r---OJ
...~-- LEFT SEQ\£NCER STATUS
"~,.---RIGHT
__~~- +5/2L1- V ON
...~~- +5/2L1- V ON
........--- +5 V COMMON
- . . . . . - - - +5 V COMMON
SEQUENCER STATUS
...~-- RIGHT SEQUENCER STATUS
...~:-- LEFT DEVICE READY
{
"'-=-:'-- LEFT
"'oE-- RIGHT
(
I
SWITCH ENABLED
SWITCH ENABLED
_-=:-.- RIGHT DEVICE READY
«
NOTES.
OJ ~ck-~/&F~~gl::O:A~~T"!f:rift~//fnfs:r'
1500058251 ,
f-
.
Figure 2.
r'2'
MEN THE CABLES ARE StiAPPED, THE LEO Fa? THE RIGHT
L.6.J SCQlENCER STA TUS SHO/IIS LEFT SEQUENCER STA TUS.
•
B1 Board LEOs
PIN 4519899 56 Sides EC 475248 25Apr88
C Copyright
.
IBM Corp. 1987, 1988
ISOLATE-5
IMAP 0210: Miscellaneous Physical Symptoms
001
(From IMAP 0205 Step 021 on page ISOLATE-4)
Do you have one of the physical symptoms listed in
Figure 31
Yes No
I
002
Select an entry from the following list and go
to the specified MAP:
/".----.......\
\.~,_../I
•
Sense data is available. Go to IMAP
0212 Step 001 on page ISOLATE-?
•
Console message is available and sense
data is not. Go to IMAP 0255 Step 001
on page ISOLATE-34.
•
Sense data and console messages are
not available. Go to IMAP 0232 Step 001
on page ISOLATE-18.
003
Find the first Item (from the top) in Figure 3 that
matches the physical symptom. Then go to the
specified reference.
Physical Symptom
Reference
Failure to power off string or controller
Page PWR-1, Entry AR (MIM, Volume R 10)
Failure to power off device logic
Page PWR-1, Entry AR (MIM, Volume R10)
Failure to power off drive motor
Page PWR-1, Entry H (MIM, Volume R 10)
Device Ready LED is on when Device
Enable/Disable switch is set to Disable
Page MAP-1, Entry CE (MIM, Volume R10)
No power at convenience outlet
Page PWR-1, Entry AM (MIM, Volume R10)
Motor braking problem (Drive coasts to stop
on power off)
Page PWR-1, Entry K (MIM, Volume R10)
Controller Ready LED not on (operator panel)
Page PWR-1, Entry BT (MIM, Volume RfO)
~,./'
!' -,
\
Device Ready LED not on (operator panel)
Page PWR-1, Entry BS (MIM, Volume R 10)
Status Pending LED is on continuously on the
storage control (3880 only)
IMAP 0252 Step 001 on page ISOLATE-33.
Figure 3.
'. "---
~._ ...7'
Physical Symptom Table
c
ISOLATE-6
PIN 4519899 56 Sides EC 475248 25Apr88
It:) Copyright IBM Corp. 1987, 1988
(
IMAP 0212: Checking For Undervo/tage
001
(From IMAP 0210 Step 002 on page ISOLATE-6)
(
(
((
Determine if there is an undervoltage condition.
Review all sense records for the failing string.
Undervoltage is determined from the symptom
code (sense bytes 22 and 23) as follows:
9FFF
DFFF
EFFO
EFFF
Device Undervoltage
Controller Undervoltage
Device Undervoltage
Controller Undervoltage
Is there an undervoltage condition?
Yes No
I
002
Go to IMAP 0215 Step 001 on page
ISOLATE-S.
003
Check the device and controller voltages and ripple.
Do the procedures referenced by page PWR-1,
Entry A in the MIM, Volume R 10 and then return
here.
Are the voltages and ripple OK.?
Yes No
I
004
Repair or replace FRUs as directed by the
procedures in the PWR MAPs.
(
Then go to RMAP 0320 Step 001 on page
REPAIR-Sto run diagnostics.
(
005
(
Go to IMAP 0215 Step 001 on page ISOLATE-S.
c
PIN 4519899 56 Sides EC 475248 25Apr88
© Copyright
IBM Corp. 1987. 1988
ISOLATE-1
IMAP 0215: Running Diagnostics
001
(From
(From
(From
(From
(From
(From
(From
I':~
\"",Y
IMAP
IMAP
IMAP
IMAP
IMAP
IMAP
IMAP
0200
0205
0212
0212
0232
0233
0252
Step
Step
Step
Step
Step
Step
Step
003
008
002
005
002
011
001
on
on
on
on
on
on
on
page ISOLATE~1)
page ISOLATE-3)
page ISOLATE-7)
page ISOLATE~7)
page ISOLATE~18)
page ISOLATE~22)
page ISOLATE~33)
Perform the following steps:
1.
Attach and prepare the MD. See page
directions.
MO~4
for
Warning: Never power the MD on or off with
the diskette inserted.
2.
Ensure that the storage control is powered on
and has completed its initial microcode load
(IML), and that the functional diskette for the
3380~JK is installed.
3.
Determine if the controller is the probable
failing unit. The probable failing unit is the
controller if one of the following conditions
exist:
•
'--
I/" ' '.
EREP Symptom Code Summary report
indicates probable failing unit: "Controller"
or "Storage Control Unit."
•
Highest priority symptom code (sense bytes
22 and 23) is Oxxx.
•
Clock LED is off or is as bright as the
voltage LEOs.
\
~
---'
Does one of these conditions exist?
Yes No
I
(1' -,
\l_::t f
002
Go to IMAP 0220 Step 001 on page
ISOLATE~12.
003
Go to IMAP 0218 Step 001 on page
ISOLATE-8
ISOLATE~9.
PIN 4519899 56 Sides EC 475248 25Apr88
10 Copyright
IBM Corp. 1987, 1989
rf.,.~
~\\
V
(
(
(
IMAP 0218: Controller Diagnostics
001
(From IMAP 0215 Step 003 on page ISOLATE-8)
The controller tests can be run in-line with
customer operations on the controller. Therefore,
controller paths and devices do not need to be
varied off line.
003
Some configurations do not use the DPS cards.
The DPS cards are used when the string is
attached to a 3880, but they are not used when the
string is attached to a 3990.
Is the string attached to a 38801
Yes No
I
Run the Controller tests:
(
1.
2.
(
From
(RUN
From
(RUN
the MD Main Menu, select option 1
DIAGNOSTICS).
the Diagnostic Menu, select option 1
CONTROLLER TESTS).
See page MD-12 for directions.
(
(
(
(
(
Did the controller diagnostics run to normal end
without an error?
Yes No
I
002
Analyze the error. Press the ENTER key on
the MD to analyze the error and to obtain a
FRU group or list of action items.
(-
c-
005
To completely test the DPS function run the
extended DPS tests. The extended DPS tests
check the correct interaction of DPS on both
controllers (A 1 and A2). These tests can be run
inline with customer operations on both controllers.
Therefore, controller paths and devices do not
need to be varied offline.
Run the extended DPS tests:
1.
2.
Answer questions and follow instructions
given by the MD. If you are asked "DO YOU
WANT TO RUN CONNECTION TESTS?" you
are advised to answer "YES." The
connection tests provide better isolation.
The connection tests do not run inline with
customer operation. Before running the
connection tests, all paths to the failing
controller should be varied offline. If this
cannot be done, answer "NO" to "DO YOU
WANT TO RUN CONNECTION TESTS ?"
(
004
Go to Step 007 in this IMAP.
When a FRU group is displayed, use the
guided FRU replacement procedures in the
REPAIR section to replace the indicated
FRUs.
Go to RMAP 0300 Step 001 on page
REPAIR-2.
3.
Press the PF key to return to the MD Main
menu.
From the MD Main menu, select
option 1 (RUN DIAGNOSTICS).
From the Diagnostic menu, select
option 2 (EXTENDED DPS TESTS).
See page MD-13 for more complete directions.
Did the extended DPS tests run to normal end
without an error?
Yes No
I
006
Analyze the error. Press the ENTER key on
the MD to analyze the error and to obtain a
FRU group or list of action items.
When a FRU group is displayed, use the
guided FRU replacement procedures in the
REPAIR section to replace the indicated
FRUs.
(Step 006 continues)
(Step 003 continues)
PIN 4519899 56 Sides EC 475248 25Apr88
10 Copyright
IBM Corp. 1987, 1988
ISOLATE-9
006 (continued)
Go to RMAP 0300 Step 001 on page
REPAIR-2.
009
Run the device tests:
1.
007
(From Step 004 in this IMAP)
2.
Press the PF key to return to the MO Main
menu.
From the Main menu, select option 1
(RUN DIAGNOSTICS).
From the Diagnostics menu, select option 3
(DEVICE TESTS).
To completely test controller operation, run device
tests on one or more devices.
3.
Vary devices offline.
See page MD-10 for directions.
Ask the customer to vary offline each of the
devices you want to test. Devices should be varied
offline on each attached system.
Did the device tests run to normal end without an
error?
Yes No
010
1.
2.
3.
Press the PF key to return to the MD Main
menu.
From the Main menu, select option a
(SET/RESET CE MODE).
Use the Set/Reset CE Mode option to set the
d,evices to be tested in CE mode.
See page MO-6 for directions.
Did CE mode set OK for all the devices specified?
Yes No
I
'--. -
Analyze the error. Press the ENTER key on
the MD to analyze the error and to obtain a
FRU group or list of action items.
Advise the customer that devices to be tested must
be placed in CE mode for the tests to run. If a
system attempts to use a device that has been
placed in CE mode, the storage control responds
with unit check and sends an Intervention Required
sense byte record to that system.
Set CE mode for the devices to be tested:
When a FRU group is displayed, use the
guided FRU replacement procedures in the
REPAIR section to replace the indicated
FRUs.
Go to RMAP 0300 Step 001 on page
REPAIR-2.
011
Do you have SENSE data from EREP or the system
console for the Qriginal failure?
Yes No
I
\'-..
012
-
Go to (MAP 0240 Step 001 on page
ISOLATE-24.
013
Using Figure 4, find the highest priority symptom
code group identified in the START section.
Go to the MAP step, MLX entry, or maintenance
procedure indicated.
.If "',
\t_,
~"
i'
008
',,--.
Go to (MAP 0225 Step 001 on page
ISOLATE-14.
(Step 009 continues)
ISOLATE-10
":,
I
If the devices are not varied offline successfully,
avoid using those devices or proceed only with
customer permission.
When a device is taken out of CE mode, an
interrupt is generated to each using system and
normal operation of the device should be resumed.
I~'
',,-,j
PIN 4519899 56 Sides EC 475248 25Apr88
~ Copyright IBM Corp, 1987, 1988
(
(-
(
c
Symptom Code Groups
Go To
9xxx, Dxxx, or Exxx
(controller or device equipment checks)
IMAP 0235 Step 001 on page ISOLATE-23
4xxx (data check)
Page MAP-1, Entry EB, in the MIM, Volume R10.
Oxxx. 2xxx. 3xxx. or Fxxx
(storage control errors)
The storage control maintenance package. MLX entry 6
1xxx. 6xxx. 7xxx. or 8xxx
The 3380 Stage 2 maintenance package
Axxx. Bxxx. or Cxxx
The 3380-0E maintenance package
Figure 4.
Symptom Code Groups
(
(
(
(
(
c
(
(
PIN 4519899 56 Sides EC 475248 25Apr88
@Copyright IBM Corp. 1987, 1988
{
ISOLATE-11
IMAP 0220: Normal Sequence Diagnostics
001
(From IMAP 0215 Step 002 on page ISOLATE-8)
(From IMAP 0232 Step 011 on page ISOLATE-19)
(From IMAP 0247 Step 002 on page ISOLATE-30)
Did CE mode set OK for all the devices specified?
Yes No
I
002
Go to IMAP 0225 Step 001 on page
ISOLATE-14.
Determine which devices to test.
If a device sequence status LED is off, use that
device to test. Otherwise, the failing device can
usually be determined from the Physical Address
Column of the EREP DASD Symptom Code
Summary or from Sense byte 4 unless the
symptom code is EBxx.
003
If the symptom code is EBxx, it may be necessary
to run diagnostics on each device on the failing
port to determine which device is failing. See
SMAP 0057 Step 001 on page START-32for
additional information.
2.
/'
(
.~
(From Step 005 in this IMAP)
Run the Normal Sequence diagnostics:
1.
3.
Press the PF key to return to the MD Main
menu.
From the Main menu, select
option 1 (RUN DIAGNOSTICS).
From the Diagnostics menu, select
option 0 (NORMAL SEQUENCE).
See page MD-1O for directions.
Vary devices offline.
Ask the customer to vary offline each of the
devices you want to test. Devices should be varied
offline on each attached system.
Did the diagnostics run to normal end without an
error?
Yes No
I
004
If the devices are not varied offline successfully,
avoid using those devices or proceed only with
customer permission.
Analyze the error. Press the ENTER key 0'1
the MD to analyze the error and to obtain a
FRU group or list of action items.
Advise the customer that devices to be tested must
be placed in CE mode for the tests to run. If a
system attempts to use a device that has been
placed in CE mode, the storage control responds
with unit check and sends an Intervention Required
sense byte record to that system.
When a device is taken out of CE mode, an
interrupt is generated to each using system and
normal operation of the device should be resumed.
Set CE mode:
1.
2.
From the MD Main menu, select option 0
(SET/RESET CE MODE).
Use the Set/Reset CE Mode option to set the
devices to be tested in CE mode.
When a FRU group is displayed, use the
guided FRU replacement procedures in the
REPAIR section to replace the indicated
FRUs.
Go to RMAP 0300 Step 001 on page
REPAIR-2.
005
Run the Normal Sequence diagnostics on other
controllers.
If the failing path is not known and any of the other
controllers have not been tested, go to Step 003 in
this IMAP. Repeat the normal sequence tests with
another controller (if possible) and on other
devices.
(If .,
,
'",-/
See page MD-6 for directions.
ISOLATE-12
PIN 4519899 56 Sides EC 475248 2SApr88
© Copyright IBM Corp, 1987, 1988
tr'
\~ ;I
(
(
006
(From Step 005 in this IMAP)
(
(
REPAIR section to replace the indicated
FRUs.
005 (continued)
- or If all controllers (or all that can be) have been
tested, go to Step 006 in this IMAP.
Some configurations do not use the DPS cards.
The DPS cards are used when the string is
attached to a 3880, but they are not used when the
string is attached to a 3990.
Go to RMAP 0300 Step 001 on page
REPAIR-2.
010
(From Step 007 in this IMAP)
Do you have SENSE data available from the original
failure?
Yes No
I
011
Is the string attached to a 3880?
Yes No
Go to IMAP 0240 Step 001 on page
ISOLATE-24.
I
007
Go to Step 010 in this IMAP.
(
012
Go to IMAP 0235 Step 001 on page ISOLATE-23.
008
To completely test the DPS function, it is necessary
to run the extended DPS tests. The extended DPS
tests check the interaction of DPS on both
controllers (A 1 and A2). These tests can be run
inline with customer operations on both controllers.
Therefore, controller paths and devices do not
need to be varied offline.
Run the Extended DPS Tests:
1.
2.
3.
(
Press the PF key to return to the MD Main
menu.
From the Main menu, select option 1
(RUN DIAGNOSTICS).
From the Diagnostic menu, select option 2
(EXTENDED DPS TESTS).
See page MD-13 for directions.
Did the Extended DPS Tests run to normai end
without an error?
Yes No
I
{
(
009
Analyze the error. Press the ENTER key on
the MD to analyze the error and to obtain a
FRU group or list of action items.
When a FRU group is displayed, use the
guided FRU replacement procedures in the
(
(-
PIN 4519899 56 Sides EC 475248 25Apr88
il::l Copyright
IBM Corp. 1987, 198B
ISOLATE-13
IMAP 0225: Trouble Setting CE Mode
001
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
IMAP 0200 Step 003 on page ISOLATE-1)
IMAP 0218 Step 008 on page ISOLATE-1O)
IMAP 0220 Step 002 on page ISOLATE-12)
IMAP 0230 Step 002 on page ISOLATE-16)
IMAP 0232 Step 004 on page ISOLATE-18)
IMAP 0245 Step 014 on page ISOLATE-28)
IMAP 0248 Step 004 on page ISOLATE-31)
IMAP 0260 Step 011 on page ISOLA TE-36)
IMAP 0265 Step 013 on page ISOLATE-39)
IMAP 0270 Step 012 on page ISOLATE-43)
IMAP 0275 Step 004 on page ISOLATE-46)
IMAP 0280 Step 004 on page ISOLATE-49)
IMAP 0285 Step 004 on page ISOLATE-51)
IMAP 0290 Step 004 on page ISOLATE-53)
RMAP 0320 Step 006 on page REPAIR-8)
The Set/Reset CE Mode diagnostic may not run if
certain errors are present in the storage control,
controller, or control interface. If the diagnostic
does not run, CE mode for the specified devices
does not change and the MD displays an error
message. See "Isolation Code Error Display" on
page MD-36.
''\..;
Analyze the error. Press the ENTER key on the MD
to analyze the error and to obtain a FRU group or
list of action items.
If the MD reports a OOxx isolation code and
suggests that you run the connection tests, run
them if possible. The connection tests do not run
in line with customer operation. Before running the
connection tests, all paths to the failing controller
should be varied offline. If this cannot be done,
answer "NO" to "DO YOU WANT TO RUN
CONNECTION TESTS ?"
When a FRU group is displayed, use the guided
FRU replacement procedures in the REPAIR section
to replace the indicated FRUs.
Go to RMAP 0300 Step 001 on page REPAIR-2.
ISOLATE-14
PIN 4519899 56 Sides EC 475248 25Apr88
© Copyright
IBM Corp. 1987, 1988
-.-'
(
{
IMAP 0228: Undervoltage With Data Checks
001
(
(From SMAP 0058 Step 005 on page START-33)
(From SMAP 0060 Step 003 on page START-34)
(From IMAP 0200 Step 003 on page ISOLATE-1)
Determine if there is an undervoltage condition.
Examine all 4XYY symptom codes to determine if
there is an undervoltage condition as follows:
YY, Bit 4
YY, Bit 5
= Device Undervoltage
= Controller Undervoltage
Is there a symptom code that indicates
undervoltage?
Yes
No
I
(
002
Go to IMAP 0230 Step 001 on page
ISOLATE-16.
003
Check the Device and Controller voltages and
ripple.
(
Do the procedures referenced by page PWR-1,
Entry A, in the MIM, Volume RIO, and then return
here.
Are the voltages and ripple OK?
Yes
No
I
004
(
(
(
Repair or replace FRUs as directed by the
procedures in the PWR section MAPs (see
the MIM, Volume RIO).
Go to RMAP 0320 Step 001 on page
REPAIR-8 to run diagnostics.
005
Go to IMAP 0230 Step 001 on page ISOLATE-16.
(
PIN 4519899 56 Sides EC 475248 25Apr98
© Copyright IBM Corp. 1997, 1999
ISOLATE-15
IMAP 0230: Diagnostics for Data Checks
001
3.
(From SMAP 0053 Step 002 on page START-29)
(From IMAP 0228 Step 002 on page ISOLATE-15)
(From IMAP 0228 Step 005 on page ISOLATE-15)
See page MD-1O for directions.
Vary devices offline.
Ask the customer to vary offline each of the
devices you want to test. Devices should be varied
offline on each attached system.
From the Diagnostics menu, select option 0
(NORMAL SEQUENCE).
Did the diagnostics run to normal end without an
error?
Yes No
I
004
Analyze the error. Press the ENTER key on
the MD to analyze the error and to obtain a
FRU group or list of action items.
If the devices are not varied offline successfully,
avoid using those devices or proceed only with
customer permission.
When a FRU group is displayed, use the
guided FRU replacement procedures in the
REPAIR section to replace the indicated
FRUs.
Advise the customer that devices to be tested must
be placed in CE mode for the tests to run. If a
system attempts to use a device that has been
placed in CE mode, the storage control responds
with unit check and sends an Intervention Required
sense byte record to that system.
Go to RMAP 0300 Step 001 on page
REPAIR-2.
When a device is taken out of CE mode, an
interrupt is generated to each using system and
normal operation of the device should be resumed.
005
Set CE mode:
If the failing path is not known and other controllers
have not been tested, repeat the normal sequence
tests with other controllers until all the controllers
(or all that can be) have been tested.
1.
2.
From the MD Main menu, select option 0
(SET/RESET CE MODE).
Use the Set/Reset CE Mode option to set the
devices to be tested in CE mode.
See page MD-6 for directions.
Did CE mode set OK for all the devices specified?
Yes No
I
002
Go to IMAP 0225 Step 001 on page
ISOLATE-14.
003
Run the Normal Sequence diagnostics:
1.
2.
Press the PF key to return to the MD Main
menu.
From the MD Main menu, select option 1
(RUN DIAGNOSTICS).
ISOLATE-16
Repeat the Normal Sequence diagnostics on other
controllers.
Did the diagnostics run to normal end without error
with all controllers?
Yes No
(
""-.---
I
006
Analyze the error. Press the ENTER key on
the MD to analyze the error and to obtain a
FRU group or list of action items.
When a FRU group is displayed, use the
guided FRU replacement procedures in the
REPAIR section to replace the indicated
FRUs.
Go to RMAP 0300 Step 001 on page
REPAIR-2.
(Step 007 continues)
PIN 4519899 56 Sides EC 475248 25Apr88
© Copyright IBM Corp. 1987, 1988
Ci
(
007
Is sense data available?
Yes No
I
008
Go to page MAP-1, Entry EB, in the MIM,
Volume R10.
009
(
Go to IMAP 0235 Step 001 on page ISOLATE-23.
(
(
(
(
c
(
((
PIN 4519899 56 Sides EC 475248 25Apr88
© Copyright
IBM Corp. 1987, 1988
ISOLATE-17
IMAP 0232: Determine the Failing Devices and Controllers with the Device Status Test
001
.4. -~\
IV
004
CE mode must be set before running the
Device Status test.
(From IMAP 0210 Step 002 on page ISOLATE-6)
(From MD 0050 Step 002 on page MD-50)
(From MD 0051 Step 016 on page MD-54)
0\
\""j
Set CE mode:
Do you need help to determine which devices and
controllers are failing?
Yes No
1.
I
2.
002
Go to IMAP 0215 Step 001 on page
ISOLATE-S.
I
See page MD-6 for directions.
The following procedure uses the Device Status
test to determine the failing paths and devices.
Run the Device Status test:
1.
Attach and IML the MD to one of the A units in
the failing string. See page MD-4 for
directions.
Warning: Never power the MD on or off with
the diskette inserted.
2.
3.
Ensure that the storage control is powered on
and has completed its initial microcode load
(lML) and that the functional diskette for the
33S0-JK is installed.
Check the channel enable/disable switches at
each attached storage control.
Is there at least one channel enabled on the
attached storage control?
Yes No
I
(Istep 004 continues)
ISOLATE-18
005
(From Steps 004 and 007 in this IMAP)
Continue to run the Device Status Test:
1.
2.
From the MD Main menu, select option 1
(RUN DIAGNOSTICS).
From the Diagnostic menu, select option 7
(DEVICE STATUS TEST).
The test should run until an error occurs or until
configuration data is displayed.
See page MD-25 for directions.
Did the test display configuration data?
Yes No
I
006
The test stopped with an isolation code
displayed.
Analyze the error. Press the ENTER key on
the MD to analyze the error and to obtain a
FRU group or list of action items.
When a FRU group is displayed, use the
guided FRU replacement procedures in the
REPAIR section to replace the indicated
FRUs.
PIN 4519899 56 Sides EC 475248 25Apr88
© Copyright
.
.~./
After CE mode is set, go to Step 005 in this
IMAP.
- or If CE mode does not set, go to IMAP 0225
Step 001 on page ISOLATE-14.
003
The Device Status test selects each device in the
string and performs a Read Home Address
operation. The Device Status test runs inline with
customer operation. Devices do not need to be
varied offline or placed in CE mode unless there
are no channels enabled to the storage path. See
"Diagnostic Menu Option 7 - Device Status" on
page MD-25 for additional information.
From the MD Main menu, select option 0
(SET/RESET CE MODE).
Use the Set/Reset CE Mode option to set
the devices to be tested in CE mode.
IBM Corp. 1987. 1989
(
006 (continued)
Go to RMAP 0300 Step 001 on page
REPAIR-2.
(
(
(
(
f
(
(
007
Review the configuration data to ensure that it is
correct. See page MD-27 for 3990 Configuration
data; see page MD-26 for 3880 Configuration data.
If the configuration data is not correct or you are
not sure, If the configuration data is not correct or
you are not sure, perform an analysis as instructed
by the MD. Correct any problems. If there were
any problems, rerun the Device Status test, then
return here.
To determine which devices and paths are
failing, use the error log.
Go to IMAP 0233 Step 001 on page
ISOLATE-20.
011
Record the devices that report a status of "0," "E"
or "I". These are failing devices.
Select one of these devices and failing paths to run
diagnostics. Then go to IMAP 0220 Step 001 on
page ISOLATE-12.
Record the device status for each device. See
"Device Status" on page MD-29.
Repeat the test on each controller. Start at
Step 005 in this IMAP on a 4-path (4 controllers)
subsystem. You will have to move the MD plug to
the other A unit in the string to run the Device
Status Test with the controllers in that A unit.
Operational devices should report a status of "0."
A status of "N" will be reported for devices that are
not attached.
Did any attached devices report a status other than
"0" through any controller?
Yes No
I
008
To determine which devices and paths are
failing, use the error log.
(
Go to IMAP 0233 Step 001 on page
ISOLATE-20.
009
Did any devices report a status of "0," "E," or "I"?
Yes No
I
010
Devices reporting a status of "B" or "S" are
busy. This may occur if the Device Status
test is run during periods of heavy usage. It
is usually a good indication that the device is
operational. However, you may want to
rerun the test to ensure that these devices
are operational.
PIN 4519899 56 Sides EC 475248 25Apr88
@ Copyright IBM Corp. 1987, 1988
(
ISOLATE-19
()
IMAP 0233: Determine the Failing Devices and Controllers with the Error Log
001
002 (continued)
Ensure that the storage control you are
going to use is powered on and has
completed its initial microcode load (IML),
and that the functional diskette for the
3380-JK is installed.
(From IMAP 0232 Step 008 on page ISOLATE-19)
(From IMAP 0232 Step 010 on page ISOLATE-19)
The error log is collected and maintained by the
storage control. The error log is an equipment
check log. Data checks are not included unless the
error log mask is changed to include them.
There is an error log for each controller.
Each error log contains a maximum of 16 entries
for errors that occurred on the string when using
the controller. Each entry contains 32 bytes (24
sense bytes and 8 bytes of additional information).
Each error log entry contains a counter. If an error
occurs that is the same as the previous error
(same controller and symptom code for format 7, or
same controller, device, and symptom code for
formats 1, 8, and 9) the counter in the last entry is
updated and a new log entry is not generated.
The error log may be transferred to the MD and
then searched, based on search parameters
entered on the MD. The error logs for all
controllers can be transferred to the MD and then
merged together into one error log. The merged
error log can then be searched for errors on any
device and C:lOy controller.
The error log is accessed by an inline diagnostic
routine that runs in the storage control under
control of the MD. Devices need not be in CE
mode.
Is the MD already attached and loaded?
Yes No
I
002
Attach and prepare the MD.
Warning: Never power the MD on or off with
the diskette inserted.
Go to Step 003 in this IMAP.
003
(From MD 0050 Step 002 on page MD-50)
(From Steps 002 and 005 in this IMAP)
Warning: If the MD diskette is from another string,
all logs must be transferred before searching any of
the logs. Otherwise, the results of the search will
be unpredictable.
Use the transfer log function to collect error
information from the error logs:
1.
2.
3.
From the MD Main menu, select option 2
(ERROR LOG FUNCTION).
From the Error Log Function menu, select
option 1
(TRANSFER LOG).
Specify your choice for the controller.
During the transfer, the most recent entry is first
and the oldest entry is last. When the transfer is
complete, the log can be searched or merged with
other transferred logs.
The transfer uses a diagnostic program that runs in
the storage control. Some problems with the
storage control, with the controller, or with the
control interface (DDC) do not allow the diagnostic
to run. The error log is not transferred, and the
MD displays a diagnostic error message.
Was the transfer completed without a diagnostic
error message?
Yes No
I
004
See page MD-4 for directions.
(Step 002 continues)
ISOLATE-20
Press ENTER to begin analysis in the MD.
Note: See "Isolation Code Error
Display" on page MD-36.
PIN 4519899 56 Sides EC 475248 25Apr88
It) Copyright IBM Corp. 1987, 1988
/"
~,
(
(~"
004 (continued)
When a FRU group is displayed, use the
guided FRU replacement procedures in the
REPAIR pages to replace the indicated FRUs.
(
Go to RMAP 0300 Step 001 on page
REPAIR-2.
(
005
The error log for the selected controller was written
on the MD diskette during the transfer. The error
log in the storage control is not reset.
(
(
Transfer error logs for all other controllers:.
1.
2.
(
c·
c
3.
From the MD Main Menu, select option 2
(ERROR LOG FUNCTION).
From the Error Log Function menu, select
option 1
(TRANSFER LOG)
Specify your choice for the controller.
most recent entries are displayed first. For
instructions, see "Search Log (Option 2)" on
page LOG-S.
Record each symptom code, device address,
controller, and age, for the errors that occurred
since the customer started having problems. If the
age is not known because the error log came from
a 3880 Storage Control, it is difficult to determine
which records are associated with the problem.
If possible, obtain information from the customer
about the nature and frequency of the problem.
This information, combined with your knowledge of
the devices in this account, can sometimes be used
to determine which records are associated with the
problem. The most recent entry in the log is
labeled ENT=01; the last entry in the log is the
oldest.
Are there any entries in the transferred logs?
Yes No
I
Repeat Step 003 in this IMAP for the other
controllers.
006
Have the customer run EREP.
If the string is a 4-path string, move the MD plug to
the other A unit in the string and repeat Step 003
in this IMAP for both controllers in the other A unit.
An EREP report will be required to determine
failing devices and controllers. Ask the
customer to run the EREP Exception reports.
Search transferred logs.
The error log can now be searched using option 2
of the error log function menu. For instructions,
see "Search Log (Option 2)" on page LOG-S.
Alternatively, a merge (option 3) can be done with
other transferred error logs and a search done on
the merged error logs. For instructions, see
"Merge Log (Option 3)" on page LOG-8.
(
(
C"
(
("
If the string is attached to a 3990 Storage Control,
merge the error logs just transferred using Error
Log Function Menu option 3. Error log entries for
all specified logs are merged in chronological
order. Records in the merged log can then be
searched and displayed without searching each
individual log.
Display each entry in each log transferred or in the
merged log using Error Log Function Menu option 2
(Search Log Function). Specify ALL records. The
PIN 4519899 56 Sides Eo 475248 25Apr88
10 Copyright
IBM Corp: 1987. 1988
Then go to SMAP 0020 Step 001 on page
START-8.
007
Prioritize the errors in the log using the following
procedure.
Note: All undervo/tage problems should be
resolved before attempting resolve other
problems.
Review the system codes for the following
undervoltage conditions:
DFFF
EFFF
EFFO
9FFF
Controller undervoltage
Controller undervoltage
Device undervoltage
Device undervoltage
(Step 007 continues)
ISOLATE-21
/.
~I
007 (continued)
011 (continued)
Do you have one of these symptom codes?
Yes
No
The procedure in IMAP 0215 Step 001 on page
ISOLATE-8uses device diagnostics to help isolate
the problem.
I
o
008
All symptom codes associated with the
string of drives should be considered.
1.
Determine the highest priority symptom
code group from the following list:
•
•
•
•
Dxxx (except low priority SCs)
Exxx
Low priority symptom codes:
DFx3 DFx9
DFxC
DFx5 DFxA
DFxD
DFx8 DFxB
9xxx
2.
Determine the highest priority symptom
code within the selected group. Select
the symptom code that occurs most often
or has the highest count.
3.
Determine which Devices and Controller
paths are reporting the selected
symptom code from the information
previously recorded.
Go to Step 011 in this IMAP.
t'
009
Check the Device and Controller voltages and
ripple. See page PWR-1, Entry A, in the MIM,
Volume RIO.
Are the voltages and ripple OK?
Yes
No
I
010
Repair or replace FRUs as directed by the
procedures in the PWR section MAPs (see
the MIM, Volume R 10).
C:
Next, go to RMAP 0320 Step 001 on page
REPAIR-8 to run diagnostics.
011
(From Step 008 in this IMAP)
(Step 011 continues)
ISOLATE-22
PIN 4519899 56 Sides EC 475248 25Apr88
@ Copyright IBM Corp. 1987, 1988
C
C:
(
IMAP 0235: Symptom Code Analysis
001
(
(
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
IMAP 0200 Step 003
Page ISOLATE-11)
IMAP 0220 Step 012
IMAP 0230 Step 009
IMAP 0240 Step 006
IMAP 0245 Step 023
IMAP 0250 Step 007
IMAP 0280 Step 009
IMAP 0285 Step 009
IMAP 0290 Step 009
•
If the HDA is suspected, the HDA cable
swap procedure should be performed to
help isolate the problem if there is another
repeat call. See page CARR-1, Entry Q, in
the MIM, Volume R 10.
•
On a second repeat call, perform all
additional action items.
on page ISOLATE-1)
on
on
on
on
on
on
on
on
page
page
page
page
page
page
page
page
ISOLATE-13)
ISOLATE-17)
ISOLA TE-25)
ISOLATE-29)
ISOLATE-32)
ISOLATE-50)
ISOLATE-52)
ISOLATE-54)
Go to RMAP 0300 Step 001 on page REPAIR-2 to
replace the recommended FRUs.
Use the MD to determine the probable failing FRUs:
If you already have the FRU list, go to step 5 in the
following procedure.
(
1.
Press the PF key to return to the MD Main
menu.
2.
From the Main menu, select option 3
(ANALYZE SYMPTOM CODE).
3.
In response to the prompt, enter the highest
priority symptom code. This code should have
been previously determined in the START
procedures. See "Analyze Symptom Code
(Main Menu Option 3)" on page MD-37 for
directions on entering symptom codes.
4.
Answer all questions and perform all steps
requested by the MD until a FRU list is
displayed. If the MD asks for additional sense
information, entering the requested sense bytes
will provide the smallest possible FRU list. If
the sense bytes are not available, bypass the
step by pressing the enter key.
5.
Perform actions indicated by the MD. When
replacing FRUs, use the following guidelines:
(
(
•
On the first call, replace all FRUs in the
primary FRU list.
•
On a repeat call, replace all FRUs in the
secondary FRU list. If there is a MAP
indicated by the MD, go to the MAP to
determine any additional actions.
PIN 4519899 56 Sides EC 475248 25Apr88
@ Copyright IBM Corp. 1987, 1988
f"
ISOLATE-23
o
IMAP 0240: Problem Analysis With Error Log Data
mode.
001
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
IMAP 0200 Step 003 on page ISOLATE-1)
IMAP 0218 Step 012 on page ISOLATE-10)
IMAP 0220 Step 011 on page ISOLATE-13)
IMAP 0250 Step 006 on page ISOLA TE-32)
IMAP 0265 Step 016 on page ISOLATE-40)
IMAP 0270 Step 016 on page ISOLATE-44)
IMAP 0270 Step 017 on page ISOLATE-44)
IMAP 0275 Step 009 on page ISOLATE-47)
IMAP 0280 Step 008 on page ISOLATE-50)
IMAP 0285 Step 008 on page ISOLATE-52)
IMAP 0290 Step 008 on page ISOLATE-54)
RMAP 0340 Step 003 on page REPAIR-13)
Is the MD already attached and loaded?
Yes No
I
;"
(
002
Attach and prepare the MD.
Warning: Never power the MD on or off with
the diskette inserted.
See page MD-4 for directions.
The error log is collected and maintained by the
storage control. The error log is an equipment
check log. Data checks are not included unless the
error log mask is changed to include them.
Ensure that the storage control you are
going to use is powered on and has
completed its initial microcode load (IML),
and that the functional diskette for the
3380-JK is installed.
There is an error log for each controller.
Go to Step 003 in this IMAP.
Each error log contains a maximum of 16 entries
for errors that occurred on the string when using
the controller. Each entry contains 32 bytes (24
sense bytes and 8 bytes of additional information).
Each error log entry contains a counter. If an error
occurs that is the same as the previous error
(same controller and symptom code for format 7, or
same controller. device, and symptom code for
formats 1, 8, and 9) the counter in the last entry is
updated and a new log entry is not generated.
003
(From Steps 002 and 005 in this IMAP)
Warning: If the MD diskette is from another string,
all logs must be transferred before searching any of
the logs. Otherwise, the results of the search will
be unpredictable.
The error log may be transferred to the MD and
then searched. based on search parameters
entered on the MD. The error logs for all
controllers can be transferred to the MD and then
merged together into one error log. The merged
error log can then be searched for errors on any
device and any controller.
The error log is accessed by an inline diagnostic
routine that runs in the storage control under
control of the MD. Devices need not be in CE
ISOLATE-24
('
If not already done, use the transfer log function to
collect error information from the error logs:
1.
When attached to a 3990 storage control, each
error log entry contains the age of the entry in
hours. When attached to a 3380 storage control,
age is not provided.
\
\0....,/
2.
3.
From the MD Main menu, select option 2
(ERROR LOG FUNCTION).
From the Error Log Function menu,
select option 1 (TRANSFER LOG).
Specify your choice for the controller.
If the error log has already been transferred, go to
Step 005 in this IMAP.
During the transfer, the most recent entry is first
and the oldest entry is last. When the transfer is
complete, the log can be searched or merged with
other logs that have been transferred.
The transfer uses a diagnostic program that runs in
the storage control. Some problems with the
storage control, with the controller, or with the
PIN 4519899 56 Sides EC 475248 25Apr88
10 Copyright
IBM Corp. 1987. 1988
(f -\
~j
(f'\
~,I
C
control interface (DOC) do not allow the diagnostic
to run. The error log is not transferred, and the
MD displays a diagnostic error message.
Was the transfer completed without a diagnostic
error message?
Yes No
I
Before returning the machine to the
customer, perform the End of Call procedure.
Go to RMAP 0350 Step 001 on page
REPAIR-15.
007
A summary of each log entry is displayed.
004
(
Press ENTER to begin analysis in the MD.
Select a symptom code for analysis.
(
When a FRU group is displayed, use the
guided FRU replacement procedures in the
REPAIR pages to replace the indicated FRUs.
Review all entries before selecting an entry for
analysis.
1.
(
Go to RMAP 0300 Step 001 on page
REPAIR-2.
005
(From Step 003 in this IMAP)
{
((
The error log for the selected controller was written
on the MD diskette during the transfer. The error
log in the storage control is not reset. Perform the
following steps:
1.
Repeat Step 003 in this IMAP for the other
controller in the A unit.
2.
If the string is a 4-path string, move the MD
plug to the other A Unit in the string and repeat
Step 003 in this IMAP for the controllers in that
A Unit.
3.
The error log can now be searched using
option 2 of the error log function menu or a
merge (option 3) can be done with other
transferred error logs and a search done on
the merged error logs.
Record each symptom code, device address,
controller, and age, for the errors that occurred
since the customer started having problems. If
the age is not known because the error log
came from a 3880 Storage Control, it is difficult
to determine which records are associated with
the problem.
If possible, obtain information from the
customer about the nature and frequency of the
problem. This information combined with your
knowledge of the devices in this account, can
sometimes be used to determine which records
are associated with the problem.
The most recent entry in the log is labeled
ENT =01; the last entry in the log is the oldest.
2.
Use the following guidelines to prioritize the
failures:
•
DFFF
EFFF
EFFO
9FFF
See "Search Log (Option 2)" on page LOG-5.
(
I
An EREP report or a system console
message with sense data is needed for
analysis. When sense data is available, go
to IMAP 0235 Step 001 on page ISOLA TE-23.
(
(
PIN 4519899 56 Sides EC 475248 25Apr88
© Copyright IBM Corp. 1987, 1988
Controller undervoltage
Controller undervoltage
Device undervoltage
Device undervoltage
Note: All undervoltage problems
should be resolved before attempting
resolve other problems.
Were error log entries found?
Yes No
006
Resolve undervoltage conditions first:
•
If there are multiple symptom codes and no
undervoltage symptom codes, determine
the highest priority symptom code group
from the following list:
ISOLATE-25
r '
{'
~'
,
Dxxx (except low priority SCs)
Exxx
Low priority symptom codes:
DFx3 DFx9
DFxC
DFx5 DFxA
DFxD
DFx8 DFxB
9xxx
•
3.
If there are multiple symptoms within the
selected group, select the symptom code
that occurs most or has the highest count.
After each error display, the search function
allows you to analyze the error by entering
"A." If "A" (Analyze the Error) is chosen, the 24
sense bytes are displayed, then analysis is
done on the error.
Follow directions from the MD until a FRU group is
displayed. When replacing FRUs, use the following
guidelines:
•
On the first call, replace all FRUs in the primary
FRU list.
•
On a repeat call, replace all FRUs in the
secondary FRU list. If there is a MAP indicated
by the MD, go to the MAP to determine any
additional actions.
•
If the HDA is suspected, the HDA cable swap
procedure should be performed to help isolate
the problem if there is another repeat call. See
page CARR-1, Entry Q, in the MIM, Volume R 10.
•
On a second repeat call, perform all additional
action items.
~ .../
Go to RMAP 0300 Step 001 on page REPAIR-2 to
replace the recommended FRUs.
(~
<\."•.,,'
ISOLATE-26
PIN 4519899 56 Sides EC 475248 25Apr88
@ Copyright IBM Corp. 1987, 1988
c
("
c
IMAP 0245: Device Not Ready
001
003 (continued)
(
(From IMAP 0205 Step 020 on page ISOLATE-4)
(From IMAP 0248 Step 002 on page ISOLATE-31)
Is the lock in the unlocked position?
Yes No
(
Device not ready is indicated when a Device Ready
light on the operator panel is off.
004
A Device Ready light is off when one or more of
the following conditions are present:
1.
Unlock the access mechanism lock.
2.
Set the actuator logic switch for the
failing device to the Disable position,
then back to the Enable position. See
page REPAIR-27.
3.
The device should come ready within 10
seconds.
•
(
•
•
•
•
•
•
•
(
(
I
Perform the following steps:
The drive associated with the device is
powered off or did not power on.
The device is powered off.
The device logic is disabled.
The ENABLE/DISABLE switch associated with
the device is in the DISABLE position.
There is a device Check 1.
There is a device Check 2.
A hardware fault in the device dropped Online.
(Online is indicated in sense format 1 or 8,
sense byte 19, bit 4.)
There is a hardware fault in the servo system.
If the device becomes ready, go to Step 011
in this IMAP.
- or Otherwise, go to Step 005 in this IMAP.
005
Check the Enable/Disable switch (on the operator
panel) for the not ready device to verify that the
switch is in the Enable position.
Is the switch in the Enable position?
Yes No
(From Step 004 in this IMAP)
Check the status of the drive (running or stopped).
Is the drive stopped?
Yes No
I
I
006
002
Go to Step 011 in this IMAP.
Set the switch to the Enable position.
007
If the device becomes ready, go to Step 011
in this IMAP.
- or Otherwise, go to Step 003 in this IMAP.
c
(
003
(From Step 002 in this IMAP)
Inspect the HDA access mechanism lock for the
device that is not ready. See page CARR-1, Entry
AB, in the MIM, Volume R10.
(
(
(Step 003 continues)
PIN 4519899 56 Sides EC 475248 25Apr98
@ Copyright IBM Corp. 1987, 1988
Does the drive belt appear to be In good condition
and in the correct position on the pulleys?
Yes No
I
008
Replace the belt. See page CARR-1, Entry E,
in the MIM, Volume R 10 for the procedure.
If this corrects the Device Not Ready
problem, go to RMAP 0320 Step 001 on page
REPAIR-B.
- or Otherwise, go to Step 009 in this IMAP.
ISOLATE-27
c
( "~
_.7
012 (continued)
Attach and prepare the MD.
009
(From Step 008 in this IMAP)
Check the fit of the drive belt cover to the belt
cover switch.
Warning: Never power the MD on or off with
the diskette inserted.
Does the belt cover switch appear to be correctly
operating?
Yes No
See page MD-4 for directions.
Ensure that the storage control you are
going to use is powered on, has completed
its initial microcode load (IML), and that the
functional diskette for the 3380-JK is
installed.
I
010
Correct the problem with the belt cover
switch.
If this corrects the Device Not Ready
problem, go to RMAP 0320 Step 001 on page
REPAIR-8.
- or Otherwise, go to Step 011 in this IMAP.
Continue with Step 013 in this IMAP.
013
(From Step 012 in this IMAP)
i"
Set CE Mode:
011
(From Steps 002, 004, 006, and 010 in this IMAP)
1.
Vary devices offline.
2.
Ask the customer to vary offline each of the
devices you want to test. Devices should be varied
offline on each attached system.
See page MD-6 for directions.
If the devices are not varied offline successfully,
avoid using those devices or proceed only with
customer permission.
Advise the customer that devices to be tested must
be placed in CE mode for the tests to run. If a
system attempts to use a device that has been
placed in CE mode, the storage control responds
with unit check and sends an Intervention Required
sense byte record to that system.
From the MD Main menu, select option 0
(SET/RESET CE MODE).
Use the Set/Reset CE Mode option to set the
devices to be tested in CE mode.
Did CE mode set OK for all the devices specified?
Yes No
I
014
Go to IMAP 0225 Step 001 on page
ISOLATE-14.
015
(From Step 018 in this IMAP)
Run the Normal Sequence diagnostics:
,,;1(- "-', \
\
" ' ..• J
When a device is taken out of CE mode, an
interrupt is generated to each using system and
normal operation of the device should be resumed.
Is the MD already attached to the A unit, powered
on, and ready for use?
Yes No
1.
2.
3.
Press the PF key to return to the MD Main
menu.
From the Main menu, select option 1
(RUN DIAGNOSTICS).
From the Diagnostics menu, select option 0
(NORMAL SEQUENCE).
I
See page MD-10 for directions.
012
(Step 012 continues)
(Step 015 continues)
ISOLATE-28
PIN 4519899 56 Sides EC 475248 25Apr1l8
@ Copyright IBM Corp. 1987, 1988
,
(-
015 (continued)
Did the diagnostics run to normal end without an
error?
Yes No
I
(
(
c
(
(
016
Press the ENTER key on the MD to analyze
the error and to obtain a FRU group.
When a FRU group is displayed, use the
guided FRU replacement procedures in the
REPAIR pages to replace the indicated FRUs.
022 (continued)
If this Is of concern, replace the
SequencerlServo/RPS card and Servo Power
Amplifier card for the affected actuator.
Next, go to RMAP 0300 Step 001 on page
REPAIR-2.
- or otherwise, go to RMAP 0350 Step 001 on
page REPAIR-15.
023
Go to IMAP 0235 Step 001 on page ISOLATE-23.
Go to RMAP 0300 Step 001 on page
REPAIR-2.
017
Has the failing device been tested with all
controllers?
Yes No
I
018
Go to Step 015 in this IMAP and repeat the
normal sequence tests with all controllers.
019
Is the Device Ready LED on now?
Yes No
I
020
Use the PWR section MAPs in the MIM,
Volume R10 to isolate the problem.
(
(
(
Go to page PWR-1, Entry BS, in the MIM,
Volume R10.
021
Is sense data available from the system (EREP,
console, ICKDSF, etc.)?
Yes No
I
(
(-
(
(-
022
It is not unusual for device ready LEDs to go
off during long periods of non-operation.
This may be due to a loss of track following.
If this happens, using the device will cause
the storage control to rezero the actuator.
This will bring the device back to the ready
state.
PIN 4519899 56 Sides EC 475248 25Apr88
@ Copyright IBM Corp. 1987, 1988
ISOLATE-29
IMAP 0247: Device Switch Enabled LED Off
001
(From IMAP 0205 Step 020 on page ISOLATE-4)
Check the front of the A unit to ensure that the
suspect device Enable/Disable switch is in the
Enable position.
Is the switch in the Enable position?
Yes No
I
002
Switch to Enable, then go to IMAP 0220 Step
001 on page ISOLATE-12 to run diagnostics.
003
Go to page MAP-1, Entry CEo in the MIM, Volume
R10.
iii' '"
\~.->.,
ISOLATE-30
PIN 4519899 56 Sides EC 475248 25Apr88
Copyright
IBM Corp. 1987, 1988
REPAIR-1
(
,
"'-J
RMAP 0300: FRU Replacement Procedures
001
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
IMAP 0218 Step 002 on page ISOLATE-9)
IMAP 0218 Step 006 on page ISOLATE-10)
IMAP 0218 Step 011 on page ISOLATE-10)
IMAP 0220 Step 004 on page ISOLATE-12)
IMAP 0220 Step 009 on page ISOLATE-13)
IMAP 0225 Step 001 on page ISOLATE-14)
IMAP 0230 Step 004 on page ISOLATE-16)
IMAP 0230 Step 006 on page ISOLATE-16)
IMAP 0232 Step 006 on page ISOLATE-19)
IMAP 0233 Step 004 on page ISOLATE-21)
IMAP 0235 Step 001 on page ISOLATE-23)
IMAP 0240 Step 004 on page ISOLATE-25)
IMAP 0240 Step 009 on page ISOLATE-26)
IMAP 0245 Step 016 on page ISOLATE-29)
IMAP 0245 Step 022 on page ISOLATE-29)
IMAP 0248 Step 006 on page ISOLATE-31)
IMAP 0248 Step 007 on page ISOLATE-31)
IMAP 0250 Step 002 on page ISOLA TE-32)
IMAP 0260 Step 008 on page ISOLATE-36)
IMAP 0260 Step 015 on page ISOLATE-37)
IMAP 0260 Step 016 on page ISOLATE-37)
IMAP 0265 Step 008 on page ISOLATE-39)
IMAP 0265 Step 015 on page ISOLATE-40)
IMAP 0265 Step 017 on page ISOLATE-40)
IMAP 0265 Step 018 on page ISOLATE-41)
IMAP 0270 Step 008 on page ISOLATE-43)
IMAP 0270 Step 014 on page ISOLATE-44)
IMAP 0270 Step 018 on page ISOLATE-44)
IMAP 0270 Step 019 on page ISOLATE-45)
IMAP 0275 Step 008 on page ISOLATE-47)
IMAP 0275 Step 012 on page ISOLATE-47)
IMAP 0275 Step 013 on page ISOLATE-48)
IMAP 0280 Step 006 on page ISOLATE-50)
IMAP 0285 Step 006 on page ISOLATE-52)
IMAP 0290 Step 006 on page ISOLATE-54)
IMAP 0295 Step 002 on page ISOLATE-55)
Page REPAIR-1)
Page REPAIR-1)
Page REPAIR-1)
RMAP 0320 Step 010 on page REPAIR-9)
RMAP 0320 Step 014 on page REPAIR-9)
RMAP 0330 Step 006 on page REPAIR-11)
RMAP 0330 Step 010 on page REPAIR-12)
RMAP 0340 Step 003 on page REPAIR-13)
RMAP 0340 Step 003 on page REPAIR-13)
1.
Find the card or cards you want to replace in
the charts on page REPAIR-4.
2.
Use the Repair Affects column to determine
what portion of the machine or string is
affected by replacement of the card.
Does the Repair Affects column indicate Controller?
Yes No
I
002
Refer to the Vary Offline column to
determine which devices the customer must
vary offline. If not already done, ask the
customer to vary offline each of the affected
devices. Devices should be varied offline on
each attached system.
c;
Place the affected devices in CE mode. Use
MD Main Menu Option O. Refer to
"Set/Reset CE Mode (MD Main Menu Option
0)" on page MD-7.
After the device or devices are placed in CE
mode, go to RMAP 0310 Step 001 on page
REPAIR-5.
003
Determine if the falling controller can be quiesced.
It is easier to Quiesce a controller than it is to have
the customer vary the failing path offline from each
attached system, but there are some limitations:
I •
I
I
I
Quiesce controller can be used on both 4-way
and 2-way subsystems. On a 2-way subsystem
the path must be varied offline by the customer
before the Quiesce Controller is executed.
I •
I
Quiesce path or controller cannot be used
when attached to a 3880 storage control.
c/
Quiesce is a function performed at the Storage
Control support facility keyboard and does not
involve system operators.
()
(Step 003 continues)
Ci
REPAIR-2
PIN 4519901 30 Sides EC 476573 3ONoYB8
© Copyright IBM Corp. 1987, 1988
C
o
003 (continued)
Can the failing controller be qulesced?
Yes No
I
(
004
Ask the customer to vary offline the affected
path on each attached system.
Go to RMAP 0310 Step 001 on page
REPAIR-5.
(
005
To locate the attached storage control and
determine the storage path, refer to the SSIO and
Storage Path labels on either the A-Unit front
operator panel or the rear Unit Emergency Power
Off panel.
To Quiesce a controller, use instructions in the
storage control maintenance package. Refer to
MLX Exit 9 on page MLX-3.
f
After the controller is quiesced, go to RMAP 0310
Step 001 on page REPAIR-5.
f
(
PIN 4519901 30 Sides EC 476573 3ONov88
@ Copyright IBM Corp. 1987, 1988
REPAIR-3
A 1 Board (Controller) Cards
Controller
Logic Card Name
Controller
A1(A3)
Controller
A2(A4)
DDC/DTB
I/O Control (IOC)
A1W2
A1V2
CDPSSO
CDPSS1 (Note 1)
Repair Affects
Vary Offline or Quiesce
A1B2
A1C2
Controller
Controller
Affected storage path
A1U2
A1T2
A102
A1E2
Controller
Controller
Affected storage path
DPS (Note 3)
A1S2
A1F2
String
Controller Sequencer
A1R2
A1G2
Controller
All devices in the string
(Devices cannot be quiesced)
Affected storage path
DHPLO
ClocklSERDES/ECC 1
SERDES 2 (Note 2)
MD Adapter
A1Q2
A1N2
A1P2
A1M4
A1H2
A1J2
A1K2
A1L4
Controller
Controller
Controller
Controller
Affected storage path
Notes:
c~
1.
Controllers in 2-path strings do not have CDPSS 1 cards.
2.
Late level machines do not have SERDES 2 cards.
3.
Strings attached to a 3990 storage control do not have DPS cards.
B1 and Cx Board (Device) Cards
Device
Logic Card Name
Left
Front
Right
Front
Left
Rear
Right
Rear
Repair Affects
Vary Offline
PES/PROM
Seq/Servo/RPS
Port/R-W
B102
B1E2
B1F2
B1K2
B1J2
B1H2
B1N2
B1P2
B1Q2
B1U2
B1T2
B1S2
Device
Device
Device
Affected devices
R/W Channel Board
Servo Power Amp
COB2
C4P1
C1B2
C4P2
C2B2
C4P3
C3B2
C4P4
Device
Device
Affected Devices
Power
B1G2
B1G2
B1R2
B1R2
Drive
Both devices in drive
B1W2
B1W3
B1B2
B1B3
B1W2
B1W3
B1B2
B183
B1W2
B1W3
B1B2
8183
B1W2
B1W3
B1B2
81B3
All devices
in the unit
All devices in the unit
A1
A2
A3
A4
Port
Port
Port
Port
Term
Term
Term
Term
i·
c
REPAIR .. 4
PIN 4519901 30 Sides EC 476573 3ONov88
@ Copyright IBM Corp. 1987. 1988
(
RMAP 0310: Replacing FRUs
001
(.
(
-
4.
Attach the ESO wrist strap to your wrist and
ground it to the machine frame of logic gate.
5.
Remove the card being replaced.
B.
Set the switches on the new card if required.
See note in additional action column of charts.
7.
Power On. Use the Power On procedure as
indicated in the Power Off/On Procedure
column of Figure 2 on page REPAIR-B.
Locate the unit in which FRUs will be replaced.
The unit can be determined as follows:
8.
Perform the Post Repair Checkout.
•
The string can be determined by the String
10 on the rear Unit Emergency Power Off
panel.
Go to RMAP 0320 Step 001 on page REPAIR-8.
•
When replacing controller logic cards, the
controller board (A 1/A2 or A3/ A4) can be
determined by the Controller 10 labels on
the rear Unit Emergency Power Off panel
and on the A-Gate logic card cover label.
•
When replacing device logic cards, the unit
containing the devic~ can be determined by
the device physical address labels on the
Single Actuator Maintenance (SAM) Control
Panel.
(From
(From
(From
(From
(From
(From
(From
RMAP
RMAP
RMAP
RMAP
RMAP
RMAP
RMAP
0300
0300
0300
0320
0320
0330
0330
Step
Step
Step
Step
Step
Step
Step
002
004
005
011
015
007'
011
on
on
on
on
on
on
on
page
page
page
page
page
page
page
REPAIR-2)
REPAIR-~)
REPAIR-3)
REPAIR-g)
REPAIR-10)
REPAIR-11)
REPAIR-12)
Follow this procedure when replacing FRUs:
1.
(
For additional information about locating the
failing string and device, refer to SMAP 0030
Step 001 on page START-14 and SMAP 0045
Step 001 on page START-21.
(
2.
Power off the affected portion of the string.
Use the Power Off procedure for the card you
are replacing, as indicated in the Power Off/On
Procedure column of Figure 2 on
page REPAIR-5.
3.
Read any additional action items.
(
The Additional Action column refers you to
important information related to the cards you
are replacing.
PIN 4519901 30 Sides EC 476573 3ONov88
© Copyright
(
IBM Corp. 1987. 1988
REPAIR-S
(~"'\
,
\
'~
A 1 Board (Controller) Cards
Controller
A1(A3)
Controller
A2(A4)
Power Off
Power Off/On
Procedure
on page
DDC/DTB
A1W2
A1B2
Controller
REPAIR-27
110 Control (l0C)
A1V2
A1C2
Controller
REPAIR-27
CDPSSO
CDPSS1 (Note 3)
A1U2
A1T2
A1D2
A1E2
Controller
Controller
REPAIR-27
REPAIR-27
DPS (Note 5)
Controller Sequencer
A1S2
A1R2
A1F2
A1G2
String
Controller
REPAIR-25
REPAIR-27
DHPLO
ClocklSERDES/ECC
SERDES 2 (Note 4)
MD Adapter
A1Q2
A1N2
A1P2
A1M4
A1H2
A1J2
A1K2
A1L4
Controller
Controller
Controller
Controller
REPAIR-25
REPAIR-27
REPAIR-25
REPAIR-27
Controller
Logic Card Name
Additional Action
Set switches (see
page REPAIR-18).
Set switches (see
page REPAIR-20).
See Note 1. Warning
See Note 2. Warning
Notes:
1.
Warning: Always power off the string (both controllers) before replacing a DPS card, to ensure that both
DPS card arrays will be correctly initialized. If one DPS card has power on and the storage array on that
card contains data that is not correct, the array on the card being replaced will be loaded with the data
that is not correct when the new DPS card is powered on.
If the problem Is Intermittent or you cannot determine which card Is faulty, replace both DPS cards. Do
not swap DPS cards.
2.
Warning: Do not swap MD Adapter cards. These cards are not the same. Swapping these cards will
result in the MD addressing the wrong storage path or storage director.
3.
Controllers in 2-path strings do not have CDPSS 1 cards.
4.
Late level machines do not have SERDES 2 cards.
5.
Strings attached to a 3990 storage control do not have DPS cards.
REPAIR-6
(
PIN 4519901 30 Sides EC 476573 3ONov88
© Copyright
IBM Corp. 1987, 1988
--"
,
c
c
(
(
B1 and Cx Board (Device) Cards
Power Off
Power OfflOn
Procedure
on page
Additional Action
81U2
Device
REPAIR-29
See Warning
81P2
8102
C282
81T2
81S2
C382
Device
Device
Device
REPAIR-29
REPAIR-29
REPAIR-29
C4P2
C4P3
C4P4
Device
REPAIR-29
81G2
81G2
81R2
81R2
Drive
REPAIR-28
81W2
81W3
8182
8183
81W2
81W3
8182
8183
81W2
81W3
8182
8183
81W2
81W3
8182
8183
Unit
Unit
Unit
Unit
REPAIR-26
REPAIR-26
REPAIR-26
REPAIR-26
Device
Logic Card Name
Left
Front
Right
Front
Left
Rear
Right
Rear
PES/PROM
8102
81K2
81N2
Seq/Servo/RPS
Port/R-W
R/W Channel 80ard
81E2
81F2
C082
81J2
81H2
C182
Servo Power Amp
C4P1
Power
A1
A2
A3
A4
Port
Port
Port
Port
Term
Term
Term
Term
MIM, Volume R10
Page CARR-1,
Entry N
Warning: Do not swap PES/PROM cards. A Defective PES/PROM card can cause data to be written off
track.
(
f
(""
c
PIN 4519901 30 Sides EC 476573 3ONovB8
o Copyright IBM Corp. 1987, 1988
REPAIR-7
RMAP 0320: Post-Repair Checkout
001
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
sense byte record to that system.
IMAP 0212 Step 004 on page ISOLATE-7)
IMAP 0228 Step 004 on page ISOLATE-15)
IMAP 0233 Step 010 on page ISOLATE-22)
IMAP 0245 Step 008 on page ISOLATE-27)
IMAP 0245 Step 010 on page ISOLATE-28)
IMAP 0260 Step 002 on page ISOLATE-35)
Page REPAIR-1)
RMAP 0310 Step 001 on page REPAIR-5)
Page REPAIR-25)
Page REPAIR-26)
Page REPAIR-27)
Page REPAIR-28)
Page REPAIR-29)
When a device is taken out of CE mode, an
interrupt is generated to each using system and
normal operation of the device should be resumed.
Is the MD already attached to the A unit, powered
on, and ready for use?
Yes No
I
004
Attach and prepare the MD.
Warning: Never power the MD on or off with
the diskette inserted.
To completely test the machine the normal
sequence of diagnostics must be ruh on one or
more devices, even if the problem that was
repaired was in the controller.
Ensure that the attached storage control you
are going to use is powered on, has
completed its initial microcode load (lML),
and that the functional diskette for the
3380-JK is installed.
I
002
Go to the Post-Repair Controller Checkout
procedure (RMAP 0330 Step 001 on page
REPAIR-11).
::;
See page MD-5 for directions.
Can the devices you want to test be taken from
customer operation?
Yes No
The repaired problem was in the controller
or controller path and device diagnostics can
not be run.
"'--.
Continue with Step 005 in this RMAP.
'",-
005
";
(From Step 004 in this RMAP)
If not already done, use MD Main menu option 0 to
put the devices to be tested Into CE mode.
See page MD-7 for directions.
003
Vary devices offline.
Old CE mode set OK for all the devices specified?
Yes No
I
Ask the customer to vary offline each of the
devices you want to test. Devices should be varied
offline on each attached system.
If the devices are not varied offline successfully,
avoid using those devices or proceed only with
customer permission.
006
Go to IMAP 0225 Step 001 on page
ISOLATE-14.
007
Run the Normal Sequence diagnostics:
,~,--- ......
i
""
Advise the customer that devices to be tested must
be placed in CE mode for the tests to run. If a
system attempts to use a device that has been
placed in CE mode, the storage control responds
with unit check and sends an Intervention Required
REPAIR-8
1.
2.
Press the PF key to return to the MD Main
menu.
From the Main menu, select option 1
(RUN DIAGNOSTICS).
PIN 4519901 30 Sides EC 476573 3ONov88
© Copyright IBM Corp, 1997, 1999
j
(
(-3.
From the Diagnostics menu, select option 0
(NORMAL SEQUENCE).
If possible, use both controllers in the A unit.
See page MD-11 for directions.
(
«
Did the diagnostics run to normal end without an
error?
Yes No
I
008
(
011
(From Step 009 in this RMAP)
Some configurations do not use the DPS cards.
The DPS cards are used when the string is
attached to a 3880, but they are not used when the
string is attached to a 3990.
For DPS card replacement information, refer to
RMAP 0310 Step 001 on page REPAIR-5.
Did you replace a DPS card?
Yes No
I
Go to Step 010 in this RMAP.
012
Go to Step 015 in this RMAP.
009
(
Go to Step 011 in this RMAP.
013
010
(From Step 008 in this RMAP)
Press the ENTER key to analyze the error and to
obtain a FRU list.
(
(
(
c
Answer questions and follow instructions given by
the MD. If you are asked "DO YOU WANT TO RUN
CONNECTION TESTS?" you are advised to answer
"YES." The connection tests provide better
isolation.
The connection tests do not run inline with
customer operation. Before running the connection
tests, all paths to the failing controller should be
varied omine. If this cannot be done, answer "NO"
to "DO YOU WANT TO RUN CONNECTION
TESTS?" . When a FRU group is displayed, use
the guided FRU replacement procedures to replace
the indicated FRUs.
Go to RMAP 0300 Step 001 on page REPAIR-2.
If the problem is not corrected, perform the
additional action indicated by the MD.
If the additional action has already been performed
or if there is no additional action indicated, go to
RMAP 0340 Step 001 on page REPAIR-13.
To completely test the DPS function run the
extended DPS tests. The extended DPS tests
check the correct interaction of DPS on both
controllers (A1 and A2).
Run the Extended DPS tests:
1.
2.
3.
Press the PF key to return to the MD Main
menu.
From the Main menu, select option 1
(RUN DIAGNOSTICS).
From the Diagnostic menu, select option 2
(EXTENDED DPS TESTS).
See page MD-14 for directions.
Old the extended DPS tests run to normal end
without an error?
Yes No
I
014
Press the ENTER key to analyze the error
and to obtain a FRU list.
When a FRU group is displayed, use the
guided FRU replacement procedures in the
REPAIR pages to replace the indicated FRUs.
Go to RMAP 0300 Step 001 on page
REPAIR-2.
If the problem is not corrected, perform the
PIN 4519901 30 Sides EC 476573 3ONov88
© Copyright IBM Corp. 1987, 1988
REPAIR-9
additional action indicated by the MD.
If the additional action has already been
performed or if there is no additional action
indicated, go to RMAP 0340 Step 001 on
page REPAIR-13.
017 (continued)
When you are sure that the switches are set
correctly, go to RMAP 0350 Step 001 on page
REPAIR-15 to end the call.
015
(From Step 012 in this RMAP)
For DDC/DTB card and I/O Control card
replacement information, refer to RMAP 0310 Step
001 on page REPAIR-5.
Did you replace a DDC/DTB card or an I/O Control
card?
Yes No
I
016
Go to RMAP 0350 Step 001 on page
REPAIR-15.
i
,,-.j
017
Run the Device Status test to check the switch
settings on the replaced card.
~/
--.
r
'.,.j
Refer to switch setting procedures:
DDCIDTB Card (A1W2 and A1B2)
For 2-path switch settings, see "DDC/DTB
Card (A1W2 and A1B2) Switch Settings
(2-Path)" on page REPAIR-18.
For 4-path switch settings, see "DDC/DTB
Card (A1W2 and A1B2) Switch Settings
(4-Path)" on page REPAIR-19.
I/O Control Card (A1V2 and A1C2)
For string 10 switch settings, see "I/O Control
Card (A1V2 and A1C2) Switch Settings (String
10)" on page REPAIR-20.
For 2-path switch settings, see "110 Control
Card (A1V2 and A1C2) Switch Settings (2-Path
String Configuration)" on page REPAIR-21.
For 4-path switch settings, see "I/O Control
Card (A 1V2 and A 1C2) Switch Settings (4-Path
String Configuration)" on page REPAIR-22.
c
(Step 017 continues)
REPAIR-10
PIN 4519901 30 Sides EC 476573 3ONov66
© Copyright
IBM Corp. 1987, 1988
c
(
C
(
RMAP 0330: Post-Repair Controller Checkout
001
005
(From RMAP 0320 Step 002 on page REPAIR-8)
Go to Step 007 in this RMAP.
Is the MD attached to the A unit, powered on, and
ready for use?
Yes No
006
(From Step 004 in this RMAP)
I
002
Attach and prepare the MD.
Warning: Never power the MD on or off with
the diskette Inserted.
See page MD-5 for directions.
Ensure that the attached storage control you
are going to use is powered on and has
completed its initial microcode load (IML),
and that the functional diskette for the
3380-JK is installed.
Continue with Step 003 in this RMAP.
003
(From Step 002 in this RMAP)
Run the controller tests.
(
c
The controller tests can be run in-line with
customer operations on the controller. Therefore,
controller paths and devices do not need to be
varied off line.
1.
2.
(
c
«
(
c
From the MD Main Menu, select option 1
(RUN DIAGNOSTICS).
From the Diagnostic Menu, select option 1
(CONTROLLER TESTS).
See page MD-13 for directions.
Did the controller diagnostics run to normal end
without an error?
Yes No
I
004
Go to Step 006 in this RMAP.
Answer questions and follow instructions given by
the MD. If you are asked "DO YOU WANT TO RUN
CONNECTION TESTS?" you are advised to answer
"YES." The connection tests provide better
isolation.
The connection tests do not run in line with
customer operation. Before running the connection
tests, all paths to the failing controller should be
varied offline. If this cannot be done, answer "NO"
to "DO YOU WANT TO RUN CONNECTION TESTS?"
When a FRU group is displayed, use the guided
FRU replacement procedures in the REPAIR section
to replace the indicated FRUs.
Go to RMAP 0300 Step 001 on page REPAIR-2.
If the problem is not corrected, perform the
additional action indicated by the MD.
If the additional action has already been performed
or if there is no additional action indicated, go to
RMAP 0340 Step 001 on page REPAIR-13.
007
(From Step 005 in this RMAP)
Some configurations do not use the DPS cards.
The DPS cards are used when the string is
attached to a 3880, but they are not used when the
string is attached to a 3990.
For DPS card replacement information, refer to
RMAP 0310 Step 001 on page REPAIR-5.
(Step 007 continues)
(Step 005 continues)
PIN 4519901 30 Sides EC 476573 3ONov88
@ Copyright IBM Corp" 1987. 1988
("
Press the ENTER key to analyze the error and to
obtain a FRU list.
REPAIR-11
007 (continued)
Did you replace a DPS card?
Yes No
I
008
Go to Step 011 in this RMAP.
009
To completely test the DPS function, run the
Extended DPS tests. The Extended DPS tests
check the correct interaction of DPS on both
controllers (A1 and A2).
Run the Extended DPS tests:
1.
2.
3.
Press the PF key to return to the MD Main
menu.
From the Main menu, select option 1
(RUN DIAGNOSTICS).
From the Diagnostic menu, select option 2
(EXTENDED DPS TESTS).
See page MD-14 for directions.
Did the Extended DPS tests run to normal end
without an error?
Yes No
I
010
Press the ENTER key to analyze the error
and to obtain a FRU list.
When a FRU group is displayed, use the
guided FRU replacement procedures in the
REPAIR section to replace the indicated
FRUs.
Go to RMAP 0300 Step 001 on page
REPAIR-2.
If the problem is not corrected, perform the
additional action indicated by the MD.
If the additional action has already been
performed or if there is no additional action
indicated, go to RMAP 0350 Step 001 on
page REPAIR-15.
011 (continued)
(From Step 008 in this RMAP)
For DDC/DTB card and //0 Control card
replacement information, refer to RMAP 0310 Step
001 on page REPAIR-5.
Did you replace a DDC/DTB card or an I/O Control
card?
Yes No
I
012
Go to RMAP 0350 Step 001 on page
REPAIR-15.
013
Run the Device status test to check the switch
settings on the replaced card.
Refer to switch setting procedures:
DDC/DTB Card (A1W2 and A1B2)
For 2-path switch settings, see "DDC/DTB
Card (A1W2 and A1B2) Switch Settings
(2-Path)" on page REPAIR-18.
For 4-path switch settings, see "DDC/DTB
Card (A 1W2 and A 1B2) Switch Settings
(4-Path)" on page REPAIR-19.
-
./
I/O Control Card (A1V2 and A1C2)
For string 10 switch settings, see "//0 Control
Card (A1V2 and A1C2) Switch Settings (String
10)" on page REPAIR-20.
For 2-path switch settings, see "I/O Control
Card (A 1V2 and A 1C2) Switch Settings (2-Path
String Configuration)" on page REPAIR-21.
For 4-path switch settings, see "I/O Control
Card (A1V2 and A1C2) Switch Settings (4-Path
String Configuration)" on page REPAIR-22.
!f ""\
~-./
When you are sure that the switches are set
correctly. go to RMAP 0350 Step 001 on page
REPAIR-15 to end the call.
011
(Step 011 continues)
REPAIR-12
'......
PIN 4519901 30 Sides EC 476573 3ONov88
© Copyright IBM Corp. 1987, 1988
(
(
(
(
RMAP 0340: End of Guided Maintenance
001
(From
(From
(From
(From
(From
(From
IMAP 0260 Step 016 on page ISOLATE-3?)
IMAP 0295 Step 003 on page ISOLATE-55)
Page REPAIR-1)
RMAP 0320 Step 010 on page REPAIR-9)
RMAP 0320 Step 014 on page REPAIR-10)
RMAP 0330 Step 006 on page REPAIR-11)
Is the machine still failing?
Yes No
Go to .. Analyze Symptom Code (Main Menu
Option 3)" on page MD-40.
•
Go to IMAP 0240 Step 001 on page ISOLATE-24.
•
I
c
c
002
Go to RMAP 0350 Step 001 on page
REPAIR-15.
003
Perform a top-card-connector check.
•
If a device problem seems to be indicated,
reJJlace all 81-board cards (for the failing
device) that were not previously replaced.
Go to RMAP 0300 Step 001 on page REPAIR-2.
•
Check the operator panel switches.
If a controller problem seems to be indicated,
replace all A 1-board cards (for the failing
controller) that were not previously replaced.
Ensure that the switches on the operator panel are
.
set to Enable.
Go to RMAP 0300 Step 001 on page REPAIR-2.
Perform voltage and ripple checks.
If not already done, perform voltage and ripple
checks on the affected controller and devices. See
page PWR-1, Entry A, in the MIM, Volume R10 to
perform voltage checks.
•
I •
I
I
I
Order the A 1 and/or 81 boards.
if EREP or sense data is available, check for
the FSCs in the following list. If you have any
of the FSCs listed, go to page MD-1, Entry C,
Option A before continuing:
If not already done, consider the following:
9F9F
•
EI08
El80 }
4x2x
4x6x
4xAx
(
If sense data is available, use MD option 3 to
analyze the highest priority symptom code.
> With failures on heads
4, 5, 6, and/or 7
Go to "Analyze Symptom Code (Main Menu
Option 3)" on page MD-40.
•
(
If the MD asked that you define a failure
boundary, you may have selected the wrong
boundary. If you have not followed the 'define
failure boundary' procedures closely, perform
the analysis again.
Go to MD 0050 Step 001 on page MD-53.
Check the affected A 1 and/or 81 board top card
connectors to ensure that they are all in place and
seated correctly. See the decal on the gate
cover(s) for correct locations.
(
If sense data is not available, perform error log
analysis.
If sense data is available, the highest priority
symptom code has been analyzed, and there
are additional symptom codes, use MD option 3
to analyze the next highest priority symptom
code.
PIN 4519901 30 Sides EC 476573 3ONov88
@ Copyright IBM Corp. 1987, 1988
Call your next level of support for assistance.
When contacting your next level of support, it is
important that the following information be readily
available.
REPAIR-13
•
Machine type and model
•
Machine serial number
'.
System type/serial number
•
Severity status
•
A telephone number where you can be reached
at the account
•
Customer problem indications (symptom codes,
message numbers, visual symptoms),
•
Parts replaced and actions previously taken
•
Sense data obtained from LOGREC printouts,
ICKDSF printouts, and console messages,
•
The type of storage control unit (3880 or 3990)
and its functional microcode including EC level.
•
Physical address of the failing device(s),
•
The physical configuration of the subsystem,
including
Whether it is a 2-path or 4-path
configuration,
Whether there are one or two strings on the
CTL-I interface,
•
If diagnostic errors occurred while running the
MD, record the IC codes and the E-bytes,
•
Field manager's name and office phone
number.
You have reached the end of the guided
maintenance procedures.
To continue with unguided problem analysis, refer
to page HELP-1 in the ECM, Volume R40. The ECM
is shipped in microfiche form with each A unit. The
microfiche is located in the MOM, Volume R30.
I"r~'
<~ .~/
/f ~
'"./
REPAIR-14
PIN 4519901 30 Sides EC 476573 3ONov88
© Copyright IBM Corp. 1987, 1988
tr"-
~_/
(
RMAP 0350: End of Call
001
(
(:
(
(
(
(
(
(
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
(From
IMAP 0240 Step 008 on page ISOLATE-26)
IMAP 0245 Step 022 on page ISOLATE-29)
IMAP 0260 Step 011 on page ISOLA TE-36)
IMAP 0265 Step 004 on page ISOLA TE-38)
IMAP 0265 Step 011 on page ISOLATE-39)
IMAP 0270 Step 004 on page ISOLATE-42)
IMAP 0280 Step 001 on page ISOLATE-49)
IMAP 0280 Step 006 on page ISOLATE-50)
IMAP 0285 Step 006 on page ISOLATE-52)
IMAP 0290 Step 009 on page ISOLATE-54)
IMAP 0290 Step 010 on page ISOLATE-54)
Page REPAIR-1)
Page REPAIR-1)
RMAP 0320 Step 016 on page REPAIR-10)
RMAP 0320 Step 017 on page REPAIR-10)
RMAP 0330 Step 010 on page REPAIR-12)
RMAP 0330 Step 012 on page REPAIR-12)
RMAP 0330 Step 013 on page REPAIR-12)
RMAP 0340 Step 002 on page REPAIR-13)
Page MD-1)
Page MD-6)
Page MD-35)
2. Select a Controller.
A controller must be selected to perform the
following procedures. Anyone of the
controllers in the string may be used to start
the end of call procedure. However, if you
have been working on a particular controller,
that controller should be selected first. The
procedure will have you select other
controllers as required.
After a controller is selected, the status of the
following conditions is displayed:
•
Devices in CE Mode
•
Write Inhibit Conditions
Write inhibit is a condition set in the
Storage Control for a storage path. It is
set when an error occurred on the path
during a write operation. It is set to
prevent a defective storage path from
destroying customer data. There are two
write inhibit conditions:
The end of call procedure prepares a subsystem
for system operation after failing conditions have
been corrected.
c
The End of Call procedure uses the End of Call
option on the MD. The End of Call option provides
for displaying and resetting the following
conditions:
•
Devices in CE Mode.
•
Write inhibit conditions.
(-
•
Fenced conditions (3990 Storage Control only).
(
Use the following procedure to prepare the
subsystem for customer use:
(
1. Select End of Call.
From the MD Main Menu, select option Z (End
of Call). The MD will prompt you for the
following steps:
(
(/
( ,-.
-"
(
Controller Write Inhibit is set for a
controller if a controller error or
storage path error is detected during
a write operation.
PIN 4519901 30 Sides EC 476573 3ONov88
~ Copyright IBM Corp. 1987, 1988
Channel Write Inhibit is set for a
channel if a channel or channel
related storage path error is detected
during a write operation.
Note: When a write inhibit condition is
set in a 4-path subsystem, it is set for both
storage paths in a cluster.
•
Fenced Conditions (3990 Storage Control
only)
Fencing is a technique used by the 3990
storage path functional microcode to
remove a failing path from system
operation. This is done to prevent
path-related failures from degrading
subsystem performance any more than
necessary. There are three types of
REPAIR-15
fencing:
storage paths in the storage path cluster.
Channel Fenced
Set when there are excessive channel
or channel related failures. Channel
fenced conditions are not displayed with
storage control status.
5. Reset Write Inhibit and Fenced Conditions for the
other controller In the A unit.
The MD asks if you want to test the other
controller for Write Inhibit and Fenced
conditions.
Storage Path Fenced
Set when there are excessive storage
path errors or when a Quiesce path is
set from the Storage Control Support
Facility. Storage path fencing is used on
4-path subsystems only.
If Write Inhibit and Fenced conditions are
detected when testing the other controller, the
MD asks if you want to reset them. These
conditions should be reset.
Controller Fenced
Set when the Quiesce Controller is used
from the Storage Control Support
Facility. Controller Quiesce can be used
on 2-path or 4-path systems.
Device Fenced
Set for a device when there are
excessive controller errors on a single
path to that device. The device remains
operational on other paths. Device
fencing is used on 4-path subsystems
only.
3. Reset CE Mode.
If devices are in CE mode, the MD asks if you
want to reset CE Mode for these devices.
CE Mode must be reset before returning
devices to normal operation. Regardless of
which controller was selected in step 2, CE
mode is reset for all devices and attached
storage paths.
4. Reset Write Inhibit and Fenced Conditions.
If Write Inhibit or Fenced conditions were
detected, the MD asks if you want to reset
them. These conditions should be reset.
The MD resets Write Inhibit and Devices
Fenced conditions in the storage director or
storage path attached to the controller
selected in step 2.
Note: When a Write Inhibit condition is reset
in a 4-path subsystem, it is reset for both
REPAIR-16
6. Test the other A unit in the string.
The MD asks you to move the MD cable to
the other A unit. Next, repeat steps 2 through
5 to reset all Write Inhibit and Fenced
conditions.
Note: This step is for 4-path strings only.
7. Check the Air Filter.
Check the primary air filter located on the
actuator power supply. Replace if necessary.
/'
8. Check all gate fans in the unit.
If a gate fan is not turning, repair or replace
as necessary, See page PWR-1, Entry N, in
the MIM, Volume R 10.
9. Advise the customer to vary the failing paths and
devices online.
Note: Resetting Write Inhibit and Fenced
conditions will not cause the failing path to
automatically come online. The customer
must vary the path online.
10. If necessary, reinitialize the volume.
If maintenance was done because at three or
more track addresses data was read
successfully with offset invoked (sense byte 7
= 48 through 4F or sense byte 7 = 58
through 58), the customer must dump the
data and initialize the volume at the medial
level before the machine is returned to
operation.
To do this, use the Device Support Facilities
INITIALIZE command with the VALIDATE and
NOCHECK parameters specified. See the
PIN 4519901 30 Sides EC 476573 3ONov88
@ Copyright IBM Corp. 19B7. 19BB
rf'
~
(
Device Support Facilities Users Guide and
Reference, GC35-0033.
(
(
f
(
(
(
(
(
(
=0 \
". ,/
PIN 4519901 30 Sides EC 476573 3ONov88
© Copyright
IBM Corp. 1997, 1989
REPAIR-17
(
o
DDC/DlB Card (A1W2 and A1B2) Switch Settings (2-Path)
Note:
For 4-path switch settings, see page REPAIR-19.
Both controllers in the string must be set with the same string address.
String Address
Set the string address switch (switch 1) on the new
card to the same settings as on the card being
replaced. Set switches 2, 3, and 4 to zero (0). If
you are not sure of the correct switch settings,
refer to the string address on the operator panel
and the following table.
After setting the string address, verify that the
string address you've set (either 0 or 1) is the
same as the string address shown on the rear Unit
Emergency Power Off panel (for an example, see
Figure 7 on page INTRO-10 (3880 subsystem) or
Figure 16 on page INTRO-24 (3990 subsystem».
,/
!
,
Note: The addressing on all the A 1W2 and A 182 cards for the string must be set the SAME. All
controllers are set to either 0 or 1. Each device must have the same controller-device address
regardless of the selection path to the device.
2-Path
String
Address
Range
OO-OF
10-lF
Top
String
String Address
Switch 1
0
closed
open
1
20-2F
0
30-3F
40-4F
1
0
50-SF
1
60-6F
0
70-7F
1
80-8F
0
90-9F
1
closed
open
AO-AF
BO-BF
0
1
closed
open
CO-CF
0
DO-DF
EO-EF
1
0
closed
open
FO-FF
1
closed
open
closed
open
closed
open
closed
open
closed
open
II
II
II
II
I
2
3
4
Switch I in this
example is set
for string 1.
Switches 2, 3, and 4
are not used (see note),
part side
of card
Bottom
Note: Switches 2, 3, and 4 are actually not used. However, you are advised to set them to the closed
position to eliminate any confusion as to their correct setting.
t"~
~,~ .
'4.../
(f'
\~]
REPAIR-18
PIN 4519901 30 Sides EC 476573 30Nov88
@ Copyright IBM Corp. 1987, 1988
c
DDCIDTB Card (A1W2 and A1B2) Switch Settings (4-Path)
Note:
(
(
For 2-path switch settings, see page REPAIR-18.
Both controllers in the string must be set for the same string address.
String Address
Set the string address switch (switch 1) on the new
card to the same settings as on the card being
replaced. Set switches 2, 3, and 4 to zero (0). If
you are not sure of the correct switch settings,
refer to the string address on the operator panel
and the following table.
After setting the string address, verify that the
string address you've set (either 0 or 1) is the
same as the string address shown on the rear Unit
Emergency Power Off panel (for an example, see
Figure 16 on page INTRO-24)
Note: The addressing on all the A 1W2 and A 1B2 cards for the string must be set the SAME. All
controllers are set to either 0 or 1. Each device must have the same controller-device address
regardless of the selection path to the device.
(
(
(
(
4 Path
string
Address
Range
0O-lF
Top
String
0
20-3F
1
40-5F
0
60-7F
I
80-9F
0
AO-BF
1
CO-DF
0
EO-FF
I
closed
open
closed
open
closed
open
closed
open
II
II
II
II
I
2
3
4
Switch 1 in this
example is set
for string 1.
Switches 2. 3. and 4
are not used (see note).
-- part side of card
(
(
String
Address
Switch I
closed
open
BoUom
Note: Switches 2, 3, and 4 are not used. However, you are advised to set them to the closed position
to eliminate any confusion as to their correct setting.
(
(
(
(0
PIN 4519901 30 Sides EC 476573 3ONov88
@ Copyright IBM Corp. 1987, 1988
«
REPAIR-19
110 Control Card (AiV2 and AiC2) Switch Settings (String 10)
The string 10 switches set the string 10 that is reported to the system in sense data. The switches for each
controller in the string must be set the same. The switches should be set to the value indicated on the
operator panel. DO NOT USE 00 or FF for a string 10.
String ID
("
to ::',
\.....f
Use the following table to set the string 10 switches
on the 110 Control card (A 1V2, A 1C2) in each
controller.
After setting the string 10, verify that the string ID
you've set is the same as the string 10 shown on
the rear Unit Emergency Power Off panel (for an
example, see Figure 7 on page INTRO-10 (3880
subsystem) or Figure 16 on page INTRO-24 (3990
subsystem)).
Top
Closed
Switch
Binary I
~
Open
1
II
I :
I :
I :
I :
I :
I :
I :
Switches I through 8:
Closed = I
Open
= 0
I
Switch 10 is not used: Set to 0 (open).
3
4
5
II
II
In this example, switches I through 8
are set to 0111 0011 = X'73'.
Switch 9 (single HDA 3380 Model CJ2)
must be set to 0 (open) for 3380-AJ4
and 3380-AK4 strings.
2
II
II
II
6
7
II
II
8
II
II
9
10
Bottom
part side
of card
,r' -,
(~j
REPAIR·20
PIN 4519901 30 Sides EC 476573 3ONov88
© Copyright
IBM Corp, 1987, 1988
1/0 Control Card (A1 V2 and A1 C2) Switch Settings (2-Path String Configuration)
Note:
(
(
For 4-path string configuration switch settings, see page REPAIR-22.
The string-configuration switches describe the configuration of the string to the storage control microcode.
These switches must be set the same for both controllers in the string.
String Configuration: Switches 1 and 2
Use the following table to set configuration
switches 1 and 2 on the 110 Control Card (A1V2,
A1C2) in each controller.
Strin~ Confi~uration
(
- Pat )
A unit only
(
(
A unit and one B unit
Switches
I
2
Closed
Open
A unit and three B units
Open
Closed
Open
Switches 3 and 4 must be set to "Closed".
Open
II
II
Closed Closed
Closed Open
A unit and two B units
(
Top
II
II
I
2
3
4
Switches I and 2 in
this example are set
for an A unit and
three B units.
Switches 3 and 4 are
not used.
Part side of card
Bottom
(
c
(
(
PIN 4519901 30 Sides EC 476573 3ONov88
«:l Copyright IBM Corp. 1987, 1988
(
REPAIR-21
(
\
\~)
1/0 Control Card (A1V2 and A1C2) Switch Settings (4-Path String Configuration)
Note:
For 2-path string configuration switch settings, see page REPAIR-21.
These switches must be set the same for each controller in the string.
String Configuration: Switches 1 and 2
Use the following table to set configuration
switches 1 and 2 on the 110 Control Card (A1V2,
A1C2) for device address range OO-OF.
Top
Closed
String Confi~uration
(4 - Pat )
Device Addresses 00 - OF
A unit only
A unit and one B unit
A unit and two B units
A unit and three B units
Switches
1
II
II
II
II
2
Closed Closed
Closed Open
Open
Open
Closed
Open
Open
1
2
3
4
Switches 1 and 2 are
set for an A unit
and three B units.
Switches 3 and 4 are
set for a second A unit
with three B units.
Bottom
-- Part side of card
/
\.
".
/
string Configuration: Switches 3 and 4
Use the following table to set configuration
switches 3 and 4 on the 110 Control card (A1V2,
A1C2) for device address range 10-1F.
String Configuration
(4 - Path)
Device Addresses 10 - IF
One A unit only
A unit and two B units
A unit and two B units
A unit and three B units
Switches
3
4
Closed Closed
Closed Open
Open
Open
Closed
Open
(: " .
./
REPAIR-22
PIN 4519901 30 Sides EC 476573 3ONov88
© Copyright IBM Corp. 1987. 1988
c-
(
Power OnlOff Procedures
For example, if you plan to replace the PES/PROM
card located at 81N2, always switch off SW69S and
CP623 through CP626. to power off the" Left Rear"
device.
Warning: Never switch off the CPs for both
actuators on the same HDA without powering off
the drive first. Machine damage will result. See
"Drive Power Off Procedure" on page REPAIR·28.
(
(
(
( -
See Figure 2 on page REPAIR-6 and Figure 2 on
page REPAIR-? to determine the affected portion of
the string (what to power off) before replacing a
specific FRU. Next, see Figure 3 on
page REPAIR-24 and Figure 2 to locate the
switches used to power off the affected portion of
the string.
Warning: Never switch off the CPs (CP603 through
CP636) for more than two actuators (devices)
without powering off the entire unit. Powering off
more than two devices may cause the voltages on
the remaining device to exceed specification which
could cause damage to the device or its associated
circuits.
Warning: If the HDA cables are swapped, all of the
logic cards that are normally used with the left
actuator in the HDA are used with the right
actuator, and the logic cards that are normally used
with the right actuator are used with the left
actuator. The Single Actuator Maintenance (SAM)
panel switches and CPs control power to the same
logic cards whether or not the cables are swapped.
The MD always calls out the appropriate logic cards
for replacement, even if the cables are swapped.
The same CPs and switches must be used to power
on or to power off specific logic cards whether or
not the cables are swapped.
Warning: The device address associated with an
actuator remains the same whether or not the
cables are swapped. While the cables are
swapped, power any specific device address off
and on with the SAM CPs and switches that
normally control the other actuator in the same
HDA.
24 Vdc Panel for A and B units
(
~
(
d) d) d) d)-Mmdl ,
CP40q.
CP405
• CP406
BLOWER
2q V
2q V
• CONTAINED IN THE A UNIT ONLY
Figure 2.
CPq07
CPQ08
FRT
RR
• 5WQ86
REMOTE
d):!f.d) ,
• cpQ 10
[NTl 1
• [pQ 1 1
[NTl 2
• 5WQ87
~
REMOTE
IA508~LL
24 Vdc Panel - A and B Units
(
{'\
(
(
PIN 4519901 30 Sides EC 476573 3ONov88
©
Copyright IBM Corp. 1987, 1988
REPAIR-23
Single Actuator Maintenance Control Panel for A and B Units
FRONT DRIVE
.J675
J676
.J677
(j
,
REAR DRIVE
.J678
FRONT
• •• .J679 MOTOR
REAR
.J681
MOTOR.J68 •••
·:::.. [J• rBl
fD1 CJ ...
l§J=OFF-=l§J • :::
J682
J683
J68~
"'-"'.
ON,
•••
SW691
SW692
··• ....•• .• .• .• ·• ..•• ·• ..••
·..... ••• • •• • ••
CP602
o
(1'622
J685
J686
J687
+5
+5
0
FRONT SW691J CP613
RIGHT
ACTUATOR
~:~
I:J
LOGIC
··• ....•• -··.... -··• ....•• -··• ....••
·.. · ·..
CI'611J
n;r-) n;r-)
'1:!1-'
REAR SW695 CP623
•
roo--
J689
J688
I
",,-j
•••
~
CP62~
J690
J691
r'
J692
(1'605
(1'606
~
~
(1'61',)
(1'616
~
~
(1'625
(1'626
~
~
(1'635
CP636
~
~
\
LEFT
ACTUATOR
::~
CI n;r-)
-w---- raJ
~
LOGI(
REAR SW696 CP63J
RIGHT
ACTUATOR
::~
(P6JIJ
bI raJ
raJ
"'"'""W-' ~
LOGI(
ISOOO12SQL
Figure 3.
Single Actuator Maintenance Control Panel
REPAIR-24
PIN 4519901 30 Sides EC 476573 3ONov88
© Copyright
IBM Corp. 1987, 1988
c
c
«
(
Warning: Never power off the MD with the
diskette inserted.
After the drives have powered on, the
Sequencer Status LED (on the B1-board) for
each actuator begins to flash. Wait for 3
minutes or until the LED for the device
becomes solid On before running diagnostics
on the device.
2.
Set CP410 and CP411, located on the 24 Vdc
panel in each A unit, to off (see Figure 2 on
page REPAIR-23).
If the string has been powered down less than
10 minutes, the 3 minute waiting period may be
bypassed by using the following procedure.
3.
Wait until all the drives on the string have
stopped.
a.
After the drives have powered on and the
Sequencer Status LED (on the B1-board) for
each actuator begins to flash, operate the
Logic Enable/Disable Switch for each
actuator, first to Disable, then back to
Enable.
b.
When the LED on the B1-board for the
actuator becomes solid On, diagnostics
may be run on the device.
c.
If the LED does not become solid On, wait 1
minute before running diagnostics.
String Power Off Procedure
1.
(
4.
Unload the MD diskette and turn off the MD.
If any of the drives do not stop, go to page
PWR-1, Entry H, in the MIM, Volume R10.
String Power On Procedure
1.
2.
Turn on the MD and load the MD diskette.
Warning: Do not return a string to the
customer until all drives have been running for
at least 3 minutes.
Warning: Never power on the MD with the
diskette Inserted.
(
3.
(
Set CP410 and CP411, located on the 24 Vdc
panel, to On (see Figure 2 on
page REPAIR-23).
Wait for all the drives on the string to power'
on.
5.
If any of the drives do not power on, run
diagnostics on the failing drive(s). Use the
procedure on page REPAIR-B.
After FRU replacement, run the diagnostics to
verify the repair and to ensure that there are
no new problems.
6.
Go to RMAP 0320 Step 001 on page REPAIR-B.
(
(
PIN 4519901 30 Sides EC 476573 3ONov88
© Copyright IBM Corp. 1987, 1988
(
REPAIR-25
If either drive does not power on, run
diagnostics on the failing drive. Use the
procedure on page REPAIR-B.
Unit Power Off Procedure
This procedure powers off all drives in the unit, but
does not power off the entire unit.
7.
1.
Set both motor switches (SW691 and SW692) on
the Single Actuator Maintenance (SAM) control
panel to Off (see Figure 3 on page REPAIR-24).
2.
Wait until both drives in the unit have stopped.
If the unit has been powered off less than 10
minutes, the 3 minute waiting period may be
bypassed by using the following procedure:
If either drive does not stop, go to page PWR-1,
Entry H, in the MIM, Volume R10.
3.
After the drives have powered on, the
Sequencer Status LED (on the B1-board) for
each actuator begins to flash. Wait for 3
minutes or until the LED for the device
becomes solid ON before running diagnostics.
a.
Set CP405 (on the 24 Vdc panel) to Off (see
Figure 2 on page REPAIR-23).
Unit Power On Procedure
1.
Set CP405 located on the 24 Vdc panel to On
(see Figure 2 on page REPAIR-23).
2.
Wait five seconds.
3.
Set the motor switch, SW691, on the Single
Actuator Maintenance control panel to On (see
Figure 3 on page REPAIR-24).
4.
Wait 25 seconds.
5.
Set the motor switch, SW692, on the Single
Actuator Maintenance control panel to On.
6.
b.
When the LED on the 81-board for the
actuator becomes solid On, diagnostics
may be run on that device.
c.
If the LED does not become solid On, wait 1
minute before running diagnostics.
/
'\
Warning: Do not return a unit to the customer
until both drives have been running for at least
3 minutes.
S.
After each FRU replacement, run the
diagnostics to verify the repair and to ensure
that there are no new problems.
9.
Go to RMAP 0320 Step 001 on page REPAIR-S.\..,/
Wait for the drives to power on.
REP"IR-26
After the drives have powered on and the
Sequencer Status LED (on the B1-board) for
each actuator begins to nash, operate the
Logic Enable/Disable Switch for each
actuator, first to Disable, then back to
Enable.
/'
PIN 4519901 30 Sides EC 476573 3ONoV88
© Copyright IBM Corp. 1987, 1988
c
(
(
Controller Power Off Procedure
Controller Power On Procedure
Warning: Setting both controller power switches to
OFF will power off all drives in the string.
1.
Set CP410 (controller A1!A3) or CP411
(controller A2!A4) located on the 24 Vdc panel
(on the 3380-AJ4 or AK4 unit) to ON (see
Figure 2 on page REPAIR-23).
2.
After FRU replacement, run the diagnostics to
verify the repair and to ensure that no new
problems exist.
3.
Go to RMAP 0320 Step 001 on page REPAIR·8.
1.
(
Set CP410 (controller A1!A3) or CP411
(controller A2! A4) located on the 24 Vdc panel
(on the 3380-AJ4 or AK4 unit) to off (see
Figure 2 on page REPAIR-23).
f
(
{
c
(
(
(
PIN 4519901 30 Sides EC 476573 3ONov88
© Copyright IBM Corp. 1987. 1988
REPAIR-27
u
c·.··
./
Drive Power Off Procedure
Warning: When powering off a drive, NEVER power
off the· CPs for either of the associated actuators
before turning off the Motor Switch. Machine
damage will result.
Warning: The drive cannot power off when either
of the logic switches associated with the motor Is
in the Disable position.
1.
Set the Front or Rear Motor switch (SW691 or
SW692) to Off and wait for the motor to stop
(refer to Figure 3 on page REPAIR-24).
Set the Front (CP602) or Rear (CP622)
to On.
5.
Wait five seconds.
6.
Set the Motor switch (SW691 or SW692) to On.
7.
Wait for the drive to power on.
If the drive does not power on, run diagnostics
on the drive. Use the procedure on page
REPAIR-8.
8.
If the drive motor does not stop, go to page
PWR-1, Entry H, in the MIM, Volume RIO.
2.
Set the Front (CP602) or Rear (CP622)
to Off.
+ 5V CP
4.
C~
After the drive has powered on, the Sequencer
Status LED (on the B1-board) for each actuator
begins to flash. Wait for 3 minutes or until the
LED for the device becomes solid ON before
running diagnostics.
+ 5V CP
3.
Set the two Logic switches for the front or rear
drive to Disable.
4.
Set the actuator CPs to Off (left to right) for
both actuators associated with the drive.
5.
Set the front motor CP (CP407) or rear motor
CP (CP408) to Off. These CPs are on the 24
Vdc Panel (shown in Figure 2 on
page REPAIR-23).
If the drive has been powered down less than
10 minutes, the 3 minute waiting period can be
bypassed by doing the following:
Drive Power On Procedure
1.
Set the motor CP (CP407 or CP408) located on
the 24 Vdc Panel to On (see Figure 2 on
page REPAIR-23).
2.
Set the actuator CPs to On (right to left) for
both actuators.
3.
Set the two Logic switches to Enable.
a.
After the Sequencer Status LED for each
actuator begins to flash, operate the Logic
Enable/Disable Switch for each actuator,
first to Disable, then back to Enable.
b.
When the LED on the B1-board for the
actuator becomes solid ON, diagnostics can
be run.
c.
If the LED does not become solid ON, wait 1
minute before running diagnostics.
Warning: Do not return a unit to the customer
until both drives have been running for at least
3 minutes.
9.
After FRU replacement, run the diagnostics to
verify the repair and to ensure that no new
problems exist.
10. Go to RMAP 0320 Step 001 on page REPAIR-8.
()
REPAIR-28
PIN 4519901 30 Sides EC 476573 3ONov88
© Copyright IBM Corp. 1987, 1988
c
(
Device Power Off Procedure
Device Power On Procedure
1.
Set the correct actuator Logic switch (SW693.
SW694. SW695 or SW696) to Disable. See
Figure 3 on page REPAIR-24.
1.
Set the actuator CPs located on the Single
Actuator Maintenance (SAM) Control panel to
On (right to left). See Figure 3 on
page REPAIR-24.
2.
Set the associated actuator CPs to Off (left to
right).
2.
Set the actuator logic switch to Enable.
3.
After FRU replacement, run the diagnostics to
verify the repair and to ensure that no new
problems exist.
4.
Go to RMAP 0320 Step 001 on page REPAIR-8.
(
(-
(
(
((
(
f_
(
(
PIN 4519901 30 Sides EC 476573 30Nov88
10 Copyright
IBM Corp. 1987, 1988
REPAIR-29
EC History of Repair Section
EC HISTORY OF P/N 4519901
EC Number
Date Of EC
EC Number
Date Of EC
475245
14Nov86
475248
25Apr88
475246
2IJu187
476573
30Nov88
475247
llSep87
Notes:
;1'
i
"'- ,~ .
C'
:i
REPAIR-30
.,:
PIN 4519901 30 Sides EC 476573 3ONovB8
<0 Copyright
IBM Corp. 1987, 1988
C)
{"
Maintenance Device (MD) Operation
(
The MD section contains instructions for attaching
and using the IBM maintenance device (MD) and
references that explain messages from the MD.
When you attach the MD to the 3380 to isolate a
problem, you should always start at page START-1
in this manual and follow the instructions on that
page. Refer to "MD Menus" on page MD-2 for a
diagram of the MD menus and options.
Use the following Entry Table to quickly find the
information in this section.
c
(
Entry
Select one of the following:
Go to page
A
Prepare the MD
MD-5
B
Set/reset CE mode (MD Main menu option 0)
MD-7
C
Run diagnostics (MD Main menu option 1)
o - Normal seQuence
1 - Controller test
2 - Extended DPS test
3 - Device test
4 - Connection test
5 - HA Scan
6 - Motor test
7 - Device status
8 - CE track repair
9 - Storage Control Status
A - Non-repeatable Runout Test
MD-10
MD-11
MD-13
MD-14
MD-15
MD-17
MD-19
MD-24
MD-26
MD-32
MD-34
MD-37
0
Error log function (MD Main menu option 2)
LOG-1
F
Analyze symptom code (MD Main menu
option 3)
MD-42
G
Review Option List (MD Main menu option R)
o - Review CE mode
1 - Review Diagnostics
2 - Review Error Log
3 - Review Repair
4 - Review HA Scan
MD-49
MD-49
MD-50
MD-50
MD-50
MD-51
H
Use diagnostic aids (MD Main menu option D)
MD-44
J
End of call (MD Main menu option Z)
RMAP 0350 Step 001
on page REPAIR-15
(
(
(
(
(
(:
c
PIN 4519903 84 Sides
EC 476581 01AugS9
\0 Copyright IBM Corp. 1987. 1988
MD-1
MD Menus
MAIN MENU
r--DIAGNOSTIC MENU
o - SET/RESET CE MODE
I 0 - NORMAL SEQUENCE (80-85)
1 - RUN DIAGNOSTICS--------J 1 - CONTROLLER TEST (80)
2 - ERROR LOG FUNCTION-----, 2 - EXTENDED DPS TEST (90)
3 - ANALYZE SYMPTOM CODE
I 3 - DEVICE TEST (81-85)
Z - END OF CALL
I 4 - CONNECTION TEST
R - REVIEW --------------, I 5 - HA SCAN (91)
D - DIAGNOSTIC AIDS----, I I 6 - MOTOR TEST (99)
I I I 7 - DEVICE STATUS (80,96)
I I I 8 - CE TRACK REPAIR (83,9C)
I I I 9 - STORAGE CONTROL STATUS (9F)
I I I A - NRRO TEST
J
r---------------------- CTLR/DPS
80
81
82
83
84
85
8A
90
9C
PIN 4519903 84 Sides
© Copyright
-
DEVICE LOGIC
SERVO
R/W DATA TRANSFER
R/W FUNCTION
SEEK EXERCISER
INCREMENTAL SEEK
EXTENDED DPS
CE TRACK REPAIR
EO 476581 01Aug89
IBM Corp. 1987, 1988
,
,
"""-
I I
I L_-ERROR LOG FUNCTION
I
1 - TRANSFER LOG
I
2 - SEARCH LOG
I
3 - MERGE LOG
I
4 - CHANGE LOG MASK
I
5 - ERASE LOG
I
I
L----REVIEW
o - DEVICES IN CE MODE
1 - LAST 4 DIAGNOSTIC ERRORS
2 - ERROR LOG
3 - LAE".:' REPAIR
4 - HA SCAN
MO-2
:/
(-
Maintenance Device (MD) Display - Reference Number Table
(
When an MD display refers to this page, find the reference number in the
following table. Go to the r~ferenced page for instructions and information.
(
(
(
(
c
REF
Title of the MD Frame
See Page
DOS
010
020
030
03S
050
051
052
090
092
093
NRRO test
Select An Option
Select controller
Set/reset CE mode
Devices in CE Mode/Default Devices
Define the failure boundary
Define the failure boundary (EBXX)
Define the failure boundary (Data Checks)
Write Inhibit
Storage Path Fenced
Fenced Devices
MD-37
MD-S2
MD-52
MD-S3
MD-54
MD-S5
MD-58
MD-61
MD-63
MD-63
MD-64
100
Diagnostic menu
MD-64
200
20S
210
215
Select controller
Invalid controller selected
Select device(s)
Invalid device selected
MD-65
MD-65
MD-65
MD-65
310
320
330
340
350
360
395
Do not forget to reset CE mode
Do you want to run connection tests?
Which controller connection do you want to test?
Controller AX connection tests ready
DPS tests cannot be run
HA Scan
Scan Complete/Scan Review
MD-65
MD-66
MD-66
MD-67
MD-67
MD-67
MD-70
400
405
410
415
420
425
430
440
460
462
465
468
480
Select one of the following error log functions
Select controller (Error Log)
Enter search arguments
Enter log number
Enter the date
Enter logs to be merged
Enter the sense formats to be logged
Error Log Mask/Transfer stopped by keyboard entry
Configuration
Configuration Data/Unknown Storage CTRL
Device status
Device type
Storage Control Status
MD-71
MD-71
MD-72
MD-72
MD-72
MD-72
MD-73
MD-73
MD-73
MD-75
MD-76
,v1D-77
MD-78
500
550
Isolation code error display
Enter the symptom code
MD-78
MD-79
605
615
Warning: HDA cables swapped
MD does not provide analysis for displayed IC
MD-79
MD-81
700
742
Select review option from fonowing list
Press Enter to analyze, or "N" to display next error
MD-81
MD-81
800
Diagnostic aids
MD-82
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987. 19S5
MD-3
REF
Title of the MD Frame
See Page
900
Control unit and MD disks are not compatible
MD-83
PIN 4519903 84 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1997, 1999
MD-4
(
f
'"
••
(
Prepare the MD
When preparing to use the IBM maintenance
device (MD) to service the 3380-JK, use the
following procedure.
1.
4.
GO TO
PSG MD-1
FOR OPERATIONAL
I NSTRUCTI ONS
Attach the MD at the rear of the A unit.
The signal cable socket is located at the right
edge of the A unit. The power convenience
outlet is located in the lower left corner of the
A unit.
(
(2.
(
Ensure that the storage control is powered on
and has completed its initial microcode load
(IML).
Note: When you attach the MD to the 3380
to isolate a problem, you should always start
at page START-1 in this manual and follow
the instructions on that page.
5.
Verify that the functional diskette for the
3380-JK is installed.
f
(
The MD will attempt to gather configuration
information about the subsystem. If the MD is
connected to the controller and no error
occurs, the next display will be the MD Main
menu (see step 7 on page MD-6).
Power on the MD.
Take the MD diskette from the holder inside the
right rear door of the A unit and insert the
diskette into the MD. Press the IML
pushbutton.
(
(
(
If an error occurs while trying to get
configuration information, the following frame is
displayed:
The MD performs a check of its functions. (This
check takes only a few seconds.) After the
checking process, the MD loads information
from the diskette and displays the following
frame:
AN ERROR OCCURRED
WHILE TRYING TO GET
CONFIGURA TI ON
INFORMATION
**DISK LOADED**
3380-JK
PN xxxxxxx
EC xxxxxx
Continue with the next step.
6.
(
The frame shows the machine type and the EC
number of the diskette. Always check this
frame to ensure that you are using the correct
3380-JK diskette.
PIN 451990384 Sides
© Copyright
(
EC 476581 01Aug89
IBM Corp. 1987, 19BB
Press the ENTER key to advance to the next
frame:
CONFIGURATION
INFORMATION IS BEING
GATHERED.
PLEASE WAIT.
Warning: Never power the MD on or off with the
diskette inserted.
3.
Press the ENTER key to advance to the next
frame:
Press the ENTER key to advance to the next
frame:
SELECT THE TYPE OF
DASD STRING:
1. 3380 AJ4/AK4
2. 3380 CJ2
MD-S
7.
You can return to the first frame of the MD
Main menu by pressing the PF key at almost
any time.
Select option 1 and press the ENTER key to
advance to the next frame:
SELECT THE TYPE OF
CONTROL UNIT:
1. 3880
2. 3990
The options available from the MD Main menu
are: ,
If the reply to this request is "1," then the MO
displays the MO Main menu.
Option
Meaning
0
Set or reset CE mode for selected
devices. CE mode must be set prior to
running device diagnostics.
See page MD-7.
If the reply to this request is "2," then the
following frame is displayed:
ARE THE CONTROLLERS
CABLED AS A 4 PATH
SUBSYSTEM?
1
, After the yes/no reply is entered, the MO
displays the MO Main menu.
*** MAIN
MENU ***
(SEE
PSG MD-3,
010)
O-SET/RESET CE MODE
I-RUN DIAGNOSTICS ...
2
Perform various functions on the error
log contained in the storage control.
See page LOG-1.
3
Accept input of symptom codes and
sense data for analysis. The MD need
not be attached to the 3380.
See page MD-42.
The MO Main menu identifies the major
functions that can be performed with the MD.
Resets CE mode and refers to the
end-of-call procedure on RMAP .0350
Step 001 on page REPAIR-15.
The second line of the MD Main Menu frame
means that entry 010 in the table on page MD-3
in this manual identifies reference material to
explain the frame.
R
The ellipsis (three dots ... ) in the lower right
corner of the frame indicates message
continuation in successive frames.
Review status from previous
operations with the MD.
See page MD-49.
0
Allow manual control of the
diagnostics. See page MD-44.
Press the ENTER key to advance to the
remaining frames of the MO Main menu:
2-ERROR LOG FUNCTION
3-ANALYZE SYMPTOM
CODE
Z-END OF CALL
/~
,
/"
i
'-.
"
"-
/,"-',
\ ""'.,7
Z
8.
Run diagnostics on selected
controllers and devices. This option
selects the diagnostic menu.
See page MO-10.
I
\..c/!
9.
Select an option from the MD Main menu by
keying the option number or letter and then
pressing the ENTER key on the MD.
1/"
'"
\'-<. .. /'
0~''''
e-,
"-_::/'
R-REVIEW (0-4)
D-DIAGNOSTIC AIDS
C
C
PIN 451990384 Sides
© Copyright
MD-6
EC 476581 01Aug89
IBM Corp, 1987, 19B8
0~'
!"i"
~/
--
-------
------
-----
(
(
(
(
Set/Reset CE Mode (MD Main Menu Option 0)
Before diagnostics can be run on a device, the
device must be in CE mode. The device
diagnostics do not run if the device is not in CE
mode.
If the devices to be tested are not varied offline
successfully, only set CE mode with customer
permission. Explain to the customer that when CE
mode for a device is reset, an interrupt is
generated to each using system and normal
operations should be resumed.
(
(If you want to review the devices that are
already in CE mode without performing a set or
reset operation, you should use MD Main menu
option R. See page MD-49.)
4.
Enter either a 1 or a 2 to select an option. from
this frame. Regardless of your selection (Set
or Reset), one of the following frames is
displayed:
SELECT CONTROLLER:
(SEE PSG MD-3,020)
-
The Set/Reset CE Mode option uses a diagnostic
program (in the storage control) to set CE mode
bits in the DPS arrays for the devices specified.
Each device in CE mode is unavailable to all
attached systems.
1.
(
(
Select option 1 to set speCified devices in CE
mode or select option 2 to reset CE mode for
specified devices.
Before a device is returned to the customer for
normal use, CE mode must be reset for that device.
Ensure that devices are varied offline before they
are set into CE mode. When a device is varied
offline, it should be varied offline to all attaching
systems. If a system attempts to use a device that
is in CE mode, the storage control responds with
unit check and indicates INTERVENTION REQUIRED in
the sense data.
(
OPTION:
1. SET CE MODE
2. RESET CE MODE
S~LECT
2.
Ask the customer to Vary offline each of the
devices you want to test or put in CE mode.
Devices should be varied offline on each
attached system. If the desired devices are not
varied offline successfully, proceed with
customer permission.
If necessary, press the ENTER key to advance
to the MD Main menu or use the PF key to
return to the MD Main menu.
The PF key always causes the MD to return to
the MD Main menu.
3.
While the MD Main Menu is displayed, select
option 0: Set/Reset CE Mode (key in 0 and
press the ENTER key.) The MD displays the
following frame:
PIN 4519903 84 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1987, 1988
or-
SELECT CONTROLLER:
DEFAULT IS: x
(SEE PSG MD-3,020)
The "x" indicates the default controller. The
default controller is the controller selected
during a previous operation. If none was used
no default is displayed.
Regardless of which controller is used, the CE
mode function (set or reset) applies to all paths
(controllers). The valid responses are:
Response
ENTER
1
2
3
4
Meaning
Default (if shown)
Controller A 1
Controller A2
Controller A3
Controller A4
MD-7
Note: Selecting controller A3 or A4 is valid
only on strings with four storage paths, and
only when the MD is attached to the A unit that
contains controllers A3 and A4. Controllers A 1
and A2 cannot be selected when the MD is
attached to the A unit that contains controllers
A3 and A4.
5.
Select an operational controller to use for
setting or resetting CE mode. Depending on
your mode selection (Set or Reset), one of the
following frames is displayed.
Setting CE Mode
ENTER 2-DIGIT ADDR.
OF DEVICECS) TO BE
SET IN CE MODE.
(SEE PSG MD-3, 030)
-
or-
Resetting CE Mode
ENTER 2-DIGIT DEVICE
ADDRESSES OR PRESS
ENTER TO RESET ALL.
(SEE PSG MD-3, 030)
CE mode may be set or reset for one or more
devices in a string. There are several ways to
enter device addresses. Two digits must be
used for each address entered.
•
Enter a single device address, "00" through
"1F. "
•
Enter multiple device addresses separated
with a single space (for example, 02 05 07
OA).
•
Enter a range of addresses separated with
the character "T" (for example, 02T07).
•
Enter multiple ranges of addresses,
separating each range with a space (for
example, 02T04 08TOB).
•
With "Reset CE Mode" selected, press the
ENTER key without specifying devices to
reset CE mode for all devices that were
previously set in CE mode.
PIN 4519903 84 Sides
EC 476581 01Aug89
@Copyright IBM Corp. 1987, 1988
6.
Specify the devices for which you want to
change CE mode. One of the following frames
is then displayed:
Setting CE Mode
CE MODE IS BEING
SET.
PLEASE WAIT.
-
or-
Resetting CE Mode
CE MODE IS BEING
"":SET •
PLEASE WAIT.
Immediately after the message is displayed, an
inline diagnostic routine is started using the
specified controller. CE mode is set or reset
for all specified devices.
After the inline diagnostic has completed
setting or resetting CE mode, one of the
following frames is displayed:
Setting CE Mode
DEVICES IN CE MODEl
OX: .123 ........... .
IX: ............... .
(SEE PSG MD-3, 035)
-
or-
Resetting CE Mode
CE MODE IS RESET
FOR ALL DEVICE
ADDRESSES.
In the preceding frame, devices 01, 02, and 03
are in CE mode. Devices 00 and 04 through OF
are represented by dots in the second line of
the display. Devices 10 through 1F are
displayed on the third line if the string has 4
paths.
MD-8
(
(
(
(
Some problems with the storage control, the
controller, or the control interface do not allow
the CE mode diagnostic to run and CE mode
can not be set. The MD displays an error
message similar to that shown below. If this
happens, analyze the error and correct the
problem before continuing.
IC**** CTL=A* DEV=**
MD DETECTED ERROR
EB=** **** **** ****
**** **** **** ****
f
PRESS ENTER TO
ANALYZE OR PF KEY
FOR MAIN MENU.
(SEE PSG MD-3. 500)
(
(
(
(
(
(
PIN 4519903 84 Sides
ro Copyright IBM Corp.
EC 476581 01Aug89
1987, 1988
MD-9
Run Diagnostic Tests (MD Main Menu Option 1)
1.
6
Press the PF key to return to the MD Main
menu.
Run the motor test on the specified
drive. See page MD-24.
(''\
,
2.
Select option 1 (Run Diagnostics) and advance
to the MD Diagnostics menu:
7
*(SEE
DIAGNOSTIC MENU *
PSG MD-3,IOO)
8
2-EXTENDED DPS TEST
3-DEVICE TEST
4-CONNECTION TEST
5-HA SCAN
6-MOTOR TEST
7-DEVICE STATUS
8-CE TRACK REPAIR
9-STORAGE CTL STATUS
9
o
Run controller and device
diagnostics on the specified
controllers and devices.
See page MD-11.
1
2
3
Run controller tests on the specified
controllers. See page MD-13.
Run the extended DPS tests on both
controllers. This option can not be
used when the string is attached to
a 3990. See page MD-14.
Locate and repair defects on the CE
cylinder and on the four diagnostic
cylinders. All home addresses on
these cylinders are rewritten.
See page MD-32.
--/
/ / ",\
/'
Collect the following status from the
Storage Control.
•
•
•
The diagnostic options and their meanings are:
Meaning
>~,-",
/-,
O-NORMAL SEQUENCE
I-CONTROLLER TEST ...
Option
" - >7
Test the controller and collect the
status of all devices in the string
without putting the devices in CE
mode. See page MD-26.
Devices in CE Mode
Device and Storage Path Fenced
conditions (3990).
Write Inhibit conditions
.....,
\~.
,7
"',
See page MD-34.
A
/"
Run the NRRO test on a specific
drive.
,/
<",'-. . /
See page MD-37.
3.
Ge ~o the referenced pages for information
about the diagnostic options.
\,~
,/'
Run device tests on the specified
devices. See page MD-15.
/'
""
',-j
4
Test for connection errors between
the MD, the specified controller and
the storage path or storage director.
See page MD-17.
5
Scan Home Address on the
specified device. See page MD-19.
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
MD-10
IBM Corp. 1987, 1988
I
,"'~
"'-> /
Diagnostic Menu Option 0 - Normal Sequence
(
c·
The normal sequence is a set of linked diagnostics
used to perform a machine checkout. The
controller and device tests in the following list are
run in sequence.
80
81
82
83
84
85
-
(
One or more controllers may be specified. If
multiple controllers are specified the normal
sequence is run on one controller and on each
device specified. An abbreviated normal sequence
is then run with the other controllers and on each
device specified. The abbreviated normal
sequence consists of diagnostics 80, 81, and 83.
Diagnostics 82, 84 and 85 are excluded. If it is
necessary to run the complete normal sequence
with multiple controllers, each controller must be
selected individually.
(
From the Diagnostic Menu, select option O.
Controlier/DPS
Device Logic
Servo
Read/Write Data Transfer
Read/Write Function
Seek Exerciser
The normal sequence runs on controllers and
devices as specified by the user.
(
1.
The MD displays one of the following frames:
•
•
•
•
•
•
(
f
To run the Normal Sequence diagnostics:
The devices to be tested must be in CE mode . .The
normal sequence runs inline while customer
operation continues on devices not specified for
testing. See the DIAG section in the MSM, Volume
R60, for information about the diagnostics.
(
SELECT CONTROLLER:
(SEE PSG MD-3, 200)
-
or-
SELECT CONTROLLER:
DEFAULT IS: x x
(SEE PSG MD-3, 200)
The "x"s shown in the preceding frame identify
the last controller or controllers used to run
diagnostics in this session. If none was used,
no default is displayed.
The valid responses are:
Response
Meaning
ENTER
Default (if shown)
Controller A 1
Controller A2
Controller A3
Controller A4
Controllers A1 and A2
Controllers A3 and A4
Controllers A 1, A2, A3 and A4
1
2
3
4
12
34
1234
Note: Selecting controller A3 or A4 is
valid only on strings with four storage
paths. The MD will instruct you when it
is necessary to move the MD plug.
("
(
(
c
PIN 4519903 84 Sides
© Copyright
EC 476581 01AugB9
IBM Corp. 1987. 1988
MO-11
2.
•
Enter your response for controller use.
Press the ENTER key without specifying
devices to accept the default devices.
The MD displays one of the following frames:
Devices to be tested must be in CE mode. If
any device specified for testing is not in CE
mode, an error message is displayed.
SELECT DEVS. (OO-OF)
(SEE PSG MD-3,210)
DEFAULT IS:
-
3.
or-
SELECT DEVS. (OO-lF)
(SEE PSG MD-3.210)
DEFAULT IS:
Specify the devices to be tested.
When the ENTER key is pressed, the
diagnostics start. The diagnostics continue
until:
Press the ENTER key to display the default
devices (if any), as shown in one of the
following frames:
DEFAULT DEVICES
a.
An error is detected, or
b.
All specified controllers in the A unit to
which the MD is attached, and all specified
devices have been tested.
ox: .123 ........... .
When controllers in both A units of a 4-path.
string are speCified, the MD stops after all
devices have been tested using the controllers
in the A-unit attached to the MD. The MD
instructs you to move the MD connector to
continue testing with the controllers in the
adjacent A-unit. Do not power off the MD when
moving the MD connector.
(SEE PSG MD-3,03S)
-
or-
DEFAULT DEVICES
OXI .123 ........... .
IX:
.•.......••...•.
(SEE PSG MD-3,03S)
Default devices (01, 02, and 03) are shown as
an example. Default devices are those that
were specified during the previous diagnostic.
If no device was tested during the current
diagnostic session, no default is displayed.
4.
If an error is detected, go to "Isolation Code
Error Display" on page MD-41.
5.
If no errors are detected, the MD displays the
following frame:
There are several ways to enter device
addresses. Two digits must be used for each
address entered.
•
Enter a single device address, "00" through
"1F."
•
Enter multiple device addresses separated
with a single space (for example, 02 05 07
OA).
•
Enter a range of addresses separated with
the character "T" (for example, 02T07).
•
Enter multiple ranges of addresses
separated with a space (for example, 02T04
DaTOS).
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
END OF NORMAL TEST
SEQUENCE. DO YOU
WANT TO RUN THE
SAME TESTS AGAIN!
6.
Respond to the question by pressing the YES
key or the NO key.
7.
Press the PF key to return to the MD Main
menu.
MD-12
f
Diagnostic Menu Option 1 - Controller Test
f
t·
The controller test (diagnostic routine 80) checks
the basic controller and DPS operation. This test
runs inline with customer operation. (When
attached to a 3990 storage control, the DPS portion
of the test is bypassed.) Devices need not be in CE
mode.
Note: Selecting controller A3 or A4 is
valid only on strings with four storage
paths. The MD will instruct you when it
is necessary to move the MD plug.
2.
To run the Controller Test:
1.
Enter your response for controller use.
From the Diagnostic menu, select option 1.
When the ENTER key is pressed, the
diagnostics start. The diagnostics continue
until:
The MD displays one of the following frames:
a.
An error is detected, or
b.
The speCified controllers have been tested
in the A unit attached to the MD.
SELECT CONTROLLER:
(SEE PSG MD-3, 200)
(
-
If controllers in both A units of a 4-path string
are specified, the MD stops after the controllers
in the A unit attached to the MD have been
tested. The MD instructs you to move the MD
connector to continue testing the controllers in
the adjacent A unit. Do not power off the MD
when moving the MD connector.
or-
SELECT CONTROLLER:
DEFAULT IS: x x
(SEE PSG MD-3, 200)
(
(
(
The "x"s identify the last controller or
controllers used to run rJiagnostics in this
session. If none was used, no default is
displayed.
3.
If an error is detected, go to "Isolation Code
Error Display" on page MD-41.
Valid responses are:
4.
If no errors are detected, the MD
following frame:
Response
Meaning
ENTER
1
2
3
4
Default (if shown)
Controller A 1
Controller A2
Controller A3
Controller A4
Controllers A 1 and A2
Controllers A3 and A4
Controllers A 1, A2, A3 and A4
12
34
1234
(
dispJay~
the
END OF CONTROLLER
TESTS. DO YOU WANT
TO RUN THE SAME
TESTS AGAIN?
5.
Respond to the question by pressing the YES
key or the NO key.
6.
Press the PF key to return to the MD Main
menu.
(.
(
{;
/
(~
(
PIN 451990384 Sides
EC 476581 01Aug89
Cl Copyright IBM Corp. 1987, 1988
MD-13
Diagnostic Menu Option 2 - Extended DPS Test
The Extended DPS test (diagnostic routine 90)
provides a complete checkout of the DPS
hardware. This test checks the interaction between
the two controllers. (The Extended DPS tests
cannot be run on a 3380-JK attached to a 3990.)
To run the Extended DPS Test:
1.
The test runs inline with customer operation.
Devices need not be in CE mode.
From the Diagnostic menu, select option 2.
When the ENTER key is pressed, the
diagnostics start running.
~
The MD displays the following frame:
DPS TESTS ARE NOW
STARTED.
The diagnostics continue until normal end or
until an error is detected. If an error occ,urs, go
to "Isolation Code Error Display" on
page MD-41.
If no errors are detected, the MD displays the
following frame:
END OF DPS TESTS.
DO YOU WANT TO RUN
THE SAME TESTS
AGAIN!
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
2.
Respond to the question by pressing the YES
key or the NO key.
3.
Press the PF key to return to the MD Main
menu.
MD-14
(
..
(--(
Diagnostic Menu Option 3 - Device Test
To run the Device Test:
The device test runs the following device
diagnostics in the sequence shown.
1.
From the Diagnostic menu, select option 3.
The MD displays one of the following frames:
((
•
•
•
•
•
81
82
83
84
85
-
Device Logic
Servo
Read/Write Data Transfer
Read/Write Function
Seek Exerciser
SELECT CONTROLLER:
(SEE PSG MD-3, 200)
-
SELECT CONTROLLER:
DEFAULT IS: x x
(SEE PSG MD-3, 200)
The controller test is not included.
(
The device test runs on one or more devices as
specified by the user.
(
One or more controllers may be specified for
running the device test. If multiple controllers are
specified, the device test is run with one controller
on each device specified. An abbreviated device
test is then run with the other controllers on each
device specified. The abbreviated device test
consists of diagnostics 81 and 83. Diagnostics 82,
84, and 85 are excluded. If it is necessary to run
the complete device test with other controllers,
each controller must be selected individually.
(:~
{-:
or-
The "x"s shown in the preceding frame identify
the last controller or controllers used to run
diagnostics in this session. If none was used,
no default is displayed.
The valid responses are:
Response
Meaning
ENTER
1
Default (if shown)
Controller A 1
Controller A2
Controller A3
Controller A4
Controllers A 1 and A2
Controllers A3 and A4
Controllers A1, A2, A3, and A4
~
The devices to be tested must be in CE mode. The
device tests run inline while customer operation
continues on devices not specified for testing. See
the DIAG section in the MSM, Volume R60, for
information about the diagnostics.
3
4
12
34
1 234
Note: Selecting controller A3 or A4 is
valid only on strings with four storage
paths. The MD will instruct you when it
is necessary to move the MD plug.
{'
Device tests will run using the specified
controller. The controller selected should be
operational. If you suspect that the controller
that you want to use is not operational, use
diagnostic option 0 (normal sequence) rather
than diagnostic option 3.
(\
("
{ --/
c
("
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. lSS7, lSSe
MD-15
2.
•
Enter your response about controller use.
Press the ENTER key without specifying
devices to accept the default devices.
The MD displays one of the following frames:
Devices to be tested must be in CE mode. If
any device specified for testing is not in CE
mode, an error message is displayed.
SELECT DEVS. COO-OF)
(SEE PSG MD-3.210)
DEFAULT IS:
-
3.
or-
SELECT DEVS.(OO-IF)
(SEE PSG MD-3.210)
DEFAULT IS:
When the ENTER key is pressed, the
diagnostics start. The diagnostics continue
until:
Press the ENTER key to display the default
devices (if any), as shown in one of the
following frames:
DEFAULT DEVICES
OX: .123 ........... .
(SEE PSG MD-3.03S)
-
DEFAULT DEVICES
OX: .123 ........... .
IX: ...•.........•••
(SEE PSG MD-3,03S)
•
Enter a single device address, "00" through
"1F."
Enter multiple device addresses separated
with a single space (for example, 02 05 07
OA).
•
Enter a range of addresses separated with
the character "r (for example, 02T07).
•
Enter multiple ranges of addresses,
separating each range with a space (for
example, 02T04 08TOS).
PIN 4519903 84 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1997, 1999
An error is detected.
b.
All devices have been tested with the
AJ4/AK4 unit attached to the MD.
4.
If an error is detected, go to "Isolation Code
Error Display" on page MD-41.
5.
If no errors are detected, the MD displays the
following frame:
There are several ways to enter device
addresses. Two digits must be used for each,
address entered.
.•
a.
If controllers in both A-units of a 4-path string
are specified, the MD stops after all devices
have been tested using the controllers in the
A-unit attached to the MD. The MD instructs
you to move the MD connector to continue
testing with the controllers in the adjacent
A-unit. Do not power off the MD when moving
the MD connector.
or-
Default devices (01, 02, and 03) are shown as
an example. Default devices are those that
were specified during the previous diagnostic.
If no device was tested during the current
diagnostic session, no default is displayed.
Specify the devices to be tested.
END OF DEVICE TESTS.
DO YOU WANT TO RUN
THE SAME TESTS
AGAIN!
6.
Respond to the question by pressing the YES
key or the NO key.
7.
Press the PF key to return to the MD Main
menu.
MD-16
«
(
Diagnostic Menu Option 4 - Connection Test
The connection test checks the connection between
the MD, the specified controller, and the storage
path or storage director. The connection test does
not run concurrently with customer operation on
the controller being tested. Before this option is
used, the path must be varied offline to all devices
from each attached system. Devices need not be
in CE mode.
To run the Connection Test:
1.
From the MD Diagnostic Menu, select option 4.
The MD displays the following frames:
******WARNING*******
CONNECTION TEST
CAN NOT RUN
CONCURRENTLY WITH ...
The MD connection test is run by the controller
sequencer (not the storage control). There are
three parts to the test.
(
(
(
(-'
•
•
•
CUSTOMER OPERATION
ON THE CONTROLLER
BEING TESTED.
MD to controller connection
Controller function
Storage control
To run the connection test. the MD will instruct you
to power the controller off and then on. During the
power-on sequence, the controller sequencer (if
operating correctly) establishes communication
with the MD and the connection test starts.
Warning: The connection test does not run
concurrently with customer operation on the
controller being tested.. Ensure that the controller
can be taken from the customer.
Warning: If any drives in the string are not
running, place the drive motor switches for those
drives to OFF because those drives might not
sequence on correctly.
SEE PSG MD-3.320 FOR
INSTRUCTIONS OR USE
THE PF KEY TO RETURN
TO THE MAIN MENU.
2.
Press the ENTER key to advance to the next
frame. The following prompt appears:
WHICH CONTROLLER
C~NNECTION DO YOU
WANT TO TEST?
(SEE PSG MD-3. 330)
The valid responses are:
Response
Meaning
1
2
3
Controller
Controller
Controller
Controller
4
A1
A2
A3
A4
Note: Selecting controller AS or A4 is
valid only on 'strings with four storage
paths. The MD will instruct you when it
is necessary to move the MD plug.
(
(/
('~
(':
("
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
MD-17
3.
Enter your response to Identify the controller to
be tested.
7.
Respond to the question by pressing the YES
key or the NO key.
The MD displays the following instructions:
8.
Press the PF key to return to the MD Main
menu.
CTLR x CONNECTION
TESTS ARE READY. DO
THE FOLLOWING (SEE
PSG MD-3, 340)
1. POWER OFF CTLR x
2. PRESS ENTER
3. WITHIN 5 SECONDS,
POWER ON cn R x.
4.
Follow the instructions on the MD.
a.
b.
Turn off the indicated controller. Use one
of the following CPs to power off and on:
•
Controller A1 or A3 - CP410 in the A
unit being tested.
•
Controller A2 or A4 - CP411 in the A
unit being tested.
Press the ENTER key. You must turn the
controller on within 5 seconds or the MD
will time out.
Press the ENTER key now to cause the MD
to start polling the controller sequencer.
c.
Turn on the indicated controller.
;'
I
When the controller is powered on, the
controller sequencer branches to a program
step that looks for the MD poll. If the controller
sequencer is started successfully and if the
sequencer sees the MD poll, communication is
established between the controller sequencer
and the MD. The connection test then starts.
5.
If an error is encountered, go to "Isolation
Code Error Display" on page MD-41.
6.
If no errors are detected, the following is
displayed:
END OF CONNECTION
TESTS. DO YOU WANT
TO RUN CONNECTION
TESTS AGAIN!
MD-18
PIN 451990384 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987. 1988
- - - - ...
-~
.
(
(
(
Diagnostic Menu Option 5 - HA Scan
HA Scan (diagnostic routine 91) reads all the home
addresses (HAs) on the specified range of cylinder
addresses. If data checks are detected, the error
information is saved until the scan is completed.
At the end of the scan, error data is summarized
and presented by the MD. The selected device
must previously be placed in CE mode. The HA
Scan diagnostic runs inline with customer
operation on devices not selected by the user.
To run the HA Scan test:
1.
From the MD Diagnostic Menu, select option 5.
The MD displays one of the following frames:
SELECT CONTROLLER:
(SEE PSG MD-3. 360)
-
(
or-
SELECT CONTROLLER:
DEFAULT IS: x
(SEE PSG MD-3. 360)
(
The x shown in the preceding frame is replaced
by 1, 2, 3,4 or a blank to identify the last
controller used to run diagnostics in this
session. If none was used, no default is
displayed.
(
(
The valid responses are:
(
Response
Meaning
ENTER
1
2
3
Default (if shown)
Controller A 1
Controller A2
Controller A3
Controller A4
4
Note: Options 3 and 4 are only valid on strings
with four storage paths when the MD is
attached to the unit that contains controllers A3
and A4.
2.
Enter your response for controller use.
The MD displays one of the following frames:
SELECT DEVICE:
(SEE PSG MD-3, 360)
-
SELECT DEVICE:
DEFAULT IS: xx
(SEE PSG MD-3. 360)
c'
('
or-
PIN 451990384 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
MD-19
!' "'
\, . /
Default device was specified during the
previous diagnostic. If no single device was
tested during the current diagnostic session, no
default is displayed.
SELECT TRACK TYPEI
1. CUSTOMER TRACKS
2. CE TRACKS
3. HA MAP TRACKS
Device addresses are in hexadecimal (00
through 1F). The response may have the
following forms:
The valid responses are:
Response
Meaning
Response
1
Customer tracks: Tracks used for
customer data.
Accepts the default device shown
for testing.
2
Specifies device OB for testing
(only one device can be tested at
a time).
CE tracks: Tracks on the CE
cylinder and on the four
diagnostic cylinders.
3
HA MAP tracks: Tracks that
contain factory-assigned skip
displacement information for the
volume.
ENTER
OB
Meaning
The device must be in CE mode.
3.
Enter your response to identify the device to be
used.
5.
The MD displays the following frame:
Automatic mode automatically continues after
errors until the scan is completed or until there
are 80 tracks in error. At the end of the scan,
error data is presented in either detail form or
summary form (at your option).
Manual mode stops on each error and displays
the error information. You can continue the
scan by pressing the ENTER key.
When an error is displayed in manual mode,
you have the opportunity to change the mode
to automatic mode. At the end of the scan,
error data is presented in either detail or
summary form (at your option).
4.
Select the mode of operation from the display.
•
If CE tracks or HA MAP tracks are
speCified, skip to step 9 on page MD-21.
If customer tracks are specified, the MD
prompts for track range information:
DO YOU HANT TO SCAN
ALL CUSTOMER TRACKS!
(SEE PSG MD--3,360)
. 6.
Respond with either the YES key or the NO key.
(If the response is YES, skip the next two
steps.)
If you respond with the NO key, you must
specify a range of cylinders to be scanned.
The MD prompts first for a beginning cylinder
number, then for an ending cylinder number.
The ending cylinder number must be higher
than, or equal to, the beginning cylinder
number.
The MD displays the following frame allowing
you to select the track type:
PIN 4519903 84 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
"
~
\ . ..,/
Respond to the track type prompt.
•
SELECT CONTROL MODE:
1. AUTOMATIC
2. MANUAL
(DEFAULT IS AUTO)
(
MD-20
'",
\
-/
(
is read, the data check can be considered
permanent. If the HA is sometimes read
without error, the data check can be
considered temporary.
The range of cylinder numbers are in
hexadecimal:
Cylinder
Number
Lowest
Highest
(-
3380-J
000
375
II
cJJLr
('
03 02
Sense
byte 6
31
I
7.
8.
Enter the three-digit hexadecimal number of the
beginning cylinder.
Enter the three-digit hexadecimal number of the
ending cylinder.
The MD displays the following frame:
C\
c
HOW MANY TIMES DO
YOU WANT TO READ
EACH HAt
(DEFAULT IS 2)
HA Scan cannot directly determine if a data
check in the home address area is permanent.
To determine that a data check is permanent,
the home address (HA) must be read multiple
times. If a data check occurs each time the HA
PIN 4519903 84 Sides
EC 476581 01Aug89
© Copyright IBM Corp_ 1987, 1988
•
6 minutes for a model AJ4 or BJ4 actuator
•
18 minutes for a model AK4 or BK4 •
actuator
Valid responses to the "How many times ... to
read" prompt are:
Head
01
After the beginning cylinder is entered, the MD
prompts for the ending cylinder number. The
ending cylinder number must be equal to or
greater than the number for the beginning
cylinder.
(/
The minimum run time for scanning all home
addresses on the volume (reading each HA one
time) is:
In this example, byte 5 contains 02 and byte 6
contains 31. The address is cylinder X"0302,"
head X "01."
<:
f--
It follows that the separation of permanent and
temporary data checks is more reliable when
the reading is repeated more. But repeating
reading takes time.
Note: The Physical Address (PA) bytes, from
the error bytes of another diagnostic or from
sense bytes 5 and 6, may be converted to
cylinder and head by using the following
diagram.
Sense
byte 5
02
(--
3380-K
000
A5F
Response
Meaning
ENTER
1
2
Default: Read each track 2 times
Read each track 1 time
Read each track 2 times
F
9.
Read each track 15 times
Enter your response for the number of times
the tracks are to be read. (If the response is
not 1, skip to step 10 on page MD-22.
If the number of times for reading the tracks is
1, the MD asks "HOW MANY TIMES DO YOU
WANT TO RETRY IF DATA CHECKS OCCUR?"
Valid responses to the "How many times ... to
retry" prompt are:
Response
Meaning
1
2
Retry each track in error 1 time
Retry each track in error 2 times
F
Retry each track in error 15 times
MD-21
10. Enter your response to select the number of
retries. (When a single read is selected, select
multiple retries to make it easier to determine
if the error is solid or intermittent.)
DO YOU WANT TO RUN
HA SCAN AGAIN 1
The scan starts. During the scan, the MD
displays the following frame:
***HA SCAN***
RUNNING ON DEVICE xx
CYLINDER
xxx
ERROR COUNT
xx
The cylinder number (hexadecimal) is updated
every 16 (X"10") cylinders. The error count is
updated each time a data check is
encountered.
The scan continues until one of the following
occurs:
•
Normal end, with no errors detected.
•
An equipment check (any isolation code
other than IC 9180). See "Isolation Code
Error Display" on page MD-41. The
following options are provided:
Analyze the error
Quit and return to the MD Main menu
Continue scanning with the next track
•
•
•
A data check occurs during manual mode
(IC 9180). See "Isolation Code Error
Display" on page MD-41. (The scan can be
continued after the error display. The
option to switch to automatic mode is
available.)
More than 80 data checks detected. (The
scan ends, and detail or summary
information may be displayed. See
following.)
Normal end with data checks detected.
(Detail or summary information may be
displayed. See following.)
o
*SCAN COMPLETE*
NO ERRORS
WERE
DETECTED
12. Respond to the question by pressing the YES
key or the NO key.
13. Press the PF key to return to the MD Main
menu.
I"'--j
14. If data checks are detected during the scan,
one of the following frames is displayed to
show the error condition:
* SCAN COMPLETE *
xx TRACKS HAD
DATA CHECKS
SEE PSG MD--3,39S ...
-
,
or-
'
_/
.
...,-/
* SCAN STOPPED *
THE DATA CHECK LOG
IN THE MD IS FULL.
SEE PSG MD-3,39S ...
The "xx" in the first frame is the number of tracks
(in hexadecimal) that had errors. The maximum
number of tracks that can be stored by the MD is
80 (X"50"). If more than 80 tracks had errors, the
scan is terminated and the second display is
generated.
After the scan is completed or stopped, a display of
the information can be selected from the Select
Display Mode menu:
SELECT DISPLAY MODE:
I-HEAD/ARM SUMMARY
2-TRACK SUMMARY
3-TRACK DETAIL
4-RUN HA SCAN
AGAIN
S-RETURN TO MAIN
MENU
11. If no errors are detected, the next two frames
are displayed:
PIN 4519903 84 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
MD-22
\
"'.
-'
ERROR COUNT = Number of errors
(hexadecimal)
Display Modes
After displaying the selected information, the MD
returns to the Select Display Mode menu and
. another mode can be selected. The following
display modes can be specified:
(
(-
Display Mode 1: Head/Arm Summary
The Head/Arm Summary mode presents a
summary (by head) of the number (in
hexadecimal) of tracks that had data checks.
The summary is presented in four displays,
one for each head arm:
= Number of errors for each head
Display Mode 2: Track Summary
The Track Summary mode presents one or
more displays with a one-line entry for each
track that had errors:
CYL=xxx
CYL=xxx
CYL=xxx
CYL=xxx
(
(
Display Mode 5: Return to Main Menu
Return to the MD Main Menu.
** ARM x **
HEAD:
xx xx xx xx
ERRORS: xx xx xx xx
ERRORS
(
Display Mode 4: Run HA Scan Again
Repeat the HA Scan test.
HD=x
HD=x
HD=x
HD=x
CNT=x
CNT=x
CNT=x
CNT=x
CYL = Cylinder number (hexadecimal)
HD
Head number (hexadecimal)
CNT = Number of errors (hexadecimal)
=
Display Mode 3: Track Detail
Track detail mode presents one display for
each track in error. There are two types of
error displays:
** DATA CHECK **
HOME ADDRESS AREA
CYL = xxx
HEAD = x
ERROR COUNT = x
c
-
or-
NO SYNC BYTE FOUND
HOME ADDRESS AREA
CYL = xxx HEAD = x
ERROR COUNT = x
= Cylinder number (hexadecimal)
CYL
HEAD
= Head number (hexadecimal)
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987. 1988
MD-23
Diagnostic Menu Option 6 - Motor Test
The motor test (routine 99) is used to isolate
problems resulting in a motor stopped condition.
The drive motor must be stopped before the test is
started. Both devices associated with the stopped
HDA and motor are required, and both should be
varied offline from the system. Controllers and
devices are selected by the user. The devices
must be in CE mode. The motor test runs inline
with customer operation on devices not selected by
the user.
To run the Motor Test;
1.
From the Diagnostic Menu, select option 6.
The MD displays the following frames:
WARNING - BOTH
DEVICES OF THE SAME
DRIVE WILL BE USED.
PROPER OPERATION ...
OF THIS ROUTINE
REQUIRES THE MOTOR
TO BE STOPPED.
SEE MIMI PWR-l 1 AO
The MD advances to one of the following
fre'"'1es when the ENTER key is pressed:
SELECT CONTROLLER:
(SEE PSG MD-3, 200)
-
or-
SELECT CONTROLLER:
DEFAULT IS: x x
(SEE PSG MD-3, 200)
The "x"s shown in the preceding frame identify
the last controller or controllers used to run
diagnostics in this session. If none was used,
no default is displayed.
The valid responses are:
Response
Meaning
ENTER
1
2
Default (if shown)
Controller A 1
Controller A2
Controller A3
Controller A4
3
4
Note; Selecting controller A31A4 is valid only
on strings with four storage paths, and only
when the MD is attached to the A unit that
contains controllers A3 and A4. Controllers A 1
and A2 cannot be selected when the MD is
attached to the A unit that contains controllers
A3 and A4.
PIN 4519903 84 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
MD-24
If
~ ..
,
2.
Enter your response for controller use.
The MD advances to one of the following
frames:
693 and 694 (if a front drive is to be tested) or
695 and 696 (if a rear drive is to be tested).
5.
After the drive motor has stopped, set the
switches indicated by the MD to the DISABLE
position to disable the device logic for both
actuators in the drive. Then press the ENTER
key to start the test. Follow instructions given
by the MD until the test is completed.
6.
If the test completes without errors, the motor
starts and the following frame is displayed:
SELECT DEVICE:
(SEE PSG MD-3, 210)
-
(
or-
SELECT DEVICE:
DEFAULT IS: xx
(SEE PSG MD-3, 210)
(
3.
(
The response to this prompt must be two
hexadecimal digits representing one of the two
devices in the drive being tested. Either device
address is permitted.
7.
Specify a single device in the drive to be
tested. The MD displays the following frame:
Respond to the question by pressing the YES
key or the NO key.
8.
Press the PF key to return to the MD Main
menu.
TURN MOTOR SWITCH
SWxxx OFF. MOTOR
MUST BE STOPPED BEFORE CONTINUING
('
END OF TEST. DO YOU
WANT TO RUN AGAIN?
To begin the test, the drive motor switch must
be off and the motor must be stopped.
The xxx part of SWxxx in the frame above is
replaced by 691 (if the front drive was identified
for testing) or by 692 (if the rear drive was
identified). See page INTRO-34 for the switch
locations.
(/
4.
Turn off the switch indicated in the display.
Then press the ENTER key to advance to the
next frame. The MD displays the following
frame:
DISABLE DEVICES aa,
bb WITH SWxxx AND
SWyyy AT SAM PANEL.
THEN PRESS ENTER.
In the preceding frame, the aa and bb are
replaced by the addresses of the devices in the
drive to be tested. The xxx and yyy parts of
SWxxx and SWyyy in the frame are replaced by
PIN 4519903 84 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
('
MD-25
Diagnostic Menu Option 7 - Device Status
The Device Status test is used for the following
purposes:
session. If none was used or if two controllers
were used, no default is displayed.
•
•
The valid responses are:
Testing controller operation.
Testing communication with each device in the
string without having to put devices in CE
mode.
Testing device enable/disable switches on the
operator panel.
Displaying string configuration information.
Displaying device status information.
•
•
•
The Device Status test may be run inline with
customer operation. Devices need not be in CE
mode unless the storage path or storage director is
disabled from all system channels. If the storage
path or storage director is disabled from all
channels, place all devices in CE mode before
running the Device Status test. Devices in CE
mode are tested the same as those not in CE
mode.
The Device Status test begins by running the
controller test (diagnostic routine 80). If no errors
are detected, the Device Status test (diagnostic
routine 96) is run.
The Device Status test selects each device on the
string and performs a Read Home Address
operation (without moving the access mechanism).
String configuration information is displayed.
To run the Device Status test:
1. From the Diagnostic menu, select option 7.
~--",\
Response Meaning
ENTER
1
2
3
4
Default (if shown)
Controller A 1
Controller A2
Controller A3
Controller A4
Note: Selecting controller A3 or A4 is valid only on
strings with four storage paths, and only when the
MD is attached to the A unit that contains
controllers A3 and A4. Control/ers A 1 and A2
cannot be selected when the MD is attached to the
A unit that contains controllers A3 and A4.
2. Enter your response for controller use.
The diagnostic routines start running .and continue
until string configuration information is displayed or
until an error is encountered.
•
If an error is encountered, go to "Isolation
Code Error Display" on page MD-41.
•
If no error is detected, the MD displays
configuration information in one of the following
series of frames. The configuration information
displayed is dependent on storage control type:
For a 3880 subsystem, see "Gonfiguration
(3880)" on page MD-27.
The MD displays one of the following frames:
SELECT CONTROLLER:
For a 3990 subsystem, see "Configuration
(3990)" on page MD-28.
(SEE PSG MD-3, 200)
-
or-
SELECT CONTROLLER:
DEFAULT IS: x
(SEE PSG MD-3, 200)
The "x" shown in the preceding frame identifies the
last controller used to run diagnostics in this
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
MD-26
IBM Corp. 1987, 1988
."
I
c:
(
Configuration (3880)
For strings attached to a 3990 storage control, refer
to page MD-28.
If no error is detected, the MD displays
configuration information as shown in the following
frames.
storage director attached to the string has a
unique SOlD.
CTLR PATH=x
x = The path that was sensed from the
controller by the storage control during a
sense controller path operation.
o = Controller A1 Path
STRING CONFIGURATION
CTLR Ax ADR=x ID=xx
3880 SDID=xx
(
1
A*
THE STRING HAS:
A*
=
AAAA-
ID=xx
xx = the physical ID of the controller (A 1 or
A2) set in the physical address switches on
the I/O control card. See page REPAIR-20.
The controller physical ID is normally put on
the 3380 operator panel at installation time
(String ID).
c
The physical ID must be set the same for all
controllers in the string.
SDID=xx
xx
The physical ID of the storage director
in the 3880. The SDID is normally put on the
3380 operator panel at installation time. Each
UNIT ONLY
AND 1 B UNIT
AND 2 B UNITS
AND 3 B UNITS
The last frame of the series asks if you want the
configuration data analyzed. To respond, press the
YES key or the NO key:
YES
The MD has you check for inconsistencies in
the collected data. If no inconsistencies are
found, the following menu is displayed. If
inconsistencies are found, the test must be
rerun to display device status.
NO
The MD displays the following menu.
ADR=x
x = the string number (O or 1) set in the
subsystem configuration switches on the
DDC/DTB card. See page REPAIR-18.
The string number is normally put on the
operator panel at installation time. All
controllers in the string must be set the same.
* = The number of units that are in the
string. One of the following lines is displayed.
DO YOU WANT AN
ANALYSIS OF THIS
CONFIGURATION DATA?
(SEE PSG MD-3, 460)
CTLR Ax
x
the controller number that was selected
to run the test.
= Controller A2 Path
3. Enter your response about the analysis of the
configuration data:
If you do not want to analyze the configuration data
or if no inconsistencies were found during the
analysis of collected data, the MD displays the
following menu:
SELECT OPTION:
1 = DISPLAY DEVICE
STATUS
2 = REVIEW CONFIG ...
3
4
= RERUN
= END
CONFIG.
=
C"
PIN 4519903 84 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
MD-27
1
=
=
3 =
2
4
=
Continue the Device Status test and display
the device status. See "Display the Device
Status" on page MD-30.
Configuration (3990)
For strings attached to a 3880 storage control, see
"Configuration (3880)" on page MD-27.
Review the configuration data just displayed.
Collect the configuration data again and
display it.
End the Device Status test without displaying
device status.
4. Select an option from the preceding menu.
5. To continue the device status test, go to "Display
the Device Status" on page MD·30.
One series of frames is for 2-path configurations
and the other series of frames is for 4-path
configurations.
2-Path Configurations
STRING CONFIGURATION
CTLR Ax ADR=x ID=xx
3990 SSID=xxxx SP=x
CTlR PATH=x
THE STRING HAS:
A*
DO YOU WANT AN
ANALYSIS OF THIS
CONFIGURATION DATA!
(SEE PSG MD-3, 460)
I
/
4-Path Configurations
STRING CONFIGURATION
CTLR Ax ADR=x ID=xx
3990 SSID=xxxx SP=x
CTlR PATH=x
SUB-STRING 0 HAS:
A*
SUB-STRING 1 HASI
A*
DO YOU WANT AN
ANALYSIS OF THIS
CONFIGURATION DATA!
(SEE PSG MD-3, 460)
(
CTLR Ax
=
X
the controller number that was selected
to run the test.
ADR=x
x = the string number (0 or 1) that is set in
the subsystem configuration switches on the
DDC/DTB card. See page REPAIR-18.
The string number is normally put on the
operator panel at installation time. All
controllers in the string must be set the same.
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1997. 1999
MD-28
(
(
ID=xx
xx
the physical 10 of the controller (A 1, A2,
A3 or A4) set in the physical address switches
on the I/O control card. See page REPAIR-20.
The controller physical 10 is normally put on
the 3380 operator panel at installation time
(String 10).
=
The physical ID must be set the same for all
controllers in the string.
SSID=xxxx
xxxx = The physical ID of the Storage Path in
the 3990. The SSIO is normally put on the
3380 operator panel at installation time. The
SSID must be set the same for all storage
paths attached to the string at installation
time.
(
CTLR PATH=x
x = The path that was sensed from the
controller by the storage control during a
sense controller path operation.
o = Controller A1
Path
(
1
2
3
Path
(
A*
= Controller A2 Path
= Controller A3
= Controller A4 Path
*=
The number of units that are in the string
or sub-string (4 path). One of the following
lines is displayed.
AAAA-
C
NO
The MD displays the following menu.
3. Enter your response about the analysis of the
configuration data.
If you do not want to analyze the configuration data
or if no inconsistencies were found during the
analysis of collected data, the MD displays the
following menu:
SElECT OPTION:
1 = DISPLAY DEVICE
STATUS
2 = REVIEW CONFIG ...
x = The storage path (SPO-SP3) connected to
the controller that was selected to run the
test. The storage path is normally put on the
3380 operator panel at installation time.
(
(
The MD has you check for inconsistencies in
the collected data. If no inconsistencies are
found, the following menu is displayed. If
inconsistencies are found, the test must be
rerun to display device status.
SP=x
(-
(':
YES
3
4
1
= RERUN
=
END
CONFIG.
=
Continue the Device Status test and display
the device status.
2
Review the configuration data just displayed.
3 = Collect the configuration data again and
display it.
4 = End the Device Status te!:!t without displaying
device status.
=
4. Select an option from the preceding menu.
5. To continue the device stat':!s test, go to "Display
the Device Status" on page MD·30.
UNIT ONLY
AND 1 B UNIT
AND 2 B UNITS
AND 3 B UNITS
The last frame of the series asks if you want the
configuration data analyzed. To respond, press the
YES key or the NO key:
c\
C::
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
MD-29
Display the Device Status
E
Equipment Check
6. To continue the device status test (option 1 from
the "Select Option" menu is selected), the
following frames are displayed:
2-Path Subsystem
X
DEVICE STATUS:
OI23456789ABCDEF
=
OX:
****************
DO YOU WANT A
DESCRIPTION OF THE
STATUS CODES?
(SEE PSG MD-3, 465)
1I.(ervention Required: Device is disabled and
not on-Line; this code has nothing to do with
whether or not a device is in CE mode.
N
Not Installed
o
Operational: An operational device normally
reports a status of "0."
S
Servo busy: If a device has been powered on
without at least one storage path channel
interface enabled, the device usually reports
a status of "S" because there is no attached
system to handle the attention interrupt. To
eliminate this problem, place the device in CE
mode (use MD Main menu Option 0) and
rerun the device status test.
x=
Device Number (O-F)
OX: Represents devices ~O-OF
4-Path Subsystem
X
DEVICE STATUS:
= OI23456789ABCDEF
OX:
IX:
****************
****************
DO YOU WANT A
DESCRIPTION OF THE
STATUS CODES?
(SEE PSG MD-3, 465)
X = Device Number
OX: Represents devices ~O-OF
1X: Represents devices 10-1F
The asterisks (*) in the top frame are replaced by
device status (for details, see step 7). The bottom
frame asks if you want the status codes (used in
the frame on the top) described. To respond, press
the YES key or the NO key:
YES
NO
Status code information (shown below) is
displayed. Press the ENTER key to continue
with the display of device type information.
The MD displays device type information.
7. Enter your response about the device status code
descriptions.
Displayed status code information is:
B
D
Display Device Type Information
8. To continue the device status test: after the
device status code description response is
completed, device type information is displayed in
one of the following series of frames. One series is
for 2 path configurations and the other series of
frames is for 4 path configurations.
2-Path Configurations
DEVICE TYPE:
FROI=xxx
FR02=xxx
FR03=xxx
FR04=xxx
DO YOU WANT AN
ANALYSIS OF THE
DEVICE TYPE DATA?
(SEE PSG MD-3, 468)
FR01 =
FR02 =
FR03
FR04 =
=
XXX
=
AJ4 or AK4 unit unit
1st B unit attached to A unit
2nd B unit attached to A unit
3rd B unit attached to A unit
Model designation (AJ4, AK4, BJ4, or
BK4) of the installed machine frame(s).
Frame members are not displayed if
they are not installed.
Device Busy
Data Check
PIN 4519903 84 Sides
EC 476581 01AugB9
© Copyright IBM Corp. 1987, 1988
MD-30
(
(
xxx =
111 The diagnostic routine was unable to
4-Path Configurations
DEVICE TYPE - SSO:
FR01=xxx
FR03=xxx
FR02=xxx
FR04=xxx
(
(
SSl:
DEVICE TYPE
FRll=xxx
FR13=xxx
FR14=xxx
FR12=xxx
DO YOU WANT AN
ANALYSIS OF THE
DEVICE TYPE DATA!
(SEE PSG MD-3, 468)
(
(
('
FR01
FR02
FR03
FR04
FR11
FR12
FR13
FR14
=
=
=
=
=
=
=
=
xxx =
(
c
YES
The MD has you check the data for
inconsistencies. If no inconsistencies are
found and no errors occur, the MD displays
the following frame.
NO
The MD displays the following frame.
determine the 3380 model. This occurs
when the diagnostic cannot select any of
the devices in the unit due to heavy
system activity.
SELECT OPTION:
1.= REVIEW
2.= RERUN TEST
3.= END
9. Enter your response about the analysis of the
device-type data.
10. Select option 1, 2, or 3.
11. Press the PF key to return to the MD Main
menu.
1st AJ4 or AK4 unit unit
1st B unit attached to 1st A unit
2nd B unit attached to 1st A unit
3rd B unit attached to 1st A unit
2nd AJ4 or AK4 unit unit
1st B unit attached to 2nd A unit
2nd B unit attached to 2nd A unit
3rd B unit attached to 2nd A unit
Model designation (AJ4, AK4, BJ4, or
BK4) of the installed machine frame(s).
Frame members are not displayed if
they are not installed.
xxx =
117 The diagnostic routine was unable to
determine the 3380 model. This occurs
when the diagnostic cannot select any of
the devices in the unit due to heavy
system activity.
SSO =
(Substring 0) is the first substring in a
4-path configuration. Substring 0
contains ports 0-3 (devices 00 - OF).
SS1 =
(Substring 1) is the second substring in
a 4-path configuration. Substring 1
contains ports 4-7 (devices 10 - 1F).
The last frame displayed asks if you want the
device type information analyzed. To respond,
press the YES key or the NO key:
PIN 4519903 84 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
MD-31
Diagnostic Menu Option 8 - CE Track Repair
The CE track repair option uses diagnostic routines
83 and 9C to perform the following functions to the
CE cylinder and the four diagnostic cylinders.
•
•
•
Test each track for data checks
Inspect each track with data checks for defects
Skip all defects found
The CE track repair routines are designed to run
inline while customer operation continues on
devices not selected by the user. The selected
device must be previously placed in CE mode. For
additional information about CE track repair, see
page MAP-1, Entry EF, in the MIM, Volume R10.
To run the CE Track Repair option:
1.
From the Diagnostic Menu, select option 8.
The MD displays one of the following frames:
SELECT
CONTROLLER~
(SEE PSG MD-3, 200)
-
or-
SELECT CONTROLLER:
DEFAULT IS: x
(SEE PSG MD-3. 200)
The "x" shown in the preceding frame is
( ,
replaced by 1, 2, 3, 4, or a blank to identify the
" ...
last controller used to run diagnostics in this
session. If none was used, no default i s ' \
displayed.
(---\
The valid responses are:
'''''-j
Response Meaning
ENTER
1
2
3
4
Default (if shown)
Controller A 1
Controller A2
Controller A3
Controller A4
CE track repair will run using the specified
controller. The controller specified should be
operational.
Note: Selecting controller A3 or A4 is valid
only on strings with four storage paths, and
only when the MD is attached to the A unit that
contains controllers A3 and A4. Controllers At
and A2 cannot be selected when the MD is
attached to the A unit that contains controllers
A3 and A4.
PIN 4519903 84 Sides
EC 476581 01AugB9
@ Copyright IBM Corp. 1987, 1988
MO-32
\
......
-
./
(-'
2.
Enter your response about controller use.
4.
If an unrecoverable equipment check error is
encountered, go to "Isolation Code Error
Display" on page MD-41.
5.
If no errors are detected and the repair process
is completed successfully, the following frame
is displayed.
The MD displays one of the following frames:
SELECT DEVICE:
(SEE PSG MD-3, 210)
(
-
or-
*
REPAIR COMPLETE *
NO DATA CHECKS
WERE DETECTED ON THE
CE TRACKS.
SELECT DEVICE:
DEFAULT IS: xx
(SEE PSG MD-3, 210)
(
Default devices are those that were specified
during the previous diagnostic. If no device
was tested during the current diagnostic
session, no default is displayed.
(
6.
Press ENTER to return to the MD Main menu.
Device addresses are in hexadecimal (00
through 1F).
(
The device to be tested must be in CE mode. If
the device specified for testing is not in CE
mode, an error message is displayed.
(
3.
( --",
~,
Specify a single device.
When the ENTER key is pressed, the CE track
repair diagnostic starts. The diagnostic
continues until normal end or until an
unrecoverable error is encountered. Several
displays appear while the routines are running.
These frames are to keep you informed of what
is happening:
* CE CYLINDER *
WRITE/READ ROUTINE
RUNNING ON DEVICE xx
(ROUTINE 83)
*RUNNING
CE TRACK
ON
REPAIR *
DEVICE XX
CYLINDER XXX HEAD X
(ROUTINE 9C)
(/
The CE track repair process may take several
minutes. The amount of time depends on the
extent of the problem. It takes a little more
than 2.0 minutes for each defect found.
C
c
(
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987. 1988
MD-33
c
o
Diagnostic Menu Option 9 - Storage Control Status
The Storage Control Status option uses diagnostic
routine 9F to collect status information from the
Storage Control. The status information collected
includes the following:
r\
'V
To run the Storage Control Status option:
1.
From the Diagnostic Menu, select option 9.
The MD displays one of the following frames:
•
•
•
•
•
Devices in CE Mode
Controller Write Inhibit conditions
Channel Write Inhibit conditions
Storage Path Fenced conditions
Device Fenced conditions
SELECT CONTROLLER:
(SEE PSG MD-3, 200)
-
or-
SELECT CONTROLLER:
DEFAULT IS: X
(SEE PSG MD-3, 200)
The Storage Control Status test is designed to run
in line with customer operation. Devices do not
need to be placed in CE Mode.
The X shown in the preceding frame identifies
the last controller used to run diagnostics in
this session. If none was used, no default is
displayed.
The valid responses are:
Response Meaning
ENTER
1
2
3
4
Default (if shown)
Controller A1
Controller A2
Controller A3
Controller A4
The storage control status test will run using
the specified controller. The controller
specified should be operational.
Note: Selecting controller A3 or A4 is valid
only on strings with four storage paths, and
only when the MD is attached to the A unit that
contains controllers A3 and A4. Controllers A 1
an... A2 cannot be selected when the MD is
attached to the A unit that contains controllers
A3 and A4.
2.
PIN 4519903 84 Sides
EC 476581 01Aug89
(
~'\
" .-/j
Enter your response for controller use.
When the ENTER key is pressed, the Storage
Control Status test starts and continues until
Storage Control Status is displayed or until an
error is encountered.
© Copyright IBM Corp. 1987, 1988
c
MD-34
(""
v
c
f
3.
If an error is encountered, go to "Isolation
Code Error Display" on page MD-41.
4.
If no error is detected, the following frame is
displayed:
7.
Press the ENTER key to determine if the
storage control has set write inhibit for the
controller selected in step 2 on page MD-34.
One of the following frames is displayed:
NO WRITE INHIBIT
CONDITIONS WERE
DETECTED FOR
CONTROLL ER Ax.
- orWRITE INHIBIT IS SET
FOR CONTROLLER Ax.
SEE PSG MD-3,090 ...
STORAGE CONTROL
STATUS
FOR
CONTROLLER A*
5.
,,"
Press the ENTER key to display status
information:
xxxx S*ID: xx
STORAGE PATH: x
CONTROLLER: Ax
SEE PSG MD-3, 480 ...
If line 1 is "3880 SDID: xx,"
xx = the Storage Director identifier.
If line 1 is "3990 SSID: xx,"
xx
the Subsystem identifier.
=
STORAGE PATH: x
x = the Storage Path (0-3)
CONTROLLER: Ax
Ax = the Controller (A 1, A2, A3, or A4)
6.
(
Controller Write Inhibit is set for a controller if
a controller error or storage path error is
detected during a functional write operatron
(diagnostics do not cause write inhibit
conditions). Write inhibit is set to prevent a
defective storage path from destroying
customer data.
Write inhibit can be reset as follows:
•
Use the MD Main Menu option Z (End of
Call -- see RMAP 0350 Step 001 on page
REPAIR·15).
•
If the MD is attached to a 3990 storage
control, use the Storage Path Restart
switch.
•
Issue the ICKDSF CONTROL command with
the ALLOWWRITE parameter specified.
See the Device Support Facilities Users
Guide and Reference, GC35-0033 and IMAP
0280 Step 001 on page ISOLATE·49.
Press the ENTER key to display devices in CE
Mode. The following frame is displayed:
2-Path Configurations
DEVICES IN CE MODE:
OX: ••••••••••••••.•
(SEE PSG MD-3, 035)
OX:
= Devices 00 - OF
4-Path Configurations
DEVICES IN CE MODE:
OX: ••••••••.•••••••
lX: ...•..•...••.•••
(
(
(SEE PSG MD-3, 035)
= Devices 00 - OF
1X: = Devices 10 - 1F
OX:
PIN 4519903 84 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
MD-35
8.
Press the ENTER key to determine if the
storage control has set write inhIbit on one or
more channels. One of the following frames is
displayed:
NO CHANNEl WRITE
INHIBIT CONDITIONS
WERE DETECTED FOR
CONTROllER Ax.
-
or-
Support Facility. Diagnostics do not cause
fenced conditions to be set.
Controller Fenced is set by a Quiesce from the
Storage Control Support Facility.
10. Press the ENTER key to display Device Fenced
conditions. One of the following frames is
displayed:
CHANNEL WRITE
INHIBIT IS SET FOR
CHANNELS: ....... .
SEE PSG MD-3.090 .. .
NO FENCED CONDITIONS
WERE DETECTED FOR
ANY DEVICES.
-
Channel Write Inhibit is set in the storage
control for a channel if a channel or channel
related storage path error is detected during a
functional write operation (diagnostics do not
cause channel write inhibit conditions). Write
Inhibit conditions may reset as described in
step 7 on page MD-35.
9.
Press the ENTER key to display Fenced
conditions (3990 4-path only).
Fencing is a technique used by the storage
path functional microcode to remove a failing
path from system operation. This is done to
prevent path related failures from degrading
subsystem performance.
•
If the string is a 2-path string, only
Controller Fenced applies.
•
If the string is a 4-path string, Storage Path
Fenced conditions are displayed as follows.
NO STORAGE PATH
FENCED CONDITIONS
WERE DETECTED FOR
CONTROllER Ax.
-
DEVICE FENCED
IS SET FOR CTLR Ax.
DEVICES ARE:
***FENCED DEVICES***
OX: ....•..••....•..
IX: 0 .•..•..••..•..•
(SEE PSG MD-3,093)
Device Fenced is set for a device when there
are excessive controller errors for a single path
to the device. The device remains operational
on other paths. In the above frame, device 10
is fenced.
Storage path controller and device fenced
conditions can be reset as 'follows:
•
•
•
•
or-
STORAGE PATH FENCED
CONDITION IS SET
FOR CONTROLLER Ax.
(SEE PSG MD-3,092)
or-
•
Use MD Main Menu option Z: End of Call.
Use the storage path restart switch at the
storage control.
Perform a RESUME operation on the
quiesced storage path or controller.
Issue a ICKDSF CONTROL command with
the CLEARFENCE parameter specified. See
the Device Support Facilities Users Guide
and Reference, GC35-0033 and IMAP 0280
Step 001 on page ISOLATE-49.
IML the storage path.
11. Press the PF key to return to the MD Main
Menu.
Storage Path Fenced is set when there are
excessive storage path errors or when a
Quiesce is set from the Storage Control
PIN 451990384 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
MD-36
/
'"
/
(
f
(
(
Diagnostic Menu Option A - Non-Repeatable Runout (NRRO) Test
The Non-Repeatable Runout Test (Routin'e 89)
measures differences in the servo position error
signal on successive disk revolutions. This is used
to indicate conditions other than normal track
runout that may adversely affect servo tracking.
The X shown in the preceding frame identifies
the last controller or controllers used to run
diagnostic routines in this session. If none
were used, no default is displayed.
The valid responses are:
This routine has been designed to run along with
customer operation on the selected devices. The
devices that are selected should not be in CE
mode.
(
(
Response Meaning
ENTER
1
2
Routine 89 will alternately test the two devices on
the selected HDA when only one device was
selected.
Routine 89 does not have a normal ending point.
The routine will run indefinitely until a fault exists
or until the routine is terminated by the user. The
minimum run time is 24 hours, or until an error
stop is presented. For more information, see the
DIAG section in the MSM, Vol. R60.
3
4
Note: Selecting controller A3 or A4 is
valid only on strings with four storage
paths. The MD will instruct you when it
is necessary to move the MD plug.
2.
This routine is not to be run unless instructed by
the maintenance information.
CAUTION
***
A ***
PHYSICAL
DEVICE
WITH 9F9F FSC
ENVIRONMENTAL
(
(
DATA CAN BE REPORTED
THROUGH OTHER
DEVICES. THE CORRECT
PHYSICAL ADDRESS
IS FOUND IN
SENSE BYTE 4.
SEE THE PSG START
SECTION SMAP 0045.
To run the Non-Repeatable Runout option:
1.
From the
~iagnostic
Enter your response for controller use.
The MD displays the following frames:
If you were sent here from the MD for FSC E180,
E108, or 9302, verify that the errors are on heads 4,
5, 6, and/or 7 (see sense byte 6). If the errors are
not on these heads, return to the MD frame that
sent you here and continue with the procedure. If
the errors are on heads 4, 5, 6, and/or 7, continue
with the non-repeatable runout (NRRO) test.
(
Default (if shown)
Controller A 1
Controller A2
Controller A3
Controller A4
Menu, select option A.
The MD displays one of the following frames:
SELECT CONTROLLER:
(SEE PSG MD-3, 005)
(
-
(
SELECT CONTROLLER:
DEFAULT IS: x
(SEE PSG MD-3, 005)
or-
3.
You are then asked to select a device with one
of the following frames:
SELECT DEVICE:
(SEE PSG MD-3, 210)
-
or-
SELECT DEVICE:
DEFAULT IS: xx
(SEE PSG MD-3, 210)
()
PIN 4519903 84 Sides
© Copyright
('
EC 476581 01Aug89
IBM Corp. 1987, 1988
MD-37
Enter a single device. Multiple addresses are
not permitted and will be ignored. If the
configuration is 2-way, then enter 00 through
OF. If the configuration is 4-way, then 00
through 1F is permitted.
WHICH ACTUATOR
POSTED THE 9F9F FSC1
(1) LEFT. (2) RIGHT
OR (3) BOTH!
Enter the response for which actuator posted
the FSC 9F9F and go to step 8.
The device should not be in CE mode.
4.
Enter your response for device address.
8.
The MD displays the following frame:
I
The MD displays the following frame:
THE NRRO TEST IS NOW
RUNNING ON DEVICES
xx AND yy
(SEE PSG MD-3. 005)
WAS FSC 9F9F POSTED
AGAINST THE HDA
UNDER TEST?
Respond to the question by pressing the YES or
NO key.
If the answer is YES, go to step 7.
If the answer is NO, go to step 5.
5.
The MD displays the following frame:
ENTER OTHER POSTED
FSC. OR. PRESS ENTER
FOR NONE.
6.
Enter your response for other posted FSCs.
If the FSC is 9302, E1S0, E10S, or none go to
step S.
If the FSC was not 9302, E10S, or E1S0 the MD
displays the following frame:
USE FSC ANALYSIS
(MENU OPTION-3) TO
ANALYZE FSC xxxx.
Press the ENTER key to return to the MD Main
menu.
7.
/-"
=
xx
even device
yy == odd device
9.
The test has no normal end and will run
continuously until the MD has determined that
a fault exists or until it is stopped by pressing
the ENTER key on the MD keyboard. The
minimum run time is 24 hours, or until an error
stop is presented.
10. If system activity is large enough that it
prevents routine 89 from collecting test data on
a continuing basis, the following frame is
displayed:
DEVICE(S) ARE BUSY.
THIS MAY BE DUE TO
HEAVY SYSTEM USAGE.
RETRY WILL CONTINUE.
The system activity is too high. Re-schedule
your maintenance activity for a less busy time.
11. If an unexpected error occurs, the following
frames will be displayed:
AN UNEXPECTED ERROR
WAS RECEIVED. TO
ISOLATE THE PROBLEM,
CHOOSE THE NORMAL ...
SEQUENCE OPTION FROM
THE DIAGNOSTIC MENU.
PRESS ENTER TO
RETURN TO MAIN MENU.
If FSC 9F9F was posted against the HDA under
test, the following frame will be displayed:
12. The CE can stop the routine by pressing the
ENTER key while the routine is running. If an
insufficient number of samples were taken for
PIN 451990384 Sides
EC 476581 01Aug89
~ Copyright IBM Corp. 1997. 1988
MD-38
rf "",
\t./
(
(
(
(
(
('
calculations the MD will display the frame in
step 22 on page MD-40. If there were enough
samples taken, the MD display will be
explained in step 23 on page MD-40. If the CE
does not stop the routine, go to the next step.
13. If the NRRO test determines that the results of
the value calculations is in excess of
permissible limits and the problem is with one
device (left or right), the frames in step 15 will
be displayed. If the NRRO test determines that
the problem is with both devices, go to step 14.
REPLACE THE HDA.
SEE CARR-I, ENTRY-A
FOR REPLACEMENT
PROCEDURE.
15. The left or right device NRRO calculations are
high. The following frames will be displayed:
f:
C'
xxxxx
= RIGHT
LEFT OR
(-.
(-
IS THE CABLE SWAP IN
PLACE!
16. Respond to the question by pressing the YES or
NO key.
If the answer is YES, go to step 18.
If the answer is NO, the following frame will be
displayed:
C
Respond to the question by pressing the YES or
NO key.
If the answer is NO, go to step 19.
If the answer is YES, go to step 14.
HAVE THE FRU'S
FOR THE FAILING
ACTUATOR BEEN
REPLACED?
Respond to the question by pressing the YES or
NO key.
If the answer is YES, go to step 21 on
page MD-40.
If the answer is NO. go to step 20,
ELECTRONICS MAY BE
CAUSING THE ERROR.
CAN YOU PERFORM A
CABLE SWAP!
(
18. The cables have been swapped. The following
frames will be displayed:
19. The following frame will be displayed:
(
(j
Pressing ENTER returns you to the main menu.
AFTER THE CABLE
SWAP!
This is the end of the NRRO routine, pressing
ENTER will return you to the main menu.
THE xxxxx device
HAS HIGH NRRO.
PERFORM THE CABLE
SWAP, THEN RE-RUN
THE NRRO TEST.
IS THE SAME DEVICE
POSTING 9F9F FSC'S
pP HIGH NRRO VALUES
BOTH BEFORE AND
14. The following will be displayed:
(
If the answer is YES, the following frame will be
displayed:
20. The following frames will be displayed:
17. Respond to the question by pressing the YES or
NO key.
If the answer is NO, go to step 19.
PIN 4519903 84 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1987, 1988
MD-39
{.
I
\s..)
REPLACE THE FRU'~
FOR THE FAILING
ACTUATOR. THEN RUN
NRRO TEST AGAIN
*
PRIMARY FRU LIST
- WARNING SEE PROCEDURE ON
REPAIR 1,ENTRY-A
NRRO TESTED NORMAL.
BE SURE THAT THE
SELECTED DEVICE IS
CORRECT. IF NEEDED,
*
COB2 RW CHANL DEV
BIF2 PORT R/W DEV
BIE2 SEQ/SERVO DEV
This is the end of the NRRO routine, pressing
ENTER will return you to the main menu.
21. The following frame will be displayed:
GO TO PSG START
SECTION SMAP 0045.
This is the end of the NRRO routine, pressing
ENTER will return you to the main menu.
"--- ....,/
24. If the samples are not indicating a defective
HDA, the the following frame will be displayed:
SUCCESSFUL TEST END.
PRESS ENTER KEY TO
RETURN TO MAIN MENU.
WATCH FOR ADDITIONAL
9F9F FSC'S AGAINST
THESE SAME DEVICE(S)
AND HDA. IF THIS ...
CONDITION CONTINUES,
GO TO PSG MD-I,
ENTRY-C, OPTION-A,
TO RERUN NRRO TESTS.
This is the end of the NRRO routine, pressing
ENTER will return you to the main menu.
"
:/
22. The MD will display the following frame:
AN INSUFFICIENT
NUMBER OF SAMPLES
HAVE BEEN COLLECTED.
(LESS THAN xxr.) ..•
PRESS TO
RESUME. OR PF TO
TERMINATE TEST. ALL
DATA WILL BE LOST.
PF will return you to the MD MAIN MENU. If
ENTER is pressed the test is resumed. Go to
step 8 on page MD-38.
23. A sufficient number of samples were taken. If
the NRRO did not test normal, the frames in
step 24, step 14 on page MD-39, or step 15 on
page MD-39 will be displayed depending on the
the calculations. If the NRRO tested normal,
the following frames will be displayed.
PIN 4519903 84 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987. 19BB
.~./
MO-40
Isolation Code Error Display
(
If an error is encountered, the following is
displayed:
IC**** CTL=A* DEV=**
RTN=** TEST=**
EB=** **** **** ****
**** **** **** ****
Isolation Code
-
CTl
Controller Number (A1, A2, A3, or A4)
PRESS ENTER TO
ANALYZE OR "N" TO
RUN NEXT CONTROLLER
(SEE PSG MD-3, 500)
DEV
Device Address (in hexadecimal)
-
RTN
Diagnostic Routine ID
EB
Error Bytes (see note)
Note: The error bytes are used by the MD for FRU
isolation. The guided maintenance procedures do
not require you to use these bytes for any purpose.
They are displayed primarily for support personnel.
..
f:
PRESS ENTER TO
ANALYZE OR "N" TO
RUN NEXT TEST.
(SEE PSG MD-3, 500)
IC
TEST Test Number within the Routine
(
Depending on the operation being performed at the
time of the error, one of the following frames will
be displayed:
If you are preparing to discuss this problem with
support personnel, you may want to record these
error byte values.
or-
or-
PRESS ENTER TO
ANALYZE OR PF KEY TO
RETURN TO MAIN MENU
(SEE PSG MD-3, 500)
To analyze an error and obtain a FRU group, press
the ENTER key only.
To test the next controller or device (if multiple
controllers or devices were specified), enter Nand
then press the ENTER key.
To return to the MD Main menu, press the PF key.
Further definition of the information contained in
these displays can be found in the DIAG section of
the MSM, Volume R60 and in the IC section of the
ECM, Volume R40.
(,
(-'
(~
(:
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
MD-41
Analyze Symptom Code (Main Menu Option 3)
contains controllers A3 and A4. Controllers A 1
and A2 cannot be selected when the MD is
attached to the A unit that contains controllers
A3 and A4.
This option can be used to get a FRU list from the
MD by entering a symptom code and some device
information. The MD does not need to be attached
to the 3380 to analyze a symptom code.
1.
From the MD Main menu, select option 3.
2.
The MD displays the fol;owing frame:
Enter your response to identify the controller.
The MD next asks for device information with
the following frame:
ENTER THE SYMPTOM
CODE.
(SEE PSG MD-3. 550)
ENTER THE FAILING
DEVICE TO ANALYZE
THE ERROR (00 TO XX)
Enter a four-character symptom code from
system sense bytes 22 and 23.
•
00 to XX indicates the valid range of devices
that may be entered:
''-,
~
The MD then displays the following frame:
"00 TO OF" for 2-path strings
"00 TO 1F" for 4-path strings
ENTER THE FAILING
CONTROLLER TO
ANALYZE THE ERROR
A device must be selected so that the MD can
determine a FRU group when device FRUs may
be involved. If multiple devices are failing,
select one of the devices. If the MD analysis
procedure needs to know if multiple devices
are failing, the appropriate questions will be
asked.
(xxxxxxxxxxxxx)
A controller must be identified so that the MD
can determine a FRU group when controller
FRUs are involved. If both controllers are
failing or if specific devices are failing through
both controllers, select either controller. If the
MD analysis procedure needs to know if both
controllers are failing, the question will be
asked.
The last line of the display (xxxxxxxxxxxxx)
displays valid controller options 1 or 2 for 2
path strings and 1, 2, 3, or 4, for 4 path strings.
The valid responses are:
3.
Enter a valid device address, "00" through "OF"
or "00" through "1F."
Next, the MD asks about the HDA Cable Swap.
If the cables are swapped, the FRU group is
affected:
ARE THE HDA CABLES
SWAPPED?
Response Meaning
1
2
3
4
Controller
Controller
Controller
Controller
A1
A2
A3
A4
If the HDA cables are swapped on the actuator
fOI vVhich you are doing analysis, press the YES
key. Otherwise, press the NO key. The MD
will display the correct FRU list for the logic
that is operating the actuator.
Note: Selecting control/er A3 or A4 is valid
only on strings with four storage paths, and
only when the MD is attached to the A unit that
PIN 4519903 B4 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987. 1988
MO-42
\~"
../
.
(
c
4.
Answer the question about the cable swap.
Other questions regarding the sense
information might be asked. Most other
questions may be bypassed by pressing the
ENTER key. If questions are bypassed, the
result is larger FRU groups.
5.
Answer all questions to the best of your ability.
When the answers to the questions have been
processed, the MD displays FRU information:
* PRIMARY
FRU
- WARNING
LIST *
SEE PROCEDURE ON
REPAIR 1,ENTRY A ...
****************
****************
****************
****************
-
(
(
and then -
SECONDARY FRU LIST
- WARNING SEE PROCEDURE ON
REPAIR 1,ENTRY A ...
****************
****************
****************
****************
{'
(
Ensure that the correct power on/power off
procedures are used. See page REPAIR-1 to
determine the correct procedure for replacing
each FRU.
(\
C:i
PIN 4519903 84 Sides
© Copyrighl
EC 476581 01Aug89
IBM Corp. 1987. 1985
MD-43
Diagnostic Aiqs (Main Menu Option DJ
Diagnostic Aids provides a method of running
individual diagnostics and allows entry of
parameters to control them. With few exceptions,
maintenance procedures do not need Diagnostic
Aids. The diagnostic menu provides most needed
functions in a much easier to use form. See "Run
Diagnostic Tests (MD Main Menu Option 1)" on
page MD-10.
Any routine may be run using the Diagnostic Aids
option, and you have full control of routine
execution. But if you enter inappropriate control
parameters for the routine being run, some very
misleading results can occur.
1.
From the MD Main menu, select option D.
The MD displays one of the following frames:
SELECT CONTROLLER:
(SEE PSG MD-3. 800)
-
or-
SELECT CONTROLLER:
DEFAULT IS: x
(SEE PSG MD-3. 800)
The "x" shown in the preceding frame
identifies the last controller used to run
diagnostics in this session. If none or two were
used, no default is displayed. The valid
responses are:
Response Meaning
ENTER
1
2
3
4
Default (if shown)
Controller A 1
Controller' A2
Controller A3
Controller A4
2.
Enter your response for controller selection.
The MD displays one of the following frames to
prompt for the ID of the routine you want to
run:
*
-
DIAGNOSTIC AIDS *
ENTER ROUTINE ID.
(Z HEX DIGITS)
or-
* DIAGNOSTIC
AIDS *
ENTER ROUTINE ID.
DEFAULT IS: xx
Selection of a routine is always required. If a
routine ID is not specified, the MD provides (as
a default value) the last routine ID specified by
the user.
To select a diagnostic routine. key in the
two-character routine ID of the routine to be
run and then press the ENTER key. The
following routine IDs can be selected:
80
81
82
83
Controller Tests
Device Logic Tests
Servo Tests
Read/Write Data Transfer Tests
84
85
8A
Read/Write Function Tests
Seek Exerciser
Incremental Seek Exerciser
90
91
93
Extended DPS Tests (see note)
HA Scan
DOC Tests (see Warning)
96
97
98
99
Device Status Test (see Warning)
Sense Utility (see Warning)
Error Log Search (see Warning)
Motor Start Test (see Warning)
9C
9D
CE Track Repair
DPS Array Display/Dump Utility (see
Warning)
Diagnostic Microcode Patch (see
Warning)
Note: Selecting controller AS/ A4 is valid only
on strings with four storage paths, and only
when the MD is attached to the A unit that
contains controllers AS and A4. Controllers A 1
and A2 cannot be selected when the MD is
attached to the A unit that contains controllers
AS and A4.
PIN 4519903 84 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
9E
MD-44
9F
(
Set/Reset/Sense CE Mode (see
Warning)
routine is selected, the default is reset to no
device selection.
Note: If the 3380-JK is attached to a
3990, the extended DPS tests (Routine
90) cannot be run.
(
Warning: An attempt to run a diagnostic
routine for a device without specifying a device
for selection causes an error. The error
message that results does not mention that this
error can be caused by user failure to specify a
device. It is the user's responsibility to ensure
that a device is selected when running a device
test under the Diagnostic Aids option.
Warning: Although any of the above routines
may be selected, routines 93, 96, 97,98,99, 90,
9E and 9F provide special purpose functions
and may produce confusing results if run under
the Diagnostic Aids option.
f
Warning: A device specified for selection must
be in CE mode. Failure to set CE mode for the
selected device results in IC 1101 (byte 1 of the
error message is the address of the device
being tested). Use MD Main option 0 to set or
reset CE mode as required.
See the routine descriptions in the DIAG
section of the MSM, Volume R60 for additional
information.
3.
{~
('
Enter your routine selection. Press the ENTER
key to rerun the default or enter the
two-character 10.
*SELECT
DIAGNOSTIC AIDS *
DEVICE(OO-XX)
OR PRESS ENTER IF
NO DEVICE IS NEEDED.
or--
*SELECT
DIAGNOSTIC AIDS *
DEVICE(OO-XX)
DEFAULT IS: xx
(j
('
("
{:
f'
C
Respond to the prompt for device selection.
Enter no more than one device address.
Next, the MD prompts for device information:
-
(
4.
.
SELECT DEVICE (DO-XX) indicates the valid
range of devices that may be selected.
XX
XX
= OF for 2 path strings
= 1F for 4 path strings
When the MD requests that a device be
selected, determine if the routine being run
requires use of a device. Routines 81, 82, 83,
84, 85, 8A, 97, 99, and 9C all require that a
device be selected.
Routines 80, 90, 93, 96, 98, 90, 9E and 9F do not
require device selection but can be run even if
a device is selected.
Once a device is selected, the MO continues to
use that device address as a default value until
a different routine is selected. When a new
PIN 451990384 Sides
EC 476581 01Aug89
(0 Copyright IBM Corp. 1987, 1988
00 - OF for 2 path strings
00 - 1F for 4 path strings
The MO next prompts for parameters:
ENTER PARAMETERS
IF REQUIRED
(SEE PSG MD-3, 800)
Each diagnostic routine except routine 9C has
default parameter values that are correct for
the routine so that the entry of control
parameters is not usually required. However,
in special cases, the user may want to enter
parameters to provide some unusual diagnostic
function. (Routine 9C is equipped with default
parameters that cause an error, preventing the
accidental running of that diagnostic routine.)
The number of parameters required and the
meaning of each parameter (except for
parameter 0) varies depending on the routine
being run.
Each parameter is two hexadecimal digits. Up
to sixteen parameters (numbered 0 to F) may
be entered consecutively with no spaces
between successive parameters. If more than
MD-45
one parameter is entered, then all the
parameters used by the routine must be
entered before the ENTER key is pressed.
(Remaining default parameter values are
ignored after the second parameter is entered.)
The first parameter (parameter 0) is the "run
control" parameter. Parameter 0 provides
control functions that are the same for all
routines. Some of the run control functions are
not appropriate for some of the routines and
can cause unexpected results. See the
descriptions of the routines (in the DIAG
section of the MSM, Volume R60) for
restrictions on the selection of the run control
parameter.
The control functions provided by parameter 0
are:
00
Allow Routine Linking, No Loop,
Halt On Error
01
Allow Routine Linking, No Loop,
Bypass Error Halt
02
Allow Routine Linking, Loop, Halt
On Error
03
Allow Routine Linking, Loop,
Bypass Error Halt
04
Inhibit Routine Linking, No Loop,
Halt On Error
05
Inhibit Routine Linking, No Loop,
Bypass Error Halt
06
Inhibit Routine Linking, Loop, Halt
On Error
07
Inhibit Routine Linking, Loop,
Bypass Error Halt
The Allow Routine Linking function allows
routine 80 to link to routine 96, and allows
routines 81 through 85 to link and run in
sequence.
PIN 4519903 84 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987. 1988
If the Inhibit Routine Linking function is
selected, or if any routine other than 80 through
85 is selected, only the selected routine is run.
The Loop function causes the selected routine
(or the linked series of routines) to be
repeated. The Loop function in combination
with the Allow Routine Linking function causes
rOtitines 81 through 85 to repeatedly run in
sequence.
If the No Loop function is used, the selected
routine or linked series of routines run only
once.
The Bypass Error Halt function allows routine
execution to continue after an error is detected.
This function may be used in combination with
the Loop function to create a scope loop.
If the Halt On Error function is selected, routine
execution stops and an error message is
displayed when an error is detected.
If only the first parameter (Parameter 0) is
entered, the default values built into the routine
being run are used for the remaining
parameters. This permits the user to specify
only the run control parameter if desired. If
more than one parameter is entered, then all
parameters applicable to the routine being run
should be entered. If too few parameters are
entered, the MD supplies X"OO" for the
remaining parameters. If too many parameters
are entered, the excess parameters are
ignored.
The meaning of parameter entries 1 through F
varies,depending on the routine being run.
See the routine description in the DIAG section
of the MSM, Volume R60.
After parameters have been entered, the MD
uses those parameters as default values until a
different routine is selected. When a new
routine is selected, the default parameter
values are reset to the built-in default values
for the new routine.
MD-46
',,- /
(
Warning: Entry of parameter values that are
incorrect or inappropriate for the routine being
run can cause some very confusing or
misleading results.
(
7.
The MD displays the following frame:
* DIAGNOSTIC AIDS *
ROUTINE ** RUNNING
ON CONTROLLER x,
DEVICE xx.
(**3E*)
See the description for the routine being run for
the parameters applicable to that routine.
(
5.
(
Press the ENTER key to use the default
parameter values, or enter the run control
parameter (parameter 0) only and use the
remaining default parameters, or enter all of
the parameters and then press the ENTER key.
Immediately after this message is displayed,
the selected inline diagnostic routine is started
using the specified controller and device.
No response is required, but the keyboard is
enabled. If any MD key is pressed while this
message is being displayed, the routine being
executed stops.
If you enter no parameters, the MD displays the
following frame. (If parameters were entered,
the MD bypasses the next two frames.)
(
The (****) on the bottom line of the above
frame is the dynamic display area for those
diagnostics that present dynamic displays.
DO YOU WANT TO LOOP
THE ROUTINE !
When the diagnostic ends without errors, the
MD displays the following frame:
Press the YES key or the NO key to respond:
YES
NO
c
6.
Enter your response about stopping on errors.
Looping is desired. The MD
displays the frame that follows.
Looping is not desired. The routine
will stop if an error occurs. The MD
bypasses the next frame shown in
this sequence.
ROUTINE xx ENDED.
DO YOU WANT TO RUN
THE SAME TEST AGAIN!
Press the YES key, the NO key, or the PF key to
respond:
Enter your response about routine looping.
YES
NO
The following frame appears only if no
parameters were entered and routine looping
was specified:
DO YOU WANT TO HALT
WHEN AN ERROR
OCCURS !
PF
Repeat the same routine.
Return to the beginning of the
Diagnostic Aids option.
Return to the MD Main menu.
8.
Respond to the question by pressing the YES
key or the NO key.
9.
Press the PF key to return to the MD Main
menu.
Press the YES key or the NO key to respond:
c
('
YES
NO
Stopping is desired when an error is
detected.
The routine is to continue when an
error is detected.
/
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
MD-47
Error Display
If an error is detected during the execution of the
diagnostic routine (and stopping on errors is
permitted), the MD displays the following frames:
r:",
\·'""'O.. ~.•/
* DIAGNOSTIC AIDS *
RTN=**
IC=****
EB=** **** **** ****
**** **** **** ****
PRESS ENTER TO RESTART OR "A" TO
ANALYZE. PF KEY TO
RETURN TO MAIN MENU.
--,,/
= Diagnostic routine ID
= Isolation code
RTN
IC
EB
= Error bytes
Note: The error bytes are used by the MD for FRU
isolation. The guided maintenance procedures do
not require you to use these bytes for any purpose.
They are displayed primarily for support personnel.
/ - ...... ,
If you are preparing to discuss this problem with
support personnel, you may want to record these
error byte values.
See the DIAG section in the MSM, Volume R60, and
the IC section in the ECM, Volume R40 for more
information about this display.
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 19B7, 19BB
MD-48
(
(
Review Option List (Main Menu Option R)
From the MD Main menu, s"llect option R.
REVIEW CE MODE (Option 0)
The MD displays the following frames:
Use the review CE mode option to display the
addresses of devices that are in CE mode. The MD
must be attached to the 3380-JK.
SELECT REVIEW OPTION
FROM FOLLOWING LIST.
(SEE PSG MD-3, 700)
O-REVIEW CE MODE ...
('
I-REVIEW
2-REVIEW
3-REVIEW
4-REVIEW
1.
DIAGNOSTICS
ERROR LOG
REPAIR
HA SCAN
From the MD review options menu, select
option O.
The MD displays one of the following frames:
SELECT CONTROLLER:
('
('
The following options are valid.
(SEE PSG MD--3, 020)
o-
-
Review CE mode
A diagnostic routine will be invoked to find
out which devices are currently set in CE
mode. See page MD-49.
SELECT CONTROLLER:
DEFAULT IS: x
(SEE PSG MD-3, 020)
1 - Review diagnostics
The x shown in the preceding frame identifies
the last controller used to run diagnostics in
this session. If none or two were used, no
default is displayed.
The last 4 diagnostic errors are saved on the
MD diskette. The most recent error is shown
first and subsequent errors can be displayed
by entering uN" (Next). See page MD-50.
('
(~:
('
(~~
(~
or-
The valid responses are:
2 - Review error log
Use Review Error Log to go directly to the
Search Log Function of the MD. For a
description of how to use the Search Log
Function, see page LOG-S.
Response Meaning
ENTER
1
2
3 - Review repair
Display the last repair specified by a
maintenance analysis procedure.
See page MD-SO.
3
4
Default (if shown)
Controller A 1
Controller A2
Controller A3
Controller A4
Note: Responses 3 and 4 are valid only on
strings with four storage paths when tile MD is
attached to the unit that contains controller A3
and A4.
4 - Review HA Scan
Display the HA Scan results from the last HA
Scan that was run. See page MD-51.
2.
c
Enter your response for controller use.
Which controller is specified makes no
difference if both controllers are operational.
3.
The MD advances to the following frame:
c'
(~
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
MD-49
CE MODE IS BEING
SENSED.
PLEASE WAIT.
Immediately after this message is displayed, an
inline diagnostic routine is started (or
restarted) using the specified controller. If the
routine is completed without errors, the MD
displays the CE mode status of the devices:
REVIEW DIAGNOSTICS (Option 1)
Use review option 1 to display the last 4 errors
detected by the diagnostic routines or by the MD.
The MD need not be attached to the 3380-JK.
1.
From the MD review options menu, select
option 1.
The MD displays the following frames:
2-path subsystem
DEVICES IN CE MOD~:
OX: .. 23 ........... .
(SEE PSG MD-3, 035)
LAST ERROR DETECTED
BY DIAGNOSTICS (OR
MD) WAS AS FOLLOWS:
/~-
-
or-
4-path subsystem
DEVICES IN CE MODE:
OX: .. 23 ........... .
IX: 01 ............. .
(SEE PSG MD-3, 035)
-
,
IC**** CTL=A* DEV=**
RTN=** TEST=**
EB=** **** **** ****
**** **** **** ****
PRESS ENTER TO
ANALYZE OR "N" TO
DISPLAY NEXT ERROR.
(SEE PSG MD-3, 742)
or-
2.
CE MODE IS RESET
FOR All DEVICE
ADDRESSES.
Some problems with the storage control, with
the controller, or with the control interface will
not allow the diagnostic routine to run and CE
mode is not sensed. The MD displays an error
message. Analyze that error and correct the
problem before continuing.
Follow the directions displayed by the MD.
,/'
"-..
/
REVIEW ERROR LOG (Option 2)
Use Review Error Log to go directly to the Search
Log Function of the MD. For a description of how to
use the Search Log Function, ~ee page LOG-S.
REVIEW REPAIR (Option 3)
Select option 3 to display the last repair (FRU
group) that was displayed. The MD does not need
to be attached to the 3380-AJ4 or AK4 unit.
The last repair displayed is a result of the last
symptom analyzed. The FRU group displayed may
be from analysis of one of the following:
•
•
•
Diagnostic error
Symptom code
Error Log Entry
LAST REPAIR
DISPLAYED BY ERROR
ANALYSIS PROCEDURES
WAS AS FOllOWS:
PIN 4519903 84 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
\,..
MD-50
((--
Following this display, the primary FRU group, the
secondary FRU group, and any additional action
are displayed.
** ARM x **
HEAD:
xx xx xx xx
ERRORS: xx xx xx xx
REVIEW HA SCAN (Option 4)
= Number of errors for each head
ERRORS
Select option 4 to review HA Scan results from the
last time the HA Scan diagnostic was run. The MD
does not need to be attached to the 3380-AJ4 or
3380-AK4 unit. One of the following frames is
displayed:
* SCAN
NO ERROR
REVIEW *
DATA STORED
PRESS PF KEY TO
RETURN TO MAIN MENU
-
or-
*xxSCAN
REVIEW *
TRACKS HAD
DATA CHECKS
SEE PSG MD-3, 395 ...
("~
c
c
c
(j
xx indicates the number of tracks, in hexadecimal,
that had errors. The maximum number of tracks
that can be stored by the MD is 80 (X"50")
After the scan is recalled from the MD diskette,
summary or detailed information may be displayed
by selecting the display mode from this menu:
SELECT DISPLAY MODE:
I-HEAD/ARM SUMMARY
2-TRACK SUMMARY
3-TRACK DETAIL
4-RUN HA SCAN
AGAIN
5-RETURN TO MAIN
MENU
Display Modes
After displaying summary or detail information, the
MD returns to this menu and another mode may be
selected.
c
Display Mode 2: TRACK SUMMARY
Presents one or more displays with a one line
entry for each track that had errors:
CYL=xxx
CYL=xxx
CYL=xxx
CYL=xxx
HD=x
HD=x
HD=x
HD=x
CNT=x
CNT=x
CNT=x
CNT=x
CYL = Cylinder number (hexadecimal)
HD = Head number (hexadecimal)
CNT = Number of errors (hexadecimal)
Display Mode 3: TRACK DETAIL
Presents one display for each track in error.
There are two types of error displays:
** DATA CHECK **
HOME ADDRESS AREA
CYL = xxx
HEAD = x
ERROR COUNT = x
-
or-
NO SYNC BYTE FOUND
HOME ADDRESS AREA
CYL = xxx HEAD = x
ERROR COUNT = x
CYL = Cylinder number (hexadecimal)
HEAD = Head number (hexadecimal)
ERROR COUNT = Number of errors
(hexadecimal)
Display Mode 4: RUN HA SCAN AGAIN
Repeat the Review HA Scan option.
Display Mode 5: RETURN TO MAIN MENU
Return to the MD Main Menu.
Display Mode 1: HEAD/ARM SUMMARY
This display mode presents a summary by
head of the number of tracks in hexadecimal
that had data checks. The summary is
presented in four displays, one for each head
arm:
{:'
PIN 4519903 84 Sides
© Copyright
(-
EC 476581 01Aug89
IBM Corp. 1987, 1988
MD-S1
010 SELECT AN OPTION
This is the MD Main Menu. Valid responses are:
o=
1
R3
SET/RESET CE MODE
Devices must be set in CE Mode before
diagnostics are run. CE mode must be reset
before they are returned to customer
operation. More than one device may be
placed in CE mode at a time.
=
R4 =
Review the last repair that was
displayed by an error analysis
procedure.
Review HA scan results from the
last time HA SCAN was run.
For additional information, see Review Option
List on page MD-49.
= RUN DIAGNOSTICS
Used to select the diagnostic option menu.
020 SELECT CONTROLLER
2 = ERROR LOG FUNCTION
Used to set the log mask, transfer the log, or
search the log previously transferred to the
MD, or erase the error log in the specified
storage path or storage director.
The selected controller is used to run the
diagnostic that sets, resets, or provides for
reviewing devices in CE mode.
3 = ANALYZE SYMPTOM CODE
Used for entering symptom codes for
analysis. The MD does not need to be
attached to the 3380.
Response Meaning
Z = END OF CALL
References end of call procedure and resets
CE mode for all devices.
o
= DIAGNOSTIC AIDS
Provides for manually controlling diagnostic
execution.
R
= Review (0 • 4)
Valid responses are:
ENTER
1
,2
3
4
Default (if shown)
Controller A 1
Controller A2
Controller A3
Controller A4
Note: Selecting controller A3 or A4 is valid only on
strings with four storage paths, and only when the
MD is attached to the A unit that contains
controllers A3 and A4. Controllers A 1 and A2
cannot be selected when the MD is attached to the
A unit that contains controllers A3 and A4.
Used to review the following:
Enter your response for controller use.
RO =
Review which devices are in CE
mode.
R1 =
Review the last 4 diagnostic
errors. Error analysis procedures
may be invoked after the error is
displayed.
R2
=
Review the last error log entry that
was previously displayed or
display the error log summary that
was previously transferred from
the storage control. Error analysis
procedures may be invoked from
the error display.
PIN 451990384 Sides
tl::l Copyright
(
EC 476581 01Aug89
IBM Corp. 1987, 1988
MD-S2
/'
(
030 SET I RESET CE MODE
(
f-
CE mode must be set for any given device before
device diagnostics can be run on that device.
Device diagnostics do not run on a device that is
not in CE mode.
Before a device is set into CE mode, the device
should be varied offline to each attaching system.
CE mode must be reset for any given device before
the device is returned to the customer or varied
on-line.
If a system attempts to use a device that is not
varied offline and is in CE Mode, the storage
control reports Intervention Required to that system
for that device. When CE mode is reset for a
device, an interrupt is generated. The interrupt
should restore normal operation.
(
(
(
(~
Setting CE Mode
ENTER 2-DIGIT ADDR.
OF DEVICE(S) TO BE
SET IN CE MODE.
(SEE PSG MD-3, 030)
There are several ways to enter device addresses.
Two digits must be used for each address entered.
•
Enter a single device address, "00" through
"1F."
•
Enter multiple device addresses separated with
a single space (for example, 02 05 07 OA),
•
Enter a range of addresses separated with the
character "l" (for example, 02T07).
•
Enter multiple ranges of addresses separated
with a space (for example, 02T04 08TOB).
•
With "Reset CE Mode" selected, press the
ENTER key without specifying devices to reset
CE mode for all devices that were previously
set in CE mode.
CE Mode will be set for the specified device(s) on
all paths. More than one device may be set in CE
Mode. All device addresses in the range 00
through OF are valid. Device addresses in the
range 10 through 1F are valid for 4-path strings.
Devices previously set in CE mode remain in CE
mode.
Resetting CE Mode
ENTER 2-DIGIT DEVICE
ADDRESSES OR PRESS
ENTER TO RESET ALL.
(SEE PSG MD-3, 030)
( ""'
j
CE Mode will reset for the selected device(s) on all
paths. More than one device may be reset. All
device addresses in the range 00 through OF are
valid. Device addresses in the range 10 through 1F
are valid for 4-path strings.
(~.
c
c
PIN 4519903 84 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1987, 1988
(-
MD-53
r"
035 DEVICES IN CE MODE/DEFAULT DEVICES
When CE Mode has been reset
The MD displays the following frame:
CE MODE IS RESET
FOR ALL DEVICE
ADDRESSES.
~i
When default devices are listed
The MD displays one of the following frames:
2-path subsystem:
DEFAULT DEVICES
OX:
•• 23 ••••••••••••
(SEE PSG MD-3, 035)
When CE Mode has been set
The MD displays one of the following frames:
2-path subsystem:
DEVICES IN CE MODE:
OX:
•• 23 ••••••••••••
-
or-
4-path subsystem:
DEFAULT DEVICES
OX:
•• 23 ••••••••••••
IX: 01 .....••....•..
(SEE PSG MD-3, 035)
(SEE PSG MD-3, 035)
-
or-
4-path subsystem:
DEVICES IN CE MODE:
OX:
In the frames above, devices 02 and 03 are
default devices. Devices 00, 01, and 04
through OF are represented by dots and are
not default devices.
'"",-_./
•• 23 .•••••••••••
IX: 01 .......•......
(SEE PSG MD-3, 035)
In the frames above, devices 02 and 03 are in
CE mode. Devices 00, 01, and 04 through OF
are represented by dots and are not in CE
Mode.
In the 4~path display, devices 02, 03, 10, and
11 are in CE Mode. Devices 00, 01, 04
through OF, and 12 through 1F are
represented by dots and are not in CE Mode.
PIN 451990384 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
In the 4~path display, devices 02, 03, 10, and
11 are default devices. Devices 00, 01, 04
through OF, and 12 through 1F are
represented by dots and are not default
devices.
Default devices are those that were specified
during the previous diagnostic. If no device
has been tested during the current diagnostic
session, there are no default devices.
MD-S4
/
,
/",
\
('
050 DEFINE THE FAILURE BOUNDARY
(
MD 0050:
(/
The DASD Symptom Code Summary report is
divided into sections by PFU. The PFUs are:
Channel, Storage Control Unit, Controller,
Multiple, Device, Volume and Unknown. The
PFU sections are included in the report by
order of importance, with the most important
PFU listed first.
The best method for defining the failure boundary
is to use the DASD Symptom Code Summary report
(see "DASD Symptom Code Summary" on
page PROG-17).
(
If the DASD Symptom Code Summary report is not
available, the failure boundary can be defined
using multiple sense records from other sources.
These sources include console messages, other
EREP reports, and the error log in the storage
control. If the problem produces a solid failure, the
failing devices and paths can usually be
determined by running the Device Status Test with
each controller.
(
(~
Is there a DASD Symptom Code Summary report
available?
Yes No
I
(-~
c
002
Go to Step 004 in this MD.
003
A DASD Symptom Code Summary report is
available:
1.
Determine the highest priority symptom code
obtained from SMAP 0025 Step 001 on page
START-12.
2.
Determine the highest priority symptom code
group. The highest priority symptom code
group includes the highest priority symptom
code and all other symptom codes with the
same first character. Include all permanent and
temporary error records.
f,'"
,/
(
the symptom code group.
001
(From RMAP 0340 Step 003 on page REPAIR-13)
If there are symptom codes listed under more
than one PFU, select the PFU listed first.
5.
Use the following guidelines to define the
failure boundary.
PFU = CHANNEL: Indicates that the probable
failing unit and failure boundary is the channel or
storage control. You should not be starting
analysis with this maintenance package.
PFU = STORAGE CONTROL UNIT: Indicates that
the controller is the probable failing unit.
Note: If the failure affect column indicates
SCUJCTLR, a PFU of Storage Control Unit indicates
that both controllers (0 and 1) on the control
interface are reporting errors.
=
PFU
CONTROLLER: Be careful if the symptom
code is DF05. DF05 can be caused by the customer
switching off controller power and may not have
anything to do with the real problem.
Look at the FAILURE AFFECT column and assign
the failure boundary in the following way. If the
FAILURE AFFECT is:
•
CTLR - the failure boundary is controller.
•
SCu/CTLR - The failure boundary is controller.
•
CTLRlDEV - The failure boundary is:
"'"
j
3.
c
4.
Examine all entries in the physical address
column for all symptom codes in the group to
identify the failing device addresses. Refer to
SMAP 0030 Step 001 on page START-14.
Determine the probable failing unit (PFU) for
PIN 451990384 Sides
EC 476581 01AugS9
© Copyright IBM Corp. 1987, 1989
Drive - if only two devices are listed in the
physical address column and both devices are
MD-55
\""'--'-'
on the same HDA.
Port - if more than one device on more than
one HDA is listed in the physical address
column, all the devices are in the same unit
and the string contains at least one B unit.
004
(From Step 002 in this MD)
If sense data Is available, determine the failure
boundary using the following procedure:
1.
Determine the highest priority Symptom Code
obtained from SMAP 0040 Step 001 on page
START-19 or from IMAP 0233 Step 003 on page
ISO LATE-20.
2.
Determine the highest priority symptom code
group. The highest priority Symptom Code
group includes the highest priority symptom
code and all other symptom codes with the
same first character. Include all permanent
and temporary error records.
Controller - if multiple devices are listed in the
physical address column and they are on more
than one port (there must be at least one B unit
in the string).
Do not know - if other.
PFU = MULTIPLE: Indicates that multiple devices
may be involved. The failure boundary is:
Device - if only one device is listed in the
physical address column.
Drive - if only two devices are listed in the
physical address column and both devices are
in the same HDA.
Port - if there are more than two devices listed
in the physical address column and all the
devices are in the same unit.
_/
3.
Examine all sense records in the symptom
code group to identify the failing device
addresses. Ref~r to SMAP 0045 Step 001 on
page START-21.
I\....- .,.-'
\,,---,/:
If sense data is not available and the device status
test indicates failing devices, the device status test
results may be used to determine the failure
boundary. See IMAP 0232 Step 001 on page
ISOLATE-18.
Do Not Know - ifother.
PFU
=
DEVICE: The failure boundary is normally
device, however, if the FAILURE AFFECT is
CTLR/DEV, EREP has determined that only one
device is failing. The failure boundary is device.
However, consider that there is a slight chance that
the controller is failing and that because of low
usage of other devices on the string, only one
device reported the error.
Use the DASD String Summary report to determine
usage.
Select the failure boundary from the following list,
then r- urn to the procedure that sent you here:
4
\'-.:.._.,1
Device
A failure of only one device or access
mechanism, or the failure of the media
associated with that device or access
mechanism, regardless of which controller
paths are failing.
Drive
A failure of two devices in a single
head-disk-assembly (HDA) regardless of
which controller paths are failing.
PFU = UNKNOWN: Indicates that the failure
boundary is unknown.
Port
A failure of two or more devices on different
HDAs and limited to a single port (1 unit)
regardless of which controller paths are
failing (see note).
Note: If only one port is in use, the
PIN 4519903 84 Sides
EC 476581 01Aug89
MD-56
~ Copyright IBM Corp. 1987. 198B
,/
determination between port and controller
cannot be made. Specify the failure boundary
as "DO NOT KNOW."
(,
(
('
Controller
A failure of two or more devices on two or
more ports and limited to a single controller
(see note).
Note: If only one controller is in use, the
determination between port and controller
cannot be made. Specify the failure boundary
as "DO NOT KNOW."
Do not know
The failure boundary cannot be determined.
(~
c
c
c\
c
PIN 4519903 84 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1985
MD-S7
051 DEFINE THE FAILURE BOUNDARY (EBXX)
MD 0051:
001
•
Substring 1 contains ports 4-7 (devices 10-1F)
(From SMAP 0057 Step 001 on page START-33)
The best method for defining the failure boundary
is to use the DASD Symptom Code Summary report
(see "DASD Symptom Code Summary" on
page PROG-17).
If the DASD Symptom Code Summary report is not
available, the failure boundary can be defined
using multiple sense records from other sources.
These sources include console messages, other
EREP reports, and the error log in the storage
control. If the problem produces a solid failure, the
failing devices and paths can usually be
determined by running the Device Status Test with
each controller.
Is there a DASD Symptom Code Summary report or
multiple sense records with an EBXX symptom
code?
Yes No
I
002
Go to Step 015 in this MD.
If devices in the range of OO-OF are failing, then
Substring 0 is failing. If devices in the range of
10-1 F are failing, then Substring 1 is failing.
•
10 = Sub-string 0
01 = Sub-string 1
•
The port is determined from XX, bits 4 - 7
SUB-STRING a
SUB-STRING 1
8
9
A
B
4
5
6
7
= Port a
(1000)
(1001)
(1010)
(1011) =
= Port
= Port
1
2
Port 3
=
(0100)
Port
(0101)
Port
(0110) = Port
(0111)
Port
=
=
4
5
6
7
Look at the EBXX symptom codes then define the
failing substrings and ports using the above
information. Use the following procedure to define
the failure boundary:
Is there more than one value of XX?
Yes No
I
003
The EBXX Symptom Code represents a failure of
the Controller/Device Port (CDP) Interface.
Defining the failure boundary correctly requires an
EREP System Exception report, a DASD Symptom
Code Summary Report, or multiple sense records.
The failure boundary cannot always be determined
from the Physical Address column in the Symptom
Code Summary, or from Sense Byte 4. The EBXX
Symptom Code must be used. The EBXX Symptom
Code contains substring select bits and port select
bits. These bits will define the failing substrings
and ports.
2-Path strings contain one substring (substring 0).
4-path strings are divided into two substrings
(substring 0 and substring 1).
•
For symptom code EBXX, the substring is
determined from XX, bits 4 - 5
004
Go to Step 008 in this MD.
005
Does the subsystem have 4 paths?
Yes No
\~'
I
006
The failure boundary is "CONTROLLER."
007
The failure boundary is "MULTIPLE PORTS."
008
(From Step 004 in this MD)
(Step 008 continues)
Substring 0 contains ports 0-3 (devices OO-OF)
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
MD-58
('
008 (continued)
Determine the failing device addresses. Look at
the Physical Address column of the Symptom Code
Summary Report or Sense Byte 4, bits 3 - 7.
(
Is there more than one device failing?
Yes No
I
019
Go to Step 020 in this MD.
020
(From Step 019 in this MD)
Is there more than one device failing?
Yes No
I
009
The failure boundary is "DEVICE."
010
Is there more than one drive failing?
Yes No
(
I
{'
011
The failure boundary is "DRIVE."
('
{
022
Determine if there is more than one drive failing. A
drive is two devices on the same HDA.
Is there more than one drive failing?
Yes No
I
012
Are there any a-units in the substring?
Yes No
I
013
f
The failure boundary is "DO NOT KNOW."
('
021
The failure boundary is "DEVICE."
014
The failure boundary is "PORT."
023
The failure boundary is "DRIVE."
024
Determine which substrings are failing. The failing
substring can be determined as follows:
•
•
015
Substring 0: If devices in the range OO-OF are
failing.
Substring 1: If devices in the range 10-1F are
failing.
(From Step 002 in this MD)
The Device Status test should have been run to
determine which devices and paths are failing.
Has the Device Status test been run?
Yes No
I
(
016
Go to IMAP 0232 Step 001 on page
ISOLATE-18.
f
017
('
Did any devices report a status of "E"?
Yes No
Determine if there are any a units in the substring.
The following question pertains to any B units in
the substring, whether they are failing or not.
Are there any failing a units in the failing
substring?
Yes No
I
025
The failure boundary is "DO NOT KNOW."
026
Determine which ports are failing. Use the following
chart:
I
(
'
018
./
The failure boundary is "DO NOT KNOW."
(Step 019 continues)
c
("
PIN 4519903 84 Sides
© Copyright
EC 476581 01 Aug89
IBM Corp. 1987, 1988
MD-59
Substring 0:
Substring 1:
Failing
Devices
Failing
Devices
Port
14-17
18-1B
1C-1F
5
6
Port
-------------00-03
0
04-07
08-0B
DC-OF
-------------4
10-13
1
2
3
7
Are all the failing devices on one port?
Yes No
I
027
The failure boundary is:
•
.. /
"CONTROLLER" in a 2-path string.
'",-
•
.,./
"MULTIPLE PORTS" in a 4-path string.
028
The failure boundary is "PORT."
\.,~
PIN 4519903 84 Sides
© Copyright
EC 476581 01 Aug89
IBM Corp. 1987, 1988
MD-60
./
(
(:
(
052 DEFINE THE FAILURE BOUNDARY (DATA CHECKS)
(
MD 0052:
001
The following procedure is used to define the
failure boundary for data checks (symptom code
4XXX) and for a no data found condition (symptom
code EDOD).
The best method of defining the failure boundary is
by using the EREP DASD Symptom Code Summary
report. See page PROG-17 for a sample of this
report.
(
(/
If this report is not available, then sense data from
other sources such as EREP Event History report,
Console messages, or ICKDSF messages should
be used.
If sense data is not available and the diagnostics
produce a solid failure, then diagnostic results can
be used.
(.~
Is an EREP Symptom Code Summary report or
sense data available?
Yes No
I
Port A failure of two or more devices on
different HDAs and limited to a single
port (1 unit), regardless of which
controller paths are failing (see note).
Note: If only one port is in use, the
determination between port and
controller cannot be made. Specify the
failure boundary as "CONTROLLER."
Controller A failure of two or more devices
on two or more ports (see note).
Do not know The failure boundary cannot be
determined.
003
Is an EREP Symptom Code Summary report
available?
Yes No
002
I
Use diagnostics and the following procedure
to define the failure boundary.
004
Examine all sense records with symptom
codes 4XXX or EDOD to identify the failing
device addresses and failing paths.
Determine the failing device and path from
sense byte 4 using the following list:
If diagnostics do not fail consistently, the
failure boundary cannot be defined.
If not already done, run the diagnostics on
different devices with each controller as
necessary
Select the failure boundary from the
following list, then return to the procedure
that sent you here.
C·
Drive A failure of two devices in a single
head-disk-assembly (HDA) regardless of
which controller paths are failing.
Device A failure of only one device or access
mechanism, or the failure of the media
associated with that device or access
mechanism, regardless of which
controller paths are failing.
/
PIN 4519903 84 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1987, 1988
Sense Byte 4
OOxx xxxx
01xx xxxx
10xx xxxx
11xx xxxx
Path
Path
Path
Path
Path
Sense Byte 4
xxOx xxxx
xx1x xxxx
String Address
String 0
String 1
Sense Byte 4
xxxO 0000
xxxO 0001
Device Address
Device 00
Device 01
0
1
2
3
MD-61
xxxO 0010
xxxO 0011
Device 02
Device 03
xxx11110
xxx1 1111
Device 1E
Device 1F
Select the failure boundary from the
following list, then return to the procedure
that sent you here.
Device A failure of only one device or access
mechanism, or the failure of the media
associated with that device or access
mechanism, regardless of which
controller paths are failing.
Drive A failure of two devices in a single
head-disk··assembly (HDA) regardless of
which controller paths are failing.
Port A failure of two or more devices on
different HDAs and limited to a single
port (1 unit) regardless of which
controller paths are failing (see note).
Note: If only one port is in use, the
determination between port and
controller cannot be made. Specify the
failure boundary as "CONTROLLER."
Controller A failure of two or more devices
on two or more ports (see note).
Do not know The failure boundary cannot be
determined.
005
Examine all entries in the physical address column
for all symptom codes 4XXX or EDOO, to identify the
failing device addresses. Refer to SMAP 0030 Step
001 on page START-14.
The physical address is presented as CC.P-DD,
where:
=
(CC) in the Physical Address column with the
string ID on the A unit operator panel.
•
To identify the controller, match the path bit (P)
in the Physical Address column with the path
character on the A unit operator panel.
•
To identify devices, match the device address
(DD) in the Physical Address column with the
device addresses on the A unit operator panel.
Define the failure boundary as follows:
:
Device
i, 0nly one device is listed in the physical
address column.
Drive
If only two devices are listed in the physical
address column and both devices are in the
same HDA.
Port
If one of the following conditions exists:
•
Multiple devices are listed in the physical
address column, all the errors are related
to one unit, and multiple controllers are
reporting errors.
•
Multiple devices are listed in the p~ysical
address column, all the errors are related
to one unit, only one controller is
reporting errors, and there is at least one
B unit in the string.
If there are no B units in the string and only
one controller is reporting errors, select a
failure boundary of controller.
Controller
If multiple devices are listed in the physical
address column and they are on more than
one port, or, there is only one controller
reporting errors and there are no B units on
the string.
CC
The physical 10 of the string
P = Storage Path (0-3)
DD
The physical ID of the device
c
=
•
To identify the string, match the controller ID
PIN 451990384 Sides
EC 4765P1 01Aug89
~ Copyright IBM Corp. 1987, 1988
"'-j
MO-62
j
(
090 WRITE INHIBIT
092 STORAGE PATH FENCED
(--
When the set/reset CE mode diagnostic resets CE
mode for all devices, it checks the storage path or
storage director that is attached to the selected
controller to determine if the controller, or any of
the attached system channels, are write inhibited.
Fencing is a technique used by 3990 storage path
functional microcode to remove a failing path from
system operation. This prevents path-related
failures from degrading subsystem performance.
(
c-
If a write inhibit condition exists, one of these
messages is displayed:
•
NOTE: WRITE INHIBIT IS SET FOR
CONTROLLER A"1~
Write inhibit is set for a controller if a controller
error is detected during a write operation.
•
NOTE: WRITE INHIBIT IS SET FOR
CONTROLLER A*
CHANNELS **
Write inhibit is set for a channel if a storage
control or channel error is detected during a
write operation.
Write inhibit prevents all further write operations
through the write inhibited controller or channel.
Storage Path Fenced is set when there are
excessive storage path errors or when a Quiesce is
set from the Storage Control Support Facility.
Diagnostics do not cause Storage Path Fenced
conditions to be set.
Do not confuse Storage Path Fenced with Control
Interface Fencing. Control Interface Fencing is a
hardware fenced condition where controller drivers
are disabled. Storage Path Fenced is a conqJtion
that exists in storage control microcode only.
Storage Path Fenced may be reset as follows:
•
MD Main Menu option Z: End of Call.
•
St(lrage Path Restart switch on the Storage
Control.
•
By performing a RESUME operation on the
quiesced storage path. See MLX Exit 8 on
page MLX-3.
Write Inhibit conditions may be reset as follows:
•
Storage path restart switch on the Storage
Control (3990 only)
•
By issuing the ICKDSF CONTROL command
with the CLEARFENCE parameter specified.
•
MD Main Menu option Z: End of Call
•
By IMUng the storage path.
•
ICKDSF CONTROL command with the
ALLOWWRITE parameter specified.
({
('
(
('
c
When a write inhibit condition is indicated for a
controller or one or more of its associated
channels, it is a good idea to check the other
controller(s) to see if a similar condition exists.
Use MD Main Menu option 0 (SET/RESET CE
MODE), select the other controller(s) and reset CE
mode for all devices. If the controller(s) or any of
its associated channels are write inhibited, this
display will be repeated.
PIN 4519903 84 Sides
EC 476581 01Aug89
~ Copyright IBM Corp. 1987, 1988
MD-63
o
093 FENCED DEVICES
Fencing is a technique used by 3990 storage path
functional microcode to remove a failing path from
system operation. This prevents path-related
failures from degrading subsystem performance.
Device Fence is set for a device when there are
excessive controller errors for a single path to the
device. The device remains operational on other
paths.
Do not confuse Device Fence with Control Interface
Fencing. Control Interface Fencing is a hardware
fenced condition where controller drivers are
disabled. Device Fenced is a condition that exists
in storage control microcode only.
Device Fence may be reset as follows:
•
MD Main Menu option Z: End of Call
•
Storage Path Restart switch on the Storage
Control
•
By performing a RESUME operation on the
quiesced storage path
•
By ICKDSF CONTROL command with the
CLEARFENCE parameter specified.
•
By IMUng the storage path
100 DIAGNOSTIC MENU
Select one of the following options:
o=
Normal Sequence
Runs the controller test on the selected
controller(s), and runs the device test on the
selected device(s). Customer operation can
continue on devices not selected.
2 = Extended DPS test (3880 Storage Control only)
Runs the extended DPS test. Devices do not
have to be in CE mode to run this test. The
extended DPS test runs inline with customer
operation.
3 = Device tests
Runs the device tests on the selected
device(s). Customer operation can continue
on devices not selected.
4 = Connection test
Checks out the communication between the
MD, the 3380 controller, and the Storage
Control. The connection test does not run
concurrently with customer operation on the
selected controller.
5 = HA Scan
Scan home address on the specified device.
(\
\
6 = Motor test
Both devices on the drive must be selected.
Customer operation can continue on devices
not selected. See page PWR-1, Entry AO, in
the MIM, Volume R10.
7 = Device Status
Tests the controller and collects the status of
all devices tn the string without putting the
devices in CE mode. The device status test
r:. rlS concurrently with customer operation.
8 = CE Track Repair
Writes all the home addresses on the CE
tracks and the diagnostic tracks. Checks
each track for defects and skips any defects
that are found. Customer operation can
continue on devices not selected.
1 = Controller test
Runs the controller test on the selected
controller(s). Devices do not have to be in CE
mode to run this test. The controller test runs
inline with customer operation.
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
MD-64
'
'-~
c
(
('
(
(
200 SELECT CONTROLLER
210 SELECT DEVICE(S)
A default controller is displayed if a controller was
specified during a previous operation. No default
is displayed if the MD has just been IMLed.
Respond with one of the following:
•
Specify device to be tested (for example, "03"
= Device 3 will be tested.)
•
Specify devices with a space in between (for
example, "02 04 OS" = Devices 2, 4 and Swill
be tested.)
•
Specify a single range of devices (xxTyy) (for
example, "02TOS" = Devices 2, 3, 4 and Swill
be tested.)
•
Specify multiple ranges of devices (xxTyy
xxTyy) (for example, "02TOS OCTOE" = Devices
2, 3, 4, and S, and devices C, D, and E will be
tested.)
•
Press the ENTER key to use the default devices
(if any). Devices previously used are displayed
as default devices. No device will be shown as
a default device if the MD has just been IMLed.
Valid responses are:
(
Response Meaning
(
(
ENTER
1
2
3
4
12
34
1234
(
(
(
(
(
(
default (if default is displayed)
Controller A1
Controller A2
Controller A3
Controller A4
Controllers A 1 and A2 (note)
Controllers A3 and A4 (note)
Controllers A1, A2, A3, and A4 (note)
Note: Multiple controllers are valid only
when using diagnostic menu option 0, 1, or 3.
Selecting controller A3 or A4 is valid only on
strings with four storage paths. The MD will
instruct you when it is necessary to move the
MD plug.
For diagnostic menu options 0, 1, or 3, more than
one controller may be specified. If controllers in
both A-units of a 4-path string are specified, the MD
stops after testing is completed with the A-unit that
is attached to the MD. The MD instructs you to
move the MD connector to continue testing with the
adjacent A-unit. Do not power off the MD when
moving the connector.
205 INVALID CONTROLLER SELECTED
The controller selected is not valid for this
subsystem configuration. For example, controller
A3 or A4 are valid only if the subsystem is a 4-path
configuration.
Press the PF key to return to the MD Main Menu.
Note: If diagnostic menu option 6 (motor
start test) or option. A (non-repeatable runout
test) was selected, only 1 device address can
be entered. Either device address associated
with the drive can be used.
215 INVALID DEVICE SELECTED
The device selected is not valid for this subsystem
configuration. For example, device 1F is valid only
if the subsystem is a 4-path configuration.
Press the PF key to return to the MD Main Menu.
310 DO NOT FORGET TO RESET CE MODE
Devices in CE mode must be reset before returning
them tl" the customer. Use the MD Main Menu
option 0 or Z to reset devices in CE mode.
(
(('~
{,.(
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1967, 1966
MD-65
320 DO YOU WANT TO RUN
CONNECTION TESTS?
330 WHICH CONTROLLER CONNECTION
DO YOU WANT TO TEST?
Warning: The connection test cannot run
concurrently with customer operation on the
controller being tested. Ensure that the controller
can be taken from the customer.
Warning: The connection test cannot run
concurrently with customer operation on the
controller being tested. Ensure that the controller
can be taken from the customer.
Warning: If one or more drives in the string are
not running, place the drive motor switches for
those drives to OFF. If this Is not done, those
drives might not sequence on correctly.
Warning: If one or more drives in the string are
not running, place the drive motor switches for
those drives to off. If this is not done, those drives
might not sequence on correctly.
The MD connection test is run by the controller
sequencer (not the storage control). There are
three parts to this test.
Valid responses are:
- MD to controller connection
- Controller function
- Storage control test
To run the connection test, the MD instructs you to
power the controller OFF and then on. During the
power-on sequence, the controller sequencer (if
operating correctly) establishes communication
with the MD, and the connection tests start.
1
2
3
4
= Controller A 1
= Controller A2
'- . /
= Controller A3
= Controller A4
Note: Selecting controller A3 or A4 is valid
only on strings with four storage paths, and
only when the MD is attached to the A unit
that contains controllers A3 and A4.
Controllers A 1 and A2 cannot be selected
when the MD is attached to the A unit that
contains controllers A3 and A4.
When running the test, follow the instructions given
by the MD.
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1985
MD-66
\....-,j
«
(
340 CONTROLLER AX CONNECTION
TESTS READY.
Respond to the following:
Warning: The connection test cannot be run
concurrently with customer operation on the
controller being tested. Ensure that the controller
can be taken from the customer.
The MD will instruct you to turn off the indicated
controller and press ENTER. Pressing ENTER
causes the MD to start polling the controller
sequencer.
(
(
(
(
The MD will next instruct you to turn on the
indicated controller. Turning on the indicated
controller causes the controller sequencer to
branch to a program step that looks for the MD
poll. If the controller sequencer is started
successfully and if the sequencer sees the MD poll.
communication is established between the
controller sequencer and the MD. The connection
test is then started.
Use the Controller Power On/Off switch on the
operator panel or one of the following CPs to
power off and on.
- CP410 for controller A 1 or A3
- CP411 for controller A2 or A4
350 DPS TESTS CANNOT BE RUN
(
360 HA SCAN
The MD is attempting to run DPS test on a 3990
Storage Control. The DPS test will not run on a
3990 Storage Control. Press the PF key to go to
the MD Main Menu.
(
SELECT CONTROLLER
Valid responses are:
1
2
3
4
= Controller Ai
= Controller A2
= Controller A3
= Controller A4
Note: Selecting controller A3 or A4 is valid
only on strings with four storage paths, and
only when the MD is attached to the A unit
that contains controllers A3 and A4.
Controllers A 1 and A2 cannot be selected
when the MD is attached to the A unit that
contains controllers A3 and A4.
SELECT DEVICE
Select one valid device address (00 through
iF)
SELECT CONTROL MODE
Valid responses are:
1
2
= Automatic (default)
= Manual
Automatic mode automatically continues on
error and presents the error data at the end
of the scan or until 80 data checks have been
detected.
Manual mode stops on each error and
displays the error information. The error can
be analyzed. or the scan can be continued
until.the end of the scan or when 80 data
checks have been detected.
If an error is detected when in manual mode,
the user is provided the option of changing to
automatic mode.
SELECT TRACK TYPE
Valid responses are:
1
2
c
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 19B5
= Customer tracks
= CE/Diagnostic tracks
MD-67
3
= HA Map tracks
On a 3380-AJ4 and 3380-8J4 volume,
•
Customer tracks are on cylinders 0
through 375 hexadecimal.
•
CE tracks are on cylinder 376
hexadecimal.
•
Diagnostic tracks are on cylinders 377
through 37A, near the guard band inner
diameter (G8ID) and servo tracks.
•
ENTER THE THREE-DIGIT HEX NUMBER OF THE
BEGINNING CYLINDER.
(This question is asked only when customer
tracks are selected.) The maximum valid
ranges (hexadecimal) are:
Beginning
Cylinder
HA Map tracks are on cylinder FFFD (-3),
near the guard band outer diameter
(G80D) and head loading zone.
Customer tracks are on cylinders 0
through A5F hexadecimal.
•
CE tracks are on cylinder A62
hexadecimal.
•
Diagnostic tracks are on cylinders A63
through A66, near the guard band inner
diameter (G8ID) and servo tracks.
•
HA Map tracks are on cylinders A6D
through A68 (-4 through -2), near the
guard band outer diameter (GBOD) and
head loading zone.
If CE/Diagnostic tracks or HA Map tracks are
selected, the diagnostic program determines
the cylinders to be tested. ,
DO YOU WANT TO SCAN ALL CUSTOMER
TRACKS?
(This question is asked only when customer
tracks are selected.) Valid responses are:
YES key
NO key
= Scan the whole pack
=
Specify a range of cylinders to
be scanned, based on the
response to the next two
messages.
Model K
000 - 375
000 - A5F
Note: The Physical Address (PA) bytes from
the error bytes of another diagnostic or from
sense bytes 5 and 6 may be converted to
cylinder and head by using the following
diagram.
Sense
byte 5
~,
I
Sense
byte 6
~
02
II
On a 3380-AK4 and 3380-BK4 volume,
•
Model J
/
Cy
JLr
03 02
,
31
Head
01
In this example, byte 5 contains 02 and byte 6
contains 31. The address is cylinder X"0302,"
head X "01."
~
,
ENTER THE THREE-DIGIT HEX NUMBER OF THE
ENDING CYLINDER.
(This question is asked only when customer
tracks are selected.) The maximum valid
ranges (hexadecimal) are:
Model J
Ending
Cylinder
000 - 375
Model
K
\ ......",'
000 - A5F
Note: The ending cylinder must be greater
than or equal to the beginning cylinder.
HOW MANY TIMES DO YOU WANT TO READ EACH
HA?
HA Scan cannot directly determine if a data
check in the home address area is
permanent. To determine that a data check
is permanent, the home address (HA) must
be read multiple times. If a data check
occurs each time the HA is read, the data
check can be considered permanent. If the
HA is sometimes read without error, the data
check can be considered temporary.
-:--"',
"
I~
7
\
I,~j'
fr'"
~J
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
MD-68
(
It follows that the separation of permanent
and temporary data checks is more reliable
when the reading is repeated more. But
repeating reading takes time.
The minimum run times for scanning all home
addresses on the volume are:
Model K
Model J
(
1 read
2 reads
6 minutes
12 minutes
18 minutes
36 minutes
Valid responses are "1" through "F":
HOW MANY TIMES DO YOU WANT TO RETRY IF
DATA CHECKS OCCUR?
Valid responses are "1" through "F":
1 = Retry 1 time for each track in error
2 = Retry 2 times for each track in error
F
= Retry 15 times for each track in error
When a single read is selected, select
multiple retries to determine if the error is
solid or intermittent.
1 = Read each HA 1 time
2 = Read each HA 2 times (default)
F
(
= Read each HA 15 times
If the response is "1," the following question
is asked:
(
({
PIN 451990384 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 19S5
MO-69
395 SCAN COMPLETE I SCAN REVIEW
One of following is displayed if data check errors
are detected during the scan operation:
** ARM X **
HEAD:
xx xx xx xx
ERRORS: xx xx xx xx
* xx
SCAN COMPLETE *
TRACKS HAD
DATA CHECKS
SEE PSG MD-3,39S ...
-
or-
* xxSCAN
REVIEW *
TRACKS HAD
DATA CHECKS
SEE PSG MD-3,395 ...
ERRORS
* SCAN -STOPPED *
THE DATA CHECK LOG
IN THE MD IS FULL.
SEE PSG MD-3,395 ...
After the scan is completed, stopped, or reviewed,
summary or detailed information can be displayed
by selecting the display mode from the "Select
Display Mode" menu:
SELECT DISPLAY MODE:
I-HEAD/ARM SUMMARY
2-TRACK SUMMARY
3-TRACK DETAIL
4-RUN HA SCAN
AGAIN
5-RETURN TO MAIN
MENU
After displaying summary or detail information, the
MD returns to the "Select Display Mode" menu and
another mode can be selected.
Display Mode 1: HEAD/ARM SUMMARY
This display mode presents a summary by
head of the number of tracks in hexadecimal
that had data checks. The summary is
presented in four displays, one for each head
arm:
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1997, 1988
'.
Display Mode 2: TRACK SUMMARY
Presents one or more displays with a one line
entry for each track that had errors:
CYL=xxx
CYL=xxx
CYL=xxx
CYL=xxx
"xx" in the previous displays indicates the number
of tracks, in hexadecimal, that had errors.
The maximum number of tracks that can be stored
by the MD is 80 (X"SO"). If more than 80 tracks had
errors, the scan is terminated with this frame:
/'"
= Number of errors for each head
HD=x
HD=x
HD=x
HD=x
CNT=x
CNT=x
CNT=x
CNT=x
CYL = Cylinder number (hexadecimal)
HD
Head number (hexadecimal)
CNT = Number of errors (hexadecimal)
=
Display Mode 3: TRACK DETAIL
Presents one display for each track in error.
There are two types of error displays:
** DA T'A CH ECK **
HOME ADDRESS AREA
CYL = xxx
HEAD = x
ERROR COUNT = x
-
or-
NO SYNC BYTE FOUND
HOME ADDRESS AREA
CYL = xxx HEAD = x
ERROR COUNT = x
\ ...-..
~/
CYL = Cylinder number (hexadecimal)
HEAD = Head number (hexadecimal)
ERROR COUNT = Number of errors
(hexadecimal)
Display Mode 4: RUN HA SCAN AGAIN
Repeat the Review HA Scan option.
Display Mode 5: RETURN TO MAIN MENU
Return to the MD Main Menu.
MD-70
4'--.....,
(
""',j
r
,
{
(
(
(-
{
400 SELECT ONE OF THE FOLLOWING
ERROR LOG FUNCTIONS
1 = Transfer Log
Transfers up to 16 error log entries for the
selected controller to the MD and stores them
on the MD diskette. Any specific entry may
be subsequently selected for analysis. The
log records are assigned numbers for
reference.
2 = Search Log
Search any error log previously transferred to
the MD with the specified search argument
and return the first (newest) entry that
matches the search argument for analysis.
3 = Merge Logs
Merge any or all of the log files previously
transferred to the MD into a single merged
log file. If the log records contain age (3990
only), the merged log will contain all the
records in age order (newest to oldest). If the
records do not contain age, the controller A1
log records are placed before the controller
A2 log records. The log records are assigned
record numbers for reference.
(
405 SELECT CONTROLLER (ERROR LOG)
Select a controller for the function to be performed.
Only the error log in the storage path or storage
director attached to the selected controller will be
affected. Valid responses are:
Response Meaning
1
2
3
4
Controller
Controller
Controller
Controller
A1
A2
A3
A4
Note: Selecting controller A3 or A4 is valid only on
strings with four storage paths, and only when the
MD is attached to the A unit that contains
controllers A3 and A4. Controllers A 1 and A2
cannot be selected when the MD is attached to the
A unit that contains control/ers A3 and A4.
Enter your response for controller use.
4 = Change Log Mask
Change the error log mask to log information
other than the default. The default is to log
sense formats 1, 7, 8 and 9 and not to log
intervention required or data check errors.
The mask can be changed to log any of the
formats (1 through 4 and 7 through 9) and to
log intervention required.
5 = Erase Log
Reset the error log in the storage control to
an empty state.
For additional information, see "Select Error Log
Function" on page LOG·2.
c\
PIN 4519903 84 Sides
CO Copyright
EC 476581 01AugB9
IBM Corp. 1987, 1988
MD-71
410 ENTER SEARCH ARGUMENTS
415 ENTER LOG NUMBER
The search of an error log can be for a specific
controller, a specific device, or a specific symptom
code (or group of codes). Except for controller, the
search can be performed on any combination of
these search parameters. The controller can be
specified only for the merged log.
Valid responses are:
Controller
Enter A1 (or 1), A2 (or 2), A3 (or 3), A4 (or 4),
or press ENTER for all controllers (merged log
only).
Device
Enter 2 digit device number or press ENTER
for all devices.
Symptom Code
Enter a valid symptom code or enter a
symptom code with "X"s (for example: DXXX
for any symptom code that starts with a "0,"
or D2XX for any symptom code that starts
with "02") or press ENTER for all symptom
codes.
See the LOG section of this manual for a
description of the messages which are a result of
the search option.
1 for Controller A1 log
2 for Controller A2 log
3 for Controller A3 log (4-path subsystems)
4 for Controller A4 log (4-path subsystems)
M for merged log
R to return to the Main Log menu
The log number refers to a log previously
transferred and stored on the MD diskette. "M"
refers to the merged log (see Item 3 in "400
SELECT ONE OF THE FOLLOWING ERROR LOG
FUNCTIONS" on page MD-71).
Entering an "R" will cause the MD to return to the
Main Log menu. For additional information. see
"Search Log (Option 2)" on page LOG-S.
' ""'-
420 ·ENTER THE DATE
The date entered will be included in the log stored
on the MD diskette.
Enter 6 characters of date information in the format
MMDDYY.
/-
425 ENTER LOGS TO BE MERGED
The numbers entered will cause the corresponding
logs previously transferred and stored on the MD
diskette to be merged (combined) into a single log
file on the MD diskette. The original log files are
unchanged. A previously merged file will be
replaced by the newly merged file.
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
MO-72
< -"
,/
",,~
(
(
430 ENTER THE SENSE FORMA TS TO BE
LOGGED
(
Press the ENTER key to accept the default values
(formats 1, 7, 8, 9).
(
(-
Enter the formats to be logged, separated with a
single space. The following options are available.
All formats to be logged must be entered. Defaults
will not be assumed.
0123-
Log
Log
Log
Log
format
format
format
format
0 errors
1 errors
2 errors
3 errors
(ProJram or system checks)
(Device check 2s)
(Storage Control check 2s)
(Storage Control check 1s)
4 - Log format 4 errors (Data checks)
7 - Log format 7 errors (Controller check 1s)
(:
('
8 - Log format 8 errors (Controller check 2s/device
check 1s)
9 - Log format 9 errors (Device check 2s)
F - Log format F errors (Storage Control error)
440 ERROR LOG MASK/TRANSFER
STOPPED BY KEYBOARD ENTRY
Error log diagnostic routine is interrupted by a
keyboard entry. Press ENTER to continue the
function from the point of interruption.
460 CONFIGURA TION
The MD displays configuration information in one of
the following series of frames. One series is for
2-path configurations and the other series is for
4-path configurations.
2-Path Configurations
STRING CONFIGURATION
CTlR Ax ADR=x ID=xx
3xxO SxID=xxxx SP=x
CTlR PATH=x
THE STRING HAS:
A*
4-Path Configurations
STRING CONFIGURATION
CTlR Ax ADR=x ID=xx
3990 SSID=xxxx SP=x
CTlR PATH=x
THE STRING HAS:
A*
SUBSTRING 0 HAS:
A*
SUBSTRING 1 HAS:
A*
CTLR Ax
x = the controller number selected to run the
test.
(
ADR=x
x
the string number (0 or 1) set in the
subsystem configuration switches on the
DDC/DTB card. See page REPAIR-18 for
2-path switch settings and page REPAIR-19 for
4-path switch settings. The string number is
normally put on the operator panel at
installation time. All controllers in the string
must be set the same.
(
=
('
ID=xx
=
~A
the physical 10 of the controller (A1, A2,
A3, or A4) set in the physical address
switches on the I/O control card. See page
REPAIR-20 for 2-path switch settings and page
REPAIR-22 for 4-path switch settings. The
PIN 4519903 84 Sides
© Copyright
(- •.
EC 476581 01AugB9
IBM Corp. 1987, 1988
MD-73
controller physical 10 is normally put on the
operator panel at installation time (STRING
10). The physical 10 (STRING 10) must be set
the same for all controllers in the string.
3xxO
The type of storage control:
xx=88 for 3880 storage control.
xx=99 for 3990 storage control.
SSID=xxxx or SDID=xx
xxxx
the physical 10 of the storage path
(SSID) in a 3990 storage control. The SSID is
normally put on the 3380 operator panel at
installation time. The SSID must be set the
same for all storage paths attached to the
string.
=
=
xx
the physical 10 of the storage director
(SOlD) in a 3880 storage control. The SOlD is
normally put on the 3280 operator panel at
installation time. Each 3880 storage director
attached to the string has a unique SOlD.
SP = x (3990 only)
x
the storage path (Q...3) that is connected
to the controller selected to run the test. The
storage path is normally put on the 3380
operator panel at installation time.
A*
A* = the number of units that are in the
string or substring (4-path). One of the
following lines is displayed:
A-UNIT ONLY
A-AND 1 B UNIT
A-AND 2 B UNITS
A-AND .3 B UNITS
:-"
Note: Substring 0 refers to those devices or
units (devices OO-OF) that are attached to ports
0-3. Substring 1 refers to those devices or
units (devices 10-1F) that are attached to ports
4-7.
For additional information, see "Diagnostic Menu
Option 7 - Device Status" on page MD-26. The MD
provides the option to have you check for
inconsistencies in the configuration information. If
you would like to check for these inconsistencies,
press the YES key to "DO YOU WANT AN
ANALYSIS OF THIS CONFIGURATION DATA?"
If you find inconsistencies, they must be corrected
and the test must be rerun to get beyond this point
in the test.
=
Note: 3880 storage controls do not have
storage paths and SP =x is not displayed.
CTLR PATH = x
x = The path sensed from the controller by
the storage control during a sense controller
path operation.
o = Controller A1 Path
1
2
3
= Controller A2 Path
= Controller A3
PIN 451990384 Sides
© Copyright
Path
= Controller A4 Path
EC 476581 01Aug89
IBM Corp, 1987. 1988
MD ..74
462 CONFIGURATION DATA/UNKNOWN
STORAGE CTRL
See MSM, Volume R60 SENSE Section, Format 1,
byte 8 for additional information about string
features. Press the ENTER key to continue.
The following frames display configuration data:
CONFIGURATION DATA
TYPE=XX FEATURES=XX
(SEE PSG MD-3, 462)
ENTER TO CONTINUE
UNKNOWN STORAGE CTRL
TYPE=XX FEATURES=XX
(SEE PSG MD-3. 462)
ENTER TO CONTINUE
(
The above frame indicates that the storage control
type is not known by the MD. If the storage control
type is not valid, the MO assumes a storage control
type of 3990.
{
(~'
(
'
_/
c
TYPE=XX
XX Storage control type.
=
For 3880, this byte is 00.
For 3990 family storage control, the type
information is determined from vital product
data in the storage control:
00=
05 =
06 =
OF =
10=
3880
3990
3990
3380
3990
Model
Model
Model
Model
2
3
CJ2
1
FEATURES=XX
XX = String features. String features is the
data returned from the controller during a
Sense Features command (command 11).
f~
f'
(~
(~
Bit
Bit Name
o
OPS installed
4-path
Second COP card installed
1
2
3
4
5
6
7
o
Single HOA
o
o
1
C""
,/
c
PIN 4519903 84 Sides
EC 476581 01Aug89
~ Copyright IBM Corp. 1987, 1988
MD-75
465 DEVICE STA TUS
DEVICE STATUS:
X = Ol23456789ABCDEF
OX: OOOOOOOONNNNNNNN
IXI ~EOOOOOONNNNNNNN
2-path subsystem
DEVICE STATUS:
X = Ol23456789ABCDEF
OX: ****************
-
or-
The above example indicates:
•
The string is a 4-path string because there are
devices in both the OX range and in the 1X
range.
•
Devices 00-07 are installed and operational.
•
Devices 10 and 12-17 are installed and
operational.
•
Devices 08-0F and 18-1F are not installed.
•
Device 11 is failing with an equipment check.
4-path subsystem
DEVICE STATUS:
X = Ol23456789ABCDEF
OX: ****************
IX: ****************
=
X
OX:
1X:
Device number (O-F)
Represents devices OO-OF.
Represents devices 10-1F (4-path
subsystems).
i'
The asterisks (*) are replaced by device status.
The device status codes are:
B
D
E
The MD provides the option to display a description
of the status codes listed above.
Device Susy
Data Check
Equipment Check
Intervention Required: Device is disabled and
not on-Line; this code has nothing to do with
whether or not a device is in CE mode.
N
Not Installed
o
Operational: An operational device normally
reports a status of "0."
s
Servo busy: If a device has been powered on
without at least one storage path channel
interface enabled, the device usually reports
a status of "S" because there is no attached
system to handle the ...ttention interrupt. To
eliminate this problem, place the device in CE
mode (use MD Main menu Option 0) and
rerun the device status test.
'"-,,,/
A status of "S" or "S" is reported if a device is
busy. This may occur if the Device Status test is
run during periods of heavy usage. It is usually a
good indication that the device is operational. For
example,
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
f
\,
468 DEVICE TYPE
Device type information is displayed in one of the
fol/owing series of frames. One series is for 2-path
configurations and the other series of frames is for
4-path configurations.
FR11 =
FR12 =
FR13 =
FR14
2nd AJ4 or AK4 unit
1st B unit attached to 2nd A unit
2nd B unit attached to 2nd A unit
3rd B unit attached to 2nd A unit
xxx =
Model designation (AJ4. AK4. BJ4, or
BK4) of the instal/ed machine frame(s).
Frame members are not displayed if
they are not instal/ed.
=
2-Path Configurations
DEVICE TYPE:
FR01=XXX
FR03=XXX
FR02=XXX
FR04=XXX
(
('
FR01 =
FR02 =
FR03 =
FR04 =
AJ4 or AK4 unit
1st B unit attached to A unit
2nd B unit attached to A unit
3rd B unit attached to A unit
xxx
Model designation (AJ4, AK4. BJ4. or
BK4) of the installed machine frame(s).
Frame members are not displayed if
they are not instal/ed.
=
(-'
(-
xxx
=?71
'"
./
(
The diagnostic routine was unable to
determine the model. This occurs when
the diagnostic cannot select any of the
devices in the unit due to heavy system
activity .
4-Path Configurations
= 71? The diagnostic routine was unable to
determine the model. This occurs when
the diagnostic cannot select any of the
devices in the unit due to heavy system
activity.
DO YOU WANT AN
ANALYSIS OF THE
DEVICE TYPE DATA!
(SEE PSG MD-3, 468)
(
(
xxx
=
SSO
Substring 0
the first substring in a
4-path configuration. Substring 0 •
contains ports 0-3 (devices 00 - OF).
SS1
Substring 1 = the second substring in a
4-path configuration. Substring 1
contains ports 4-7 (devices 10 - 1F).
Note: Models AJ4 and BJ4 have a large letter J on
the HDAs. Models AK4 and BK4 have a large letter
K on the HDAs.
The number of B units in the string is determined
from the string configuration switches on the I/O
control card. See page REPAIR-20 for 2-path
switch settings and page REPAIR-22 for 4-path
switch settings.
DEVICE TYPE - 550:
FR01=XXX
FR03=XXX
FR02=XXX
FR04=XXX
(/
DEVICE TYPE - SSI:
FRll=XXX
FR13=XXX
FR12=XXX
FRI4=XXX
DO YOU WANT AN
ANALYSIS OF THE
DEVICE TYPE DATA?
(SEE PSG MD-3, 468)
FR01 =
FR02 =
FR03 =
FR04 =
1st AJ4 or AK4 unit
1st B unit attached to 1st A unit
2nd B unit attached to 1st A unit
3rd B unit attached to 1st A unit
PIN 4519903 84 Sides
© Copyright
(--,
EC 476581 01Aug89
IBM Corp_ 1997. 1999
MD-77
480 STORAGE CONTROL STATUS
500 ISOLATION CODE ERROR D/SPLA Y
Storage control status identifies the physical 10 of
the storage control or storage director and the
physical path attached to the controller:
Select the specific MD display from the following:
xxxx S*ID: xx
STORAGE PATH: x
CONTROLLERI Ax
SEE PSG MD-3, 480 ...
If line 1 Is "3880 SOlD: xx"
xx = Storage Director identifier.
PRESS ENTER TO ANALYZE OR liN" TO RUN NEXT
CONTROLLER
An error has occurred during a controller test. The
user has the option to analyze it or to continue
running on the next controller (if both controllers
were selected).
If line 1 is "3990 SSID: xx"
xx = Subsystem identifier.
If both controllers were selected, entering "N" will
start the controller diagnostics for the untested
controller, For additional information, see page
MD-41.
STORAGE PATH: x
x = Storage Path (0-3)
PRESS ENTER TO ANALYZE OR liN" TO RUN NEXT
DEVICE
CONTROLLER: Ax
Controller (A 1, A2, A3, or A4)
Ax
An erm:- has occurred during a device test. The
user has the option to analyze it or to continue
running on the next device (if another device was
selected).
=
Status of conditions in the storage control is
displayed in the following frames.
•
Devices in CE mode
•
Write Inhibit conditions
•
Fenced conditions (3990 only)
Entering liN" will start the device diagnostic tests
for remaining specified devices. For additional
information, see page MD-41.
PRESS ENTER TO ANALYZE OR PF" KEY TO
RETURN TO MAIN MENU
II
Storage path fenced conditions
Device fenced conditions
An error has occurred and user has the option of
analyzing it or returning to the MD Main Menu.
''
!/'
For additional information, see Diagnostic menu
option 9 (Storage Control Status) on page MD-34.
PIN 4519903 84 Sides
EC 476581 01Aug89
MD-78
© Copyright IBM Corp. 1987, 1988
,;t ,
'"
I
"
(
(-
550 ENTER THE SYMPTOM CODE
605 WARNING: HDA CABLES SWAPPED
Enter a valid symptom code. Symptom codes that
are not valid are rejected and this message is
displayed again. Symptom codes are obtained
from the EREP System Exception Report, from the
Symptom Code Summary, or from sense bytes 22
and 23. Valid symptom codes for the 3380-JK are:
The MD detected that the HDA cables are
swapped. Review the following to determine how
the hardware and maintenance package are
affected.
•
4xxx, 9xxx, Dxxx, Exxx,
•
7xxx - In some cases, when the functional
microcode is not able to identify the failing
device type, 3380 stage 1/stage 2 sense data is
generated. For these cases, 7xxx symptom
code entry is permitted and the appropriate
3380-JK FRUs are displayed.
(
(
•
('
(
~\
2xxx - In other cases, when errors are
propagated to the storage control, the storage
control generates 2xxx symptom codes. For
these cases, 2xxx symptom code entry is
permitted and the appropriate 3380-JK FRUs
are displayed.
/
DEVICE ADDRESSES
A device address always identifies the same
actuator and the data that is accessed by that
actuator whether or not the cables are
swapped.
OPERATOR PANEL DEVICE READY LIGHTS AND
SWITCHES
Swapping cables on a 3380 HDA does not
change the relationship between the device
address, or physical actuator, and the
operator panel Enable/Disable switch and
Ready LED for that device.
Therefore, there is no need to apply a label to
the operator panel as was done for 3380-DE
devices.
MD LOGIC CARD CALLOUT
The MD always calls out the correct group of
logic cards whether or not the cables are
swapped.
('
('
SINGLt: ACTUATOR MAINTENANCE (SAM) PANEL
SWITCHES AND CIRCUIT PROTECTORS (CPs)
The SAM panel switches and CPs are always
associated with the same logic cards. When
the cables are swapped, the left set of device
logic cards and its associated CPs and
switches control the right actuator in the HDA,
and vice versa. Always ensure that you use
the correct switches for the logic card you
plan to replace.
(/"
('
POWER CARD LEDs
('
The Sequencer Status LEOs on the power
cards are reversed when the cables are
swapped. See Figure 1 on page MD-80. The
other LEOs retain their original meaning (do
not reverse).
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
MD·79
B1 Board Power Card LEOs I Cable Swap
;'~~\
[
,
""_/
REAR DRIVE
FRONT DRIVE
NOTE.
FRONT DRIVE DO£S
NOT exIST IN MODEl..
CJ2.
I
YA
VB
I
Y.J
I
r",
II
VC
I
YK
I
ye:
VD
I
YM
I
I
YL
I
J
F
K •L
YE
I
YN
I
YG
I
YP
I
VH
I
YR
I
I
YQ
I
M
S
T
r--',
\",y
U
Bl-BOARD (REF) - - . - j
~I
~~I
Bl-BOARD (FRONT DRIVE - G2 COMMON POWER CARD
REAR DRIVE
- R2 COMMON POWER CARD) - - - - - "
~
,~
~
~
~
\--.....
I
~
I
~OA
I
I
I
\~'
I
NORMALCHDA CABLES NOT SWAPPED)
HDA CABLES SWAPPED
BEFORE
AFTER
THE HDA CABLE SWAP.
REFERENCE TO THE FOLLOWING,
THE
LEOS
MAKE
THE HDA CABLE SWAP, THE SAME LEOS MAKE
REFERENCE TO THE FOLLOWING.
!r------IJJ
"'-=-':-- LEFT SEQUENCER STATUS
1!JIIo~~-
I0Il;:-:--
~,---
+5/2"" V ON
RIGHT SEQUENCER STATUS
. ..,;:;......;:....-- +5/2"" V ON
+5 V COMMON
~--
...~-- RIGHT SEQUENCER STATUS
+5 V COMMON
Bl'oE--- LEFT SEQUENCER STATUS
,
_~:--
LEFT DEVICE READY
I]]
LEFT DEVICE READY
_-::-:'-- LEFT SWITCH ENABLED
LEFT SWITCH ENABLED
fIIIIo-::;-- RIGHT SWITCH ENA"LED
_-=--
__€ : - - RIGHT DEVICE READY
_€:--RIGHT DEVICE READY
RIGHT SWITCH ENABLED
NOTES.
OJ~~,&l~f~~l'f.o~~A~~T"::~J!f//11-Jst;T
1$000$82511
Figure 1.
'21
/lll£N THE CABLES ARE SHAPPED, THE LED FOR THE RIGHT
u:J5E'Ql.ENCER STATUS SHOHS LEFT SEQUCNCCR STATUS.
•
B1 Board Power Card LEOs I Cable Swap
PIN 4519903 84 Sides
«:> Copyright IBM Corp.
EC 476581 01Aug89
1997, 1999
,/
MO-80
(-
(
615 MD DOES NOT PROVIDE ANALYSIS
FOR DISPLA YEO IC
742 PRESS ENTER TO ANALYZE, OR 'N'
TO DISPLA Y NEXT ERROR
For more information on these ICs (Isolation
Codes), see the DIAG section of the MSM, Volume
R60 for a description for the diagnostic that you are
running.
A diagnostic error has been recalled from the MD
diskette while using the Review option. You have
the option to analyze it or to display the previous
error that was stored on the MD diskette.
700 SELECT REVIEW OPTION FROM
FOLLOWING LIST
Up to four diagnostic errors are saved on the MD
diskette for review. The most recent error is
displayed first. Entering "N" displays the previous
error saved. Valid responses are:
Review CE mode
A diagnostic routine is invoked to find out
which devices are currently set in CE mode.
(
(
(:'
c
Review diagnostics
The last 4 diagnostic errors are saved on the
MD diskette. The most recent error is shown
first and previous errors can be displayed by
entering UN" (next). Maintenance analysis
can be invoked from the error display.
Response
Meaning
ENTER
Analyze the error
Display the previous error saved
N
For additional information, see "MD Main Menu
Option R" on page MD-49.
Review error log
Display the error log summaries that were
previously transferred to the MD diskette.
Maintenance analysis can be invoked from
the summary display.
This is the same function as the Search Log
Function. For additional information, see
"Search Log (Option 2)" on page LOG-5.
Review repair
Display the last repair specified by a
maintenance analysis procedure.
(j
C:
PIN 4519903 84 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987. 1988
MD-81
800 DIAGNOSTIC AIDS
SELECT CONTROLLER
Valid responses are:
1.
=
Controller
1
2 = Controller
3 = Controller
4
Controller
=
A1
A2
A3
A4
Note: Selecting controller A3 or A4 is valid
only on strings with four storage paths, and
only when the MD is attached to the A unit
that contains control/ers A3 and A4.
Control/ers A 1 and A2 cannot be selected
when the MD is attached to the A unit that
contains controllers A3 and A4.
ENTER ROUTINE 10
Valid responses are:
80 = Controller/DPS tests
81 = Device logic tests
82
Servo tests
83 = R/W data transff'r tests
=
84 =
85
8A
90 =
Extended DPS tests (3880 only)
= HA Scan (see note 1)
= DDC Tests (see note 1d)
= Device status test (see note 1a)
= Sense Utility (see note 2)
=
98
Error log search (see note 1b)
99 = Motor Start Test (see note 1a)
9C
CE track repair routine
9D
DPS Array Dump Routine (3880 only see note 2)
=
=
9E
Routines 91, 93, 96, 98, 99, 9F, CO, and D3
should not be run with diagnostic aids:
Use MD Main Menu option 1 to run 96
and 99.
b. Use MD Main Menu option 2 to run 98.
c. Use MD Main Menu option 0 to run
9F.
d. Routines 93, CO and D3 should be run
from the storage control.
a.
2.
Routines 97, 9D and 9E are for product
engineering and normally should not be
run without assistance from product
engineering.
For additional information, refer to the DIAG
section in the MSM, Volume R60.
SELECT DEVICE
Select one device address. Valid responses
are:
R/W function tests
= Seek Exerciser tests
= Incremental seek test
91
93
96
97
Notes:
= Diagnostic Microcode Patch Routine
"00" through "OF" for 2-path strings
''~O''
through "1F" for 4-path strings
ENTER: No device is needed
ENTER PARAMETERS IF REQUIRED
See the DIAG section in the MSM, Volume
R60 to determine diagnostic parameters. Each
parameter is entered as two hexadecimal
digits. Up to sixteen parameters can be
entered with no spaces between each
parameter.
Press the ENTER key to enter default
parameters. You will be asked the following:
(see note 2)
9F = Set/reset CE mode (see note 1c)
CO = 3380 A04! AA4 Controller Tests (see
note 1d)
D3
3380 A04/AA4 DDC Tests (see note 1d)
=
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1967. 1969
MD-82
(
DO YOU WANT TO LOOP THE ROUTINE?
Press the YES key or the NO key to respond:
NO = Run routine to completion or until
stopped by an error.
YES
Loop routine and continue looping
until stopped by keyboard entry.
=
If the response is "YES," you will be asked
the following:
DO YOU WANT TO HALT WHEN AN ERROR
OCCURS?
(
Press the YES key or the NO key to respond:
YES = When an error occurs, stop.
NO = When an error occurs, continue
looping.
(
900 CONTROL UNIT AND MD DISKS ARE
NOT CaMPA TlBLE
Make sure the diskettes being used are the correct
level for the machine being tested.
This message indicates that the storage control
diskette is down-level. Pressing the ENTER key
allows you to continue with the maintenance
procedures. However, be aware that there can be
functional differences between the two diskettes
and the supporting documentation.
Contact your next level of technical support if
necessary.
If other options are needed, you will have to enter
parameters. See the DIAG section in the MSM,
Volume R60.
(
{
",
/
('
c
PIN 451990384 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
MO-83
.
\, .:...
,,/ -',,-
EC History of Maintenance Device Section
EC HISTORY OF P/N 4519903
EC Number
Date Of EC
EC Number
Date Of EC
475245
14Nov86
475248
25Apr88
475246
21Ju187
476573
30Nov88
475247
llSep87
476581
01Aug89
\""--,,,;1
Notes:
r/"""""
PIN 4519903 84 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
MD-84
(
Error Log Function (MD Main menu Option 2)
(
The storage control contains error logs for the
most recent errors sent to the system. These logs
may be transferred to the MD, then searched and
(
Entry
(
(
Select One of the Following
Go to page
A
Introduction
LOG-1
B
Select Error Log Function
LOG-2
C
Transfer Log (Option 1)
LOG-4
0
Search Log (Option 2)
LOG-5
E
Merge Log (Option 3)
LOG-8
F
Change Log Mask (Option 4)
LOG-9
G
Erase Log (Option 5)
LOG-10
Figure 1.
•
When a log is full, the oldest entry is overlaid.
•
•
The error log functions are transfer, merge,
search, erase, and change log mask.
•
Transfer, erase, and change log mask are
performed by an inline diagnostic (routine 98)
and do not impact customer operation (devices
do not need to be placed in CE mode).
•
The merge and search functions are done
within the MD. The MD does not have to be
connected to a string for these functions.
•
The 3990 storage paths will log a relative time
with each record. The age of the record
relative to the time it was transferred to the MD
can then be determined and displayed.
Relative time is not logged when attached to a
3880 storage control.
....
(
LOG Section Entry Table
Introduction
•
{
analyzed by the MD. Use the following entry table
to quickly find the information in this section.
•
Each attached storage path or storage director
contains 2 error logs, one for each attached
string.
With the default error log mask, the error logs
are equipment check logs. They contain
entries for all format 1, 7,8 and 9 (sense byte 7
= 1x, 7x, 8x, or 9x) sense byte records that are
sent to the system. They do not contain data
check records (sense format 4).
Each log contains up to sixteen 32-byte entries
for errors that occurred on the associated path,
24 sense bytes and 8 bytes of additional
information. Only the first 24 bytes of the 32
sense bytes generated by a 3990 are put in the
error log when attached to a 3990. All 24 bytes
that are generated go into the error log when
attached to a 3880.
c··
(.
C·
PIN 4519905 11 Sides EC 475248 25Apr88
@ Copyright IBM Corp. 1987. 1988
LOG-1
Select Error Log Function
From the MD Main menu, select option 2.
The MD displays the following frames:
SELECT ONE OF THE
FOLLOWING ERROR LOG
FUNCTIONS:
(SEE PSG MD-3, 400)
1. TRANSFER LOG
2. SEARCH LOG
3. MERGE LOGS
4. CHANGE LOG MASK
5. ERASE LOGS IN THE
CONTROL UNIT.
PF FOR MAIN MENU
Press the ENTER key to display the next frame.
These frames show the functions that can be
performed with the error log.
1. TRANSFER LOG
Transfers all log records from the specified
storage path or storage director to the MD.
The log in the storage path or storage
director is not changed or erased. The
storage area in the MD for the selected
storage path or storage director will be
erased and new data (if any) will be stored on
the MD diskette. See "Transfer Log (Option
1)" on page LOG-4.
2. SEARCH LOG
Any of the logs which have previously been
transferred to the MD, or the merged log, can
be searched for specified entries. The search
can be made by controller (storage path),
device, or symptom code. The selected
records will be displayed one at a time for
examination or analysis. See "Search Log
(Option 2)" on page LOG-5.
LOG-2
3. MERGE LOGS
Selected logs which have been previously
transferred to the MD can be merged together
into a storage area on the MD diskette. A
previously merged log is erased and new
merged log data is stored. If the log records
contain age (3990 storage control only) the
merged log puts all records in sequential
order, with the newest records appearing first
in the merged log. See "Merge Log (Option
3)" on page LOG-8.
4. CHANGE LOG MASK
Change the logging algorithm from the default
values. The logging algorithm controls the
selection of sense data for logging. The
choices are:
•
•
•
•
•
•
•
•
•
•
Log
Log
Log
Log
Log
Log
Log
Log
Log
Log
format 0 errors
format 1 errors (default)
format 2 errors
format 3 errors
format 4 errors
format 7 errors (default)
format 8 errors (default)
format 9 errors (default)
format F errors
Intervention Required
\., ../
See "Change Log Mask (Option 4)" on
page LOG-9.
5. ERASE LOGS IN THE CONTROL UNIT
Erase the log in the specified storage path or
storage director. See "Erase Log (Option 5)"
on page LOG-10.
PF FOR MAIN MENU
Press the PF key to return to the MD Main
menu.
Enter the option number for the error log function
you want to perform. Go to the reference shown.
PIN 451990511 Sides EC 475248 2SApr88
©l Copyright IBM Corp. 1987, 1988
f',
C."'....",
(
Error Log Diagnostic Error
{.
(
Some problems with the storage control, the
controller, or the control interface do not allow the
error log diagnostic to run. The MD displays an
error message. You must analyze the error and
correct the problem before continuing.
When an error is detected, the MD displays the
following frames:
Label
Meaning
IC
CTL
DEV
RTN
TEST
Isolation code
Controller number (A1, A2, A3, or A4)
Device address (in hexadecimal)
Diagnostic routine ID
Test number within the routine
Error bytes
EB
IC**** CTl=A* DEV=**
RTN=** TEST=**
EB=** **** **** ****
**** **** **** ****
(
The labels in the frame have the following
meanings:
PRESS ENTER TO
ANALYZE OR PF KEY TO
RETURN TO MAIN MENU.
(SEE PSG MD-3, 500)
«
Note: The error bytes are used by the MD for FRU
isolation. The bytes are displayed here primarily for
support personnel.
If you are preparing to discuss this problem with
support personnel, you may want to record these
error byte values.
(
('
(
(
(
PIN 451990511 Sides EC 475248 25Apr88
© Copyright
IBM Corp. 1987, 1988
LOG-3
()
', "\
C
.
Transfer Log (Option 1)
The Transfer Log function transfers all error log
entries from the selected storage path or storage
director to the MD and stores then on the MD
diskette. Each storage path or storage director
path log is stored in a separate log file on the MD
diskette. The log file on the MD diskette is erased
before the transfer begins.
Warning: If the MD diskette is from another string,
all logs must be transferred before searching any of
the logs. otherwise, the results of the search will
be unpredictable.
When the transfer log function is initially selected,
the MD displays the following:
WHICH CONTROLLER
ERROR LOG DO YOU
WANT TO TRANSFER?
(SEE PSG MD-3, 405)
The response to this display selects the log in the
storage path or storage director attached to the
controller indicated in the response.
To review all errors for a given device or to
determine if multiple controllers are failing, the
error log in each storage path or storage director
should be transferred. They can then be merged
into a single log using the merge function.
The valid responses are:
Controller
Controller
Controller
Controller
o
The MD next displays this frame:
ENTER THE DATE:
MMDDYY
(SEE PSG MD-3,420)
The date entered (maximum of 8 characters) will
be stored with the transferred log data on the MD
diskette.
ERROR LOG TRANSFER
NOW RUNNING ON
CONTROLLER Ax
PLEASE WAIT.
xx
Immediately after this display, a diagnostic routine
starts to transfer the error log from the storage
path or storage director through the specified
controller to the MD. The diagnostic runs until all
error log records for the string (up to 16) have been
transferred (about 2 minutes for a full log).
The "xx" in the bottom line of the frame is the
transfer count. Each time an error log entry is
transferred, the number (in hexadecimal) is
updated. The transfer begins with the most recent
entry and continues with successively older entries.
When the transfer of the error log is completed, the
MD displays one of the following frames. The
frame on the top indicates entries were transferred.
The frame on the bottom shows that no entries
were found in the error log selected for transfer.
*TRANSFER COMPLETE*
CONTROLLER Ax ERROR
LOG IS STORED.
Response Meaning
1
2
3
4
,J
A1
A2
A3
A4
Note: Selecting controller A3 or A4 is valid only on
strings with four storage paths, and only when the
MD is attached to the A unit that contains
controllers A3 and A4. Controllers A 1 and A2
cannot be selected when the MD is attached to the
A unit that contains controllers A3 and A4.
-
or-
*TRANSFER COMPLETE*
CONTROLLER Ax ERROR
LOG IS STORED.
NO ENTRIES FOUND.
(\
u.
""",,:r' /
When the transfer is complete, press the ENTER
key to return to "Select Error Log Function" on
page LOG-2.
Press the PF key to return to the MD Main menu.
Enter your response to select the error log.
LOG-4
PIN 451990511 Sides EC 475248 25Apr88
Copyright IBM Corp.
(
1987, 1988
LOG-11
,_/
(
EREP Reports
(
The EREP section contains descriptions of the EREP
System Exception reports and the EREP Event
History report. Select the report you want to review
from the following table.
(
Entry
(
(
Select One of the Following
Go to page
A
Introduction
PROG-1
B
EREP System Exception Reports
System Error Summary (Part 2) Report
DASD Sub')ystem Exception
DASD String Summary
DASD Informational Messages
DASD Data Transfer Summary
DASD Symptom Code Summary
PROG-1
PROG-2
PROG-4
PROG-8
PROG-10
PROG-11
PROG-16
C
EREP Event History Report
PROG-20
Figure 1.
PROG Section Entry Table
EREP System Exception Reports
(
(
(
(
(
The EREP (Environmental Recording, Editing, and
Printing) System Exception reports (SYSEXN) are
used to support the 3380. The specific reports used
are:
•
•
•
•
•
•
System Error Summary (part 2)
DASD Subsystem Exception Report
DASD String Summary
DASD Informational Messages
DASD Data Transfer Summary
DASD Symptom Code Summary
The preferred information for isolating problems is
contained in these reports. This maintenance
package uses the DASD Symptom Code Summary
report to isolate equipment checks and the DASD
Data Transfer Summary report to isolate data
checks. Both of these reports are part of the EREP
System Exception Reports. Refer to the
Environmental Recording, Editing, and Printing
Program User's Guide, GC28-1378.
The customer should normally run the System
Exception on a daily basis. If you are isolating
temporary 1/0 errors, it is normally sufficient to use
the daily reports provided by the customer. If you
are isolating a permanent 1/0 error, you should ask
the customer to run the EREP System Exception
reports (SYSEXN) for 33XX devices.
Warning: Do not attempt to off-load device
statistics when running EREP (SYSEXN) if devices
or paths are failing. A device or path problem can
prevent EREP from successfully collecting statistics
and the EREP job will not complete successfully.
To prevent off-loading statistics, a working data set
must be created from EROS (Error Recording
Dataset) and then EREP must be run against the
working data set.
To run EREP for a specific problem, two steps are
required:
1.
Create a working data set. Have the customer
lJse the following parameters: PRINT=NO,
ACC=Y, ZERO=N, TYPE=O, and
TABSIZE = 999K.
2.
Run EREP against the working data set and
print. Have the customer use the following
parameters: SYSEXN=Y, HIST, ACC=N,
TABSIZE=999K, and DEV=(3380).
("
PIN 4519907 23 Sides EC 475248 25Apr88
@ Copyright IBM Corp. 1987, 1988
PROG-1
Sy~tem
the next action to be performed. This report
supplies the necessary information for service
using the ICKDSF program routines. For
media maintenance, refer to:
Error Summary (Part 2) Report
The System Error Summary (Part 2) report lists all
the permanent DASD errors. Permanent errors are
those that the DASD subsystem and/or the
operating system were unable to correct with the
defined recovery procedure.
Storage Subsystem Library: Maintaining
IBM Storage Subsystem Media, GC26-4495
Device Support Fat;ilities: Primer for the
User of IBM 3380 Direct Access Storage,
GC26-4498
See Figure 2 for an example of this report.
How to Use This Report
The report may be used as follows:
•
An error record may be found in this report if
the time of the failure is known or the job name
is known.
•
When the error record is located, the probable
failing unit (PFU) will determine which report to
examine next:
If the PFU is VOLUME,
The DASD Data Transfer Summary report can
be used for the volume defined to determine
<1>
<2>
TIME
JOBNAME
DATE 148/87
08:39:33:5Z
09:15:54:97
13:58:08:25
13:58:08:30
14:38:29:95
14:38:Z9:95
14:38:30:08
14:38:30:1Z
14:38:30:Z5
14:38:30:75
15: 21: 35: 28
15:21:36:3Z
19:58:45:86
19:58:49:15
19:58:49:29
19:58:50:35
DASDOO01
DASDOO01
MIN8C3
MIN8C3
ABC8CO
ABC8C1
ABC8CO
ABC8C1
ABCSC1
ABC8CO
ABC 980
ABC 980
ABC8C3
ABC8CO
ABC8C3
ABcaco
If the PFU is CONTROLLER, DEVICE, or MULTIPLE,
The maintenance package for that device
should be used. The DASD Symptom Code
Summary report will give symptom codes,
physical ID, and sense information for use in
the maintenance package. Use the PFU and
physical ID to identify the unit in the Symptom
Code Summary report.
If the PFU is CHANNEL or SCU,
Enter the maintenance package for the
channel or storage control.
<7> REPORT DATE 155 87
PERIOD FROM 148 87
149 87
TO
SYSTEM ERROR SUMMARY
(PART Z)
<3>
<4>
PHYSICAL
CPU
ID
A
B
A
B
A
B
B
A
B
B
B
A
A
A
A
A
ZO-60-xx
ZO-60-xx
05.x-03
05.x-03
05.X-00
04.X-Ol
05.X-00
05.X-Ol
04.X-01
04.X-00
ZO-60-00
20-60-00
04.X-03
04.X-00
05.X-03
04.X-OO
<5>
TYPE
<6>
PHYS
ADDR
<8>
ERROR
PATH
3380
3380
3380
3380
3380
3380
3380
3380
3380
3380
3380
3380
3380
3380
3380
3380
099Z
099Z
08C3
08C3
08CO
08Cl
08CO
OSCI
08C1
08CO
0980
0980
OSC3
08CO
08C3
oaco
Zl-099Z
Zl-0992
24-08C3
24-08C3
24-08CO
20-0SCl
Z4-08CO
24-08C1
ZO-08C1
20-08CO
Zl-0980
Zl-0980
Zo-08C3
ZO-08CO
Z4-0SC3
ZO-08CO
<9>
/'~
<11>
PROB
FAILING UNIT
<10>
VOL
ERROR DESC
SAS99Z
SAS99Z
SASSC3
SAS8C3
SAS8CO
SAS8CI
SAS8CO
SAS8C1
SAS8Cl
SAS8CO
SAS980
SAS980
SAS8C3
SASSCO
SASSC3
SASSCO
PERM
PERM
PERM
PERM
PERM
PERM
PERM
PERM
PERM
PERM
PERM
PERM
PERM
PERM
PERM
PERM
EQUIP
EQUIP
EQUIP
EQUIP
EQUIP
EQUIP
EQUIP
EQUIP
EQUIP
EQUIP
EQUIP
EQUIP
OATA
DATA
DATA
DATA
CHK
CHK
CHK
CHK
CHK
CHK
CHK
CHK
CHK
CHK
CHK
CHK
CHK
CHK
CHK
CHK
CONTROLLER
CONTROLLER
UNKNOHN
UNKNOHN
MULTIPLE
MULTIPLE
MULTIPLE
MULTIPLE
MULTIPLE
MULTIPLE
DEVICE
DEVICE
VOLUME
VOLUME
VOLUME
VOLUME
***************************************************************KKKKKKK**********************
CPU
A
B
Figure 2.
MODEL
3081XA
3081XA
SERIAL NUMBER
021170
ZZ1170
Sample of System Error Summary (Part 2) Report
PROG-2
(/"""'
PIN 4519907 23 Sides EC 475248 25Apr88
~ Copyright IBM Corp. 1987, 1988
"
(
(
(
(
(
(
Field Descriptions
<1> TIME
The TIME field is the date and time the error
was detected or logged.
<2> JOB NAME
The JOB NAME field is the name of the job
that was being performed when the error
occurred.
<3> CPU
(
(
(
("
The alphabetic identifier of the CPU
(processor) that received the error records.
At the bottom of the report, all processor
alphabetic identifiers are given with their
model and serial numbers.
<4> PHYSICAL 10
The PHYSICAL 10 field contains the physical
10 of the unit affected by the failure. This field
should be used to identify the physical failing
unit(s).
The physical 10 format for the 3380 is
CC.P.DD (3380-JK) and SS-CC-DD (device
types prior to 3380-JK), where:
SS = Storage Control/storage director 10
CC = A one byte controller string 10
P = Path
DO = Device 10
<5> TYPE
The TYPE field is the device type of the
Probable Failing Unit: 3380 for any model.
<6> PHYSICAL ADDRESS
(
(
{
(
The PHYSICAL ADDRESS (PCUA) field is
obtained from the outboard recorder (OBR)
record.
When there are multiple paths to a device,
the first path defined by the system is called
the primary path (primary channel, primary
control unit, and physical 10). All paths to a
device are defined during system generation
of the operating system.
<7> REPORT PERIOD
The REPORT PERIOD lines show the day and
the year the report was requested and the
days and the year requested in the EREP
keyword input parameters.
<8> ERROR PATH
The error path (SCUA) field is obtained from
the OBR record.
Note: The physical address (PCUA) and the
error path (SCUA) are four and six character
fields respectively. The channel and/or the
control unit address will be different when the
program selects the device through a path
other than the primary path. An example of
the meaning of the PCUA and SCUA is:
PCUA = 0197 and SCUA = 0497, where 1
and 4 are the primary and secondary
channels, 9 is the control unit, and 7 is
the device physical 10 (the device portion
is the low-order byte, bits 5-7).
<9> VOLUME
The VOLUME field is the volume in use at the
time of the error.
<10> ERROR DESCRIPTION
The ERROR DESCRIPTION field is a
description of the type of error that was
detected and recorded by EREP.
<11> PROBABLE FAILING UNIT
The PROBABLE FAILING UNIT field identifies
the unit which is most probably the source of
the failure. The EREP program does an
analysis relative to error records in the error
recording data set (EROS). The probable
failing unit in this section is a result of that
analysis.
<12> CPU ... MODEL ... SERIAL NUMBER
The model and serial number of the CPU
(processor) alphabetic character used in the
CPU field.
(
('
(
PIN 4519907 23 Sides EC 475248 25Apr88
~ Copyright IBM Corp. 1987. 1988
PROG-3
DASD Subsystem Exception Report
How to Use This Report
The Subsystem Exception DASD report lists
accumulated permanent and temporary errors
relative to the DASD subsystems. Outboard
recorder (OBR) and miscellaneous data recorder
(MDR) records are analyzed by the program for
symptom code, unit identification, type of error, and
other information. The accumulated errors are
given for each unit in the probable failing unit
category from channel to volume. Within each
section, the probable failing units are sequenced
by most permanent errors and then by most
temporary errors.
The 3380-JK temporary data checks and temporary
seek errors are only printed on this report when
the error rate exceeds a threshold in the storage
control. Therefore, any 3380-JK reporting read or
seek errors in this report needs servicing.
This report highlights problems related to disk
storage operation and should be used to identify
problems and to determine if previous corrective
action fixed a problem.
If the span of error records in the report covers
more than three days, a message is printed at the
top of the report. A report that spans a broad
period of time may not provide the most accurate
probable failing unit indication, because corrective
action may have been taken.
See Figure 3 on page PROG-5 for an example of
this report.
First determine the probable failing unit (PFU), then
check for the following:
If the PFU Is VOL,
The DASD Data Transfer Summary report
should be used. This report supplies the
necessary information for service using the
ICKDSF program routines. For media
maintenance, refer to:
Storage Subsystem Library: Maintaining
IBM Storage Subsystem Media, GC26-4495
Device Support Facilities: Primer for the
User of IBM 3380 Direct Access Storage,
GC26-4498
If the PFU is CTLR, DEV, or MULT,
The DASD Symptom Code Summary report
should be used. The symptom codes, unit
identifier, and other error information can be
used to enter the maintenance package.
If the PFU is CHAN or SCU,
Enter the storage control or channel
maintenance package to identify the failing
unit.
If a single device shows more than one PFU,
Determine which error has the highest
priority by using the procedures in the START
section of this manual.
,'£
\"
\
PROG-4
'""
~',
:,,)
-'
PIN 4519907 23 Sides EC 475248 25Apr88
CO Copyright
IBM Corp. 1987, 1988
r
\ \
~,_""I
(
<1>
SUBSYSTEM EXCEPTION
DASD
<2>
B-BUS OUT PARITY CHK
I-INVOKED OFFSETS
PROBABLE
FAILING
UNIT
<3>
<4>
<13>
REPORT DATE 155 87
PERIOD FROM 148 87
TO
149 87
C-CHECK DATA CHK D-DISKETTE CHK
S-SERVICE INFORMATION MESSAGE
FAILURE
AFFECT
<5>
-IMPACT OF TEMPORARY ERRORS-PHYSICAL -TOTALS-- EQU
ADDRESS
PERM TEMP
CHK
SKS
RD
OVRN OTHER
<7>
<8>
<9>
CPU
<6>
-USAGE1000
MB.
SKS
READ
<10>
***********************************************************************************************************
SCU
(
(
0010123.1
3990.02
SCU
CTLR 02. O-XX
<11> 33So-JK
SCU/CTLR
XX-83-XX
3380-DE
SCU/CTLR
DEV
41.X-02
3380-JK
DEV
VOL
ABC123
33S0-JK
DATAXFR
TOTAL
0010123.1
1
1
A
TOTAL
02.o-XX
4
4
A
TOTAL
XX-S3-XX
A
+
+
+
+
A
TOTAL
41.0-02
A
TOTAL
62.0-05
+
-+-
6
6
6
6
NlA
34
N/A
2S0
1
1
1
1
N/A
48
N/A
320
1
1
1
1
10
10
+
+
10
10
+--+-
N/A
5
N/A
120
-+-
N/A
50
NlA
240
************************************************************************************************************
o UNIT(S) EXCLUDED DUE TO LIMITS
CPU
A
MODEL
30S1XA
SERIAL NUMBER
021170
<12>
*
(
ENTRIES HITH AN ASTERISK INDICATE THAT DASDID CARDS HERE NOT
FOUND FOR THE UNIT.
NOTE: IMPACT OF TEMPORARY ERRORS IS THE NUMBER OF TIMES ERROR
THRESHOLD HAS BEEN EXCEEDED.
NOTE: BLANK ENTRIES INDICATE ZERO VALUES OR NOT APPLICABLE.
NlA MEANS NOT AVAILABLE.
NOTE: ZERO ENTRIES INDICATE RECORDS EXIST IN EREP REPORTS BUT
THRESHOLDS HERE NOT EXCEEDED.
Figure 3.
Sample of DASD Subsystem Exception Report
Field Descriptions
< 1> MESSAGE AREA
This space at the top of the report is used for
messages specific to this subsystem
exception report. For example: **WARNING**
REPORT SPANS MORE THAN 3 DAYS. PFU
ANALYSIS MAY BE IN ERROR.
(
The warning is given when the error input
records are collected over three days or
more. When the time is over three days, the
probable failing unit analysis could be in
error.
(
PIN 4519907 23 Sides EC 475248 25Apr88
@ Copyright IBM Corp. 1987, 1988
<2> B-BUS OUT PARITY CHECK
C-CHECK DATA CHK
D-DISKETIE CHK
I-INVOKED OFFSETS
S-SERVICE INFORMATION MESSAGE
These are definitions of the suffixes for the
counters that may appear in the column
identified in the IMPACT OF TEMPORARY
ERRORS column under the heading of
"OTHER."
<3> PROBABLE FAILING UNIT
The Probable Failing Unit (PFU) is the unit
which is most probably the source of the
failure. EREP identifies the PFU based on the
failure affect and the units reporting errors.
The seven PFUs assigned by EREP are as
follows:
PROG-S
/
CHAN
SCU
CTLR
MULT
DEV
VOL
UNK
Channel (channel, program, or
CPU)
Storage Control (storage control
function which resides between a
channel(s) and a controller(s».
Controller (function that provides
, an interface between one storage
control and a group of devices).
Identifies the failure as being
common to MULTIPLE addresses
on the same string.
Device (addressable unit)
Volume (data on volume)
Unknown
The line for each PFU describes its identifier,
the Failure Effect, and the total error counts
attributed to the respective combination of
PFU and Failure Effect. Usage counts for this
line are not available (N/A) since the total
usage of the unit is not determined.
<4> PFU Identifier
All units in each PFU category that had errors
are listed with probable failing unit identifiers.
The PFU identifier identifies the probable
failing unit. The PFU identifier contains the
address of the primary path to the failing unit.
It is not necessarily the address of the failing
path or the path that the error was reported
on. To determine the physical 10 of the failing
path, use the PHYSICAL ADDRESS column.
If the PFU is SCU - Storage Control,
the PFU identifier is:
SS.X-XX
3880-3 with 3380-JK
attachment feature
SS-XX-XX 3880 except Model 3880-3
with 3380-JK feature
SEQNUM.P 3990
SEQNUM.P 3380-CJ2
Where:
SS
=
storage control/storage
director 10
SEQNUM
P
/
-
= a 7-digit manufacturer's
sequence number
the storage path within the
unit (storage path 0-3)
=
If the PFU is CTLR - Controller,
the PFU identifier is:
CC.P-XX 3380-JK
XX-CC-XX device types prior to 3380-JK
Where:
CC =
P =
controller 10, and
storage path
If the PFU is MULT - Multiple,
the PFU identifier is:
CC.X-XX 3380-JK
XX-CC-XX 3380-DE
Where CC = controller 10
'<..
_1
If the PFU is DEV - Device,
the PFU identifier is:
/
---,
CC.X-DD 3380-JK
XX-CC-DD devices prior to 3380-JK
Where:
CC =
DD =
controller 10, and
physical device 10.
-,,-
If the PFU is VOL - Volume,
the PFU identifier is "nnnnnn," the
volume serial number from the
OBR/MDR device-dependent VOLID
field.
<5> FAILURE AFFECT
The FAILURE AFFECT field defines the
function or machine area affected by the
failure. The following ten failure affects are
used for the 3380-JK:
CHAN/SCU Channel/Storage Control Unit
SCU
Storage Control Unit
SCu/CTLR Storage Control UniVControlier
(DOC Interface)
CTLR
Controller
CTLR/DEV Controller/Device Interface
if\
I~,
PROG-6
PIN 4519907 23 Sides EC 475248 25Apr88
© Copyright
IBM Corp. 1987, 1988
(
(
(
MULTIPLE Failure common to more than one
device
DEV
Device
SEEK
Servo mechanism or associated
logic
DATAXFR Read/Write heads or associated
logic
UNK
Unknown
<6> CPU
The alphabetic identifier of the processor
(CPU) that received the error records. At the
bottom of the report, all processor alphabetic
identifiers are given with their model and
serial numbers.
<7> PHYSICAL ADDRESS
(
(
The PHYSICAL ADDRESS field is the physical
ID of the unit affected by the failure. This field
should be used to identify the physical failing
unit(s) and as a means of locating information
on other EREP reports.
The physical address format for the 3380 is
CC.P.DD (3380-JK) and SS-CC-DD (device
types prior to 3380-JK), where:
EQU CHK Temporary Equipment Checks
Temporary Seek Checks
SKS
The sum of the temporary data
RD
checks during reading which were
corrected by retrying plus the sum
of the temporary data checks
during reading which were
corrected by ECC.
OVRN
Counts of overruns which are
system retried (only applicable to
a PFU of CHAN)
All other temporary errors. The
OTHER
types are identified by the letter
suffix. In the case of multiple error
types, multiple letters follow the
counter.
<10> USAGE
The usage figures are in units of one
thousand for seeks and megabytes for data
checks.
< 11 > Device Type
The device type of the Probable Failing Unit:
(
(
occurred in the device, but sense data may
still exist in EREP.
SS = Storage Control/storage director ID
CC = A one-byte controller string ID
P = Path
DD = Device ID
<8> TOTALS (PERMITEMP)
PERM TOTALS: This is the count of
permanent errors recorded against the unit
and totaled for the Probable Failing Unit for
the given FAILURE AFFECT.
TEM P TOTALS: This is the sum of the counts
shown under IMPACT OF TEMPORARY
ERRORS.
3380
338o-DE
3380-JK
3380-CJ
for models A04, AA4, and 804
for models AD4. AE4. 8D4. and
BE4
for models AJ4. AK4. BJ4. and BK4
for model CJ2
<12> CPU ... MODEL ... SERIAL NUMBER
The model and serial number of the !
processor (CPU) alphabetic character used in
the CPU field.
Note: "XA" following the CPU model number
indicates that the processor complex was
running in 370-XA mode.
<9> IMPACT OF TEMPORARY ERRORS
These fields indicate the number of temporary
errors when the count exceeds a limit value.
A zero (0) entry indicates that records exist in
EREP but their rate did not exceed threshold.
A blank entry indicates that no error condition
PIN 4519907 23 Sides EC 475248 25Apr88
l!::l Copyright
IBM Corp. t987, 1988
< 13 > REPORT PERIOD
The REPORT PERIOD lines show the day and
the year the report was requested and the
day and the year of the earliest and latest
record.
PROG-7
DASD String Summary
How to Use This Report
The OASD String Summary report lists all usage for
any device reporting errors or usage during the
reporting period.
The usage information in this report can help you
determine whether a failure affect reported in the
Subsystem Exception DASO report is unique to a
particular device or is common to more than one
device in the same controller string.
The DASO String Summary can also help you
determine the Storage Control Unit/Controller
paths involved in a failure. The information at the
beginning of each string contains the physical 10
for all the SCU/Controller paths that reported either
usage or errors.
~""",
I
\'-<,:.-;,:/
Information is grouped by CTLR 10 if the
Subsystem 10 (SSIO) is not aVailable. If the SSIO is
available, information is grouped by SSIO.
See Figure 4 for an example of this report.
DASD STRING SUMMARY
REPORT DATE 155 87
PERIOD FROM 148 87
.
TO 149 87
REPORT INCLUDES ALL DASDHITH PHYSICAL IDS OR A VALID DASDID
ERROR TYPES
SEEK
CTLR
EQU. /SEEK IDATA
ACC MEGABYTES
SSID SCUID
ID
DEV
VOLUME
CHKS
XFER
XI000
READ
MEGABYTES
HRITTEN
HlVERIFY
**************•••••••••***************************************************~**********
<1>
<2>
<3>
ALL DASD PROCESSED FOR EXCEPTION REPORT
615
3017
<3>
***************************************••••••••••••••••••••************************
<4>
CPU
A
B
MODEL
3081XA
3081XA
SERIAL NUMBER
021170
221170
<4>
Figure 4. Sample of DASD String Summary Report
I
,:
"'~"
c"'_ ,/
Note: The counts for "SEEK ACC X1000"
(thousands of seek accesses), "MEGABYTES
READ," and "MEGABYTES WRITTEN
WIVERIFY" are six digit positions. If the
space is exceeded, the count is divided by
tooo and a "K" suffix is added to the number.
If the count exceeds 99,999,000, "99999K" will
be printed.
PROG-8
P/N-4519907 23
Sides EC 475248
© Copyright
25Apr88
IBM Corp. 1987, 1988
c
(
(
Field Descriptions
The string summary is in four parts as follows:
(
<1> Part One
The first part shows usage data for each
unique combination of vo/t)me, SSID, and
physical 10. The physical 10 is presented by
SCUID, string 10 and device fields.
Failure Affects of CTLR, CTLR/OEV, and OEV
are grouped under the heading of EQU CKS;
SEEK and DATA TRANSFER errors are noted
under their own headings.
(
Failure Affects of CTLR, CTLR/OEV, and OEV
are grouped under the heading of EQU CKS;
SEEK and DATA TRANSFER errors are noted
under their own headings.
The usage data for each volume/physical 10
appears under three possible headings as
follows:
•
•
•
Seek Accesses x 1000
Megabytes Read
Megabytes Written with VERIFY
<2> Part Two
The second part (right hand part of the report)
shows failure affects and usage data for each
unique combination of volume and device 10.
The usage data for each volume/physical 10
appears under three possible headings as
follows:
•
•
•
Seek Accesses x 1000
Megabytes Read
Megabytes Written with VERIFY
< 3 > Part Three
The third part of the report shows the usage
statistics for all OASO with a valid physical 10
processed for the Subsystem Exception
report, regardless of failure affect or whether
any failures were reported.
<4> Part Four
The fourth part of the summary shows all
CPUs processed for the exception report, by
letter identifier, model number, and serial
number.
(
(
f
(
(
(.
PIN 4519907 23 Sides EC 475248 25Apr88
«:> Copyright
(
IBM Corp. 1987, 1988
PROG-9
I
'.
\tj
DASD Informational Messages
The DASD Informational Messages report provides
information about conditions in the DASD
subsystem. The messages are sequenced by
physical ID and symptom code. See Figure 5.
The messages for the 3380-JK are:
•
•
•
DASD INFORMATIONAL MESSAGES
<1>
PHYSICAL
ID
<2>
SYMPTOM
CODE
Threshold logging complete for seek checks.
Threshold logging complete for data checks
without offset.
Device fenced from storage path.
REPORT DATE 304 87
PERIOD FROM 304 87
TO
304 87
<3>
<4>
COUNT
MESSAGE
,,.._ ........
,,
***********************************************************************************
20.2-09
40.1-00
62.2-05
Figure 5.
0001
0002
0225
3
6
1
THRESHOLD LOGGING COMPLETE FOR SEEK CHECKS
THRESHOLD LOGGING COMPLETE FOR DATA CHECKS HITHOUT OFFSET
DEVICE FENCED FROM STORAGE PATH
Sample of DASD Informational Message Report
Field Descriptions
<1> PHYSICAL 10
For a 3380-JK, the physical ID field is the ID of
the unit that recorded the error. The format
is: CC.P-DD, where CC equals the
hexadecimal ID for the controller, P equals
the hexadecimal storage path. and DD equals
the hexadecimal ID for the device.
<3> COUNT
The number of records of this type for the
defined storage control.
<4> MESSAGE
./
This field contains the messages for each
physical ID.
<2> SYMPTOM CODE
The Symptom Code field is generated in the
storage control. Symptom codes for 3380-JK
are:
•
0001 = Threshold Logging complete for
seek checks
•
0002 = Threshold Logging complete for
data checks without offset
•
For the Device Fenced from Storage Path
message, the symptom code field
contains the device address (CUU)
instead of a symptom code.
' ......-
"
,
-.. ~
.'Wc
.
/1 '
I
'<1./
PROG·10
PIN 4519907 23 Sides EC 475248 25Apr88
© Copyright
IBM Corp. 1987, 1988
(
(
«
DASD Data Transfer Summary
The DASD Data Transfer Summary report lists
permanent and excessive temporary data check
errors. Temporary errors are listed only when a
specified error rate, which is monitored in the
storage control functional microcode, is exceeded.
DASD DATA TRANSFER SUMMARY
PROBABLE FAILING UNIT - VOLUME
(
The report lists each volume that had data checks
reported in the Subsystem Exception DASD report.
Refer to Figure 6 and the field descriptions on
page PROG-12.
<8> REPORT DATE 034 86
PERIOD FROM 032 86
TO
032 86
<9>
SENSE COUNTS
TEMPORARY
OFFSET INVK THRESHOLD
PERM
NO
YES LOGGING
***********************************************************************************
<1> SEQUENCE BY VOLUME LABEL, PHYSICAL ADDRESS, HEAD, CYLINDER
<2> UNITADDRESS 03C3
DEVTYPE 3380-JK
VOLUME
<3>
CPU 8
PHYSICAL ADDRESS 22.X-03 ?
<4>
<5>
<6>
FAILURE AT ADDRESS: CYLINDER 0793 HEAD 05
00003013 03193541 03190005 08C02131 00010015 00F44160
140004EO 00031905 LAST SENSE AT: 060/88 15:00:40:11
o
o
8
o
FAILURE AT ADDRESS: CYLINDER 0793 HEAD 10
00003010 83193A41 0319000A 07412220 00010015 00f44160
140004F2 0003190A LAST SENSE AT: 060/86 15:00:48:39
o
o
2
o
FAILURE AT ADDRESS: CYLINDER 0087 HEAD 01
00003000 07570143 00590001 OCA64200 0100B3E6 00064380
LAST SENSE AT: 032/86 15:38:35:31
o
1
o
IS
FAILURE AT ADDRESS: CYLINDER 0095 HEAD 02
00001000 075F0243 005F0002 06B64200 00010000 00064340
LAST SENSE AT: 032/86· 15:39:44:36
o
o
o
8
FAILURE AT ADDRESS: CYLINDER 0105 HEAD 04
00003001 47690443 00690004 06AE4200 01000000 00054340
LAST SENSE AT: 032/86 15:47:13:08
o
1
o
16
FAILURE AT ADDRESS: CYLINDER 1648 HEAD 10
00001001 47706A43 0670000A 06AE4200 01000000 00054340
LAST SENSE AT: 032/86 15:47:13:08
o
o
o
6
<2> UNITADDRESS 0287
DEVTYPE 3380-JK
<3>
CPU A PHYSICAL ADDRESS 42.X-07
<4>
<5>
<6>
(
RAS3C3
VOLUME HSMBOO
********************************************************************************
MODEL
SERIAL NUMBER
<7> CPU
8
NOTE
NOTE
NOTE
(
Figure 6.
3081XA
021170
CYLINDER/HEADIBLOCK NUMBERS ARE DECIMAL VALUES
UNITADDRESS IS THE LOGICAL ADDRESS OF THE DEVICE
? FOLLOWING THE PHYSICAL ADDRESS DENOTES MULTIPLE
PHYSICAL UNITS HAD ERRORS NITH THIS VOLUME LABEL
Sample of DASD Data Transfer Summary Report
(
(
PIN 4519907 23 Sides EC 475248 25Apr88
@ Copyright IBM Corp. 1987, 1988
PROG-11
Figure 6 on page PROG-11 shows:
•
Device 03 on string 22, has exceeded the
temporary data check (OFFSET:i:: YES) threshold
ten times (eight times on head 05 and two
times on head 10):
Cylinder 793 head 05 - Eight temporary data
checks on this track caused the threshold
to be exceeded. The track had 8 temporary
errors.
Cylinder 793 head 10- This track had two
temporary data checks recorded while the
device was in logging mode.
•
•
Sense byte 4 bits 0 and 1 tell which path the
error occurred on:
Sense Byte 4· . Path
Path
Path
Path
Path
OOxxxxxx
01xx xxxx
10xx xxxx
11xx xxxx
0
1
2
3
/.',
\,.... j"
Field Descriptions
<1>
SEQUENCE BY VOLUME, PHYSICAL
ADDRESS, HEAD, AND CYLINDER.
Entries are in sequence by volume, physical
address, head, and cylinder.
Device 07 on string 42, has exceeded the
temporary data check (OFFSET = NO) threshold
two times.
<2>
UNITADDRESS ... DEVTYPE ... VOLUME
Cylinder 87 head 1 - One temporary data
check on this track caused the threshold to
be exceeded. The track had a total of 16
temporary errors.
UNITADDRESS is the key word used by
Device Support Facilities to identify the
device. The device is the logical address of
the volume reporting the error.
Cylinder 95 head 2 - This track had eight
temporary errors recorded while the device
was in logging mode.
DEVTYPE contains the device type:
3380
3380-DE
Cylinder 105 head 4 - One temporary data
check on this track caused the threshold to
be exceeded. The track had a total of 17
temporary errors.
Cylinder 1648 head 10 - This track had six
temporary errors recorded while the device
was in logging mode.
•
Sense byte 21 shows the Physical 10 of the
Storage Control.
•
Sense byte 14 shows the Controller String 10 in
format 4 and 5 sense.
3380-JK
3380-CJ
VOLUME is the customer assigned volume
label.
<3>
CPU ... PHYSICAL ADDRESS
The CPU ·field identifies which processing "mit
supplied records for this report.
The PHYSICAL ADDRESS field identifies the
device physical address. The physical
address format is CC.P-DD, where:
CC
P
DD
PROG~12
for models A04, AA4, and B04
for models AD4, AE4, BD4, and
BE4
for models AJ4, AK4, BJ4, and BK4
for model CJ2
= A one-byte Controller String 10
= Path
= Device 10
PIN 4519907 23 Sides EC 475248 25Apr88
(;) Copyright IBM Corp. 1987, 1988
\
"'<",
./
.
(
(
(
(--
<4> FAILURE AT ADDRESS
The failure at address field defines the logical
cylinder and head entries (in decimal
numbers) for the recorded errors. The
hexadecimal cylinder and head number can
be determined from sense bytes 5 and 6
using the following diagram.
Sense
byte 5
(
02
II
cJJLr
(
03 02
Sense
byte 6
31
I
<8> REPORT PERIOD
The REPORT PERIOD lines show the day and
year the report was requested and the day
and year of the earliest and latest records.
< 9 > SENSE COUNTS
See Figure 7 for an explanation of the sense
count data:
PERM
TEMPORARY OFFSET INVK NO
TEMPORARY OFFSET INVK YES
THRESHOLD LOGGING
Head
01
In this example, byte 5 contains 02 and byte 6
contains 31. The address is cylinder X"0302,"
head X "01."
< 5 > Sense Records
(
(
The last sense record is given for each
cylinder and head address recorded for each
volume reporting errors. Sense records are
listed in sequence by head and cylinder:
•
•
For a 3990 or 3380-CJ2 subsystem, 32
bytes of sense data are displayed: 24
bytes of sense data on this line and 8
bytes preceding the words" LAST SENSE
AT" on the next line.
•
The format of the record (4) is in byte 7.
•
The symptom code is in sense bytes 22
and 23.
(
(
(
(
For a 3880 subsystem, 24 bytes of sense
. data are displayed on this line.
<6> LAST SENSE AT
The LAST SENSE AT field is the date and time
of the last sense record received for the
specified cylinder/head.
<7> CPU ... MODEL ... SERIAL NUMBER
The CPU, MODEL, SERIAL NUMBER field
identifies the processing units that supplied
records to this report.
PIN 4519907 23 Sides EC 475248 25Apr98
© Copyright IBM Corp. 1987. 1988
PROG-13
DASD DATA TRANSFER SUMMARY
PROBABLE FAILING UNIT - VOLUME
REPORT DATE 034 87
PERIOD FROM 032 87
TO
034 87
*************************************************
U_ 'Ms
=l~
to Hnd the """""" of . .
SENSE COUNTS
TEMPORARY
OFFSET INVK THRESHOLD
PERM
NO
YES LOGGING
1(1(1(1(1(1(1(11
If*************************
~~t ~
data checks that occurred in the report period.
Use this colunn to find the number of times the temporary _ _ _- I
data check threshold was exceeded in the report period.
The error records associated with numbers in this column
contain the volume ID and the device address of the
device that exceeded the threshold. The track that caused
the threshold to be exceeded is not necessarily the track
with the highest error rate.
Use this column to find the number of temporary data checks - - - - - - - '
recovered with offset.
Use this column to find the number of temporary data checks
that were logged.
---------1
-,,--j
The THRESHOLD LOGGING field is a count of the temporary data
checks for the specified head/cylinder. This count was
recorded when the device was in logging mode. To determine
the total number of data checks logged for a given device.
add the values in the offset NO column to the values in the
Threshold Logging column.
Figure 7.
Headings for DASD Data Transfer Summary Report
Understanding Temporary Data Check Reporting
Thresholds (3880)
Use Figure 7 on page PROG-13 for reference.
For 3880 storage controls, there is a temporary
data check threshold counter for each device in
each storage director. If the threshold is exceeded,
the storage director goes into logging mode for
data checks and logs the next 24 data checks that
occur on the device. If the data checks are caused
by the device or media (not the path), both storage
directors will usually exceed threshold and report
data checks. If the data check problem is path
related, usually only one storage director will
exceed threshold and data checks will be reported
for multiple devices (depending on usage).
Data checks recovered without offset have
thresholds in the storage control. Occasional
temporary data checks are expected and do not
appear in the EREP reports unless the threshold is
exceeded.
When a threshold is exceeded, the RD column of
the System Exception DASD report contains a
count of the number of times the threshold was
exceeded. See Figure 3 on page PROG-5.
Thresholds (3990)
For 3990 storage controls, there is a temporary
data check threshold counter for each device,
regardless of path. If the threshold is exceeded, the
storage control goes into logging mode for data
checks that occur on the device regardless of path.
PROG-14
PIN 4519907 23 Sides
EC 475248 25Apr88
@ Copyright IBM Corp. 1987, 1988
(
Data Check Reporting
(
(
When a threshold is exceeded, and the storage
director or storage control goes into logging mode,
the first sense records logged will update the count
in the DASD Data Transfer Summary report,
OFFSET INVK - NO column. An additional 23 data
checks are logged and the count in the
THRESHOLD LOGGING column is updated for each
one. At the completion of logging mode (24 data
checks logged), a message is presented to the
operator, a Format 0 sense record is logged, and a
message is presented in the DASD Informational
Messages report (symptom code 0002).
Temporary Data Checks without Thresholds
Temporary data checks with offset active and
temporary data checks with symptom codes 4xCx
do not have thresholds and are all logged and
reported. Temporary data checks with offset active
update the count in the OFFSET INVK - YES column
and data checks with symptom codes 4xCx, update
the count in the OFFSET INVK - NO column.
((
(--
(
(
(
(
(
PIN 4519907 23 Sides EC 475248 25Apr88
C Copyright
IBM Corp. 1987, 1988
PROG-15
DASD Symptom Code Summary
This report provides information required for
hardware servicing. Each sense record reported in
the exception report is listed by PFU, symptom
code, and physical ID.
The symptom code is to be used with the 3380-JK
service procedures in the START section of this
manual. The report allows you to locate the
failures noted in the DASD Subsystem Exception
report and it allows you to note the symptom code
and first sense record for each failure. Data
checks (symptom codes 4XXX), which appear in the
DASD Data Transfer Summary, also appear here,
for use when hardware repair is required.
Use Figure 8 on page PROG-17 with the field
descriptions that follow.
',-- j
/"
PROG-16
PIN 4519907 23 Sides EC 475248 25Apr88
@ Copyright IBM Corp. 1987, 1988
<6>
REPORT DATE 034 86
PERIOD FROM 032 86
TO
032 86
<1> SEQUENCE BY PROBABLE FAILING UNIT
<2> SYMPTOM
CODE
<3>
PHYSICAL OCCURRENCES FAILURE
ID
PERM/TEMP
AFFECT
DATE AND TIME OF
FIRST OCCURRENCE
LAST OCCURRENCE
SENSE FROM FIRST OCCURRENCE
0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 2 2 2 2
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3
PHYSICAL
ERROR
ADDRESS
SSID-STRING
PATHS
CPU(S}
DEVICE
TYPE
<4>
*******************************************************************************
(
(
(
<1> PROBABLE
PROBABLE
SEQUENCE
<2> OFOO *
<3>
<4>
FAILING UNIT: C H A N N E L - - - - - - - - - - - - - - - - - - - - FAILING UNIT: STORAGE CONTROL - - - - - - - - - - - - - - - - BY SCUID, SYMPTOM CODE - - - - - - - - - - - - - - - - - - - - - - - 05.
0
1
CHAN/SCU
032/86 01:02:47:49
032/86 01:02:47:49
3880
04000000 4SC42000 00000000 00000000 00000000 OOOSOFOO
05.
0005-X
12-0285 A
PROBABLE FAILING UNIT: CONTROLLER--------------------------------SEQUENCE BY CTLID, SYMPTOM CODE - - - - - - - - - - - - - - - - - - - - - - E2CO *
80.o-xx
1
0
SCU/CTLR
032/86 10:46:50:27
032/86 10:46:50:27
3380-JK
10008280 24000183 81100100 COOOOOOO 0804C868 0007E2CO
80.0-04
0004-1
07-0794
A
PROBABLE FAILING UNIT: DEVICE----------------------------------------SEQUENCE BY CTLID, DEVID, SYMPTOM CODE - - - - - - - - - - - - - - - - - - - - - 9302 *
40.X-00
8
0
DEV
032/86 01:14:14:01
032/86 01:14:18:42
3380-JK
10000040 001E3C96 89041E04 58020427 00005838 00049302
40.0-00
0004-0
07-0780
A
40.1-00
0007-0
17-0780
B
PROBABLE FAILING UNIT: V O L U M E - - - - - - - - - - - - - - - - - - - - SEQUENCE 8Y CTLID, DEVID, SYMPTOM CODE - - - - - - - - - - - - - - - - - - - - - - 43AO *
42.X-02
0
2
DATAXFER
032/86 15:21:50:71
032/86 18:33:51:07
3380-JK
00003000 02360A43 0036000A 00034210 00000000 000643AO
42.0-02
0006-0
02-0282
A
(
(
(
4000
*
42.X-07
3380-JK
0
06
4300
*
42.X-07
3380-JK
0
17
DATAXFER
032/86 15:21:50:71
032/86 18:33:51:07
00003000 47690443 OOOBOOOA 06B64200 00010000 00054300
42.1-07
0005-0
12-0287
AB
4340
*
42.X-07
3380-JK
0
8
DATAXFER
032/86 15:21:50:71
032/86 18:33:51:07
00001000 07SF0243 OOOBOOOA 06864200 00000000 00064340
42.0-07
0006-0
02-0287
A
4380
!If
42.X-07
3380-JK
0
16
DATAXFER
032/86 15:21:50:71
032/86 18:33:51:07
00003000 07180B43 0018000B OCA64200 01000000 00064380
42.0-07
0006-0
02-0287
A
DATAXFER
032786 15:21:50:71
032/86 18:33:51:07
00003000 47706a40 FF6DDED8 00FF4200 02000000 00054000
42.1-07
0005-0
12-0287
A
*******************************************************************************
<5> CPU
A
B
MODEL
SERIAL NUMBER
3081XA
3081XA
021170
221170
NOTE: SYMPTOM CODES HITH AN ASTERISK ARE COUNTED AS ERRORS IN EXCEPTION REPORT
NOTE: PHYSICAL ID OF N/A MEANS THERE HERE NO DASDID CARDS
Figure 8. Sample of DASD Symptom Code Summary Report
(
PIN 4519907 23 Sides EC 475248 25Apr88
~ Copyright IBM Corp. 1987. 1988
PROG-17
Field Descriptions
<1> SEQUENCE BY PROBABLE FAILING UNIT
The DASD Symptom Code Summary is
divided into seven separate sections by
probable failing unit (PFU). The PFU assigned
by EREP can be used to help define the
failure boundary. The seven PFUs are:
•
•
•
•
•
•
•
Channel
Storage control
Controller
Multiple devices
Device
Volume
Unknown
If there are no entries for a unit, nothing is
printed in the report for that PFU.
<2> SYMPTOM CODE ... PHYSICAL ID ...
OCCURRENCES ... FAILURE AFFECT ...
DATE AND TIME
The SYMPTOM CODE field contains the
symptom code from bytes 22 and 23 of the
sense data. The sense data, collected and
assembled by the storage control, documents
the conditions that surround an error. For
details of the individual sense bytes, see the
SENSE section in the ECM, Volume R40.
The asterisk (*) indicates that the record is
not generated by logging mode.
The PHYSICAL 10 is the physical address of
the primary path to the failing unit. It is not
necessarily the address of the failing path or
the path that the error was reported on. To
determine the physical 10 of the failing path,
use the PHYSICAL ADDRESS column. The
physical address format for the 3380 is
CC.P-DD (3380-JK) and SS-CC-DD (device
types prior to 3380-JK), where:
SS
CC
P
DD
PROG-18
= Storage Control/Storage
Director
= A one byte Controller String 10
= Path
= Device 10
The OCCURRENCES field contains a count of
permanent (PERM) and temporary (TEMP)
errors.
The FAILURE AFFECT field defines the
function or machine area affected by the
failure. The following ten failure affects are
used for the 3380-JK:
=
CHAN/SCU
Channel/Storage Control Unit
SCU
= Storage Control Unit
SCu/CTLR = Storage Control Unit/Controller
(DOC Interface)
=
CTLR
Controller
CTLR/DEV = Controller/Device Interface
MULTIPLE
Failure common to more than
=
one device
= Device
= Servo mechanism or associated
logic
DATAXFR = Read/Write heads or
associated logic
UNK
= Unknown
DEV
SEEK
The DATE AND TIME field contains the date
and time of the first and last occurrence of
the symptom code.
<3> DEVICE TYPE ... SENSE BYTES
The DEVICE TYPE field is the machine type
(device) affected by the failure:
3380
3380-DE
3380-JK
3380-CJ
for models A04, AA4, and B04
for models AD4, AE4, BD4, and
BE4
for models AJ4, AK4, BJ4, and BK4
for model CJ2
\-...-
-,-/
The SENSE BYTES fields contain the first
sense record received for the symptom code.
The sense record contains 24 bytes for 3880
attachment and 32 bytes for 3990 attachment.
<4> PHYSICAL ADDRESS ... SSID·STRING ...
ERROR PATH ... CPU(S)
The PHYSICAL ADDRESS field is the physical
10 of the unit affected by the failure. This field
should be used to identify the physical failing
unit(s). The physical address format is
CC.P-DD, where:
n '-'",
/J
PIN 4519907 23 Sides EC 475248 25Apr88
@ Copyright IBM Corp. 1987, 1988
c
(
(
(-'
(
(
CC
P
DD
= A one-byte Controller String 10
= Path
= Device 10
The SSID-STRING identifies the Storage
control and String address where:
SSID
STRING
= Storage control physical 10
= Controller string address
(0 or 1)
The ERROR PATH field identifies the path
returning the sense information and does not
necessarily identify the failing unit. In 370
mode this will be a 4 digit address, in
XA-mode it will be a 2-digit CHPID (Channel
Path Identifier) and a 4-digit Device number.
Use the PHYSICAL ADDRESS to identify the
failing unit.
The CPU(S) field defines the processing units
which have EREP records for the device.
<5> CPU, MODEL, SERIAL NUMBER
The CPU, MODEL, SERIAL NUMBER field
indicates all the processing units that
supplied records to this report.
<6> REPORT PERIOD
The REPORT PERIOD lines show the day and
the year the report was requested and the
day and the year of the earliest and latest
record.
(
(
(
f
(
(
«
f
(
PIN 4519907 23 Sides EC 475248 25Apr88
~ Copyright IBM Corp_ 19B7, 19BB
PROG-19
(
'''L,/
c
Event History Report
The System Exception reports are normally all that
is required to isolate a problem. If additional
information is required, an Event History report
(EVENT) may be run against the working data set.
If this is done, have the customer use the following
parameters:
This report provides a one-line summary of each
entry in the System Error Record Data Set (ERDS).
Selection parameters can be used to select records
by device type, date, and time. Refer to
Environmental Recording, Editing, and Printing
Program User's Guide and Reference, GC28-1378.
EVENT=Y, HIST, ACC=N,
TABSIZE=999K,
DEV=(3380), CUA=(xxx-xxx)
where xxx-xxx is the device address (CUA) range
of the string.
EVENT HISTORV (S/370XAI
<2>
TIME
DATE
<3>
<4> <5>
<6>
<7> <8>
JOSNAME
RECTVP CP CUA
DNO
SSVS ID REASON
DEVT
CHD CSW
CRW CHP
SCSW
GAM3C2
GAM3C2
GAM3C2
GAM3C2
GAM3C2
OBRTMP
OBRTMP
OBRPRM
OBRPRM
OSRPRM
ASVNCH
ASVNCH
ASVNCH
OBRPRM
OBRPRM
OBRTMP
OBRTMP
08RTMP
OBRTMP
MDRDAS
3380 03 ID OE40
3380 03 ID OE40
3380 03 ID OE40
3380 03 31 OEOO
3380 03 31 DEOO
3380-CJ RAS9E2
3380-CJ RAS9E2
3380-CJ RAS9E3
3580 13 31 O~OO
3380 13 31 0600
3380 03 ID OEOO
3380 03 10 OEOO
3380 03 31 0600
3380 03 31 0600
3380 13
°
<9>
SPID
PSW-MCH IPROG-EC
SENSE
04 06
REPORT DATE
PERIOD FROM
<14>
TO
SNID
<10>
RCVRVXIT
COMP/MOD CSECTID ERROR-ID·
08 10
12 14
16 18
20 22
VOLUME
SEEK
ESW
338 87
338 87
338 87
<12>
<13>
SO
CT
338 87
II 33 52
II 33 52
11 33 52
II 33 55
II 33 57
II 37 10
II 37 13
II 37 14
13 12 12
13 12 12
13 12 13
13 12 13
13 12 13
13 12 37
13 13 23
43
60
99
09
53
76
49
77
25
36
21
24
64
46
16
Figure 9.
N/A
N/A
N/A
GAM321
GAM321
GAM320
GAM320
GAM321
GAM321
N/A
oA
-B
0A
-S
08
oS
oS
"S
0A
"8
"8
-S
oA
08
08
03C2
03C2
03C2
03C2
03C2
09E2
09E2
09E3
0321
0321
0320
0320
0321
0321
0320
10000063 02030190 0120lE04
10000063 02030190 0120lE04
10000063 02030290 01201E04
10000063 02000190 01201E04
10000063 02000190 0120lE04
00 I 0 1 000 00008FEO 2380001F
00101000 00008FEO 2380000F
00901000 00008FEO 430000lF
10000063 41030890 81Z0lE04
1000006341193890 8120lE04
10008263 00193D83 81100130
10008263 0003008381100130
10000063 01193890812011004
10000063 010S0190 S1201E04
00001063 40007969 0000E4E8
08500507
08500407
08500407
08100007
08100007
FF000804
FF000604
FF000404
49800817
49800817
00000000
00000000
A8020407
A8020407
00000000
0000004B
0000004B
00000048
00000049
00000049
23013AEO
23013AEO
23013AEO
00004818
00004818
OF04CA08
OF04CA08
00008F18
00008FIS
04370000
00F49350
00F49350
00F49350
00F49310
00F49310
00332S0 1
00332620
00EE2COI
00799349
00799249
0078E130
0078E130
00789302
00789302
00790000
040DOCEO
040DOCEO
040DOCEO
040DOCEO
040DOCEO
05100201
05100200
05100201
RAS321
RAS321
RAS3Z0
RAS320
RAS321
RAS321
RAS320
00000301
00000301
00000302
00000001
00000001
F I 003300
FI003300
FIOOEEOO
3 8
793 8
793 13
3 13
793 8
5 I
63.0
63.0
63.0
63.0
63.0
63.1
63.1
63.0
63.0
63.0
63.0
63.1
Sample Event History Report
Comments about the sample Event History Report:
Field Descriptions
•
The first five lines could be produced by
3380-JK devices attached to a 3380-CJ2 or a
3990 storage control.
•
The next three lines were produced by
3380-CJ2 storage control errors.
•
The next six lines were produced by 3380-JK
devices attached to a 3880 storage control.
•
The last line shows one MDRDAS record type
for a 3380-JK device attached to a 3880 storage
control (the remaining MDRDAS records are
not shown).
<1> TIME
The Event History report lists each of the
logged errors ordered by time (hour, minute,
second, hundredths of a second) provided by
the error recovery procedure (ERP) of its
associated operating system. This time is
usually not the time of error, but (in most
cases) the time of recording. If any of the
attached operating systems have incorrect
time-or-day clocks, relationships may not be
initially understandable until this difference is
taken into account.
<2> JOBNAME
The JOBNAME field indicates the job name
being processed at the time of the failure.
<3> RECTYP
The record type (RECTYP) field indicates
whether the outboard recorder (OBR) record:
PROG-20
PIN 4519907 23 Sides EC 475248 25Apr88
~ Copyright IBM Corp. 1987, 19S5
.
c
(
(
OBRPRM
A permanent error.
OBRTMP
A temporary error. The system error
recovery procedures permitted a
successful recovery from the error.
ASYNCH
Asynchronous data records contain the
32-byte service information message
(SIM) generated by the storage control.
(
(-
MDRDAS
Miscellaneous data records: the
statistical data (number of seeks and
number of bytes read) for the device
reported by the storage director to the
system.
<4> CP
The CP field indicates the system (CPU)
where the error was logged.
(
(-
The SENSE field contains the sense data
collected for an OBR or MDR record. The
high-order hex character in byte 7 indicates
the format for 3380 OBR and MDR records.
See the SENSE section in the ECM, Volume
R40 to identify the individual bits. MDR
records do not have their first eight bytes (0
through 7) printed.
Note: 24 sense bytes are reported when
attached to a 3880 storage control. 32 sense
bytes are reported when attached to a 3990
storage control.
<10> VOLUME
When the 3380-JK is attached to a 3880
storage control, the VOLUME field contains
the volume identifier (VOLlD) of the volume.
When the 3380-JK is attached to a 3990
storage control,
•
For Format 4 errors and record type
"MDRDAS," the VOLUME field contains
the volume identifier (VOLlD).
•
For other than Format 4 errors and record
type "MDRDAS," the VOLUME field
contains sense bytes 24 through 27.
<5> CUA
The control unit address (CUA) field contains
the primary channel and unit address (PCUA)
used by the operating system to select the
device. This CUA may not be the CUA used to
select this device at the time of failure.
For the 3380, the unit's position is the
physical 10 of the device.
<6> DEVT
The DEVT field indicates the type of device
that this record was generated for. Device
models are not identified.
c
<9> SENSE
<11> SEEK
When the 3380-JK is attached to a 3880
storage control, the SEEK field contains the
cylinder and head address for record types
"OBRPRM" and "OBRTMP."
When the 3380-JK is attached to a 3990
storage control,
<7> CMD
The command (CMD) field contains the
command that was being executed when the
error occurred.
•
For Format 4 errors, the SEEK field
contains the cylinder and head address
for record types "OBRPRM" and
"OBRTMP."
•
For other than Format 4 errors, the SEEK
field contains sense bytes 28 through 31
for record types "OBRPRM" and
"OBRTMP,"
<8> CSW
(
The channel status word (CSW) field indicates
the CSW that existed when the error
occurred.
('
PIN 4519907 23 Sides EC 475248 25Apr88
@ Copyright IBM Corp. 1987, 1988
PROG-21
<12> SD
For 3380-JK devices, the SO field contains the
physical identifier of the string and the
storage path that the string is connected to
(CC.P). For previous models of the 3380, the
SO field contains the storage director physical
identifier.
Note: This field is not used for storage
control ASYNCH data records.
<14> SNID
The SNIO (sense path group 10) indicates the
10 that was sent back from the storage
director in reply to a Sense Path Group 10
command.
CPU address
CPU ID number
CPU model number
(,
~/
<13> CT
Not used for 3380-JK devices.
\~ .... --/
(~
.
/f ",
~/
PROG-22
PIN 4519907 23 Sides EC 475248 25Apr88
@ Copyright IBM Corp. 1987, 19BB
r)',
\rlj
(
c
EC History of Program Section
(
(
EC HISTORY OF P/N 4519907
EC Number
Date Of EC
EC Number
14Nov86
475245
475246
21Ju187
475247
llSep87
475248
Date Of EC
25Apr88
Notes:
f~
((-'
(
('
(-
PIN 4519907 23 Sides EC 475248 25Apr88
~ Copyright IBM Corp. 19B7, 19BB
PROG-23
'~_-/
(11'-
\,4./
(
by its position. All bit positions should be
considered when tflere is a combination of
bits.
Symptom Code to FRU Charts
(
The charts on the following pages indicate the
FRUs that can cause a specific symptom code (SC)
to be presented.
(
The FRUs are listed in the order of probability that
could cause the symptom code.
The FRU lists in this section are for reference only.
DO NOT use these lists to make the final
determination on which FRUs to replace.
(
{
'
j
c
c
('
(
(
(
Use the MD Main Menu menu Option 3. The MD
uses additional information supplied by you and
normally provides better FRU isolation than these
lists. See "Analyze Symptom Code (Main Menu
Option 3)" on page MD-42.
Most of the symptom codes are described in detail
in the ECD section of the ECM, Volume R40. Tables
on the SC-FRU pages include page references to
the ECD section. If an ECD page reference is not
listed for the symptom code you're interested in,
see page HELP-1 in the HELP section of the MSM,
Volume R60.
How to Use the SC-FRU Charts
3. Find the FRUs
Look across the chart to find the FRUs that
are associated with the controller and/or
device that is failing. Again, the MD should
be used for FRU isolation if at all possible.
In the SC-FRU charts, primary FRUs are in
bold; secondary FRUs are in italics.
Note: Sometimes a defective FRU m a
control/er can cause a symptom code
associated witfl a device. In this case,
primary FRUs listed for the device
should be suspected as the cause of
the symptom code. Secondary FRUs
listed for the device or controller can
also cause the symptom code.
4. Determine the probability
The FRUs are listed in the order of their
probability of causing the symptom code.
See the following example:
1. Find the major symptom code group
The major symptom code group is listed at
the top of each chart. The major symptom
code groups are in numerical (hexadecimal)
sequence.
2. Find the value for X, XX, or XXX in the chart.
The value for X, XX, or XXX within any major
symptom code group is given in the first
column ot the chart. This value is shown in
two different forms:
•
•
c:\
If the XXX or bit column contains an asterisk
(*) the MD will definitely provide better
isolation than the SC FRU charts.
=
Actual value
X, XX, or XXX, equals the
actual value of the low-order part of the
symptom code.
Bit = X, XX, or XXX, is expressed in
binary form where each bit is represented
PIN 4519909 50 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1997, 1988
Controller
A2
Al
(A4)
(A3l
AIQ2,AIN2 AIH2.AlJ2
AIV2,A1U2 AIC2,A1D2
If the failing controller is A 1 or A3, the order
of probability (highest to lowest) is as follows:
1.
2.
A1Q2 and A1N2 are primary FRUs:
A1Q2
A1N2
3.
A 1V2 and A 1U2 are secondary FRUs:
A1V2
4.
A1U2
SC-FRU-1
SC=4XYY (4X= Format 4 Byte 07, YY= Format 4
B)~e
23)
"~'~,,
'",,-,J
Data
Chec~s
(SC
= 40VV
through 47VVl
Description
Byte
22)
4X (Sense
HA Area; Data Check
40*
Count Areai Data Check
41*
Key Area; Data Check
42*
Data Area; Data Check
43*
44** - HA Area; No Sync Byte Found
Count Area; No Sync Byte Found
45**
46** - Key Area; No Sync Byte Found
47** - Data Area; No Sync Byte Found
48 through 4F - not used
4XYY
YY (Sense Byte 23)
Bits 0, 1
00
Recovered withoutECC
01 - Correctable with first level ECC
10 - Correctable w/second level ECC
11 - Permanent, not correctable
Offset Active
Bit 2
Bit 3 - Always 0
Bit 4 - Device undervoltage
Controller undervoltage
Bit 5
Offset recovery bypassed
Bit 6
Track following lost
Bit 7
Notes:
* For details about SC 40XX through 43XX, see page ECD-185 in the £CM, Volume R40.
** For details about SC 44XX through 47XX, see page ECD-183 in the £CM, Volume R40.
---
VV Bits
0123 4567
Description
Controller
Al
A2
(A3)
(A41
xxIx OOxx Offset active
*
-
--
Front Devices
Rear Devices
Right
Right
Left
Left
O,4,8,C
1,5,9,D
2,6,A,E
3,7,B,F
BID2,BIE2 BIK2,BIJ2 BIN2,BIP2 BIU2,BlT2
COB2
CIB2
C2B2
C3B2
See MIM Volume RIO, page MAP-I, Entry EB
BID2,BIF2 BIK2,BIH2 BIN2,BIQ2 BIU2,BIS2
xxxx lxxx Device
undervoltage
xxxx xlxx Controller
undervoltage
See MIM Volume RIO, page PWR-l, Entry P
before replacing FRUs.
AIQ2,AIR2 AIH2,AIG2
See MIM Volume RIO, page PWR-l, Entry 0
before replacing FRUs.
BID2,BIE2 BIK2,BIJ2 BIN2,BIP2 BIU2,BlT2
COB2
CIB2
C2B2
C3B2
xxxx OOxl loss of track
* following
See MIM Volume RIO, page MAP-I, Entry EB
xxOx OOxO All other
* data check
conditions
AIQ2,AIN2 AIH2,AIJ2 COB2,BIF2 CIB2,BIH2 C2B2,BIQ2 C3B2,BIS2
8IG2,C182 8IG2,C082 8IR2,C382 8IR2,C282
81D2,81£2 B1K2,81J2 81N2,81P2 BW2,8H2
See MIM Volume RIO, page MAP-I, Entry EB
* A more detailed analysis can be obtained by using the MD to analyze the SC.
Figure 1.
SC=4XVV (4X = Format 4 Byte 07, VV
PIN 4519909 50 Sides
© Copyright
AIV2
EC 476581 01Aug89
IBM Corp. 1987, 1988
AIC2
==
Format 4 Byte 23)
SC-FRU-2
,/
"
('
(','
(j
SC=90XX (XX= Format 1 Byte 10)
Device Power status (SC
= 90XXl
Description
Bit
Front Devices
Rear Devices
Right
Left
Right
Left
O,4,8,C
2,6,A,E
3,7,B,F
1,5,9,D
2 No air flow
B1E2.B1G2 B1J2,B1G2 B1P2,B1R2 B1T2,B1R2
BIJ2
BIE2
BH2
BIP2
*
(ECD-13)
See MIM Volume RIO, page PWR-l, Entry EI
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 9020, see page ECD-13 in the ECM, Volume R40.
(
Figure 2.
Controller
A1
A2
(A3l
(A4l
SC=90XX (XX = Format 1 Byte 10)
{,'
SC=91XX (XX= Format 1 or 9 Byte 11)
c'
Device Check-2 status (SC
c~
c
{
Bit
Description
0
2
Device Sequencer
check
(ECD-lft)
RPS check
4
(ECD-16)
HDA Cable Swap check
*
Controller
A1
A2
(A3l
(A4l
No
No
6
(ECD-18)
Read/Write check
(ECD-365)
Power Card check
No
7
(ECD-19)
Funnel Parity check
'
/
= 91XXl
5
(ECD-20)
No
* A more detailed analysis can
NOTE: For details about symptom
Figure 3.
SC=91XX (XX= Format 1 or 9 Byte 11)
PIN 4519909 50 Sides
© Copyright
Front Devices
Rear Devices
Right
Right
Left
Left
1,5,9,D
2,6,A,E
3,7,B,F
O,4,8,C
B1E2,B1D2 B1J2,B1K2 B1P2,B1N2 B1T2,BIU2
BIF2,BIG2 BIH2,BIG2 BIQ2,BIR2 BIS2,BIR2
further action - go to RMAP 340 on page REPAIR-13.
B1D2,B1E2 B1K2,B1J2 B1N2,B1P2 B1U2,B1T2
B1F2,COB2 B1H2,CIB2 B1Q2,C2B2 B1S2,C3B2
BIG2,C4PI BIG2,C4P2 BIR2,C4P3 BIR2,C4P4
See MIM Volume RIO, page MAP-I, Entry DJ
B1E2,BIF2 BIJ2,B1H2 B1P2,B1Q2 B1T2,B1S2
BIG2
B1G2
B1R2
B1R2
further action - go to RMAP 340'on page REPAIR-13.
B1Q2
B1F2
BIH2
B1S2
Volume
RIO,
page
MAP-I,
Entry
See MIM
EA
BIJ2,BIH2
B1P2,B1Q2
B1T2,BlS2
BIE2,BIF2
B1G2
B1G2
B1R2
BIR2
further action - go to RMAP 340 on page REPAIR-13.
BIF2,B1E2 BIH2,BIJ2 BIQ2,BIP2 B1S2,B1T2
BIG2
BIR2
BIG2
BIR2
further action - go to RMAP 340 on page REPAIR-13.
be obtained by using the MD to analyze the SC.
code 9lXX, see page fCD-nnn in the ECM, Volume R40.
EC 476581 01Aug89
IBM Corp. 1987. 19B8
SC-FRU-3
SC=92XX (XX= Format 9 Byte 12)
Read/Write status 1 (SC
Bits
0123 4567
= 92XXl
Description
Controller
Front Devices
Rear Devices
Right
Right
Al
A2
Left
Left
(A3)
(A4)
2,6,A,E
0,4,8,C
1,5,9,D
3,7,B,F
OIXI XXXI Write Mode
AIN2.AIQ2 AIJ2.AIH2 COB2,BIF2 CIB2.BIH2 C2B2,B1Q2 C3B2.B1S2
Verify error
B1E2.BID2 B1J2,BIK2 BIP2,B1N2 B1T2.BIU2
*
(ECO-367)
See ",IM Volume RIO, page MAP-I. Entry EA
(ECO-375)
COB2,CI82 C1B2,C082 C2B2,C382 C3B2.C2B2
llXX XXXI Read Data
AIQ2
AIH2
B1Q2
Cable error
8IF2
B1H2
81S2
(ECO-375)
(ECD-367)
(ECO-367)
See ",1", Volume RIO. page MAP-I, Entry EC
XXXX XXXX Read/Write
BIF2,COB2 BIH2,CIB2 BIQ2,C2B2 B1S2,C3B2
Channel check
(ECO-376)
See ",1", Volume RIO, page MAP-I, Entry ED
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 92XX, see page (ECO-376) in the EC"', Volume R40.
Figure 4.
SC=92XX (XX= Format 9 Byte 12)
SC=93XX (XX= Format 9 Byte 13)
Read/Write status 2 (SC
= 93XXl
Bit
Description
0
False Read/Write
check
(ECO-365)
Padding check
Controller
A1
A2
(A3)
(A4)
See
Rear Devices
Front Devices
Right
Left
Right
Left
0,4,8,C
2,6,A,E 3,7,B,F
1,5,9,D
BIH2
BIQ2
BIF2
B1S2
Volume RIO, page MAP-I, Entry ED
I
BIF2,B1E2 BIH2,8IJ2 B1Q2,8IP2 B1S2,B1T2
B1U2 .
8IN2
8102
BIK2
(ECO-378)
See ",1", Volume RIO, page MAP-l, Entry EA
2 Read/Write Sequence A1U2,A1T2 A102,AIE2 B1F2
BIQ2
BIH2
B1S2
check
(ECO-380)
See ",IM Volume RIO, page MAP-I, Entry ED
3 Index/Cell check
B1F2,B1E2 B1H2,B1J2 BIQ2,B1P2 B1S2,B1T2
8102
8IK2
BIN2
B1U2
(ECO-382)
See MIM Volume RIO, page MAP-l, Entry EA
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 93XX, see page ECD-nnn in the ECM , Volume R40.
",1",
*
Figure 5 (Part 1 of 2).
PIN 4519909 50 Sides
SC=93XX (XX= Format 9 Byte 13)
EC 476581 01Aug89
@ Copyright IBM Corp. 19B7, 19BB
SC-FRU-4
(
Read/Write status 2 (SC
Bit
f
f-::
(
4
5
6
7
= 93XXl
Description
Controller
Al
A2
(A4)
(A31
Servo Cable check
(ECD-383)
Write Overrun check
(ECD-381t)
Read/Write Servo
check
(ECD-385)
Head Address
Register parity
(ECD-387)
check
See MIM
See MIM
See MIM
See MIM
Front Devices
Rear Devices
Right
Left
Right
Left
2,6,A,E
3,7,B,F
O,4,8,C
1,5,9,D
BID2.BIF2 BIK2.BIH2 BIN2.BIQ2 BIU2.B1S2
BIE2.8IK2 B1J2.81D2 B1P2,81U2 BIT2,8IN2
8IH2.8IJ2 8IF2,8IE2 BIS2,B1T2 8IQ2.BIP2
Volume RIO, page MAP-I, Entry DF
BIF2
BIH2
BIQ2
BlS2
Volume R1O, page MAP-I. Entry EA
BIE2,B1D2 BIJ2.BIK2 B1P2.BIN2 B1T2.BIU2
C4Pl.81F2 C4P2,BIH2 C4P3,81Q2 C4P4.8IS2
Volume RIO, page MAP-I, Entry DN
B1Q2
BIF2
BIH2
BlS2
Volume RIO, page MAP-I. Entry ED
NOTE: For details about symptom code 93XX. see page ECD-nnn in the
Figure 5 (Part 2 of 2).
ECM~
Volume Rita.
SC=93XX (XX= Format 9 Byte 13)
SC=94XX (XX= Format 9 Byte 14)
Read/Write status 3 (SC
Bit
Description
0
Check Inhibit check
(ECD-388)
Read/Write Channel
Status parity check
(ECD-389)
Read Transmit/Select
Alternate check
(ECD-390)
Read/Write Control
Cable check
(ECD-391J
I
(
(
(:
2
3
Controller
Al
A2
(A4)
(A31
See MIM
See MIM
See MIM
See MIM
Rear Devices
Front Devices
Right
Right
Left
Left
3,7,B,F
O,4,8,C
1,5,9,D
2,6,A,E
BlF?,COB2 B1H2,C1B2 BlQ2.C2B2 BlS2,C3B2
Volume RIO, page MAP-I, Entry EA
COB2.BIF2 CIB2,BIH2 C2B2.BlQ2 C3B2,B1S2
Volume RIO. page MAP-I. Entry EA
B1Q2
B1S2
B1F2
B1H2
Volume RIO. page MAP-I. Entry EA
B1F2.COB2 B1H2.C1B2 B1Q2.C2B2 B1S2.C3B2
Volume RIO. page MAP-I, Entry EA
NOTE: For details about symptom code 93XX. see page ECD-nnn in the
Figure 6.
c
= 94XX)
ECM~
Volume Rita.
SC=94XX (XX= Format 9 Byte 14)
PIN 4519909 50 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1987. 1988
SC-FRU-5
c
c
SC=95XX (XX= Format 1 Byte 15)
Checkpoint Log: Checks 10 through 3A CSC
xx=
Description
10
Index failure during Idle
13
Cell Pulse failure during Idle
15
DCP Servo Wri te Inhibit detected
inactive during Rezero
16
17
18
lA
2F
31
34
*
39
3A
= 9510
through 953A)
Front Devices
Rear Devices
Right
Right
Left
Left
1,5,9,D
2,6,A,E
O,4,8,C
3,7,B,F
B1D2,B1E2 B1K2,BIJ2 BIN2,B1P2 BIU2,B1T2
C082, CftPI CI82,Clt-P2 C282,Cft.P3 C382,CftPt,.
(ECO-2I) MIM Volume RIO, page MAP-I, Entry DJ
B1D2,B1E2 BIK2,B1J2 BIN2,BIP2 BIU2,B1T2
C082,Cft.PI CI82,Cft.P2 C282,C4P3 C382,Cft.Pft.
(ECO-22) MIM Volume RIO, page MAP-I, Entry DJ
BIE2,B1D2 B1J2,BIK2 BIP2,BIN2 B1T2,B1U2
Cft.PI
C4P2
Cft.P3
Cft.PIt(ECO-23) MIM Volume RIO, page MAP-I, Entry DN
DCP Servo Write Inhibit or Write Inhibit BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
found active at end of Rezero
C4PI,8IF2 Cft.P2,8IH2 Clt-P3,8IQ2 Cft.P4,8IS2
(ECO-24) MIM Volume RIO, page MAP-I, Entry DN
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BIU2
Write Inhibit active during a
Load HAR command
Cft.PI
Cft.P2
Cft.P3
Cft.Pft.
MIM Volume RIO, page MAP-I, Entry DN
Half Track active at end of Rezero
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BIU2
C4PI
C4P2
C4P3
C4P4
(ECO-25) MIM Volume RIO, page MAP-I, Entry DN
DCP found busy with Not Write Inhibit
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,B1U2
Cft.PI,8IF2 Cft.P2,8IH2 Ct,.P3,8IQ2 C4Pt,.,8IS2
(ECO-26) MIM Volume RIO, page MAP-I, Entry DN
Motor turning without the relay
BIG2,BIE2 BIG2,BIJ2 BIR2,BIP2 B1R2,B1T2
picked during power-on reset (POR)
BIJ2
B1E2
B1T2
BIP2
(ECO-28) MIM Volume RIO, page PWR-I, Entry AO
B1E2,BID2 B1J2,BIK2 BIP2,BIN2 BIT2,B1U2
Actuator moved too far during Safe
C4PI
C4P2
Cft.P3
C4P4
(ECO-29) MIM Volume RIO, page MAP-I, Entry DN
Either a DCP detected error or a
BID2,BIE2 B1K2,BIJ2 B1N2,BIP2 BIU2,B1T2
Check-2 Reset command
C082
CI82
C282
C382
(ECO-30) MI", Volume RIO, page MAP-I, Entry DJ
DCP in an error state or wrong command
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
issued during Safe
No further action -(ECO-31)
go to RMAP 340 on page REPAIR-13.
Drive motor slowing down
BIG2,B1E2 B1G2,B1J2 B1R2,BIP2 BIR2,B1T2
BIJ2
B1E2
B1T2
B1P2
(ECO-32) MIM Volume RIO, page PWR-I, Entry AD
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 95XX, see page ECD-nnn in the ECM, Volume R40.
Figure 7 (Part 1 of 13).
SC=95XX (XX= Format 1 Byte 15)
PIN 4519909 50 Sides
EC 476581 01Aug89
© Copyright
IBM Corp. 1997, 1999
SC-FRU-6
c
(-
c
(:'
c
c
(
(
Checkpoint Log: Checks 3B through 4F (SC = 953B through 954F)
XX=
Description
Front Devices
Rear Devices
Right
Right
Left
Left
1,5,9,D
2,6,A,E
3,7,B~F
O~4~8~C
3B The DCP either did not take or complete BIE2,B1D2 BIJ2,B1K2 B1P2,B1N2 BIT2,B1U2
the Safe command
No further action -(£CO-34)
go to RMAP 340 on page REPAIR-13.
B1G2,B1E2 B1G2,BIJ2 B1R2,BIP2 BIR2,BIT2
3F Motor relay is stuck active during
a power-on reset (POR) or a power off
BIJ2
B1E2
B1T2
BIP2
(£CO-35) MIM Volume RIO, pa e PWR-I, Entry AO
41 DCP Track Follow bit active when not
BIE2,B1D2 BIJ2,B1K2 BIP2,B1N2 BIT2,B1U2
expected during Rezero
C4P4
C4Pl
C4P2
C4P3
(£CO-36) MIM Volume R1O, page MAP-I, Entry DN
44 DCP error during Rezero
B1E2,BID2 B1J2,BIK2 B1P2,BIN2 B1T2,B1U2
1(
C4Pl,81F2 C4P2,81H2 C4P3,81Q2 C4P4,81S2
(£CO-37) MIM Volume RIO, page MAP-I, Entry DN
45 DCP Servo Check 2 bit active during
B1E2,BID2 B1J2,BIK2 BIP2,BIN2 BIT2,BIU2
Rezero
C4Pl,81F2 C4P2,81H2 C4P3,81Q2 C4P4,81S2
(£CO-38) MIM Volume R1O, page MAP-I, Entry DN
46 DCP Busy found inactive during Rezero
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,B1U2
C4Pl,81F2 C4P2,81H2 C4P3,81Q2 C4P4,81S2
(£CO-39) MIM Volume R1O, page MAP-I, Entry DN
47 Servo Check 2 active prior to issuing
BIE2,B1D2 BIJ2,BIK2 B1P2,BIN2 BIT2,BIU2
C't.p 1,8 IF2 C4P2,81H2 C4P3,81Q2 C4P4,81S2
Rezero command
(£CO-40) MIM Volume RIO, page MAP-I, Entry DN
48 AGC lost during Rezero
BID2,BIE2 BIK2,BIJ2 BIN2,BIP2 BIU2,BIT2
C082
C182
C3B2
C282
(£CO-41) MIM Volume RIO, page MAP-I, Entt;y DJ
49 Track following not found at end of
BIE2,BID2 B1J2,BIK2 BIP2,BIN2 BIT2,BIU2
Rezero
C4PI,BIF2 C't.p2,BIH2 C4P3,81Q2 C4P4,8IS2
(£CO-42) HIM Volume RIO, page MAP-I, Entry DN
4C Rezero failed to start after DCP command BIE2,BID2 BIJ2,BIK2 B1P2,BIN2 B1T2,B1U2
issued
C4PI,8IF2 C4P2,BIH2 C4P3,8IQ2 C4P4,BIS2
(£CO-43) HIM Volume RIO, page MAP-I, Entry DN
4D DCP Half Track not active when expected B1E2,B1D2 B1J2,B1K2 B1P2,BIN2 B1T2,B1U2
during Rezero
C4PI,8IF2 C4P2,8IH2 C4P3,8IQ2 C4P4,BIS2
(£CO-H) HIM Volume RIO, page MAP-I, Entry DN
4F DCP Busy active detected before Rezero
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BlU2
command
No further action -(£CO-45)
go to RMAP 340 on page REPAIR-13.
1( A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 95XX, see page ECD-nnn in the £CM, Volume R40.
Figure 7 (Part 2 of 13).
SC=95XX (XX= Format 1 Byte 15)
PIN 4519909 50 Sides
EC 476581 01Aug89
(
c
© Copyright
IBM Corp. 1987, 1988
SC-FRU-7
Checkpoint Log: Checks 50 through 5C (SC = 9550 through 955C)
Front Devices
Rear Devices
Right
Right
Left
Left
l,5,9,D
2,6,A,E
3,7,B,F
O,4,8,C
50 Seek routine took too long
BIE2~BID2 BIJ2.BIK2 BIP2.BIN2 BlT2,BIU2
C4Pl
C4P2
C4P3
C4P4
(ECO-46) MIM Volume RIO, page MAP-I, Entry DN
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BlT2,BIU2
52 Guard band detected during Seek
C4P2
C4P3
C4P4
C4Pl
(ECO-46) MIM Volume RIO, page MAP-I, Entry DN
BIE2,BID2 BIJ2.BIK2 BIP2,BIN2 BIT2,BIU2
53 DCP Busy not active at beginning of
C4P3
C4P4
C4Pl
C4P2
Seek command
(ECO-47) MIM Volume RIO, page MAP-I, Entry DN
BIE2,BID2 BIJ2.BIK2 BIP2,BIN2 BlT2.BlU2
54 DCP Error bit detected active during
C4P2
C4P3
C4P4
C4Pl
* Seek
(ECO-48) MIM Volume RIO, page MAP-I. Entry DN
55 DCP detected failure during Seek
BIE2.BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
C4P3
C4P4
C4Pl
C4P2
(ECO-49) MIM Volume RIO. page MAP-I, Entry DN
56 Invalid DCP command for Seek operation
BIE2.BID2 BIJ2,BIK2 BIP2,BIN2 BIT2.BIU2
C4P4
C4Pl
C4P2
C4P3
(ECO-50) MIM Volume RIO, page MAP-I, Entry DN
BIE2,BID2 BIJ2,BlI(2 BIP2,BIN2 BlT2,BIU2
57 AGC signal inactive during Seek
C4Pl
C4P2
C4P3
C4Pft.
(ECO-51) MIM Volume RIO. page MAP-I. Entry DN
58 Not track following after fourth
BIE2,BID2 BIJ2,BIK2 BIP2.BIN2 BIT2,BIU2
settling delay during Seek
C4P4
C4Pl
C4P2
C4P3
(ECO-52) MIM Volume RIO, page MAP-I, Entry DN
59 DCP Track Follow and Offset not active
BIE2.BID2 BIJ2,BIK2 BIP2,BIN2 BlT2,BIU2
at beginning of Seek
No further action -(ECO-53)
go to RMAP 340 on page REPAIR-13.
SA DCP Busy before or after Seek
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BlT2,BIU2
No further action -(ECO-54)
go to RMAP 340 on page REPAIR-13.
5B DCP fails to accept Seek command or
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
DCP Hang condition detected
No further action -(ECO-55)
go to RMAP 340 on page REPAIR-I3.
5C DCP Busy bit goes inactive during Seek
BIE2.BID2 BIJ2,BIK2 BIP2,BIN2 BlT2.BIU2
No further action -(ECO-56)
go to RMAP 340 on page REPAIR-13.
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 95XX, see page ECD-nnn in the ECM, Volume R40.
XX=
Description
Figure 7 (Part 3 of 13).
SC=95XX (XX= Format 1 Byte 15)
PIN 4519909 50 Sides
EC 476581 01Aug89
~ Copyright IBM Corp. 1987, 1988
SC-FRU-8
c
(
\I
,
~~
,/"
c
\
(
('
(
XX=
SD
{'
(
c
(~
(
"
--j
(
c(
Checkpoint Log: Checks 5D through 6A (SC
= 955D
through 956Al
Description
Front Devices
Rear Devices
Right
Left
Right
Left
O,4,8,C
1,5,9,D
2,6,A,E
3,7,B,F
DCP Half Track bit in wrong state during BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
CftPI
CftP2
CftP3
CftPft
MIM Volume RIO, page MAP-I, ENTER DN
Seek
(ECD-57)
SE DCP Track Following bit active too soon BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
during Seek
CftPI
CftP2
CftP3
Cit-Pit(ECD-58) MIM Volume RIO, page MAP-I, ENTER DN
SF Write Inhibit or DCP Servo Write Inhibit BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
in wrong state
BIF2,C4PI BIH2,Clt-P2 BIQ2,Clt-P3 8IS2,C4P4
(ECD-59) MIM Volume RlO, page MAP-I, ENTER DN
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
60 DCP'Offset bit in wrong state during
Offset
C4PI
C4P2
Clt-P3
C4P4
(ECD-60) MIM Volume RlO, page MAP-I, Entry DN
62 Invalid command in DCP command register BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
for Offset
C4PI
Clt-P3
Cit-PitC4P2
(ECD-6I) MIM Volume RIO, page MAP-I, Entry DN
63' DCP Busy bit active before issuing
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
Offset
No further action -(ECD-61)
go to RMAP 340 on page REPAIR-13.
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
64 DCP Error active during Offset routine
CftPI
CftP2
C4P3
Cit-Pit(ECD-62) MIM Volume RlO, page MAP-I, Entry DN
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
65 DCP detected failure during Offset
C4Pl
Clt-P2
Clt-P3
C4P4
(ECD-63) MIM Volume RIO, page MAP-I, Entry DN
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
66 DCP fails to accept Offset command or
DCP Hang condition detected
No further action -(ECD-64)
go to RMAP 340 on page REPAIR-13.
67 DCP Busy bit not active at beginning of BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
Offset command
No further action -(ECD-65)
go to RMAP 340 on page REPAIR-13.
68 AGC signal inactive during Offset
BID2,BIE2 BIK2,BIJ2 BIN2,BIP2 BIU2,B1T2
C282
C382
C082
CI82
(ECD-66) MIM Volume RIO, page MAP-I, Entry DJ
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
6A DCP Track Following bit in wrong state
during Offset
C4PI
Clt-P3
C4PItClt-P2
(ECD-67) MIM Volume RIO, page MAP-I, Entry DN
A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 95XX, see page ECD-nnn in the ECM, Volume RitO
*
*
Figure 7 (Part 4 of 13).
SC=95XX (XX= Format 1 Byte 15)
PIN 4519909 50 Sides
EC 476581 01Aug89
© Copyright
IBM Corp. 1987. 19S8
SC-FRU-9
Checkpoint Log: Checks 6B through 7A (SC
XX=
= 956B
through 957Al
Description
Front Devices
Rear Devices
Left
Right
Right
Left
2,6,A,E
3,7,B,F
O,"t,8,C
1,5,9.D
6B Offset routine took too long
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
CitPI
CltP2
CltP3
CitPit
a:C0-68J MIM Volume RIO, page MAP-I, Entry DN
6C Write Inhibit signal or DCP Servo Write BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
Inhibit bit in wrong state
CitPI
CltP2
CltP3
C4PIt
(£CO-69) MIM Volume RIO, page MAP-I, Entry DN
6D DCP Track Follow and Offset inactive at BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BIU2
beginning of Offset
CitPI
CltP2
CltP3
CitPit
(£CO-70) MIM Volume RIO, page MAP-I, Entry DN
6E Write Inhibit signal or DCP Servo Write BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BIU2
Inhibit active at the end of Zero Offset C4PI
CltP2
C4PIt
C4P3
(£CO-71 ) MIM Volume RIO, page MAP-I, Entry DN
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
6F DCP Busy bit active at end of Offset
CitPI
CltP2
C4P3
CitPit
(£CO-72) MIM Volume RIO, page MAP-I, Entry DN
74 DCP Error bit detected active during
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
* search
No further action -(£CO-73)
go to RMAP 340 on page REPAIR-13.
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
75 DCP Check 2 bit detected active during
search
No further action -(£CO-71t)
go to RMAP 340 on page REPAIR-13.
76 DCP Busy bit active before issuing
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
Search command
No further action -(£CO-75)
go to RMAP 340 on page REPAIR-13.
77 Record Ready Interrupt is active too
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
long during Search
No further action -(£CO-76)
go to RMAP 340 on page REPAIR-13.
78 AGC signal detected inactive during
BID2,BIE2 BIK2,BIJ2 BIN2,BIP2 BIU2,B1T2
C082,C4PI CI82,CltP2 C282,CltP3 C382,CltPit
Search
(£CO-77) MIM Volume RIO, page MAP-I, Entry DJ
79 RPS Failed To Lock detected during
BID2,BIE2 BIK2,BIJ2 BIN2,BIP2 BIU2,BIT2
C082,CltPI CI82,CltP2 C282,C4P3 C382,C4P4
Search
(£CO-78) MIM Volume RIO, page MAP-I, Entry DJ
7A RPS check active during Search
BID2,B1E2 BIK2,BIJ2 BIN2,BIP2 BIU2,B1T2
C082,ClfPI CI82,C4P2 C282,C4P3 C382,C4P4
(£CO-79) MIM Volume RIO, page MAP-I, Entry DJ
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 95XX, see page ECD-nnn in the £CM, Volume R40.
Figure 7 (Part 5 of 13).
SC=95XX (XX= Format 1 Byte 15)
1'\
i,.t.J
PIN 4519909 50 Sides
EC 476581 OtAug89
© Copyright IBM Corp. 19S7, 19S5
SC-FRU-10
(
Checkpoint Log: Checks 7B through 88 (5C
xx=
(
("
(j
(
= 957B
through 9588)
Description
Front Devices
Rear Devices
Right
Right
Left
Left
2,6,A,E
3,7,B,F
O,4,8,C
1,5,9,D
BID2,BIE2 B1K2,BIJ2 B1N2,BIP2 B1U2,B112
7B Index/Record Ready Interrupt check
during Search
COB2,C4Pl CIB2,C4P2 C2B2,C4P3 C3B2,C4P4
(ECO-80) MIM Volume RIO, page MAP-I. Entry DJ
7C Time between Record Ready interrupts too B1D2,B1E2 B1K2,B1J2 B1N2,B1P2 B1U2,B1T2
short during search
COB2,C4Pl CIB2,C4P2 C2B2,CftP3 C382,CftPft
(ECO-81) MIM Volume RIO, page MAP-I, Entry DJ
7D Time between Record Ready interrupts too B1D2,B1E2 B1K2,B1J2 B1N2,B1P2 B1U2,B1T2
long during search
C082,CftPl C182,CftP2 C282,CftP3 C382,CftPft
I--(ECO-82) MIM Volume RIO, page MAP-I, Entry DJ
B1E2,B1D2 BIJ2,B1K2 B1P2.B1N2 B112.B1U2
7F Servo Check 2 active before issuing
Search command
No further action -(ECO-83)
go to RMAP 340 on page REPAIR-13.
B1E2,B1D2 B1J2.B1K2 B1P2.B1N2 B112,B1U2
80 Received a command that does not match
the HDA model
*
MIM Volume R10. page MAP-I. Entry DP
(ECO-8ft) MIM Volume R1O. page MAP-I. Entry DI
B1F2.B1E2 B1H2,B1J2 B1Q2,B1P2 B152,B112
82 Read/Write mode active during Safe
B1R2,
B1G2
BIG2
B1R2
No further action -(ECO-8S)
go to RMAP 340 on page REPAIR-13.
B1E2.B1D2 BIJ2,B1K2 B1P2.B1N2 B112,B1U2
83 Invalid Warm-Up command
B1Q2
B1F2
B1H2
B152
No further action -go to RMAP 340 on page REPAIR-13.
84 Either an invalid command or parameter
B1E2.B1F2 B1J2.B1H2 B1P2,B1Q2 B112,B152
was received
81K2
BIN2
81U2
8102
No further action -(ECO-86)
go to RMAP 340 on page REPAIR-13.
B1E2,B1D2 B1J2.B1K2 B1P2.B1N2 B112,B1U2
85 Received a command that does not match
* the HDA Model
(ECO-87) MIM Volume R1O, page MAP-I, Entry DI
86 Lost track following during
B1E2.B1D2 B1J2,B1K2 B1P2.B1N2 B1T2,BlU2
Data Check Recovery
CftP4
CftP1
CftP2
CftP3
(ECO-88)
MIM Volume R1O. page MAP 1, Entry DN
87 Check Reset command fails to reset Busy B1E2,B1D2 B1J2,B1K2 B1P2,B1N2 B112,B1U2
B1Q2
B1H2
B152
B1F2
No further action -(ECO-89)
go to RMAP 340 on page REPAIR-l3.
88 Invalid or illegal command received with BIE2,BID2 BIJ2,B1K2 B1P2,B1N2 B112,B1U2
Device Busy
B1F2,81G2 B1H2,81G2 B1Q2,81R2 B152.81R2
No further action -(ECO-89)
go to RMAP 340 on page REPAIR-13.
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 95XX, see page ECD-nnn in the ECM, Volume R40.
Figure 7 (Part 6 of 13).
SC=95XX (XX= Format 1 Byte 15)
PIN 4519909 50 Sides
EC 476581 01Aug89
© Copyright
IBM Corp. 1987. 1988
SC-FRU-11
Checkpoint Log: Checks 8A through A6 (SC
xx=
8A
Description
DCP Sample Interrupt out of
specification during Idle
= 958A
through 95A61
Rear Devices
Right
Left
3,7,B,F
2,6~A,E
O~4~8~C
1~5~9~D
B1E2,B1D2 B1J2,B1K2 B1P2,B1N2 B1T2,B1U2
Front Devices
Right
Left
No further action -(ECO-90)
go to RMAP 340 on page REPAIR-13.
8B DCP Sample Interrupt signal not changing B1E2,B1D2 B1J2,B1K2 B1P2,B1N2 B1T2,B1U2
changing states during Idle
No further action -(ECO-91)
go to RMAP 340 on page REPAIR-U.
B1E2,B1D2 B1J2,B1K2 B1P2,B1N2 B1T2,B1U2
8D Sequencer failed to receive required
B1H2
B1Q2
B1S2
data from the controller-device port
B1F2
(CDP)
No further action -(ECO-92J
go to RMAP 340 on page REPAIR-13.
B1E2,B1D2 B1J2,B1K2 B1P2,B1N2 B1T2,B1U2
8E Received a command that does not match
3(
the HDA model
(ECO-93) MIM Volume RIO, page MAP-I, Entry DI
8F Both guard bands detected; actuator
B1E2,B1D2 B1J2,B1K2 B1P2,B1N2 BlT2,BlU2
seems to oscillate
C4Pl
CftP2
C4P3
CftPft
(ECO-94) MIM Volume RIO, page MAP-I, Entry DN
B1E2,B1F2 B1J2,B1H2 B1P2,B1Q2 B1T2,B1S2
92 Command gate stuck active
81U2
8102
81K2
81N2
r-No further action -(ECO-95)
go to RMAP 340 on page REPAIR-13.
B1E2,B1G2 B1J2,B1G2 B1P2,B1R2 BlT2,B1R2
Al No air pressure detected during Motor
81Q2
Start
81F2
81H2
81S2
(ECO-96J MIM Volume RIO, page PWR-l, Entry I
A2 Motor relay not active during Power On
B1G2,B1E2 B1G2,B1J2 B1R2,B1P2 B1R2,BlT2
B1J2
B1E2
B1T2
B1P2
(ECO-97) MIM Volume RIO, page PWR-l, Entry AD
A3 No air pressure detected during Power On B1E2,B1G2 B1J2,B1G2 B1P2,B1R2 B1T2,B1R2
B1F2
B1Q2
81S2
81H2
(ECO-98) MIM Volume RIO, page PWR-l, Entry I
A4 Motor switch off after motor started
B1G2,B1E2 B1G2,B1J2 B1R2,B1P2 B1R2,B1T2
3(
during Power On
B1J2
B1E2
BlT2
B1P2
(ECO-99) MIM Volume RIO, page PWR-l, Entry AC
B1G2,B1E2 B1G2,B1J2 B1R2,B1P2 B1R2,B1T2
A5 Drive Motor switch momentarily off
3(
during Search / Idle
B1J2
B1E2
BlT2
B1P2
(ECO-lOO) MIM Volume RIO, page PWR-l, Entry AC
A6 Motor relay active at beginning of
B1G2,B1E2 B1G2,B1J2 B1R2,B1P2 B1R2,B1T2
Motor Start
B1P2
B1J2
B1E2
BlT2
(ECO-101) MIM Volume RIO, page PWR-l, Entry AO
3( A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 95XX, see page ECD-nnn in the ECM~ Volume RftO.
Figure 7 (Part 7 of 13).
SC=95XX (XX= Format 1 Byte 15)
PIN 4519909 50 Sides
EC 476581 01Aug89
© Copyright
IBM Corp. 1987, 1988
SC-FRU-12
('
I~
"<....
(
Checkpoint Log: Checks A7 through B4 (SC : 95A7 through 95B4)
xx:
Description
(
A7
DCP detected error in Motor Start
Decompress
(ECD-IOZ)
A8
AGC not active with motor turning at
more than 3150 RPM
(ECD-I03)
A9
DCP detected error or timeout during
Compress
(ECD-I05)
AC
DCP detected error in Motor Start
Decompress
(ECD-I06)
AD
DCP detected error during Crash Stop
Recovery routine
(ECD-I07)
AE
(
DCP Timeout during Crash Stop Recovery
(ECD-I08)
'"
~/
BO
DCP detect~d error during a Seek of the
Crash Stop Recovery
('
(ECD-I09)
Bl
DCP detected error after park prior to
starting the motor
(
(ECD-I11)
B2
(
Motor or Soft Start Thermal or Belt
Guard switch open
(ECD-IIZ)
B3
B4
B1G2,B1E2 B1G2,B1J2 B1R2,B1P2 B1R2,B1T2
BIJ2
BIE2
B1T2
B1P2
MIM Volume RIO, page PWR-I, Entry AD
B1E2,C4P1 BIJ2,C4P2 B1P2,C4P3 B1T2,BlU2
B1D2
BIK2
BIN2
BIU2
MIM Volume RIO, page MAP-I, Entry DC
BIE2,C4Pl BIJ2,C4P2 BIP2,C4P3 B1T2,C4P4
B1U2
BID2
B1K2
B1N2
MIM Volume RIO, page MAP-I, Entry DC
B1E2,B1D2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
No further action -go to RMAP 340 on page REPAIR-13.
B1E2,B1D2 B1J2,B1K2 BIP2,B1N2 B1T2,B1U2
CltPI
C4PZ
CltP3
CltPt;.
MIM Volume RIO, page MAP-I, Entry DN
B1E2,C4P1 B1J2,C4P2 B1P2,C4P3 B1T2,C4P4
B1D2
B1K2
B1N2
B1U2
MIM Volume RIO, page MAP-I, Entry DC
BIE2,C4P1 BIJ2,C4P2 B1P2,C4P3 BIT2,C4P4
B1U2
B1D~
B1K2
BIN2
MIM Volume RIO, page MAP-I, Entry DC
B1G2,B1E2 BIG2,B1J2 B1R2,B1P2 B1R2,BIT2
BIJ2
BIE2
B1T2
BIP2
MIM Volume RIO, page PWR-I, Entry F
BIFZ
Motor switch off at beginning of Power
On or during Idle
*
* A more
MIM Volume RIO, page MAP-I, Entry DC
BIE2,B1G2 B1J2,B1G2 B1P2,B1R2 B1T2,BIR2
No Air Pressure detected
(ECD-113)
(
MIM Volume RIO, page MAP-I, Entry DC
ncp Command bit active when not expected B1E2,C4P1 B1J2,C4P2 B1P2,C4P3 B1T2,C4P4
during Compress
BID2
BIK2
B1N2
BlU2
(ECD-IOlt)
AA
Front Devices
Rear Devices
Right
Right
Left
Left
O,4,8,C
1,5,9,D
2,6,A,E
3,7,B,F
B1E2,C4P1 B1J2,C4P2 B1P2,C4P3 B1T2,C4P4
B1D2
B1K2
B1N2
B1U2
(ECD-IH)
BIHZ
BIQZ
815Z
MIM Volume RIO, page PWR-I, Entry I
BIG2,BIE2 BIG2,BIJ2 BIR2,B1P2 BIR2,BIT2
BIJ2
B1E2
B1T2
B1P2
MIM Volume RIO, page PWR-l, Entry E
detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 95XX, see page ECD-nnn in the ECM 1 Volume Rt;.O.
(~
Figure 7 (Part 8 of 13).
SC=95XX (XX= Format 1 Byte 15)
PIN 4519909 50 Sides
EC 476581 01Aug89
~ Copyright IBM Corp. 1987, 1988
SC-FRU-13
,~
I
~;
Checkpoint Log: Checks B5 through CO (SC
xx=
= 95B5
Description
B5
AGC still active with motor off
B6
(ECD-II5)
Drive motor slowing down during Idle
B7
B8
B9
BA
BB
BC
BD
BE
BF
CO
(ECD-116)
DCP detected error or servo failure
during Power On
(ECD-l17)
Brake test failed at beginning of the
Power On
(ECD-118)
Soft Start relay active when not
expected
(ECD-119)
Soft Start relay not active at beginning
of Motor Start
(ECD-120)
Soft Start relay active at beginning of
Motor Start
(ECD-121)
DCP detected external error after any
Sequencer-to-DCP command
(ECD-122)
DCP detected internal error after any
Sequencer-to-DCP command
(ECD-123)
DCP Hang after any Servo command
(ECD-124)
DCP Command bit active when not expected
during Decompress
(ECD-125)
Servo Inhibit active during
Data Check Recovery
(ECD-127)
through 95CO)
Front Devices
Rear Devices
Right
Right
Left
Left
1,5,9,D
2,6,A,E
3,7,B,F
O,4,8,C
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
BIF2
BIH2
BIQ2
B152
MIM Volume R1O, page PWR-I, Entry E
BIG2,BIE2 BIG2,BIJ2 BIR2,BIP2 BIR2,BIT2
BIJ2
BIE2
BIT2
BIP2
MIM Volume RIO, page PWR-I, Entry AO
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
Clt-P2
Clt-P3
Cit-Pit
CltPl
MIM Volume RIO, page MAP-I, Entry DN
BIG2,BIE2 BIG2,BIJ2 BIR2,BIP2 BIR2,BIT2
BIJ2
BIE2
BIT2
BIP2
MIM Volume RIO, page PWR-I, Entry AO
BIG2,BIE2 BIG2,BIJ2 BIR2,BIP2 BIR2,BIT2
BIJ2
BIE2
BIT2
BIP2
MIM Volume RIO, page PWR-I, Entry AO
BIG2,BIE2 BIG2,BIJ2 BIR2,BIP2 BIR2,BIT2
BIJ2
BIE2
BIT2
BIP2
MIM Volume RIO, page PWR-I, Entry AO
BIG2,BIE2 BIG2,BIJ2 BIR2,BIP2 BIR2,BIT2
BIJ2
BIE2
BIT2
BIP2
MIM Volume RIO, page PWR-I, Entry AO
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
No further action -go to RMAP 340 on page REPAlR-13.
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
No further action -go to RMAP 340 on page REPAIR-13.
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
No further action -go to RMAP 340 on page REPAIR-I3.
BIE2,C4Pl BIJ2,C4P2 BIP2,C4P3 BIT2,C4P4
BID2
BIK2
BIN2
BIU2
MIM Volume RIO, page MAP-I, Entry DC
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
BIF",COB2 BIH2,CIB2 BIQ2,C2B2 BIS2,C3B2
Cit-PI
C4P2
C4P3
C4P4
No further action -go to RMAP 340 on page REPAIR-13.
NOTE: For details about symptom code 95XX, see page ECD-nnn in the ECM
Figure 7 (Part 9 of 13).
SC=95XX (XX= Format 1 Byte 15)
PIN 4519909 50 Sides
EC 476581 01Aug89
1
\,- /
Volume Rlt-O.
SC-FRU-14
~ Copyright IBM Corp. 1987, 1988
;'"i'
I
\--
Checkpoint Log: Checks Cl through CA (SC
xx=
Description
('
Cl
("
('
4[~
= 95Cl
through 95CAl
Front Devices
Right
Left
O,4,8,C
1,5,9,D
Rear Devices
Right
Left
2,6,A,E
3,7,B,F
DCP IML failure
BIE2.BID2 BIJ2.BIK2 BIP2.BIN2 BIT2.BIU2
BIF2.B1G2 BIH2.BIG2 BIQ2.BIR2 B1S2.BIR2
No further action -(ECD-128)
go to RMAP 340 on page REPAIR-l3.
C2 DCP BAT failure
BIE2.BID2 BIJ2.BIK2 BIP2.BIN2 BIT2.BlU2
No further action -(ECD-129)
go to RMAP 340 on page REPAIR-l3.
C3 AGC not active 18 seconds after the
BIE2.BID2 BIJ2.BIK2 BIP2.BIN2 B1T2,BlU2
motor was started
No further action -(ECD-130)
go to RMAP 340 on page REPAIR-l3.
C4 DCP detected failure during the handling BIE2.BID2 BIJ2,BIK2 BIP2,BIN2 B1T2.BlU2
of Sweep commands
Cft-PI.B IF2 Cft-P2,BIH2 Cft.P3,BIQ2 Cft.Pft-.BIS2
(ECD-131) MIM Volume RIO, page MAP-I, Entr"y DN
C5 External timer from other device not
BIJ2.BIE2 BIE2.BIJ2 B1T2,BlP2 BIP2,B1T2
running
B lK2,8 102 B102,BIK2 B1U2.BIN2 BIN2.B1U2
No further action -(ECD-132)
go to RMAP 340 on page REPAIR-l3.
C6 Motor Run relay not active after Soft
BIG2.BIE2 BIG2.BIJ2 BIR2.BIP2 BIR2.BIT2
BIE2
BIT2
BIP2
Start relay dropped
BIJ2
(ECD-133) fUM Volume RIO, page PWR-l, Entry AO
C7 Sector Compare not found during
BIE2.BID2 BIJ2,BIK2 BIP2.BIN2 B1T2,BlU2
BIF2,COB2 BIH2,CIB2 BIQ2.C2B2 B1S2,C3B2
Data Check Recovery
Cft-P2
C4P3
Cft-P4
Cft-PI
No further action -(ECD-134)
go to RMAP 340 on page REPAIR-13.
BIE2,BIF2 BIJ2,BIH2 BIP2,BIQ2 BIT2,B1S2
C8 Offset command not saved prior to
BIN2
Data Check Recovery
BID2
BIK2
BlU2
No further action -(ECD-135)
go to RMAP 340 on page REPAIR-13.
C9 Sector Compare stuck on during
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
Data Check Recovery
BIF2,COB2 BIH2,CIB2 BIQ2,C2B2 B1S2,C3B2
C4PI
C4P2
Cft-P3
C4P4
No further action -(ECD-136)
go to RMAP 340 on page REPAIR-13 .
CA DCP Busy detected during
BIE2.BID2 BIJ2.BIK2 BIP2,BIN2 B1T2.BlU2
BIF2,COB2 BIH2,CIB2 BIQ2,C2B2 B1S2.C3B2
Data Check Recovery
C4P2
C4P3
C4P4
Cft-PI
No further action -(ECD-137)
go to RMAP 340 on page REPAIR-13.
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 95XX, see page ECD-nnn in the ECM, Volume Rft.O.
Figure 7 (Part 10 of 13).
SC=95XX (XX= Fonnat 1 Byte 15)
PIN 4519909 50 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
SC-FRU-15
Checkpoint Log: Checks CB through D6 (SC = 95CB through 95D6)
Description
XX=
CB
Rear Devices
Front Devices
Left
Right
Left
Right
1,5,9,D
3,7,B,F
O,4,8,C
2,6,A,E
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BlT2,BIU2
BIF2,COB2 BIH2,CIB2 BIQ2,C2B2 B1S2,C3B2
RPS lock inactive during
Data Check Recovery
C4PI
C4P2
C4P3
C4P4
No further action -go to RMAP 340 on page REPAIR-13 .
DCP detected Checksum error during
BIE2,BID2 BIJ2,BIK2 BIP2,B1N2 BIT2,BIU2
DCP IML or DCP BAT
No further action -(ECO-139)
go to RMAP 340 on page REPAIR-13.
Guard band detected at cylinder 0 during BIE2,BID2 BIJ2,BIK2 BIP2.BIN2 BlT2,BIU2
a minus cylinder Seek command
C4PI
C4P2
C4P3
C4P4
(ECO-HO) MIM Volume RIO, page MAP-I. Entry DN
Abnormal termination of
BIE2,BIF2 BIJ2,BIH2 BIP2,BIQ2 BlT2.B1S2
BIN2
B1U2
BIK2
Data Check Recovery
BID2
No further action -(ECO-H!)
go to RMAP 340 on page REPAIR-13.
GBOD not found after a Seek to a minus
BIE2,BID2 BIJ2,BIK2 BIP2.BIN2 BIT2,BIU2
cylinder
CftPI
CftP2
CftP3
CftPft
(ECO-H2) MIM Volume RIO, page MAP-I, Entry DN
Servo failure during Sweep or Heat Up
BIE2.B1D2 BIJ2.BIK2 BIP2,BIN2 BlT2.BIU2
routines of Power On
CftPI,BIF2 CftP2,BIH2 CftP3,BIQ2 CftP4,BIS2
(ECO-H3) MIM Volume RIO, page MAP-I, Entry DN
DCP timeout doin3 a compress during
BIE2.C4Pl BIJ2.C4P2 BIP2,C4P3 BlT2.C4P4
Motor Start
BIK2
BIN2
BIU2
B1D2
(ECO-H4) MIM Volume RIO, page MAP-I, Entry DC
DCP timeout failure during Sweep
BIE2,BID2 BIJ2,BIK2 B1P2,B1N2 BlT2,BIU2
(ECO-138)
CC
CD
CE
CF
Dl
*
D2
D3
C4PI,BIF2 CftP2,BIH2 C4P3,BIQ2 C4Pft.,BIS2
page MAP-I, Entry DN
D4 Park failed before starting motor during BIE2,C4Pl BIJ2,C4P2 BIP2.C4P3 BIT2.C4P4
Power On
BID2
BIK2
BIN2
BIU2
(ECD-146) MIM Volume RIO. page MAP-I. Entry DC
D5 DCP detected error at the beginning of
B1E2.C4Pl BIJ2.C4P2 BIP2.C4P3 BlT2.C4P4
Motor Start
BIK2
BIN2
B1D2
B1U2
(ECO-147) MIM Volume RIO. page MAP-I, Entry DC
D6 Motor sensed off, but still turning at
BIE2.B1G2 BIJ2.B1G2 B1P2.BIR2 BIT2.B1R2
more than 3440 RPM
B IF2 ,B102 BIH2.BIK2 BIQ2.BIN2 BIS2.B1U2
(ECD-149) MIM Volume RIO, page PWR-l, Entry H
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 95XX, see page fCD-nnn in the ECM, Volume Rft.O.
(ECO-HS)
Figure 7 (Part 11 of 13).
PIN 4519909 50 Sides
© Copyright
MIM Volume RIO,
SC=95XX (XX= Format 1 Byte 15)
EC 476581 01Aug89
IBM Corp. 1987, 1988
SC-FRU-16
Ie'
(
Checkpoint Log: Checks D7 through F2 (SC
XX=
(
('
D7
Unexpected Offset active during
Data Check Recovery
D8
Index Not Found during
Data Check Recovery
D9
Device Selected dropped during
Data Check Recovery
DA
DCP error detected during
Data Check Recovery
DB
(..
,-.
~
DC
.
(
Description
DD
(
DF
FI
F2
= 95D7
through 95F2)
Front Devices
Rear Devices
Left
Right
Right
Left
O,4,8,C
2,6,A,E
3,7,B,F
1,5,9,D
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
BIF2,COB2 BIH2,CIB2 BIQ2,C2B2 BlS2,C3B2
No further action -(ECD-150)
go to RMAP 340 on page REPAIR-I3.
BID2,BIE2 BIK2,Bl.J2 BIN2,BIP2 BIU2,BIT2
COB2
CIB2
C2B2
C3B2
(ECD-151) MIM Volume RIO, page MAP-I, Entry DJ
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
BIF2,COB2 BIH2,CIB2 BIQ2,C2B2 BlS2,C3B2
No further action -(ECD-152)
go to RMAP 340 on page REPAIR-I3.
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
BIF2,COB2 BIH2,CIB2 BIQ2,C2B2 BlS2,C3B2
No further action -(ECD-152)
go to RMAP 340 on page REPAIR-13.
Unexpected sector compare during
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
Data Check Recovery
BIF2,C082 BIH2,CIB2 BIQ2,C2B2 BlS2,C3B2
No further action -(ECD-153)
go to RMAP 340 on page REPAIR-13.
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
DCP Check detected during
Data Check Recovery
BIF2,COB2 BIH2,CIB2 BIQ2,C2B2 BlS2,C3B2
No further action -(ECD-154)
go to RMAP 340 on page REPAIR-13.
Command gate missing during
BIF2,BIE2 BIH2,BIJ2 BIQ2,BIP2 BlS2,BlT2
Data Check Recovery
COB2,BID2 CIB2,BIK2 C2B2,BIN2 C3B2,BlU2
Controller AI/A3: AIQ2, AIU2, AIN2
Controller A2/A4: AIH2, AID2, AIJ2
(ECD-155) MIM Volume RIO, page MAP-I, Entry ED
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
DCP timeout failure during
C4P3
Crash Stop Recovery
C4P4
C4Pl
C4P2
(ECD-156) MIM Volume RIO, page MAP-I, Entry DN
BIE2,C4Pl BIJ2,C4P2 BIP2,C4P3 B1T2,C4P4
GBOD not detecteJ after Park during
Power Off
BID2
BIK2
BIN2
BlU2
(ECD-172) MIM Volume RIO, page MAP-I, Entry DC
Drive motor slowing down during sweep
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
8IG2
BIR2
BIR2
B1G2
(ECD-173) MIM Volume RIO, page PWR-I, Entry E
NOTE: For details about symptom code 95XX, see page ECD-nnn in the ECM, Volume R40.
(
"
Figure 7 (Part 12 of 13).
PIN 4519909 50 Sides
© Copyright IBM Corp,
SC=95XX (XX= Format 1 Byte 15)
EC 476581 01Aug89
1987, 1988
SC-FRU-17
Checkpoint Log: Checks F3 through FC (SC
xx=
= 95F3
Description
F3
AGC signal stuck active, motor not
turning during power-on reset (POR)
(£CO-174)
F4
Motor slowing, AGC active, motor relay
not picked during power-on reset (POR)
(£CO-175)
F6
No Air Pressure detected during Idle
F8
F9
FC
(£CO-I77 )
Sector compare not detected during
Sector Search
(£CO-178)
RPS Failed To lock detected during
load HAR command
(£CO-179)
Disable switch detected active during
any Search command
(£CO-180)
through 95FC)
Front Devices
Rear Devices
Right
Left
Right
Left
O,4,8,C
1,5,9,D
2,6,A,E
3,7,B,F
B1D2,B1E2 B1K2,B1J2 B1N2,B1P2 B1U2,BIT2
COB2,C4Pl C182,C4P2 C282,C4P3 C3B2,C4P4
-MIM Volume RIO, page MAP-I, Entry DJ
B1E2,B1D2 B1J2,B1K2 B1P2,B1N2 BIT2,B1U2
8162
81G2
81R2
BIR2
MIM Volume RIO, page PWR-I, Entry E
B1E2,B1G2 B1J2,B1G2 B1P2,BIR2 BIT2,B1R2
81Q2
81F2
81H2
8152
MIM Volume RIO, page PWR-I, Entry I
B1£2,B1D2 B1J2,BIK2 BIP2,BIN2 BIT2,B1U2
C4P2
C4P3
C4P4
C4Pl
page
MAP-I,
MIM Volume RIO,
Entry DN
BID2,BIE2 B1K2.B1J2 B1N2.B1P2 B1U2.BIT2
COB2,C4Pl C182.C4P2 C2B2,C4P3 C3B2,C4P4
MIM Volume RIO, page MAP-I, Entry DJ
B1G2,BIE2 B1G2,BIJ2 BIR2,BIP2 BIR2.BIT2
BIQ2
BIF2
BIH2
B152
MIM Volume RIO, page MAP-I. Entry CE
/'~'"
NOTE: For details about symptom code 95XX. see page ECD-nnn in the £CM, Volume R40.
Figure 7 (Part 13 of 13).
SC=95XX (XX= Format 1 Byte 15)
.j
--.
PIN 4519909 50 Sides
EC 476581 01Aug89
@ Copyright IBM Corp. 1987, 1988
SC-FRU-18
./
(.
c
SC=97XX(XX= Format 1 Byte 17)
Servo Control Check (SC = 9700 through 9754)
('
c
c
c
( .'.
Description
Rear Devices
Left
Right
2,6,A,E
3,7,B,F
00 There is no DCP error code in byte 17
B1E2.B1D2 B1J2.B1K2 B1P2.B1N2 B1T2,B1U2
8IF2,8IG2 8IH2.8IG2 8IQ2.BIR2 BIS2.BIR2
No further action -(ECD-122)
go to RMAP 340 on page REPAIR-13.
13 Motion command not accepted while busy
BIE2,BID2 BIJ2.B1K2 B1P2.BIN2 B1T2,BlU2
No further action -(ECD-122)
go to RMAP 340 on page REPAIR-13.
14 Disable Write Inhibit command
B1E2.BID2 BIJ2.BIK2 BIP2,B1N2 B1T2.B1U2
received while not offset
;;0 further action -(ECD-122)
go to RMAP 340 on page REPAIR-13.
20 Voice coil offset error too large
B1E2.BID2 BIJ2.BIK2 BIP2,BIN2 BIT2,B1U2
C4Pl,81F2 C4P2.BIH2 C4P3.8IQ2 C4P4.BIS2
*
(ECD-122) MIM Volume RIO. page MAP-I, Entry DC
21 Voice coil voltage or current integrate BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2.BIU2
C4Pl.BIF2 C4P2,BIH2 C4P3,BIQ2 C4P4,BIS2
* error too large
(ECD-122) MIM Volume RIO. page MAP-I. Entry DC
BIE2,BID2 BIJ2,BIK2 BIP2,B1N2 B1T2,B1U2
23 Excessive average VCM current
C4Pl.81F2 C4P2.BIH2 C4P3.8IQ2 C4P4.BlS2
(ECD-122) MIM Volume RIO, page MAP-I. Entry DC
BIE2,BID2 BIJ2.BIK2 BIP2.BIN2 B1T2,BlU2
51 Windage exceeds the allowable limit
C4Pl,81F2 C4P2,8IH2 C4P3,BIQ2 C4P4.8IS2
(ECD-122) MIM Volume RIO, page MAP-I, Entry DC
52 PES exceeds the allowable limit during
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
track following
Cft.PI,BIF2 Cft.P2,BIH2 Cft.P3,BIQ2 Cft.Pft..BIS2
(ECD-122) MIM Volume RIO, page MAP-I, Entry DN
53 AGC is not active
B1D2.B1E2 BIK2.BIJ2 B1N2,BIP2 BIU2.B1T2
BIG2.COB2 BIG2.CIB2 BIR2.C2B2 BIR2.C3B2
*
(ECD-122) MIM Volume RIO. page MAP-I, Entry DJ
54 locate command received when not track
B1E2.B1D2 B1J2.B1K2 BIP2.B1N2 B1T2,B1U2
following or offset
No .further action -(ECD-122)
go to RMAP 340 on page REPAIR-13.
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 97XX. see page ECD-nnn in the ECM, Volume R40.
XX=
Figure 8 (Part 1 of 3).
Front Devices
Right
Left
0,4,8,C
1,5,9,D
SC=97XX (XX= Format 1 Byte 17)
(
c
PIN 4519909 50 Sides
EC 476581 C1Aug89
© Copyright IBM Corp. 1987. 1988
SC-FRU-19
Servo Control Check (SC
xx=
= 9755
through 9783)
Description
Offset command received when not track
following or offset
55
(ECD-122)
Locate command outside the minimum or
maximum cylinder
56
(ECD-122)
AGC gain correction voltage reaches the
limit
57
(ECD-122)
58
Excessive position estimator error
(ECD-122)
Timeout during Locate, Offset, Sweep or
Rezero
59
(ECD-122)
Unable to locate guard band during
Rezero
5A
(ECD-122)
5B
PES P has too much offset
(ECD-122)
5C
PES Q has too much offset
(ECD-122)
5D
Incorrect device type
*
5E
*
81
(ECD-122)
Unable to move away from GBOD
(ECD-122)
IML fails checksum
(ECD-122)
83
Instruction storage parity error
(ECD-122)
*
Front Devices
Rear Devices
Right
Right
LeTt
LeTt
O,4,8,C
1,5,9,D
2,6,A,E
3,7,B,F
B1E2,B1D2 B1J2,B1K2 B1P2,B1N2 B1T2,B1U2
No further action -go to RMAP 340 on page REPAIR-13.
B1E2,B1D2 B1J2,B1K2 B1P2,B1N2 B1T2,B1U2
No further action -go to RMAP 340 on page REPAIR-I3.
B1E2,B1D2 B1J2,B1K2 B1P2,B1N2 B1T2,B1U2
Cft-PIt
CltP2
CltP3
CltPI
MIM Volume RlO, page MAP-I, Entry DN
B1D2,B1E2 B1K2,BIJ2 B1N2.BIP2 B1U2,BIT2
C4P4
C4P1
C4P2
C4P3
MIM Volume RIO, page MAP-I, Entry DN
B1D2,B1E2 B1K2.B1J2 B1N2,B1P2 B1U2,B1T2
C4P1
C4P2
C4P3
C4P4
MIM Volume RIO. page MAP-I. Entry DN
B1D2,B1E2 B1K2,BIJ2 B1N2,BIP2 BIU2,B1T2
CltPI,BIF2 CltP2.BIH2 CltP3,BIQ2 C4PIt,BIS2
MIM Volume RIO. page MAP-I, Entry DN
B1D2.B1E2 B1K2,B1J2 B1N2,BIP2 B1U2.B1T2
COB::>
C2B2
CI82
C3B2
MIM Volume RIO, page MAP-I, Entry DJ
BID2,B1E2 BIK2,BIJ2 BIN2,BIP2 BIU2,B1T2
COB2
CIB2
C2B2
C3B2
MIM Volume RIO, page MAP-I, Entry DJ
B1E2.B1F2 BIJ2,BIH2 BIP2,B1Q2 B1T2,B1S2
BIN2
BIU2
B1D2
B1K2
MIff Volume RlO, page MAP-I, Entry DI
BIE2,C4Pl B1J2.C4P2 BIP2,C4P3 B1T2.C4P4
BID2
B1K2
B1N2
BIU2
ffIM Volume RIO, page MAP-I. Entry DC
B1E2,B1D2 BIJ2,BIK2 BIP2.B1N2 B1T2,B1U2
BIF2,BIG2 BIH2.BIG2 BIQ2.BIR2 BIS2,8IR2
No further action -go to RMAP 340 on page REPAIR-I3.
BIE2,B1D2 B1J2,B1K2 B1P2.BIN2 B1T2,B1U2
8IF2,8IG2 8IH2,81G2 8IQ2.BIR2 81S2,8IR2
No further action -go to RMAP 340 on page REPAIR-I3.
A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 97XX. see page ECD-nnn in the ECM, Volume RltO.
Figure 8 (Part 2 of 3).
SC=97XX (XX= Format 1 Byte 17)
PIN 4519909 50 Sides
EC 476581 01AugB9
© Copyright IBM Corp. 1987, 1988
('
SC-FRU-20
/
(
Servo Control Check (SC
(
through 97FFl
Description
XX=
84
= 9784
Front Devices
Rear Devices
LeTt
Right
Right
LeTt
2,&,A,E
O,4,8,C
1,S,9,D
3,7,B,F
BIE2.BID2 BIJ2,BIK2 BIP2.BIN2 BIT2,BIU2
B1F2.B1G2 B1H2.B 1G2 B1Q2,BIR2 B1S2,B1R2
Data storage parity error
(ECO-122)
85
BIE2,BID2 BIJ2.BIK2 BIP2.BIN2 B1T2,BlU2
BIF2,81G2 81H2,81G2 81Q2,B1R2 BlS2,BIRZ
Data bus parity error
(ECO-1ZZ)
86
Time interval between interrupts is too
short
(ECO-IZZ)
87
*
FC
BID2,BIE2 BIK2.BIJ2 BIN2,BIP2 BIU2,BIT2
81FZ,8IGZ 8IHZ.81GZ 81QZ.81R2 81SZ,81RZ
No further action -go to RMAP 340 on page REPAIR-13.
No further action -go to RMAP 340 on page REPAIR-13.
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2.BlU2
BIG2,C08Z BIG2,C1BZ BIR2,CZ8Z BIR2,C3BZ
BIQZ
BIFZ
81HZ
81S2
(ECO-1ZZ) MIM Volume RIO, page MAP-I. Entry DJ
Multiply failure in Basic Assurance Test BIE2,BID2 BIJ2.BIK2 BIP2,BIN2 B1T2,BlU2
81FZ,BIGZ B1HZ.BIGZ B1QZ,B1RZ BlSZ,BIRZ
Time interval between interrupts is too
long
(ECO-1ZZ)
FD
No further action -go to RMAP 340 on page REPAIR-13.
BIE2.BID2 BIJ2,BIK2 BIP2.BIN2 B1T2,BlU2
BIF2 .B1GZ B1HZ,BIGZ B1QZ.BIRZ 81SZ,8IRZ
Illegal command
(ECO-IZZ)
88
No further action -go to RMAP 340 on page REPAIR-13.
Compute failure in Basic Assurance Test
No further action -go to RMAP 340 on page REPAIR-13.
BIE2,BID2 BIJ2.BIK2 BIP2.BIN2 BIT2,BlU2
B1FZ,BIGZ 8 IH2,B 1GZ 8IQZ.B1RZ BISZ,BIRZ
No further action -go to RMAP 340 on page REPAIR-13.
~
(ECO-IZZ)
FE
Checksum failure in Basic Assurance Test BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2.BlU2
B1FZ.BIGZ BIHZ,81GZ BIQZ.BIRZ 8ISZ,B1R2
(ECO-IZZ)
FF
('/
('
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2.BlU2
8IFZ,B1GZ BIHZ.BIGZ BIQZ,BIRZ BISZ,B1RZ
Power-on reset (POR) code hung
(ECO-IZZ)
No further action -go to RMAP 340 on page REPAIR-13.
* A more
detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 97XX, see page ECD-nnn in the ECM, Volume R40.
Figure 8 (Part 3 of 3).
{
No further action -go to RMAP 340 on page REPAIR-13.
SC=97XX (XX= Format 1 Byte 17)
"\
/'
«
c
PIN 4519909 50 Sides
EC 476581 O1AUgB9
© Copyright IBM Corp. 1987. 1988
SC-FRU-21
SC=98XX (XX= Format 9 Byte 18)
Read/Write Channel Checks (SC = 98XX)
Description
Controller
Front Devices
Rear Devices
Right
Left
Right
Al
A2
Left
2,6,A,E
(A3l
(A41
O,4,8,C
1,5,9,D
3,7,B,F
4C Transition check
AIQ2,AIN2 AIH2,AIJ2 COB2,BIF2 CIB2,BIH2 C2B2,BIQ2 C3B2,B1S2
AIV2,AIW2 AIC2,A182 8102,C182 81K2,C082 81N2,C382 81U2,C282
*
(ECD-399)
See MIM Volume RIO, page MAP-I, Entry EA
4E Transition check
AIQ2,AIN2 AlH2,AIJ2 COB2,BlF2 CIB2,BIH2 C2B2,BIQ2 C3B2,B1S2
A1V2,AIW2 AIC2,A182 8102,C182 81K2,C082 81N2,C382 81U2,C282
*
(ECD-399)
See MIM Volume RIO. page MAP-I. Entry EA
COB2,BIF2 CIB2.BIH2 C2B2.BIQ2 C3B2,B1S2
XX Read/Write Channel
* check
(ECD-376)
See MIM Volume RIO, page MAP-I. Entry EA
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 98XX. see page ECD-376 in the ECM, Volume R40.
XX=
Figure 9.
SC=98XX (XX= Format 9 Byte 18)
/
SC=9CXX (XX= Format 1 Bytes 11-15)
'
DDC Hot Bits (SC = 9CXX)
Controller
Front Devices
Rear Devices
Right
Right
Al
A2
Left
Left
(A3l
O,4,8,C
1,5,9,D
2,6,A,E
3,7,B,F
(A4l
AIU2,AIV2 AID2.AIC2 BIF2,BIG2 BIH2.BIG2 BIQ2,BIR2 B1S2.BlR2
XX Bits always on,
Format I or 9 sense A1T2
AlE2
* assembly selection
(ECD-1)
See MIM Volume RIO. page MAP-I. Entry CD
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 9CXX, see page ECD-I in the ECM, Volume R40.
XX=
Figure 10.
Description
SC=9CXX (XX= Format 1 Bytes 11-15)
l'
\'-.J
PIN 4519909 50 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
SC-FRU-22
SC=9DXX (XX= Format 1 Byte 15)
Checkpoint Log (SC = 9DXXl
(
XX=
Description
XX
Rear Devices
Front Devices
Right
Left
Right
Left
O,4,8,C
1,5,9,D
2,6,.A.E
3,.7,.B,.F
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 BIT2,BIU2
BIF2,C4Pl BIH2,C4P2 BIQ2,C4P3 BIS2,C4P4
B1G2
B1G2
BIR2
BIR2
Seek incomplete
Use FRUs for symptom code 95XX with the
same value of XX. If the value XX is not
listed for symptom code 95XX, use the
(ECO-ll) HIH Volume RIO, page MAP-I, Entry DN
FRUs shown here.
3( A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 9DXX, see page ECD-ll in the ECH~ Volume R40.
3(
(-/
Figure 11.
SC=9DXX (XX= Format 1 Byte 15)
SC=9E10 (Format 1)
Description
c
9ElO
I
Figure 12.
False end of track detected by
system program.
Warning: Data loss has occurred
or may occur. Replace HDA and all
FRUs listed.
Front Device
Right
left
COB2
ClB2
AlF2
AlH2
HDA
HDA
Rear Device
Right
left
C3B2
C2B2
AlQ2
AlS2
HDA
HDA
SC=9E10 (Format 1)
C'
c\
PIN 4519909 50 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1997. 1989
SC-FRU-23
SC=9FXX (XX= Format 1 or 9 Byte 07)
Format and Message (SC
XX=
Description
10
Intervention
Required or Device
End op with no
device checks
= 9FI0
through 9F97)
Controller
Al
A2
(A3)
(A4)
(ECO-3)
11
13
14
*
15
*
16
18
19
IB
*
IC
*
IF
97
No further action -
go to RMAP 340 on page REPAIR-13.
BIF2,BIE2 BIH2,BIJ2 BIQ2,BIP2 B152,B1T2
BID2,BIG2 BIK2,B1G2 B1N2,BIR2 B1U2,BIR2
Device Status 1 not
as expected
(ECO-3)
Rear Devices
Front Devices
Right
Right
Left
Left
1,5,9,D
2,6,A,E
3,7,B,F
O,4,8,C
BIF2,BIE2 BIH2,BIJ2 BIQ2,BIP2 B152,B1T2
BIN2
BIU2
BID2
BIK2
No further action -
go to RMAP 340 on page REPAIR-13.
BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,B1U2
C2B2
COB2
C3B2
C1B2
(ECO-S)
See MIM Volume RIO, page MAP-I, Entry DJ
Interrupt cannot be A1U2,A1T2 AID2,AIE2 BIF2,BIE2 B1H2,BIJ2 B1Q2,BIP2 BI52,B1T2
. reset
(ECO-6)
No further action - go to RMAP 340 on page REPAIR-13.
Device does not
A1U2,A1T2 A102,AIE2 B1F2,BIG2 BIH2,BIG2 BIQ2,BIR2 BI52,BIR2
respond to selection AIN2,AIV2 AIJ2,AIC2 B1E2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
AIS2
AIF2
(fCD-6)
See MIM Volume RIO. page MAP-I. Entry CD
Device Check 2 or
BIG2,BIE2 BIG2,BIJ2 BIR2,BIP2 BIR2,B1T2
Set Sector
BID2,BIF2 BIK2,81H2 BIN2,81Q2 BIU2,BlS2
Incomplete
(ECO-7)
No further action - go to RMAP 340 on page REPAIR-13.
BIF2,BIE2 BIH2.BIJ2 BIQ2,BIP2 BI52,BIT2
Device Status 1 has
an invalid combinaBID2,BIG2 BIK2,BIG2 BIN2,BIR2 BIU2,8IR2
tion of bits
(ECIl-8)
No further action - go to RMAP 340 on page REPAIR-13.
Device not ready to
BIF2,BIE2 BIH2,BIJ2 BIQ2,BIP2 B152,B1T2
do customer work
B1D2,BIG2 B1K2,81G2 B1N2,BIR2 BIU2,BIR2
(ECO-9)
No further action - go to RMAP 340 on page REPAIR-13.
Missing device
AIU2,A1T2 AID2,A1E2 BIF2,BIE2 BIH2,BIJ2 B1Q2,BIP2 B152,8 H2
A1V2
B102,BIG2 BIK2,BIG2 BIN2,BIR2 B1U2,BIR2
address bit at
A1C2
selection
(ECO-I0)
See MIM Volume RIO, page MAP-I. Entry CD
BIE2,BIG2 BIJ2,B1G2 BIP2,BIR2 BIT2,BIR2
Drive Motor switch
sensed off while
B1F2
BIQ2
B152
BIH2
running
(ECO-IV
See MIM Volume RIO, page PWR-l, Entry AC
Offset Active cannot
B1D2,B1E2 BIK2,BIJ2 BIN2,BIP2 BIU2,B1T2
be reset
(ECO-12)
See MIM Volume RIO, page MAP-I. Entry DJ
Head address
COB2,BIF2 CIB2,BIH2 C2B2,BIQ2 C3B2,B152
miscompare
B1E2
BIJ2
BIP2
BIT2
(fCO-362)
See MIM Volume RIO, page MAP-I, Entry EA
Index missing
/
*
A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code 9FXX, see page ECD-nnn in the ECM, Volume R40.
Figure 13 (Part 1 of 2).
SC=9FXX (XX= Format 1 or 9 Byte 07)
PIN 4519909 50 Sides
EC 476581 01Aug89
SC-FRU·24
@ Copyright IBM Corp. 1987, 1988
/f"
I
\"
(
Format and Message (SC
= 9F9A
xx=
Description
9A
Track physical
address miscompare
while oriented
(ECD-363)
9E
Cylinder address
miscompare
Controller
Al
A2
(A3l
(A4l
Front Devices
Rear Devices
Right
Left
Right
Left
O,4,8,C
1,5,9,D
2,6,A,E
3,7,B,F
COB2,B1E2IC1B2,B1J2 C2B2,B1P2IC3B2,BIT2
BIF2
B1H2
B1Q2
B1S2
See MIM Volume RIO, page MAP-I, Entry EA
BID2,B1E2IB1K2,BIJ2 B1N2,BIP2IB1U2,B1T2
C4P1
C4P2
C4P3
C4P4
See MIM Volume RIO, page MAP-I, Entry DJ
(
~'
(.
(ECD-364)
...
(
through 9F9El
I
I
NOTE: For details about symptom code 9FXX, see page ECD-nnn in the ECM
Figure 13 (Part 2 of 2).
SC=9FXX (Format 1 and 9)
Ie=;
Device Undervoltage (SC
9F
FF
Volume R40.
SC=9FXX (XX= Format 1 or 9 Byte 07)
C'
xx:
2
= 9FFF)
Description
Non-repeatable
runout
Controller
Al
A2
(A3l
(A4l
Front Devices
Right
Left
O,4,8,C
1,5,9,D
Rear Devices
Left
Right
2,6,A,E
3,7,B,F
See SMAP 0058: Determining the Failing FRUs
on page START-34.
Device undervoltage
(ECD-I82)
BID2,B1F2 BIK2,BIH2 B1N2,B1Q2 BlU2,BlS2
See MIM Volume RIO, page PWR-I, Entry P
NOTE: For details about symptom code 9FFF, see page ECD-182 in the ECM 2 Volume R40.
Figure 14.
SC=9FXX (Format 1 and 9)
(
('
PIN 4519909 50 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
SC-FRU-2S
\
"'--
SC=D1XX (XX= Format 7 Byte 11)
Connection Check Alert and Power status (SC
XX=
41
*
51
3(
82
*
= DlXXl
Description
Controller
Al
A2
(A31
(A41
Controller String 1: AlW2,AlN2 AlB2,AlJ2
Connection check
AlV2,AlR2 AlC2,AlG2
alert and power on
AIQ2,A1U2 AIH2,A1D2
Front Devices
Right
Left
O,4,8,C
l,S1'9,D
/'
Rear Devices
Right
Left
21'6,A,E
31'71'Bl'F
I
A1S2,AIT2 AIF2,AIE2
No further action -- go to RMAP 340 on page REPAIR-13.
Controller String 1 : A1S2,AlF2 AlF2,A1S2
Connection check
alert with DPS 1
unconditional
reserve release and
power on
No further action -- go to RMAP340 on page REPAIR-13.
Controller String 0: AlW2,AlN2 AIB2,AlJ2
AIV2,AIR2 AIC2,AIG2
Connection check
alert and power on
AIQ2,A1U2 AIH2,A1D2
A1T2
'",--/
AIE2
No further action -- go to RMAP 340 on page REPAIR-13.
A2 Controller String 01 AlS2,AIF2 AlF2,A1S2
Connection check
alert with DPS 0
unconditional
reserve release and
power on
No further action -- go to RMAP 340 on page REPAIR-13.
* A more detailed analysis can be obtained by using the MD to analyze the SC.
*
Figure 15.
'-..--_/
SC=D1XX (XX= Format 7 Byte 11)
~-
PIN 4519909 50 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
SC-FRU-26
"
('
(:
SC=D2XX (XX= Format 7 Byte 12)
Controller String 0 Check-l status 1 (SC
Bit
(
c
c
(:
Description
= D2XXl
Controller
Al
A2
(A3l
(A41
Front Devices
Right
Left
O,4,8,C
l,S,9,D
0-2 I/O Control 1 Checks
:I(
Bi ts 0-2 Equa 1 :
001 laC ROS Bit
AlR2.AlV2 AlG2.AlC2
:I(
Parity Check
A1W2
A1B2
(ECO-21O)
No further action -AlV2,A1W2 AlC2.A1B2
OIOIIOC Transfer
:I(
Clock Check
(ECD-211)
No further action -OIIIIOC Detected DDC AlW2.AIV2 AIB2.AIC2
:I(
Bus nut Parity
Check (ECD-213)
See MIM Volume
100 laC Register
AIV2,AIR2 AlC2,AlG2
:I(
Data Bus Parity AIW2
AIB2
Check
(ECD-2H)
No further action -AIV2,AIW2 AlC2,AIB2
101lIOC Register 1
:I(
Pa ri ty Check
(ECD-2IS)
No further action -110 IOC Register 3
AlV2,AlU2 AIC2,AlD2
:I(
Parity Check
AlT2
AIE2
(ECD-2I7)
No further action -III IOC Controller
AIV2,AIW2 AlC2,AIB2
:I(
Selection Check AIQ2
AIH2
(ECD-2I8)
See MIM Volume
3 DDC Bus Out Parity
AIW2
AIB2
:I(
Check
(ECD-220)
See MIM Volume
AIN2,AIQ2 AIJ2,AlH2
4 Controller Clock
:I(
Check
AIW2
AIB2
(ECD-22I)
See MIM Volume
5 Controller Sequencer AlR2,AIV2 AIG2,AIC2
Check
AIN2
AIJ2
(ECD-222)
No further action -AIV2,AIW2 AIC2,AlB2
6 DDC Bus In
pa ri ty check
(ECD-223)
No further action -7 IOC Card Check-l
AIV2,AIR2 AIC2,AIG2
check
AIW2
AIB2
(ECD-22S)
No further action --
*
*
*
* A more
Rear Devices
Left
Right
2,6,A,E
3,7,B"F
go to RMAP 340 on page REPAIR-13.
go to RMAP 340 on page REPAIR-13.
R10 Page MAP-I. Entry AA
go to RMAP 340 on page REPAIR-13.
go to RMAP 340 on page REPAIR-13.
go to RMAP 340 on page REPAIR-13.
RIO Page MAP-I, Entry AA
RIO Page MAP-I. Entry AA
RIO Page MAP-I. Entry BA
go to RMAP 340 on page REPAIR-13.
go to RMAP 340 on page REPAIR-13.
go to RMAP 340 on page REPAIR-13.
detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code D2XX, see page ECD-nnn in the ECM Volume R40.
1
Figure 16.
SC=D2XX (XX= Format 7 Byte 12)
(
c'
PIN 4519909 50 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987. 1988
SC-FRU-27
SC=D3XX (XX= Format 7 Byte 13)
Controller String 0 Check-l status 2 CSC
Bit
Description
0
DDC Tag Out
sequence check
(ECD-226)
3(
I
3(
3
3(
4
3(
Extended Command,
Tag sequence check
(ECD-228)
Controller Gate DDC
Drivers check
(ECD-230)
RCC Sequence check
(ECD-23I)
= D3XX)
Controller
Front
Al
A2
Left
(A4)
(A3)
O,4,8,C
AIW2,AIV2 AIB2,AIC2
AIN2
AIJ2
See MIM Volume RIO,
AIW2,AIV2 AIB2,AIC2
AIR2
AIG2
See MIM Volume RIO,
AIW2,AIN2 AIB2,AIJ2
AIV2
AIC2
See MIM Volume RIO,
AIW2,AIR2 AIB2,A1G2
See MIM Volume RJO,
Devices
Right
1,5,9,D
Rear Devices
Right
Left
2,6,A,E
3,7,B,F
page MAP-I, Entry AA
page MAP-I, Entry AA
page MAP-I, Entry AA
page MAP-I, Entry AA
3( A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code D3XX, see page ECD-nnn in the ECM, Volume R40.
Figure 17.
SC=D3XX (XX= Format 7 Byte 13)
\.,; ..
\ ......-
PIN 4519909 50 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
SC-FRU-28
,/
_/
(/
SC=D4XX (XX= Format 7 Byte 14)
(
(
(
Controller string I Check-I status I (SC
Bit
Description
= D4XXl
Controller
Al
A2
(A31
(A4)
Front Devices
Right
Left
1,5,9,D
O.4.8,C
Rear Devices
Left
Right
2,6,A,E
3.7,B,F
0-2 I/O Control I Checks
3(
Bits 0-2 Equal:
001 IOC ROS Bit
AIR2,AIV2 AIG2,AIC2
3(
parity check
AIB2
AIW2
(ECO-2I0)
No further action -- go to RMAP 340 on page REPAIR-13.
AIV2,AIW2 AIC2,AIB2
OlOIIOC Transfer
3(
Clock check
(ECO-2I11
No further action -- go to RMAP 340 on page REPAIR-I3.
011 IOC Detected DDC AIW2,AIV2 AIB2,AIC2
3(
Bus Out parity
check
(ECO-2I3)
No further action -- go to RMAP 340 on page REPAIR-I3.
100 lac Register
AIV2,AIR2 AIC2,AIG2
3(
Data Bus
AIW2
AIB2
parity check
(ECO-214)
No further action -- go to RMAP 340 on page REPAIR-I3.
AIV2,AIW2 AIC2,AIB2
1011 lac Register 1
3(
parity check
(ECO-2IS)
No further action -- go to RMAP 340 on page REPAIR-I3.
AIV2,AIU2 AIC2,AID2
110 IOC Register 3
3(
parity check
AIT2
AIE2
(ECO-2I71
No further action -- go to RMAP 340 on page REPAIR-I3.
AIV2,AIW2 AIC2,AIB2
111 IOC Controller
3(
Selection check AIQ2
AIH2
(ECO-2I8)
See MIM Volume RIO, page MAP-I, Entry AA
3 DDC Bus Out
AIW2
AIB2
3(
parity check
(ECO-220)
See MIM Volume RIO, page MAP-I, Entry AA
4 Controller Clock
AIN2,AIQ2 AIJ2,AIH2
3(
check
AIB2
AIW2
(ECO-2211
See MIM Volume RIO, page MAP-I, Entry BA
5 Controller Sequencer AIR2,AIV2 AIG2,AIC2
3(
check
AIN2
AIJ2
(ECO-222)
No further action -- go to RMAP 340 on page REPAIR-I3.
AIV2,Al\'I2 AIC2,AIB2
6 DDC Bus In
3(
parity check
(ECO-223)
No further action -- go to RMAP 340 on page REPAIR-I3.
7 IOC Card Check-I
AIV2,AIR2 AIC2,AIG2
3(
check
AIW2
AIB2
(ECO-22S)
No further action -- go to RMAP 340 on page REPAIR-I3.
3( A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code D4XX, see page ECD-nnn in the ECM, Volume R'tO.
Figure 18.
SC=D4XX (XX= Format 7 Byte 14)
PIN 4519909 50 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
SC-FRU-29
SC=D5XX (XX= Format 7 Byte 15)
Controller String 1 Check-l Status 2 (SC
Bit
0
*
I
*
3
*
4
*
Description
DDC Tag Out
sequence check
(ECD-Z26)
Extended Command,
Tag sequence check
(ECD-Z28)
Controller Gate DDC
Drivers check
(ECD-Z30)
RCC sequence check
(ECD-Z3I)
= D5XXl
Controller
Front
Al
A2
Left
(A41
O,4,8,C
(A3l
AIW2,AIV2 AIB2.AIC2
AIN2
AIJ2
See MIM Volume RlO,
AIW2,AIVZ AIB2,AICZ
AIRZ
See MIM Volume RIO,
AIW2,AIN2 AIB2.AIJ2
AIC2
AIV2
See MIM Volume RIO.
AIW2.AIR2 AIB2.AIG2
See MIM Volume RIO,
Devices
Right
1,5,9,D
Rear Devices
Right
Left
2,6,A,E
3,7,B,F
page MAP-I, Entry AA
pa~
MAP-I, Entry AA
page MAP-I, Entry AA
page MAP-I, Entry AA
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code D5XX. see page ECD-nnn in the ECM. Volume
Figure 19.
R~O.
SC=D5XX (XX= Format 7 Byte 15)
PIN 4519909 50 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987. 1988
SC-FRU-30
SC=D6XX (XX= Format 7 Byte 16)
Controller Sequencer Detected Check-l Errors--String 0 (SC
( "\
./
c
XX=
Description
02
Sequencer and/or
IOC register failure
(ECD-233)
*
04
*
10
*
c
12
c
18
(
,
*
*
IA
*
/
Controller
Al
A2
(A31
(A41
AIV2.AIR2 AIC2.AIG2
= D602
Front Devices
Right
Left
O,4,8,C
1,5,9,D
through D636)
Rear Devices
Right
Left
2,6,A,E
3,7,B,F
No further action -- go to RMAP 340 on page REPAIR-13.
Sequencer counter
AIG2
AIR2
failed
(ECD-234)
No further action -- go to RMAP 340 on page REPAIR-l3.
Controller command
AIV2.AIW2 A1C2.AIB2
active With a DPS
A1U2.A1T2 A1D2.AIE2
command on DDC
(ECD-23S)
Bus Out
No further action -- go to RMAP 340 on page REPAIR-13.
Controller command. AIV2.AIS2 AIC2.AIF2
DPS command. or
AIF2.AIW2 A1S2.AIB2
Selected stuck
(ECD-237)
active
No further action -- go to RMAP 340 on page REPAIR-13.
Timeout routine:
AIV2
A1C2
Allow Selection off
and Device Selected
(ECD-2ft.O)
on
No further action -- go to RMAP 340 on page REPAIR-13.
Controller timeout: AIU2.AIT2 AID2,AIE2 BIF2.BIG2 BIH2.BIG2 BIQ2.BIR2 B1S2,BIR2
DDC Tag Out = Ol~
AIV2.AIW2 AIC2.AIB2
CDP/DPS Tag In = 01
(ECD-2ft.2)
See MIM Volume RIO. page MAP-I. Entry CD
Controller command
AIV2
AIC2
active without DDC
Tag Out being in a
command gate state
(ECD-2ft.3)
No further action -- go to RMAP 340 on page REPAIR-13.
22 DPS command active
AIV2.A1S2 AIC2,AIF2
with a non-DPS
AIF2.AIW2 A1S2.AIB2
• command on the DDC
Bus Out lines
(ECD-2ft.4)
No further action -- go to RMAP 340 on page REPAIR-13.
20
*
c
(
*
32
Reset allegiance.
AIV2,AIU2 AIC2,AID2
Start conditions
A1T2
AIE2
not correct
(ECD-246)
No further action -- go to RMAP 340 on page REPAIR-13.
34
Reset allegiance,
AIV2.AIW2 AIC2,AIB2
invalid device
address received
from storage control
(ECD-2ft.7 )
No further action -- go to RMAP 340 on page REPAIR-13.
Timeout or Sense
AIW2.AIV2 AIB2.AIC2
Fault log,
nnc Tag Out lines in
invalid state
(ECD-2ft.9)
No further action -- go to RMAP 340 on page REPAIR-13.
*
*
36
*
(~
* A more
detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code D6XX, see page ECD-nnn in the ECM, Volume R40.
(~'
Figure 20 (Part 1 of 2).
SC=D6XX (XX= Format 7 Byte 16)
PIN 4519909 50 Sides
EC 476581 01Aug89
(0 Copyright IBM Corp. 1987, 1988
SC-FRU-31
Controller Sequencer Detected Check-l Errors--String 0 {SC
xx=
Description
Controller
Al
A2
(A31
(A41
70 Invalid Tag State
AIW2,AIV2 AIB2,AIC2
on the DDC Tag Out
(ECD-2S0)
lines
No further action
72 Internal lID'
AIW2,AIR2 AIB2,AIG2
AIV2
AIC2
* register failure
(ECD-251)
No further action
74 Invalid command on· AIW2,AIR2 AIB2,AIG2
DDC Bus Out
AIV2
AIC2
* the
lines
(ECD-252)
No further action
CE During device power AIV2,AIR2 AIC2,AIG2
on, controller
* failed
to take
(ECD-256)
control
No further action
* A more detailed analysis can be obtained by
NOTE: For details about symptom code D6XX, see
Figure 20 (Part 2 of 2).
= D670
Front Devices
Right
Left
O,4,8,C
1,5,9,D
through D6CEJ
Rear Devices
Right
Left
2,6,A,E
3,7,B,F
"-_ ../
-- go to RMAP 340 on page REPAIR-13.
/-.\
-- go to RMAP 340 on page REPAIR-13.
-- go to RMAP 340 on page REPAIR-13.
-- go to RMAP 340 on page REPAIR-13.
using the MD to analyze the SC.
page ECD-nnn in the ECM, Volume R40.
!/
"
/'
'.
SC=D6XX (XX= Format 7 Byte 16)
"-'"<
_/
''''-..t -, ../
PIN 4519909 50 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987. 1988
SC-FRU-32
f
SC=D8XX (XX= Format 7 Byte 18)
(
Controller Sequencer Detected Check-1 Errors--Str.ng 1 (SC
xx=
Description
02
Sequencer and/or
IOC register failed
(ECO-233)
>I:
04
>I:
10
>I:
12
c\
>I:
18
>I:
('
(
through D870)
Front Devices
Right
LeTt
0,4,8,C
1,5,9,D
Rear Devices
Right
LeTt
2,6,A,E
3,7,B,F
No further action -- go to RMAP 340 on page REPAIR-13.
Sequencer counter
A1R2
A1G2
failed
(ECO-231t)
No further action
A1V2,A1W2 A1C2,A1B2
Controller command
A1U2,A1T2 A1D2,AIE2
active with a DPS
command on nnc
(ECO-235)
Bus Out
No further action
Controller command, A1V2,A1S2 A1C2,A1F2
nps command, or
AIF2,AIW2 AlS2,AlB2
selected stuck
(ECO-237)
active
No further action
Timeout routine,
A1V2
A1C2
Allow Selection off
Device Selected on
(ECO-2ft-O)
No further action
-- go to RMAP 340 on page REPAIR-13.
-- go to RMAP 340 on page REPAIR-13.
-- go to RMAP 340 on page REPAIR-13.
-- go to RMAP 340 on page REPAIR-13.
lA
Controller timeout: A1U2,A1T2 A1D2,A1E2 B1F2,B1G2 B1H2,B1G2 B1Q2,B1R2 B1S2,B1R2
A1V2,AIW2 A1C2,AIB2
nnc Tag Out = 011
CDP/DPS Tag In = 01
(ECO-2lt2)
See 141M Volume RIO, page MAP-I, Entry CD
20
Controller command
A1V2
A1C2
active without
DDC Tag Out in a
command gate state
(ECO-2ft-3)
No fUrther action -- go to RMAP 340 on page REPAIR-l3.
DPS command active
A1V2,A1S2 A1C2,A1F2
with a non-DPS
AIF2,AIW2 AlS2,AIB2
command on the nDC
Bus Out lines
(ECO-2ft-ft.)
No further acti~n -- go to RMAP 340 on page REPAIR-13.
>I:
>I:
(
Controller
A1
A2
(A3)
(A4)
A1V2,A1R2 A1C2,A1G2
= D802
22
>I:
A1V2,AlU2 A1C2,A1D2
Reset alle~iance,
Start cond1tions
A1T2
A1E2
not correct
(ECO-2ft-6)
No further action -- go to RMAP 340 on page REPAIR-l3.
34 Reset allegiance,
A1V2,A1W2 A1C2,A1B2
>I:
invalid device
address received
from storage control
(ECO-2ft.7)
go to RMAP 340 on page REPAIR-l3.
No further action
36 Timeout or Sense
A1W2,A1V2 A1B2,A1C2
>I:
Fault log,
DDC Tag Out lineu in
invalid state
(ECO-2ft.9)
No further action -- go to RMAP 340 on page REPAIR-B.
>I: A more detailed analysis can be obtained by using the MD to analyze the SC.
32
>I:
(
("
--
NOTE: For details about symptom code n8XX, see page ECD-nnn in the
(:
Figure 21 (Part 1 of 2).
SC=D8XX (XX= Format 7 Byte 18)
PIN 4519909 50 Sides
EC 476581 01Aug89
ECM~
Volume Rita.
(
© Copyright
IBM Corp. 1987, 19Ba
SC-FRU-33
Controller Sequencer Detected Check-l Errors--string 1 (SC = D872 through D8CE)
xx=
Description
70
Controller
Al
A2
(AJ)
(A4)
AlW2,AlV2 AlB2,AIC2
Invalid Tag State
on the DDC Tag Out
lines
(£CO-250)
No further action
AIW2,AIR2 AIB2,AIG2
72 Internal 'ID'
)(
register failure
A1V2
AIC2
(£CO-251)
No further action
74 Invalid command on
AIW2,AIR2 AlB2,AIG2
)(
the DDC Bus Out
A1V2
AIC2
lines
(£CD-252)
No further action
CE During device power AIV2,AlR2 AlC2,AlG2
on, controller
failed to take
(£CD-256)
control
No further action
* A more detailed analysis can be obtained by
NOTE: For details about symptom code DaXX, see
Figure 21 (Part 2 of 2).
SC=DEXX (XX
Front Devices
Right
Left
O,4,8,C
1,5,9,D
Rear Devices
Right
Left
J,7,B,F
2,6,A,E
-- go to RMAP 340 on page RfPAIR-13.
-- go to RMAP 340 on page REPAIR-13.
-- go to RMAP 340 on page REPAIR-13.
/'
-- go to RMAP 340 on page REPAIR-13.
using the MD to analyze the SC.
page fCD-nnn in the £CM~ Volume R40.
SC= D8XX (XX= Format 7 Byte 18)
= Format 7 Byte 11)
Controller Check Alert and Power Status (SC = DE2X)
XX=
Description
2X
Storage director
detected that both
RCC sequences were
not successful
(£CO-189)
Controller
A2
Al
(AJ)
(A4)
AlW2,AIR2 AlB2,AlG2
Front Devices
Right
Left
O,4,8,C
l,5,9,D
Rear Devices
Right
Left
2,6,A,E
J,7,B,F
See MIM Volume RIO, page MAP-I, Entry AA
NOTE: For details about symptom code DE2X, see page ECD-la9 in the £CM, Volume R40.
Figure 22.
SC=DEXX (X= Format 7 Byte 11 Bits 4 & 5)
PIN 4519909 50 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
SC-FRU-34
'
(
SC=DFXY (X= Format 7 Byte 10 Bits 0-3, Y= Byte 7 Bits 4-7)
Format 7 Message (SC
(~/
XV=
Description
10
Storage control
detected DDC Bus In
parity check
(ECO-205)
Storage control
detected DDC Sync In
check
(ECO-204)
DDC Tag In check
*
20
*
40
*
c
c
( -".
,
c
4[
= DFIO
(ECO-203)
80
through DFX4l
Controller
Front
Al
A2
Left
O,4,8,C
(A3l
(A4l
AIW2,AlV2 AIB2,AIC2 81F2
AIN2,A1U2 AIJ2,A1D2
A1T2
AIE2
See MIM Volume RIO,
AIW2,AlV2 AlB2,AlC2
AIN2
AIJ2
See MIM Volume RIO,
AlW2,AIV2 AlB2,AIC2 81F2
A1S2,A1U2 AIF2,A1D2
A1T2,AIN2 AIE2,AIJ2
AIF2,AIR2 AlS2,A1G2
See MIM Volume RIO,
Devices
Right
l,5,9,D
81H2
Rear Devices
Left
Right
2,6,A,E
3,7,B,F
81Q2
81S2
page MAP-I, Entry AB
page MAP-I, Entry AD
BIQ2
81H2
BlS2
page MAP-I, Entry AB
Connection check
AlW2
alert is on solid
(ECO-202)
Al\.,2
RCC I sequence not
successful
(ECO-188)
AlB2
See MIM Volume RIO, page MAP-I, Entry AB
Xl
AIB2
See MIM Volume RIO, page MAP-I, Entry AA
X2 RCC 1 And RCC 2
AlW2,AIR2 AIB2,AIG2
sequences not
successful (ECO-189)
See MIM Volume RIO, page MAP-I, Entry AA
BIQ2
43 Storage control
AIW2,A1V2 A1B2,AIC2 BIF2
81S2
81H2
detected DDC Tag-In A1S2,A1U2 A1F2,A1D2
* check
A1T2,AIF2 AIE2,AlS2
AIN2,AIJ2 AIJ2,AIN2
AIR2
AIG2
See MIM Volume RIO, page MAP-I, Entry AB
AlV2,A1W2 A1C2,AlB2
X3 Invalid Tag In
during selection
A1S2,A1U2 A1F2,A1D2
* sequence
A1T2,AIN2 AIE2,AIJ2
AIF2
AlS2
(ECO-190)
See MIM Volume RIO, page MAP-I, Entry AA
X4 Storage control
AlW2
AIB2
detected a Check-l
check during an RCC
sequence
(ECO-191)
See MIM Volume RIO, page MAP-I, Entry AA
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code DFXY, see page ECD-nnn in the EeN, Volume R40.
Figure 23 (Part 1 of 2).
SC= DFXY (X= Format 7 Byte 10 Bits 0-3, Y= Byte 7 Bits 4-7)
PIN 4519909 50 Sides
EC 476581 01Aug89
(
© Copyright IBM Corp. 1987, 1988
SC-FRU-35
Format 7 Message (SC = DFX5 through DFXD)
Description
XV=
X5
*
X6
*
X7
*
X8
*
Invalid DDC
selection response,
timeout waiting for
a response, or
controller power
(ECO-192)
missing
Missing End Op,
transfer was
complete
(ECO-193)
Missing End Op,
transfer Was
incomplete
(ECO-19S)
Invalid Tag In for
an immediate command
sequence
Controller
Al
A2
(A4)
(A3)
Front Devices
Right
Left
O,4,8,C
1,5,9,D
Rear Devices
Right
Left
2,6,A,E
3,7,B,F
A1J2,AIC2
AIB2,AIG2
AIH2,A1D2
AIE2,AIF2
See MIM Volume RIO, page PWR-I, Entry
AIW2,AIU2 AIB2,AID2
A1T2,AIN2 AIE2,AIJ2
AlS2
AIF2
No further action -- go to RMAP 340 on page
AIU2,A1T2 AID2,A1E2
A1W2,A1N2 A1B2,A1J2
AIS2
A1F2
No further action -- go to RMAP 340 on page
A1S2,AIV2 A1F2,A1C2
AIU2,A1T2 AID2,AIE2
AIW2,A1N2 AI82,A1J2
AlS2,A1Q2 A1F2,A1H2
See MIM Volume RIO, page MAP-I, Entry
A1U2,A1T2 AID2,A1E2 B1F2
BIQ2
B1H2
AIV2
AIC2
No further action -- go to RMAP 340 on page
A1S2,AIV2 AIF2,AIC2
AIW2
AIB2
No further action -- go to RMAP 340 on page
AUI2,AIU2 AIB2,AID2
A1S2,AIN2 AIF2,A1J2
AIQ2
A1H2
No further action -- go to RMAP 340 on page
AIW2
AIB2
See MIM Volume RIO, page MAP-I, Entry
This is a status code.
No further action -- go to RMAP 340 on page
AIN2,AIV2
AIW2,AIR2
AIQ2,A1U2
A1T2,AIS2
AA
REPAIR-l3.
REPAIR-13.
(ECO-196)
AA
X9 Invalid Tag In for
B1S2
an extended command
* sequence
(ECO-197)
REPAIR-13 .
XA Storage control
timed out
* microcode
on deselection
(ECO-198)
REPAIR-13.
XB No selection
response after poll
* interrupt
(ECO-199)
REPAIR-I3.
XC Permanent path error
*
(ECO-200)
AA
XD Controller not
3(
available on
REPAIR-13.
disconnected command
(ECO-20l)
chain
3( A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code DFXY, see page ECD-nnn in the ECM 1 Volume R~O.
I
Figure 23 (Part 2 of 2).
/---...."
/
SC=DFXY (X= Format 7 Byte 10 Bits 0-3, Y= Byte 7 Bits 4-7)
,--1f --\
"'.;
PIN 4519909 50 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
SC-FRU-36
('
SC=DFXX (Format 7)
Format 7 Message (SC
(~'
c
XX=
Description
FF
Controller
undervoltage
(ECO-323)
c
(
(
Front Devices
Left
Right
O,4,8,C
1,5,9,D
Rear Devices
Left
Right
2,6,A,E
3,7,B,F
See HIH Volume RIO, page PWR-I, Entry 0
SC=9FXX (Format 7)
SC=EOXX (XX= Format 8 Byte 10)
DTI/XES Register (SC
Bit
I
c
Controller
Al
A2
(A3l
(A41
AlQ2,AIR2 AIH2,AIG2
NOTE: For details about symptom code DFFF, see page ECD-323 in the ECH, Volume R40.
Figure 24.
c
= DFFF)
*
Description
Tag In check
= EOXXl
Controller
Front Devices
Right
Al
A2
Left
O,4,8,C
1,5,9,D
(A31
(A41
AIW2,AlV2 AlB2,AIC2 BIF2
BIH2
A1S2,A1U2 AID2,A1D2
Rear Devices
Left
Right
2,6,A,E
3,7,B,F
81Q2
81S2
A1T2,AIF2 AIE2,A1S2
AIN2,AIR2 AIJ2,A1G2
AIJ2
AIN2
(ECO-203)
See HIH Volume RIO, page MAP-I, Entry AB
2 Sync In check
AIW2,AIV2 AIB2,AlC2
AIN2
AIJ2
*
(ECO-204)
See HII1 Volume RIO, page MAP-I, Entry AB
AIW2,AIV2 AIB2,AIC2 81F2
BIQ2
3 DDC Bus In
BIH2
81S2
AIN2,A1U2 AIJ2,AI02
* parity check
A1T2
AIE2
(ECO-20S)
See I1II1 Volume RIO, page MAP-I, Entry AB
* A more detailed analysis can be obtained by using the MD to analyze the SC.
Figure 25.
SC=EOXX (XX= Format 8 Byte 10)
('
(
(
c
PIN 4519909 50 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
SC-FRU-37
SC=E1XX (XX= Format 8 Byte 11)
Controller Fault Log A (SC= ElXX)
Rear Devices
Controller
Front Devices
Right
Right
Left
Left
Al
A2
(A3)
(A4)
O,4,8,C
l,5,9,D
2,6,A,E
3,7,B,F
AIQ2,AIW2 AIH2,A1B2 COB2,BID2 ClB2,BlK2 C2B2,BIN2 C3B2,BIU2
0 DHPlO Delta
3E
Frequency check
A1V2,A1H2 A1C2,A1J2 BIF2,B1K2 BIH2,B102 BlQ2,B1U2 B1S2,B1H2
(ECD-31S)
See MIM Volume RIO, page MAP-I, Entry DF
AIQ2,AIN2 AIH2,AIJ2 BID2,BIF2 BIK2,BIH2 BIN2,B1Q2 BIU2,B1S2
1 DHPlO Non-Drive
3E
check
AIV2
AIC2
B1K2
B102
B1U2
BIH2
(ECD-316)
See MIM Volume RIO, page MAP-I, Entry DF
2 Read/Write
AIQ2,AIH2 AlH2,AIJ2 COB2,ClB2 CIB2,COB2 C2B2,C3B2 C3B2,C2B2
3E
Data Cable check
BIF2
B1H2
BIQ2
BlS2
(ECD-318)
See MIM Volume RIO, page MAP-I, Entry EC
AIQ2,AIV2 AlH2,AIC2 COB2,BIF2 ClB2,BlH2 C2B2,BIQ2 C3B2,B1S2
3 Data Valid check
AIH2
A1J2
(ECD-3I9)
See MIM Volume RIO, page MAP-I, Entry DF
AIQ2,AIN2 AlH2,AIJ2 BID2,BIF2 BIK2,BIH2 BIN2,B1Q2 BIU2,B1S2
4 DHPlO Failed to
3E
BIH2
lock check
AIV2
AIC2
BIK2
B102
B1U2
(ECD-320)
See MIM Volume RIO, page MAP-l, Entry DF
AIQ2,AIV2 AIH2,AIC2
5 DHPlO Multiple
Select check
AIN2
AIJ2
(ECD-321)
No further action - go to RMAP 340 on page REPAIR-13.
AIQ2,A1W2 AIH2,AIB2 COB2,BIF2 CIB2,BIH2 C2B2,BIQ2 C3B2,B1S2
6 No Read Data check
AIV2,AIH2 AIC2,AIJ2
(ECD-322)
See MIM Volume RIO, page MAP-I, Entry DF
3E A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: Fo'r details about symptom code ElXX, see page ECD-nnn in the ECM, Volume Rft.O.
Bit
Figure 26.
Description
SC=E1XX (XX= Format 8 Byte 11)
PIN 4519909 50 Sides
EC 476581 01Aug89
© Copyright IBM Corp, 1987, 1988
SC-FRU-38
,~-"-'\
"
/
/
"
SC=E2XX (XX= Format 8 Byte 12)
Controller Fault Log B (SC= E2XXl
Bit
0
Description
Controller
Al
A2
(A31
(A41
SERDES Control check AIV2.AIN2 AIC2,AlJ2
AIW2
*
Front Devices
Right
Left
O,4,8,C
l,5,9,D
AIB2
No further action -- go to RMAP 340 on page REPAIR-l3.
(ECO-325)
1
*
ClOCK/SERDES/ECC
AlN2.AIW2 AIJ2.AIB2
card check
(ECO-326)
No further action -- go to RMAP 340 on page REPAIR-l3.
2
SERDES Path check
*
*
AIW2.AlN2 AlB2.AIJ2
(ECO-328)
See MIM Volume RIO. page MAP-I. Entry AA
A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code E2XX, see page ECD-nnn in the
Figure 27.
C..-\
c
c
c
c
ECM~
Volume R40.
SC=E2XX (XX= Format 8 Byte 12)
SC=E3XX (XX= Format 8 Byte 13)
Controller Fault Log C (SC= E3XXl
Bit
1
3
4
*
6
7
*
Description
Controller
Al
A2
(A4)
(A31
Controller Sequencer AIR2,AIV2 AIG2.AIC2
Check-2 check
(ECO-329)
*
5
c
Rear Devices
Right
Left
2,6,A,E
3,7,B,F
Multiplexer Input
parity check
Rear Devices
Right
Left
2,6,A,E
3,7,B,F
No further action -- go toRMAP 340 on page REPAIR-l3.
AIU2,AlT2 AlD2,AlE2
AlV2
AIC2
No further action -- go to RMAP 340 on page REPAIR-13.
(ECO-330)
CDP Register 3
parity check
Front Devices
Right
Left
O,4,8,C
1,5,9,D
AIV2,A1U2 AIC2,A1D2
A1T2
(ECO-331)
AIE2
No further action
--
go to RMAP 340 on page REPAIR-l3.
AIV2,AIN2 AIC2,AIJ2
IOC Card Check-2
check
(ECO-332)
No further action -- go to RMAP 340 on page REPAIR-l3.
Precompensation
AIN2,AIV2 AlJ2,AIC2
check
(ECO-333)
No further action -- go to RMAP 340 on page REPAIR-l3.
Write Gap 3 Control AIV2,AIW2 AlC2,AlB2
check
(ECO-334)
No further action -- go to RMAP 340 on page REPAIR-l3.
A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code E3XX. see page ECD-nnn in the
Figure 28.
ECM~
Volume R40.
SC=E3XX (XX= Format 8 Byte 13)
PIN 4519909 50 Sides
EC 476581 01Aug89
~ Copyright IBM Corp. 1987, 1988
SC-FRU-39
SC=E5XX (XX= Format 8 Byte 15)
Controller Fault Log E (SC= E5XXl
Bit
0
I
2
3
4
5
6
Description
Controller
Al
A2
(A4)
(A31
DPS Array check
AlS2,AlF2 AIF2,A1S2
AIW2,AIR2 A182,AIG2
(ECD-338)
No further action
DPS Internal check
A1S2,AIR2 AIF2,AIG2
(ECD-340)
No further action
A1S2,AlF2 AIF2,A1S2
DPS Compare check
AIW2,AIR2 A182,AIG2
(ECD-34!)
No further action
AIF2,A1S2 A1S2,AIF2
DPS Controller To
Controller
connection check
(ECD-342)
No further action
DPS Storage Address A1S2,AIR2 AIF2,AIG2
Register check
(SAR check)
(ECD-343)
No further action
A1S2,AIF2 AIF2,A1S2
DPS Internal
Register check
AIW2,AIR2 A182,AIG2
(ECD-34ft)
No further action
DPS Alternate check A1S2,AIF2 AIF2,A1S2
(Wait/Lock Check)
AIW2,AIR2 A182,AIG2
(ECD-346)
No further action
Front Devices
Right
Left
l,5,9,D
O,4,8,C
Rear Devices
Right
Left
2,&,A,E
3,7,B,F
-- go to RMAP 340 on page REPAIR-13.
-- go to RMAP 340 on page REPAIR-13.
-- go to RMAP 340 on page REPAIR-13.
-- go to RMAP 340 on page REPAIR-13.
-- go to RMAP 340 on page REPAIR-13.
-- go to RMAP 340 on page REPAIR-13.
-- go to RMAP 340 on page REPAIR-13.
NOTE: For details about symptom code E5XX, see page ECD-nnn in the ECM, Volume RltO.
Figure 29.
SC=E5XX (XX= Format 8 Byte 15)
PIN 4519909 50 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
SC-FRU-40
SC=E6XX (XX= Format 8 Byte 16)
(
(
Controller Fault Log F (SC= E6XX)
Bit
Description
Controller
A1
A2
(A3)
(A4)
Front Devices
Right
O, ... ,8,C
1,5,9,D
J,..~ft
0
c
~.
c
c;
DTB Bus Out
AIW2,AIV2 AIB2,AIC2
parity check
'*
(ECO-3lt7)
See MIM Vol ume RlO, page
A1N2,AU/2 A1J2,A1B2
I DTB Bus In
parity check
A1U2,AIT2 A1D2,A1E2
'*
AIV2
AIC2
(ECO-3lt8)
See MIM Volume RIO, page
2 DTB Control check
AIW2
A1B2
(£CO-3lt9)
See MIM Volume RlO, page
3 Read/Write Gate DDC AlW2,AlN2 AlB2,AIJ2
Drivers check
AIV2
AIC2
(£CO-350)
See MIM Volume RIO, page
'* A more detailed analysis can be obtained by using the MD
NOTE: For details about symptom code E6XX, see page ECD-nnn
Rear Devices
Right
Left
2,6,A,E
3,7,B,F
Figure 30.
(
4[-
MAP-I, Entry AA
MAP-I, Entry AA
MAP-I, Entry AA
to analyze the SC.
in the £CM, Volume Rita.
SC=E6XX (XX= Format 8 Byte 16)
SC=E7XX (XX= Format 8 Byte 17)
Controller Fault Log G (SC= E7XXl
Bit
(
MAP-I, Entry AA
Description
Controller
Al
A2
(A3)
(A4)
Front Devices
Left
Right
1,5,9,D
O,4,8,C
Rear Devices
Right
Left
2,6,A,E
3,7,B,F
AlU2,AIT2 AID2,AIE2
Controller to CDP
Card check
AIV2
AlC2
'*
(£CO-353)
No further action -- go to RMAP 340 on page REPAIR-13.
2 Port Response check AIU2,AlT2 AID2,AlE2
'*
(£CO-357)
See MIM Volume RlO, page MAP-I, Entry CG
AIU2,AIT2 AID2,AIE2
7 Port Degate check
'*
(£CO-361)
No further action -- go to RMAP 340 on page REPAIR-l3.
'* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code E7XX, see page ECD-nnn in the ECM, Volume RItO.
0
Figure 31.
SC=E7XX (XX= Format 8 Byte 17)
(
(
(;
PIN 4519909 50 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
SC-FRU-41
c
SC=E8XX (XX= Format 8 Byte 18)
Register 'OF' Error Codes (SC= E801 through E813)
xx=
01
*
Description
Index not active
when expected
Front Devices
Rear Devices
Controller
Right
Right
Al
A2
Left
Left
(A4)
O,4,8,C
1,5,9,D
2,6,A,E
3,7,B,F
(A3l
AIN2,AIV2 AIJ2,AIC2 BIE2,BIF2 BIJ2,BIH2 BIP2,BIQ2 B1T2,BlS2
AIU2,AIT2 AID2.AIE2 BID2
BIU2
BIK2
BIN2
AIQ2
AIH2
(ECD-278)
No further action -- go to RMAP 340 on page REPAIR-13.
02 Index active when
AIN2,AIV2 AIJ2,AIC2 BIE2.BIF2 BIJ2,BIH2 BIP2.BIQ2 B1T2,B1S2
AIU2,AIT2 AID2,AIE2 BID2
BIU2
BIK2
BIN2
* not expected
AIQ2
AIH2
(ECD-279)
No further action -- go to RMAP 340 on page REPAIR-l3.
03 Segment Boundary not AIN2,A1V2 AIJ2,AIC2 Blf~,BlD2 B1J2,B1K2 BIP2,BIN2 B1T2,BlU2
AID2,AIE2 BIF2.BIK2 BIH2,B1D2 BIQ2,B1U2 BlS2.BlN2
* active when expected A1U2,A1T2
AIQ2
AIH2
(ECD-281)
No further action -- go to RMAP 340 on page REPAIR-13.
A1N2,A1V2 A1J2,A1C2 B1E2,B1D2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
04 Segment Boundary
active when not
A1U2,A1T2 AID2,AIE2 BIF2,BIK2 BIH2,B1D2 BIQ2,B1U2 BlS2,BlN2
* expected
AIQ2
AIH2
(ECD-282)
No further action -- go to RMAP 340 on page REPAIR-13.
05 Segment boundary was A1N2,AIV2 AIJ2,AIC2 BIE2,B1D2 BIJ2,B1K2 BIP2,BIN2 B1T2,BlU2
not active after
A1U2,A1T2 A1D2,A1E2 B1F2
BIQ2
B1S2
B1H2
* clock
started
AIQ2
AIH2
(ECD-28'f.)
No further action -- go to RMAP 340 on page REPAIR-13.
06 Address Mark or
AIN2,AIV2 AIJ2,AIC2 B1E2.BID2 BIJ2,B1K2 B1P2,BIN2 B1T2,BIU2
AIU2,AlT2 AID2.AIE2 BIF2
BIQ2
Index Not Active
BIH2
B1S2
* when expected
AIQ2
AIH2
(ECD-28S)
No further action -- go to RMAP 340 on page REPAIR-13.
07 Oriented not active AIV2,AIW2 AIC2,AIB2
when expected
(ECD-287)
No further action -- go to RMAP 340 on page REPAIR-13.
10 First Byte Ready
AIN2,AIV2 AIJ2,AIC2
active when not
(ECD-288)
expected
go to RMAP 340 on page REPAIR-13.
No further action
13 Special after Index AIV2,Al\'l2 AIC2.AIB2
command
No further action -- go to RMAP 340 on page REPAIR-13.
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code E8XX. see page ECD-nnn in the ECM, Volume R'f.O.
/~-"",
--
Figure 32 (Part 1 of 3).
SC=E8XX (XX= Format 8 Byte 18)
PIN 4519909 50 Sides
EC 476581 01Aug89
~ Copyright IBM Corp. 1987, 1988
SC-FRU-42
C~;
c
c
Register 'OF' Error Codes (SC= E814 through E85l)
(
XX=
Description
Controller
Al
A2
(Al)
(A4)
Front Devices
Left
Right
Op4 p8 pC
Ip5 p9 pD
Rear Devices
Right
Left
lp7 pBpF
2p6 pApE
14
(.
(
(-
(/
( -,
j
((
(
(
(
(
Address Mark active AlQ2,AIN2 AIH2,AIJ2
when not expected
AlV2
AIC2
(ECD-289)
No further action -- go to RMAP 340 on page REPAIR-I3.
15 Lock PLO To Servo
AlV2,AIQ2 AIC2,AIH2
not active when
expected
(ECD-290)
No further action -- go to RMAP 340 on page REPAIR-13.
17 Write Address Mark
AIN2,AlV2 AlJ2,AIC2
active when not
expected
(ECD-292)
No further action -- go to RMAP 340 on page REPAIR713.
18 Lock PLO To Data not A1V2,AIQ2 A1C2,AIH2
active when expected AIN2
AIJ2
(ECD-293)
No further action -- go to RMAP 340 on page REPAIR-I3.
19 'Any Check' and 'Any AlW2,AIN2 AIB2,AIJ2
Check Duplicated'
AIV2
AIC2
mismatch
(ECD-294)
No further action -- go to RMAP 340 on page REPAIR-13.
20 CDP Tag In At valid AIU2,AIT2 AID2,AIE2 BIF2,B1G2 BIH2,BIG2 BIQ2,BIR2 BlS2,BlR2
BIJ2
BlP2
B1T2
* when not expected AIN2,AIV2 AIJ2,AIC2 BIE2
(ECD-29S)
See MIM Volume RIO, page MAP-I, Entry CD
21 In Reset Allegiance AIU2,AIT2 AlD2,AIE2 BIF2,BIG2 BlH2,BlG2 BIQ2,BIR2 BlS2,BIR2
the device AIN2,AIV2 AIJ2,AIC2 BIE2
BlJ2
BIP2
B1T2
* command,
returned incorrect
(ECD-296)
address
See MIM Volum~ RIO, page MAP-I, Entry CD
22 CDP Tag In was not
AlU2,AlT2 AlD2,A1E2 B1F2,BIG2 BlH2,BIG2 BlQ2,BIR2 BlS2,BIR2
at Selected Null,
BlP2
B1T2
B1J2
* is now at Selected AIN2,A1V2 AIJ2,AIC2 BIE2
(ECD-298)
Null
See MIM Volume RIO, page MAP-I, Entry CD
51 Expected Device
BlE2,BIF2 BIJ2,BIH2 B1P2,BIQ2 B1T2,B1S2
Checkpoint Log value
did not occur
(ECD-299)
No further action -- go to RMAP 340 on page REPAIR-13.
BIE2,BIF2 BIJ2,BIH2 BIP2,BIQ2 B1T2,B1S2
52 Expected Device
Checkpoint Log value
did not occur
(ECD-299)
No further action -- go to RMAP 340 on page REPAIR-13.
53 Expected Device
B1E2,BIF2 B1J2,BIH2 BIP2,BIQ2 B1T2,B1S2
Checkpoint Log value
did not occur
(ECD-299)
No further action -- go to RMAP 340 on page REPAIR-13.
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code E8XX, see page ECD-nnn in the ECM, Volume R40.
Figure 32 (Part 2 of 3).
SC= E8XX (XX= Format 8 Byte 18)
PIN 4519909 50 Sides
EC 476581 01Aug89
(
(
(
(
(~
© Copyright IBM Corp. 1987. 19BB
SC-FRU-43
Register 'OF' Error
C~des
XX=
Description
55
Expected Device
Checkpoint Log value
did not occur
(ECD-299)
56
Pad in Progress
active when not
(ECD-299)
expected
(SC= E855 through E8FF)
Controller
A2
Al
(A3l
(A41
Rear Devices
Front Devices
Right
Right
Left
Left
2,6 .. A,E
3J>7,B,F
O.. 4 .. 8 .. C l .. 5,9,D
BlE2,BIF2 BlJ2,BlH2 BlP2,BIQ2 B1T2,B1S2
No further action -
go to RMAP 340 on page REPAIR-l3.
BIE2,BIF2 BIJ2,BIH2 BIP2.BIQ2 B1T2.B1S2
No further action -
go to RMAP 340 on page REPAIR-l3.
/"
'.
57
58
Expected Device
Checkpoint Log value
did not occur
(ECD-299)
Expected Device
Checkpoint Log value
did not occur
(ECD-299)
80- Undervoltage Active
FF ORed with code 00-7F
BlE2.BIF2 BlJ2.BIH2 BlP2.BIQ2 B1T2.B1S2
No further action -
'<-.
go to RMAP 340 on page REPAIR-l3.
BIE2.BIF2 BIJ2.BIH2 BIP2.BIQ2 B1T2.B1S2
No further action - go to RMAP 340 on page REPAIR-l3.
See corresponding entry for Symptom Code E800 - E87F.
Also see entry for Symptom Codes EFFF and DFFF.
NOTE: For details about symptom code E8XX. see page ECD-nnn in the
Figure 32 (Part 3 of 3).
ECM~
Volume
~-
.-/
R~O.
SC=E8XX (XX= Format 8 Byte 18)
\ ..
PIN 4519909 50 Sides
-"",
EC 476581 01Aug89
<0 Copyright IBM Corp. 1987. 1988
SC-FRU-44
-
(
(
(
(
SC=EBXX
XX Bits 0-3 = Byte 16 Bits 4-7
XX Bits 4,5 = Byte 17 Bits 5,6
XX Bits 6,7 = Byte 14 Bits 6,7
CDP Card Detected Error (SC= EBXX)
Bit
Description
Controller
A1
(A3)
A2
(A4)
Front Devices
Left
Right
O,4,8,C
1,5,9,D
Rear Devices
Left
Right
2,6,A,E
3,7,B,F
*
String Configuration A1U2,AIS2 AID2,AIF2 BIF2**
B1H2**
BIQ2**
B1S2**
Bits 0-3:
AIV2
AIC2
BIG2,B1E2 BIG2,BIJ2 BIR2,BIP2 BIR2,BIT2
Number of units that
B1D2
B1K2
BIN2
BIU2
are cabled into the
string. Set by IOC
See MIM Volume RIO, page MAP-I, Entry CD
card switches 1-4.
0,1 Configuration bits
Device Range OO-OF:
00 = A unit only
01 = A and 1 B unit
10 = A and 2 B units
11 = A and 3 B units
2,3 Configuration bits
Device Range IO-IF:
00 = A unit only
01 = A and 1 B unit
10 = A and 2 B units
11 = A and 3 B units
4,5 Substring Select
(bits 4 and 5)
10 = Substring 0
01 = Substring 1
00 = Invalid
11 = Invalid
6,7 CDP Port Select
(bits 6 and 7)
456 7
1 o 0 0 = Port 0
1 o 0 1 = Port 1
1 0 1 0 = Port 2
1 o 1 1 = Port 3
0 1 0 0 = Port 4
0 1 0 1 = Port 5
o 1 1 0 = Port 6
0 1 1 1 = Port 7
Note: String ID can be determined by Sense Byte 3.
** A defective card in one device can cause failures in
other devices on the same port. See MIM Volume RIO,
page MAP-I, Entry CD, before replacing cards.
The CDP port in error (port 0 through 7) is indicated
by the combination of bits 4 and 5 (Substring Select)
and bits 6 and 7 (CDP Port Select).
* A more
(
detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code EBXX, see page ECD-351 in the ECM, Volume R40.
Figure 33.
SC = EBXX
(
(
(
(
PIN 4519909 50 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
SC-FRU-4S
SC=ECXX (XX= Format 8 Bytes 11,12,13, 18)
DDC Hot Bits CSC= ECXX)
Bit
Description
i
XX= Bits always active
on, Format 8 sense
assembly
*
Controller
Al
A2
(A3)
CA41
AlV2,AlS2 AlC2,AlF2
AlU2,AlT2 AlD2,AlE2
AlN2,AIW2 AlJ2,A182
AIMft.,AIQ2 AILft.,AIH2
(£CO-257)
Front Devices
Right
Left
O,4,8,C
l,S,9,D
Rear Devices
Right
Left
2,6,A,E
3,7,B,F
No further action -- go to RMAP 340 on page REPAIR-I3.
NOTE: For details about symptom code ECXX, see page ECD-257 in the £CM, Volume Rft.O.
Figure 34.
SC=ECXX (XX= Format 8 Bytes 11,12,13,18)
SC=EDXX (XX= Format 8 Byte 09)
"
./
DDC Bus In (SC= EDOl through EDIO)
XX=
00
01
06
07
08
on
10
*
Description
Controller
Front Devices
Rear Devices
Right
Right
Al
A2
Left
Left
(A3)
l,5,9,D
2,6,A,E
3,7,B,F
(A41
O,4,8,C
An unexpected end-op AlW2,A1V2 AlB2,AlC2
AIN2,AIQ2 AIJ2,AIH2
response received
(£CO-260)
No further action -- go to RMAP 340 on page REPAIR-I3.
nnc command overrun AIW2
AIB2
(£CO-268)
No further action -- go to RMAP 340 on page REPAIR-13.
Invalid command code A1S2,AIV2 AlF2,AlC2
AIB2
AIW2
(£CO-270)
See MIM Volume R1O, page MAP-I, Entry AA
nnc data overrun
A1W2,A1N2 AlB2,AIJ2
AIV2
AIC2
(£CD-271)
See MIM Volume RIO, page MAP-I, Entry AA
HAR modifier overrun A1W2
AlB2
on set HAR oriented
(£CO-273)
No further action -- go to RMAP 340 on page REPAIR-I3.
Data missing in a
count area
Check 2 detected in
controller
For further analysis, go to the MIM page
MAP-I, Entry EN.
A1N2,AIW2
AlS2,AIV2
A1U2,A1T2
AIQ2,AIR2
AlJ2,AIB2
AIF2,AIC2
A1D2,Al£2
AIH2,AIG2
(£CO-274)
No further action -- go to RMAP 340 on page REPAIR-13.
A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code EnXX, see page ECD-nnn in the £CM, Volume Rft.O.
*
Figure 35 (Part 1 of 2).
SC= EDXX (XX= Format 8 Byte 09)
if "
!\tj
PIN 4519909 50 Sides
© Copyright
EC 476581 01Aug89
SC-FRU-46
IBM Corp. 1987, 1988
/
DDC Bus In (SC= EDII through EDlB)
XX=
11
*
('
13
*
16
17
(~
(,
18
19
*
1B
*
Description
Controller
Front Devices
Rear Devices
Right
A2
Right
Al
Left
Left
(A3)
(A4)
O,4,8,C
1,5,9,D
2,6,A,E
3,7,B,F
Device Check 1 on
A1U2,AIT2 A102,AIE2 BIF2,BIG2 BlH2,BlG2 BIQ2,BIR2 B1S2,BIR2
AIN2,AIV2 AIJ2,AIC2 BlE2,BlD2 BIJ2,BlK2 BIP2,BIN2 B1T2,BlU2
deselection
AIS2
AIF2
(ECO-275)
See MIM Volume RIO, page MAP-1, Entry CD
CDP Hang condition
AIU2,AIT2 AID2,AIE2 BIF2,BlG2 BIH2,BIG2 BIQ2,BIR2jB1S2,BIR2
caused by late or
AlN2,AlV2 AIJ2,AlC2 BlE2,BlD2 BlJ2,BlK2 BIP2,BlN2 BIT2,BlU2
missing response
A1S2
AIF2
(ECO-305)
See MIM Volume RIO, page MAP-I, Entry CD
Data check on start AIN2,AIV2IAIJ2,AIC21
I
I
I
read/write
(ECO-307)
No further action - go to RMAP 340 on page REPAIR-l3.
Any check, but
AIV2,AIN2 AIC2,AIJ2
controller collected AIQ2,AIW2 AIH2,A1B2
no status
(ECO-308)
No further action - go to RMAP 340 on page REPAIR-13.
Sync Out/In Tag
AlW2,AIN2 AIB2,AIJ2
counts not equal
AIV2,A1R2 A1C2,AZG2
(ECO-31O)
No further action - go to RMAP 340 on page REPAIR-l3.
Device dropped to
A1U2,A1T2 A102,A1E2 BIF2,BIG2 BIH2,BIG2 BIQ2,BIR2 B1S2,BIR2
null
A1N2,AZV2 AZJ2,A1C2 BIE2,BID2 BIJ2,BIK2 BIP2,BIN2 B1T2,BlU2
AIF2
AlS2
(ECO-311)
See MIM Volume RIO, page MAP-l, Entry CD
Index found during
AIU2,AIT2 AID2,AIE2 BIE2,BIF2 BIJ2,BIH2 BIP2,BIQ2 BIT2,B1S2
defect skip in HA
AIN2,AIQ2 AIJ2,AIH2 B102
BIK2
BIN2
B1U2
(ECO-312)
No further action - go to RMAP 340 on page REPAIR-l3.
*
A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code EDXX, see page ECD-nnn in the ECM 1 Volume R40.
Figure 35 (Part 2 of 2).
SC= EDXX (XX= Format 8 Byte 09)
(
(:
(
(
(
(-
PIN 4519909 50 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 19B7, 19BB
SC-FRU-47
SC=EEXX (XX= Format 8 Byte 14)
Controller Fault Log D (SC= EEXXl
Bit
Description
contrrller
Al
A2
(A31
(A4)
1
Front Devices
Left
Right
O,4,8,C
1,5,9,D
Rear Devices
Left
Right
2,6,A,E
3,7,B,F
T
Device Check I:
(bits O. 1. 2. 3)
indicates the device
on the port in error
Note: Device ID can be determined by
sense byte 4, bits 3 through 7.
0 Device 0.4.8.C Check
1 Device 1.5.9.D Check
2 Device 2.6.A.E Check
3 Device 3.7.B.F Check
4.5 Isolation 0.1:
A1U2.A1T~IA1D2.A1E2 B1F2,B1H2IB1H2,B1F2 B1Q2.B1S2IB1S2,B1Q2
(bits 4 and 5)
AIV2,AlS2 AIC2,AIF2 8162
8162,
81R2
BIR2
OO=DPS Tie Break Chk
01=Clock Check
See MIH Volume RIO. page MAP-I. Entry CD
IO=Port Card Check
11=CDP Interface Chk
6,7 CDP Port Select:
Note: String ID can be determined by sense byte 3.
(bits 6 and 7)
Substring 0:
The CDP port in error (port 0 through 7) is indicated
Byte 17. Bi t ' 5 = 1
by the combination of sense byte 17. bits 5 and 6, and
Devices 00 - OF
the XX portion of symptom code EEXX, bits 6 and 7.
00 = Port 0
01 = Port 1
10 = Port 2
11 = Port 3
6,7 CDP Port Select:
(bits 6 and 7)
Substring 1:
Byte 17. Bit 6 = 1
Devices 10 - ' IF
00 = Port 4
01 = Port 5
10 = Port 6
11 = Port 7
* A more detailed analysis can be obtained by using the MD to analyze the SC.
NOTE: For details about symptom code EEXX, see page ECD-335 in the ECM, Volume R40.
*
,I
,/
Figure 36.
SC=EEXX (XX= Format 8 Byte 14)
,
MASTER INDEX-6
('
(
(
(','
(
;.
'\
./
EC 'See ECM index
EO 'See ECM index
EE 'See ECM index
EF 'See ECM index
F1 'See ECM index
F2 'See ECM index
F3 'See ECM index
F4 'See ECM index
F6 'See ECM index
F8 'See ECM index
F9 'See ECM index
FC 'See ECM index
checkpoint error 'See MIM index
checkpoint error reporting and collection 'See
MIM index
checkpoint error, reporting and collection 'See
MIM index
checkpoint freeze detected not active during
diagnostic command (8B) 'See ECM index
checkpoint log
checks A7 through B4
SC = 95A7 through 95B4 SC-FRU-12
checks BS through co
SC = 95BS through 9SCO SC-FRU-13
checks CB through 06
SC = 95CB through 9506 SC-FRU-15
checks C1 through CA
SC = 95C1 through 9SCA SC-FRU-14
checks 07 through F2
SC
9507 through 95F2 SC-FRU-16
checks F3 through FC
SC = 95F3 through 9SFC SC-FRU-17
checks 10 through 3A
SC = 9510 through 953A SC-FRU-6
checks 3B through 4F
SC = 953B through 954F SC-FRU-6
checks 50 through 6A
SC = 9550 through 956A SC-FRU-8
checks SO through 5C
SC = 9550 through 955C SC-FRU-7
checks 6B through 7A
SC = 956B through 957 A SC-FRU-9
checks 7B through 88
SC = 957B through 9588 SC-FRU-10
checks 8A through A6
SC = 958A through 9SA6 SC-FRU-11
SC = 90XX SC-FRU-23
10 through FF SC-FRU-23
checkpoint log byte 15 'See MSM index
checks - service 'See MIM index
capacitor 'See MIM index
=
(
(\
C'
PIN 4519911 39 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
diode 'See MIM index
rectifier 'See MIM index
checks, read/write channel board voltages 'See
MIM index
checks, safety 'See Index of INST
checks, voltage 'See MIM index
circuit card connector assignments 'See MIM
index
circuit protectors INTRO-35
clock check analysis procedure MAP 'See MIM
index
clock LEOs INTRO-30
clock ~>,nchronization
to cell boundary 'See MSM index
to sync byte 'See MSM index
clock/SEROES/ECC card check 'See ECM index
coil, magnet 'See MIM index
command gate missing during data check
recovery 'See ECM index
command gate stuck active 'See ECM index
command transfers during track processing 'See
MSM index
commands 'See MSM index
compare device status 'See Index of INST
complete did not occur for power sequence 'See
MIM index
complete relay 'See MIM index
complete string removal procedure 'See Index of
INST
condition code 3 - console message ISOLATE-42
conditions, airflow error 'See MIM index
conditions, drive motor error 'See MIM index
sense bit indications
format 1, byte 10, bits 3 through 7 'See MIM
index
format 1, byte 11, bit 3 'See MIM index
conditions, drive motor switch error 'See MIM
index
configuration MO-73
configuration data MO-75
configuration response verification 'See Index of
INST
configuration switch
4-path, nondisruptive 'See Index of INST
configuration, device numbering (PSG Vol. R05,
entry G) 'See MIM index
configuration, string address switches
(2-path) REPAIR-18
configuration, string address switches
(4-path) REPAIR-19
configurations
MASTER INDEX-7
subsystem INTRO-3
3380-J string and port INTRO-14
3380-J 2-path string INTRO-14
3380-JK port INTRO-17
3380-JK 4-path string INTRO-15
3990 subsystem INTRO-14, INTRO-17
configurations, subsystem INTRO-12
configurations, 3380-JK string 'See Index of INST
connect the ac line cord 'See Index of INST
connection check alert 'See ECM index
connection check alert and power status
SC = D1XX SC-FRU-26
connection check alert and power status, byte
11 'See MSM index
connection test MD-17, MD-64
connector assignments, circuit card 'See MIM
index
connector, MD 'See MIM index
connectors
maintenance device (MD) INTRO-11, INTRO-25
connectors, HDA - front drive 'See MIM index
connectors, HDA - rear drive 'See MIM index
connectors, tailgate 'See MIM index
console message - condition code 1 ISOLATE-38
console message - condition code 3 ISOLATE-42
console message - intervention
required ISOLATE-35
console message - missing channel end and device
end IS 0 LATE-46
console message - volume fenced from storage
path ISOLATE-51
console message - write inhibited ISOLATE-49
console messages - miscellaneous ISOLATE-53
console messages and EREP INTRO-37
console messages, IMAP reference
table ISOLATE-34
contamination, voice-coil 'See MIM index
control cable checkout procedure 'See MIM index
control card switch settings REPAIR-20
control interface 'See MSM index
control mode 'See MSM index
fencing INTRO-6
line descriptions 'See MSM index
transfer mode 'See MSM index
control interface analysis 'See MIM index
control interface analysis MAP 'See MIM index
control interface check-2 MAP 'See MIM index
control interface test isolation codes MAP 'See
MIM index
control panel, single actuator
maintenance INTRO-34, 'See MIM index
PIN 4519911 39 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
control panel, single actuator maintenance
(SAM) 'See MIM index
control, power sequence 'See MIM index
control, power summary diagram 'See MIM index
controller INTRO-11, MD-55
Ax connection tests ready MD-67
fault log A SC-FRU-38
fault log B SC-FRU-39
fault log C SC-FRU-39
fault log D SC-FRU-48
fault log E SC-FRU-40
fault log F SC-FRU-41
fault log G SC-FRU-41
o check 1 status SC-FRU-27, SC-FRU-28
o register 1F SC-FRU-31
1 check 1 status SC-FRU-29, SC-FRU-30
controller A1 CDP 0 and 1 pin connections 'See
MIM index
controller A1 CDP 2 and 3 pin connections 'See
MIM index
controller A 1 CDP 4 and 5 pin connections 'See
MIM il1dex
controller A1 CDP 6 and 7 pin connections 'See
MIM index
controllerA2 CDP 0 and 1 pin connections 'See
MIM index
controller A2 CDP 2 and 3 pin connections 'See
MIM index
controller A2 CDP 4 and 5 pin connections 'See
MIM index
controller A2 CDP 6 and 7 pin connections 'See
MIM index
controller A3 CDP 0 and 1 pin connections 'See
MIM index
controller A3 CDP 2 and 3 pin connections 'See
MIM index
controller A3 CDP 4 and 5 pin connections 'See
MIM index
controller A3 CDP 6 and 7 pin connections 'See
MIM index
controller A4 CDP 0 and 1 pin connections 'See
MIM index
controller A4 CDP 2 and 3 pin connections 'See
MIM index
controller A4 CDP 4 and 5 pin connections 'See
MIM index
controller A4 6 and 7 pin connections 'See MIM
index
controller check alert and power status
SC = DEXX SC-FRU-34
controller checkout, post-repair REPAIR-11
MASTER INDEX-8
"'- ... ;:0/
'"
i
..7
~
\...
/
«
(~'
c
c
c
c
c
c
controller clock check 'See ECM index
controller command active with a OPS command
on the DOC bus out 'See ECM index
controller command active without DOC tag out
being in a command gate state 'See ECM index
controller command, OPS command or selected
stuck active 'See ECM index
controller connection test MO-66
controller dc power safety inspection, models AJ4
and AK4 'See MIM index
controller diagnostics ISOLATE-9
controller failed to set or reset long busy
latch 'See ECM index
controller fault log A
SC= E1XX SC-FRU-38
controller fault log A, byte 11 'See MSM index
controller fault log B
SC = E2XX SC-FRU-39
controller fault log B, byte 12 'See MSM index
controller fault log C
SC = E3XX SC-FRU-39
controller fault log C, byte 13 'See MSM index
controller fault log 0
SC = EEXX SC-FRU-48
controller fault log 0, byte 14 'See MSM index
controller fault log E
SC E5XX SC-FRU-40
controller fault log E, byte 15 'See MSM index
controller fault log F
SC = E6XX SC-FRU-41
controller fault log F, byte 16 'See MSM index
controller fault log G
SC = E7XX SC-FRU-41
controller fault log G, byte 17 'See MSM index
controller functions 'See MSM index
controller gate DOC drivers check 'See ECM
index
controller not available on disconnected command
chain 'See ECM index
controller not connected correctly
not connected correctly ISOLATE-55
controller not ready ISOLATE-32
controller path control 'See MSM index
controller sequencer function 'See MSM index
control function 'See MSM index
monitor function 'See MSM index
controller power on/off 'See Index of INST,
REPAIR-27
controller power on/off switches INTRO-8,
INTRO-22, 'See MIM index
controller power supplies 'See Index of INST
=
c~
(~
(
c
(
(
PIN 4519911 39 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
controller power supply 'See MIM index
controller power supply (A1/A2) 'See MIM index
controller power supply circuit
protectors INTRO-32
controller power supply CPs INTRO-32
controller ready LEOs INTRO-8, INTRO-22, 'See
MIM index, 'See Index of INST
controller selection 'See MSM index
controller sequencer and/or IOC register
failure 'See ECM index
controller sequencer check 'See ECM index
controller sequencer check-2 'See ECM index
controller sequencer counter failed 'See ECM
index
controller sequencer detected check-1
errors--string 0
SC = 0602 through 0636 SC-FRU-31
SC = 0670 through 06CE SC-FRU-31
controller sequencer detected check-1
errors--string 1
.
SC = 0802 through 0870 SC-FRU-33
SC = 0872 through 08CE SC-FRU-33
controller sequencer microcode detected
check-1 'See ECM index
controller sequencer microcode detected
error 'See ECM index
controller sequencer transfers control 'See MSM
index
controller string 0 check-1 status 1
SC = 02XX SC-FRU-27
controller string 0 check-1 status 2
SC = 03XX SC-FRU-28
controller string 1 check-1 status 1
SC
04XX SC-FRU-29
controller string 1 check-1 status 2
SC = 05XX SC-FRU-30
controller to COP card check 'See ECM index
controller under voltage 'See ECM index
controller voltage regulator card 'See MIM index
controller voltage regulator card LEOs INTRO-32
controller 0 or 1 check-1 status, byte 12 or 14 'See
MSM index
controller 0 or 1 check-1 status, byte 13 or 15 'See
MSM index
controller 0/1 connection check alert 'See ECM
index
controller 0/1 OPS unconditional reserve
release 'See ECM index
controller-to-device port interface check
procedure 'See MIM index
=
MASTER INDEX-9
/'
controller/drive power-on sequence chart 'See
MIM index
controller, A gate 'See MIM index
controller, power-on sequence chart 'See MIM
index
controller, select MD-52
convenience outlet 'See IV'IM index
convenience outlet distribution 'See MIM index
front 'See MIM index
rear 'See MIM index
conversion of a joined 2-path configuration to
4-path 'See Index of INST
conversion to 4-path procedure, start 'See Index
of INST
conversion to 4-path, summary 'See Index of INST
conversion 2-path to 4-path feature 'See Index of
INST
corrective action determination with sense
data START-17
corrective action with EREP START-8
count area format 'See MSM index
cover alignment 'See Index of INST
covers, A and B units 'See MIM index
CP failure analysis, tripped 'See MIM index
CPs, controller power supply INTRO-32
CPs, 24 Vdc panel INTRO-32
CP410 location 'See Index of INST
cross reference
entries from other MLMs MLX-2
exits to other MLMs MLX-3
MLM MLX-1
CTL-I
cable and terminator connect 'See Index of
INST
cable connect to 3880., 2-path, 2nd string 'See
Index of INST
cable connect to 3990, 2-path 'See Index of
INST
cable connect to 3990, 2-path, 2nd string 'See
Index of INST
cable connect to 4-path 'See Index of INST
cable routing, 2-path 'See Index of INST
cable routing, 4-path 'See Index of INST
CTL-I cable connect to 3880, 2-path 'See Index
of INST
second string cable connecting to 4-path 'See
Index of INST
second string connecting to 2-path,
4-path 'See Index of INST
CTL-I analysis MAP 'See MIM index
CTL-I check 1 'See MIM index
PIN 4519911 39 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
CTL-I check 1 for 3990 MAP 'See MIM index
CTL-I check 2 for 3990 MAP 'See MIM index
CTL-I check-2 MAP 'See MIM index
CTL-I test isolation codes MAP 'See MIM index
CTL-I, recable the 'See Index of INST
cylinder address miscompare 'See ECM index
cylinder seek routine 8A MAP 'See MIM index
cylinders INTRO-26
\--- /
DASD
data transfer summary PROG-11
event history report PROG-20
informational messages PROG-10
symptom code summary PROG-16
DASD string summary PROG-8
DASD subsystem exception report PROG-4
data and fill characters
reading 'See MSM index
writing 'See MSM index
data capacity INTRO-26
data check 'See ECM index
data check - cards and HDA 'See MIM index
data check - define failure boundary 'See MIM
index
data check - drive or device 'See MIM index
repair strategy 'See MIM index
data check analysis 'See MIM index
data check detection and correction INTRO-30
data check on start read/write 'See ECM index
data check recovery routine, expected device
checkpoint OK or failed 'See ECM index
data check recovery routine, pad-in-progress active
when not expected 'See ECM index
data checks
message SC-FRU-24, SC-FRU-49
SC = 40YY through 47YY SC-FRU-2
data checks on customer tracks 'See MIM index
data checks on the CE tracks 'See MIM index
data checks, diagnostics for ISOLATE-16
data checks, planning the repair action
for START-23
data detector check 'See ECM index
data tracks INTRO-26
data transfer summary PROG-1, PROG-11
data valid check 'See ECM index
DCP BAT failure 'See ECM index
MASTER INDEX-10
/
'
\--,
/
/
\'-l./
c
(
(
(
-,\
j
c
(. ' j
,\
(~\
C:,
_/
( "'
/
(
'
..
{~.
DCP busy bit active at the end of the offset 'See
ECM index
DCP busy bit active before issuing offset
command 'See ECM index
DCP busy bit active before issuing rezero
command 'See ECM index
DCP busy bit active before issuing search
command 'See ECM index
DCP busy bit active either at the beginning or end
of the seek 'See ECM index
DCP busy bit detected inactive during rezero 'See
ECM index
DCP busy bit goes inactive during seek 'See ECM
index
DCP busy bit not active at the beginning of offset
command 'See ECM index
DCP busy bit not active at tile beginning of seek
command 'See ECM index
DCP busy detected during data check
recovery 'See ECM index
DCP check detected during data check
recovery 'See ECM index
DCP command bit active when not expected during
compress routine 'See ECM index
DCP command bit active when not expected during
decompress 'See ECM index
DCP detected checksum error during DCP IML or
DCP SAT 'See ECM index
DCP detected error after park prior to starting the
motor 'See ECM index
DCP detected error at the beginning of motor
start 'See ECM index
DCP detected error doing a decompress of the
motor start 'See ECM index
DCP detected error during a seek of crash stop
recovery 'See ECM index
DCP detected error or hang during the sticky crash
stop recovery 'See ECM index
DCP detected error or it failed to start the rezero
·command 'See ECM index
DCP detected error or servo failure during power
on 'See ECM index
DCP detected error or timeout during a compress
routine 'See ECM index
DCP detected error or timeout during a
decompress 'See ECM index
DCP detected external error after any sequencer to
DCP command 'See ECM index
DCP detected failure during offset 'See ECM index
DCP detected failure during rezero routine 'See
ECM index
PIN 4519911 39 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
DCP detected failure during seek 'See ECM index
DCP detected failure during the handling of sweep
commands 'See ECM index
DCP detected internal error after any sequencer to
DCP command 'See ECM index
DCP error bit detected active during a 'go around'
search 'See ECM index
DCP error bit detected active during offset 'See
ECM index
DCP error bit detected active during seek 'See
ECM index
DCP error detected during data check
recovery 'See ECM index
DCP fails to accept offset command or DCP hang
condition detected 'See ECM index
DCP fails to accept seek command or DCP hang
condition detected 'See ECM index
DCP found busy with not write inhibit 'See ECM
index
DCP half track bit active at the end of the
rezero 'See ECM index
DCP half track bit in wrong state during seek 'See
ECM index
DCP half track bit not active when expected during
rezero routine 'See ECM index
DCP hang after any servo command 'See ECM
index
DCP IML failure 'See ECM index
DCP in an error state or wrong command issued
during safe routine 'See ECM index
DCP offset bit in wrong state during offset 'See
ECM index
DCP sample interrupt signal not changing states
during idle routine 'See ECM index
DCP sample interrupt signal out of specification
during idle 'See ECM index
DCP servo check 2 bit active during rezero 'See
ECM index
DCP servo check 2 bit detected active during a 'go
around' s.earch 'See ECM index
DCP servo write inhibit bit detected inactive during
rezero 'See ECM index
DCP servo write inhibit or write inhibit detected
active at the end of the rezero 'See ECM index
DCP timeout doing a compress during the motor
start routine 'See ECM index
DCP timeout during the sticky crash stop
recovery 'See ECM index
DCP timeout failure during crash stop
recovery 'See ECM index
DCP til'Yleout failure during sweep 'See ECM index
MASTER INDEX-11
DCP track following bit active too soon during
seek 'See ECM index
DCP track following bit active when not expected
during rezero routine 'See ECM index
DCP track following bit in wrong state during offset
routine 'See ECM index
DOC
bus in SC-FRU-46
CTL-I test failed MLX-2
hot bits SC-FRU-22, SC-FRU-46
interface check MLX-2
interface check FRUs MLX-2
DOC Bus In
SC = ED01 through ED10 SC-FRU-46
SC= ED11 through ED1B SC-FRU-46
DOC bus in parity check 'See ECM index
DOC bus out parity check 'See ECM index
DOC command overrun 'See ECM index
DOC data overrun 'See ECM index
DOC hot bits
SC = 9CXX SC-FRU-22
SC = ECXX SC-FRU-46
DOC interface
See control interface
DOC tag-out sequence check 'See ECM index
DDC/DTB card switch settings
2-path REPAIR-18
4-path REPAIR-19
definitions 'See Index of INST
deselection 'See MSM index
detection and correction, data check INTRO-30
detector, air flow 'See MIM index
determine corrective action with EREP START-8
determining the failing FRUs START-34
determining the failing ports (EBxx)START-33
determining the probable failing unit START-16
device
check-2 status SC-FRU-3
level selection INTRO-6, INTRO-18
power status SC-FRU-3
status MD-73. MD-75, MD-76
status test MD-64
test MD-15
tests MD-64
type MD-77
device address
device address range 'See Index of INST
device addressing switches 'See Index of INST
device cards, replacement procedures REPAIR-2
device characteristics summary INTRO-26
device check active 'See ECM index
PIN 451991139 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
device check 1, selected 'See ECM index
device check 1, unselected 'See ECM index
device check-1 on selection 'See ECM index
device check-2 'See ECM index
device check-2 or set sector incomplete 'See ECM
index
device check-2 status
SC = 91XX SC-FRU-3
device check-2 status byte 11 'See MSM index
device does not respond to selection 'See ECM
index
device dropped to null 'See ECM index
device enable/disable switch INTRO-8, INTRO-22
device enable/disable switch test 'See Index of
INST
device enable/disable switches 'See MIM index
device error 'See ECM index
device functions 'See MSM index
device interface 'See MSM index
control part 'See MSM index
controller ports 'See MSM index
line descriptions 'See MSM index
read/write part 'See MSM index
tag in 'See MSM index
tag out 'See MSM index
device level selection INTRO-6, INTRO-18
device logic power supply 'See Index of INST
device logic switches 'See MIM index
failure modes 'See MIM index
device not ready ISOLATE-27
device not ready to do customer work 'See ECM
index
device numbering (see PSG Vol. R05, entry
G) 'See MIM index
device power 'See Index of INST
device power on/off REPAIR-29
device power status
SC = 90XX SC-FRU-3
device power status byte 10 'See MSM index
device ready LEOs INTRO-8, INTRO-23, 'See MIM
index, 'See Index of INST
device selected dropped during data check
recovery 'See ECM index
device selection 'See MSM index
device sequence status LED INTRO-31
device sequencer 'See MSM index
device sequencer check 'See ECM index
device sequencer status LED off ISO.LATE-31
device status .MD-76
device status routine, run 'See Index of INST
device status test 'See Index of INST, MD-26
MASTER INDEX-12
(\
{
\
,/
(
" "
/'
()
c
c
c
(~
(
(
Device status test, determining failing devices and
controllers ISOLATE-18
device status 1 not as expected 'See ECM index
device status 1, byte 19 'See MSM index
device status 2, byte 20 'See MSM index
device status-1 has an invalid combination of
bits 'See ECM index
device status, compare 'See Index of INST
device switch enabled LED off ISOLATE-30
device type MD-77
device under voltage 'See ECM index
device undervoltage
SC = 9FFF SC-FRU-25
DHPLO delta frequency check 'See ECM index
DHPLO failed to lock check 'See ECM index
DHPLO multiple select check 'See ECM index
DHPLO non-drive check 'See ECM index
diagnostic
menu option 5 MD-19
diagnostic aids MD-82
diagnostic cylinders INTRO-27, INTRO-28
diagnostic D3 error stops MAP 'See MIM index
diagnostic error stops MAP 'See MIM index
diagnostic options, select MD-10
diagnostic routine descriptions
routine CO - controller test 'See MSM index
routine D3 - DDC tests 'See MSM index
routine 80 - controller tests 'See MSM index
routine 81 - device logic tests 'See MSM
index
routine 82 - servo tests 'See MSM index
routine 83 - read/write data transfer tests 'See
MSM index
routine 84 - read/write function tests 'See
MSM index
routine 85 - seek exercifer 'See MSM index
routine 9C - CE track repair 'See MSM index
routine 9D - DPS array display/dump
utility 'See MSM index
routine 9E - patch diagnostic microcode 'See
MSM index
routine 9F - set/reset/sense CE mode 'See
MSM index
routine 90 - DPS test 'See MSM index
routine 91 - home address scan 'See MSM
index
routine 93 - DDC tests 'See MSM index
routine 96 - device status test 'See MSM index
routine 97 - sense utility 'See MSM index
routine 98 - error log search 'See MSM index
routine 99 - motor start test 'See MSM index
PIN 4519911 39 Sides
4:l Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
diagnostic run procedure
controller 'See Index of INST
device 'See Index of INST
diagnostic runtime 'See Index of INST
diagnostic, invalid command on the DDC bus out
lines 'See ECM index
diagnostic, invalid diagnostic command or
subcommand received 'See ECM index
diagnostic, invalid extended subcommand
received 'See ECM index
diagnostic, invalid tag state on the DDC tag out
lines 'See ECM index
diagnostic, lock PLO to servo not active when
expected 'See ECM index
diagnostic, sequencer counter failure 'See ECM
index
diagnostic, sUb-routine EO failed during diagnostic
X'80' 'See ECM index
diagnostics
main menu INTRO-38
menu option A MD-37
menu option a MD-11
menu option 1 MD-13
menu option 2 MD-14
menu option 3 MD-15
menu option 4 MD-17
menu option 6 MD-24
menu option 7 MD-26, MD-73, MD-75, MD-76,
MD-77
menu option 8 MD-32
menu option 9 MD-34
running INTRO-38
running with 2 MDs 'See Index of INST
tests MD-2
diagnostics for data checks ISOLATE-16
diagnostics, controller ISOLATE-9
diagnostics, normal sequence ISOLATE-12
diagnostics, running ISOLATE-8
diagram, main menus MD-2
diagram, power control summary 'See MIM index
diagrams, power supply 'See MIM index
diagrams, subsystem with 3880, mod 3 or mod
23 'See Index of INST
diagrams, subsystem with 3990 'See Index of INST
differences between 3380-JK and 3380-DE
AC distribution 'See Index of INST
actuator swapped label 'See Index of INST
attachment to a 3880 'See Index of INST
attachment to a 3990, 2-path 'See Index of
INST
MASTER INDEX-13
attachment to a 3990, 4-path 'See Index of
INST
back panel 'See Index of INST
baseplate shipping locks 'See Index of INST
cable labels 'See Index of INST
device address range 'See Index of INST
device addressing switches 'See Index of INST
device logic power supply 'See Index of INST
device power 'See Index of INST
digital servo 'See Index of INST
end covers 'See Index of INST
fixed A and B gates 'See Index of INST
format 5 'See Index of INST
format 9 'See Index of INST
frame designations 'See Index of INST
front and rear covers 'See Index of INST
gate thermals 'See Index of INST
interframe connector gate 'See Index of INST
jack pads 'See Index of INST
multiburst ECC 'See Index of INST
nondisruptive installation of a second 4-path
string 'See Index of INST
nondisruptive installations of B units 'See
Index of INST
operator panel 'See Index of INST
string address switch 'See Index of INST
string IDs 'See Index of INST
tailgate 'See Index of INST
UEPO control 'See Index of INST
undervoltage detection 'See Index of INST
2-path or 4-path configurations 'See Index of
INST
differential pressure check, HDA manifold 'See
MIM index
digital servo 'See Index of INST
diode check 'See MIM index
disable
switch, device enable/ INTRO-8, INTRO-22
switches, logic enable/ INTRO-35
disable switch detected active during any search
command 'See ECM index
disconnect, mainline power cable procedure 'See
MIM index
disk layout INTRO-29
disk surface organization 'See MSM index
diskette compatibility MD-d3
diskette ordering procedure PRE-2
display or analyze next error MD-81
distribution 'See MIM index
+ 24 Vdc 'See MIM index
convenience outlet 'See MIM index
PIN 451991139 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
50 hertz 'See MIM index
60 hertz 'See MIM index
distribution, +5 Vdc from +24 Vdc 'See MIM
index
distribution, convenience outlet 'See MIM index
DLS (r .... vice level selection) INTRO-6, INTRO-18
documentation ordering procedure PRE-1,
INTRO-39
DPS
description INTRO-6
DPS (3990)
description INTRO-18
DPS alternate check (wait/lock check) 'See ECM
indexc
DPS array check 'See ECM index
DPS array not in.itialized 'See ECM index
DPS cleanup checks on channel or system
reset 'See ECM index
DPS command active with a noo-DPS command on
the DOC bus out lines 'See ECM index
DPS compare check 'See ECM index
DPS controller-to-controller connection
check 'See ECM index
DPS functions 'See MSM index
DPS internal check 'See ECM index
DPS internal register check 'See ECM index
DPS storage address register check 'See ECM
index
DPS test MD-14
DPS tests cannot be run MD-67
DPS, reason for 'See MSM index
drive 'See MSM index
drive ac power assembly check, 50 Hz, except
Japan 'See MIM index
drive ac power assembly check, 60 Hz (50 Hz
Japan) 'See MIM index
drive ac 60 hertz and 50 hertz Japan 'See MIM
index
drive actuator power supply 'See MIM index
drive belt resistance check 'See MIM index
drive belt resistance range 'See MIM index
drive belt, HDA 'See MIM index
drive motor 'See MIM index
drive motor brake 'See MIM index, 'See MSM
index
drive motor error conditions 'See MIM index
drive power problem analysis 'See MIM index
sense bit indications 'See MIM index
sense byte 10 analysis 'See MIM index
sense byte 11 analysis 'See MIM index
(,/
-~,
"-.
,'
.~./
MASTER INDEX-14
(
c
c
c
c
C',
drive motor slowing down during idle 'See ECM
index
drive motor slowing down during sweep 'See
ECM index
drive motor slowing down or checks active 'See
ECM index
drive motor switch error conditions 'See MIM
index
drive motor switch error conditions MAP 'See
MIM index
Drive motor switch momentarily off during search
or idle routine 'See ECM index
drive motor switch sensed off 'See ECM index
drive motor switch sensed on at the beginning of
diagnostic routine 99 'See ECM index
drive motor switch sensed on at the end of
diagnostic routine 99 'See ECM index
drive motor/brake power sequence 'See MIM
index
drive motor/brake sensing 'See MIM index
drive motors, HDA 'See MIM index
drive padding 'See MSM index
drive power on/off REPAIR-28
drive power-on/off, front 'See MIM index
drive power-on/off, rear 'See MIM index·
drive to CDP card check 'See ECM index
drive, S gate 'See MIM index
drive, power-on sequence chart 'See MIM index
DTS bus in parity check 'See ECM index
DTS bus out parity check 'See ECM index
DTS control check 'See ECM index
DTI/XES register SC-FRU-35, SC-FRU-37
SC = EOXX SC-FRU-37
durations, installation 'See Index of INST
during device power on, controller failed to take
control 'See ECM index
dynamic path selection, (see DPS) INTRO-6,
INTRO-18
/
(
ESXX error 'See ECM index
ECC bytes
reading from disk 'See MSM index
writing 'See MSM index
ECC status, byte 16 and 17, format 4 'See MSM
index
ECXX.bits always active on, format 8 sense
assembly 'See ECM index
c
PIN 4519911 39 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
Either a DCP detected error or the check-2 reset
command 'See ECM index
either an invalid command or parameter was
received 'See ECM index
emergency power off cable 'See Index of INST
emergency power off switch test 'See Index of
INST
emergency power off switch test 2-path 'See
Index of INST
emergency power off, unit - rear 'See MIM index
emergency switch, unit INTRO-S, INTRO-10,
INTRO-22, 'See MIM index
enable/disable switch test 'See Index of INST
enable/disable switch, device INTRO-8, INTRO-22
enable/disable switches, device 'See MIM index
enabl~1 disable switches, logic INTRO-35
encoder circuit description 'See MSM index
end cover 'See Index of INST
end covers 'See Index of INST
end of call REPAIR-15
end of guided maintenance REPAIR-13
end op with transfer count equal to zero 'See
ECM index
end op with transfer count not equal to zero 'See
ECM index
enter log number MD-72
enter logs to be merged MD-72
entries from other MLMs MLX-2
entries to the INST section 'See Index of INST
entry G, (PSG Vol. R05) , for subsystem
configuration 'See MIM index
entry page, MD reference 'See MIM index
entry table
INTRO section INTRO-1
entry table - repair procedures REPAIR-1
entry table, MAP section 'See MIM index
EPO cable 'See Index of INST
equipment specifications, miscellaneous INTRO-2
erase error log LOG-10
EREP
and console messages INTRO-37
reports PROG-1
EREP release level 'See MSM index
EREP reports PROG-1
EREP reports, guidelines for periodic review
of START-6
error
display, isolation code MD-78
function stopped MD-73
sense formats MD-73
error conditions, airflow 'See MIM index
MASTER INDEX-1S
error conditions, drive motor 'See MIM index
error correction code logic check 'See ECM index
error display, isolation code MD-41
error log INTRO-38
error log data, problem analysis with ISOLATE-24
error log function LOG-1
error log functions MD-71
error log mask LOG-9
error, checkpoint 'See MIM index
error, enter the date MD-72
error, reporting and collection - checkpoint 'See
MIM index
ESD-sensitive parts 'See MIM index
event history report PROG-20
exits to other MLMs MLX-3
extended command, tag sequence check 'See
ECM index
extended DPS test MD-14, MD-64
external timer from other device not running 'See
ECM index
extra RCC required 'See ECM index
failing devices and controllers, using Device Status.
test ISOLATE-18
failing devices and controllers, using the error log
to determine ISOLATE-20
failing ports (EBxx), determining START-33
failing strings and devices, locating START-21
failure boundary definitions 'See MSM index
controller 'See MSM index
device 'See MSM index
drive 'See MSM index
port 'See MSM index
unknown 'See MSM index
failure determination, subsystem START-4
failure, tripped CB or CP analysis 'See MIM index
failures
intermittent INTRO-38
multiple ports START-33
single ports ST ART-33
solid INTRO-38
fan, gate 'See MIM index
fault log A SC-FRU-38
fault log B SC-FRU-39
fault log C SC-FRU-39
fault log D SC-FRU-48
fault log D , active check-1 bits in 'See MIM index
PIN 4519911 39 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987. 1988
fault log E SC-FRU-40
fault log error isolation MAP 'See MIM index
fault log F SC-FRU-41
fault log G SC-FRU-41
fault symptom code (see symptom code)
feature codes, selective INTRO-2
features, attachment, storage control 'See Index
of INST
feedthrough path 'See MSM index
fenced .devices MD-64
fenced, volume ISOLATE-51
fencing
control interface INTRO-18
subsystem INTRO-18
field 'See MSM index
filter assembly, absolute 'See MIM index
first byte ready active when not expected 'See
ECM index
fixed A and B gates 'See Index of INST
format and message
SC = 9F10 through 9F97 SC-FRU-24
SC = 9F9A through 9F9E SC-FRU-24
format and message, byte 7 'See MSM index
format 5 'See Index of INST
format 6 sense data summary 'See MSM index
format 7 message
SC = DFFF SC-FRU-37
SC = DFX5 through DFXD SC-FRU-35
SC = DF10 through DFX4 SC-FRU-35
format 9 'See Index of INST
format/message
SC = EF81 through 8A SC-FRU-49
format, byte, and bit table headings 'See MSM
index
four path INTRO-18
frame
bolting the frames together 'See Index of INST
frame designations 'See Index of INST
from +24 Vdc, +5 Vdc distribution 'See MIM
index
front drive connectors, HDA 'See MIM index
front drive power-on/off 'See MIM index
front drive power-on/off procedures 'See MIM
index
front operator panel (3880 subsystem) INTRO-7
front operator panel (3990) INTRO-20
front operator panel labels (3880) INTRO-9
front view 'See MIM index
FRU bit 0 'See ECM index
FRU bit 1 'See ECM index
FRU replacement INTRO-38
MASTER INDEX-16
I~
C
("
C
~"'\
~
(";
FRU replacement procedures REPAIR-2
FSC (see symptom code)
functions
controller 'See MSM index
device 'See MSM index
DPS 'See MSM index
drive 'See MSM index
RPS 'See MSM index
servomechanism 'See MSM index
funnel parity check 'See ECM index
(':
('/
C'
c
c
( ":
-"/
(
'"
/
gaps 'See MSM index
gate fan 'See MIM index
gate thermals 'See Index of INST
GBOD not detected after park during power
off 'See ECM index
GBOD not found after a seek to a minus
cylinder 'See ECM index
ground check, baseplate 'See Index of INST
ground isolation 'See Index of INST
groups, cover (A and B units) 'See MIM index
guard band detected at cylinder 0 during a minus
cylinder seek command 'See ECM index
guard band detected during seek 'See ECM index
guard bands 'See MSM index
guard, belt 'See Index of INST
guardband INTRO-27
guided maintenance, end of REPAIR-13
guidelines for periodic review of EREP
reports ST ART-6
c
(
ha scan MD-19, MD-67
ha scan, review MD-51
half wave voltage ripple measurement 'See MIM
index
HAR bus parity check 'See ECM index
HAR modifier overrun on set HAR oriented 'See
ECM index
hardware, scheduling START-35
HDA
actuator lock 'See Index of INST
actuator unlock 'See Index of INST
pulley lock 'See Index of INST
PIN 4519911 39 Sides
© Copyright
f:
EC 476581 01Aug89
IBM Corp. 1987, 1988
pulley unlock 'See Index of INST
HDA actuator locking lever, lock/unlock 'See MIM
index
HDA air pressure check 'See MIM index
HDA cable swap 'See MIM index
HDA cable swap check 'See ECM index
HDA cables swapped MD-79
HDA connectors - front drive 'See MIM index
HDA connectors - rear drive 'See MIM index
HDA drive belt 'See MIM index
HDA drive motors 'See MIM index
HDA manifold, differential pressure check 'See
MIM index
HDA shipping lock 'See MIM index, 'See Index of
INST
HDA temperature check 'See Index of INST
HDA temperature conversion chart 'See MIM
index
HDI procedure 1 'See MIM index
air plenum attachment 'See MIM index
air plenum removal 'See MIM index
front HDA removal 'See MIM index
HDA drive belt installation 'See MIM index
pivot rear HDA base casting 'See MIM index
rear drive power-off and actuator lock 'See
MIM index
rear HDA drive belt removal 'See MIM index
rear HDA power-on procedure 'See MIM index
rear HDA removal 'See MIM index
tools and supplies 'See MIM index
voice-coil magnet removal 'See MIM index
voice-coil magnet replacement 'See MIM
index
HDI procedure 2 _'See MIM index
air ,.Ienum replacement 'See MIM index
front HDA power-on procedure 'See MIM index
HDA installation 'See MIM index
rear HDA power-on procedure 'See MIM index
tools and supplies 'See MIM index
voice-coil magnet removal 'See MIM index
voice-coil magnet replacement 'See MIM
index
head address miscompare 'See ECM index
head address register parity check 'See ECM
index
head and disk assembly 'See MIM index
access mechanisms 'See MSM index
front HDA power-off procedure 'See MIM index
HDA actuator lock procedure 'See MIM index
HDA removal 'See MIM index
head addresses 'See MSM index
MASTER INDEX-17
head-to-disk interference 'See MIM index
mounting bolt torque 'See MIM index
mounting surface cleaning 'See MIM index
packing/unpacking instructions 'See MIM
index
plastic shipping bag 'See MIM index
read/write channel board 'See MIM index
rear HOA power-off procedure 'See MIM index
removal 'See MIM index
replacement 'See MIMindex
shipping lock 'See MIM index
shipping lock installation 'See MIM index
temperature 'See MIM index
temperature check 'See MIM index
temperature conversion chart 'See MIM index
temperature strip 'See MIM index
tools and supplies 'See MIM index
voice coil interference 'See MIM index
voice coil-to-VCM interference 'See MIM index
weight 'See MIM index
head layout INTRO-29
head offset value, byte 15, format 4 'See MSM
index
head-to-disk interference 'See MIM index
home address 'See MSM index
hose and air plenum replacement, procedure
2 'See MIM index
hot bits, DOC SC-FRU-22, SC-FRU-46
how to use the checklists 'See Index of INST
I/O control card switch settings REPAIR-20
IC analysis MO-81
ICKOSF error messages 'See MIM index
ICKOSF run information 'See Index of INST
10 string switches REPAIR-20
10, logic board pins 'See MIM index
10, storage director INTRO-9, INTRO-11, 'See
Index of INST
10, string INTRO-9, INTRO-11, INTRO-23
10, subsystem 'See Index of INST
identification of logic board pins 'See MIM index
identify the A1/A2 A unit 'See Index of INST
IDs, string 'See Index of INST
in reset allegiance command, the device returned
an encorrect address 'See ECM index
index active when not expected 'See ECM index
PIN 4519911 39 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
index detection 'See MSM index
index failure during idle 'See ECM index
index found during defect skip in HA 'See ECM
index
index missing 'See ECM index
index not active when expected 'See ECM index
index not found during data check recovery 'See
ECM index
indexlcell check 'See ECM index
index/record ready interrupt check during search
routine 'See ECM index
index, location section 'See MIM index
indicators (See LEOs)
operator panel INTRO-20
panel, operator INTRO-?
inhibit or reset active 'See ECM index
inspection, physical ISOLATE-2
install end cover 'See Index of INST
installation duration goals 'See Index of INST
installation power problems 'See Index of INST
installation procedure, start 'See Index of INST
integration test, sUbsystem 'See Index of INSi
interface check FRUs, DOC MLX-2
interface procedure, COP cable 'See MIM index
COP interface repair 'See' MIM index
controller-to-device port interface repair 'See
MIM index
interface repair, COP 'See MIM index
interface repair, controller-to-device port 'See
MIM index
repair, COP interface 'See MIM index
interframe connector gate 'See Index of INST
intermediate filter 'See MIM index
intermediate filter check 'See MIM index
intermittent failure definition 'See MSM index
intermittent failures INTRO-38, 'See MSM index
internal '10' register failure 'See ECM index
internal machine safety inspection 'See MIM
index
internal mpchine safety inspection, front 'See MIM
index
internal machine safety inspection, rear 'See MIM
index
interrupt not resettable 'See ECM index
interrupts, device INTRO-8, INTRO-22
intervention required - console
message ISOLATE-3S
INTRO section entry table INTRO-1
introduction INTRO-1, 'See MSM index, 'See Index
of INST
invalid command code 'See ECM index
MASTER INDEX-18
,/ '\
(
f
(
(
('
(/
(
(~
invalid command in DCP command register for
offset 'See ECM index
invalid command in DCP command register for
seek routine 'See ECM index
invalid command on the DOC bus out lines 'See
ECM index
invalid DOC selection response or timeout 'See
ECM index
invalid device MD-65
invalid or illegal command received with the device
busy 'See ECM index
invalid tag in during selection sequence 'See
ECM index
invalid tag in for an extended command
sequence 'See ECM index
invalid tag in for an immediate command
sequence 'See ECM index
invalid tag state on the DOC tag out lines 'See
ECM index
invert the operator panel '.see Index of INST
IOC card check-1 'See ECM index
IOC card check-2 'See ECM index
IOC controller selection check 'See ECM index
IOC detected DOC bus out parity check 'See ECM
index
IOC register data bus parity check 'See ECM
index
IOC register 1 parity check 'See ECM index
IOC register 3 parity check 'See ECM index
IOC ROS bit parity check 'See ECM index
IOC transfer clock check 'See ECM index
lOCO 'See Index of INST
isolate a problem START-2
isolate section entry table ISOLATE-1
isolate the problem ISOLATE-1
isolation code
1110 'See ECM index
1120 'See ECM index
1150 'See ECM index
isolation code (lC) 'See MSM index
isolation code error display MD-41, MO-78
jack pad 'See Index of INST
jack screws 'See Index of INST
Japan, 50 hertz, primary power box - A unit
MIM index
joined A units 'See Index of INST
PIN 4519911 39 Sides
EC 476581 01 Aug89
© Copyright IBM Corp. 1987, 1988
'See
K pulley shipping lock
'See Index of INST
labels
cables 'See Index of INST
CP410 'See Index of INST
device address 'See Index of INST
operator panel 'See Index of INST
pink/blue 'See Index of INST
serial number 'See Index of INST
spare machine serial number 'See Index of
INST
unit not installed 'See Index of INST
labels (3880)
front operator panel labels (3880) INTRO-9
front operator panel labels (3990) INTRO-23
rear unit emergency power off panel labels
(3990) INTRO-25
unit emergency power off panel, rear
(38BO) INTRO-10
LED
+1.7 Vdc power supply 'See Index of INST
+5 Vsp power supply 'See Index of INST
AC power-on LEO 'See Index of INST
controller 'See Index of INST
controller ready 'See Index of INST
device ready 'See Index of INST
rear panel 'See Index of INST
verify +5 and +1.7 off 'See Index of INST
LEO continuously on, status pending ISOLATE-33
LEO off, device sequencer status ISOLATE-31
LED off, device switch enabled ISOLATE-30
LEOs 'See MIM index
ac power-on 'See MIM index
controller ready 'See MIM index
device ready 'See MIM index
LEOs, device ready 'See MIM index
LEOs, A1 board voltage and clock INTRO-30
LEOs, B1 board power card INTRO-31
LEOs, controller ready INTRO-8,INTRO-11,
INTRO-22, INTRO-25
LEOs, controller voltage regulator card INTRO-32
LEOs, device ready INTRO-B, INTRO-23
MASTER INDEX-19
c
--I
LEOs, location of INTRO-7, INTRO-20
LEOs, logic board INTRO-30
LEOs, phase INTRO-31
Legend 'See MIM index
leveling pad 'See Index of INST
leveling pads 'See Index of INST
line cord connector, ac 'See Index of INST
LOC section entry page 'See MIM index
how to use the LOC section entry page 'See
MIM index
LOC section index 'See MIM index
local/remote switch, SW487 'See Index of INST
locate 'See Index of INST
locate the failing machine with console
messages START-36
locating 8 unit first 'See Index of INST
locating the A-unit first 'See Index of INST
locating the A-unit only 'See Index of INST
locating the failing string and devices START-14
locating the failing strings and devices START-21
location (LOC) section index 'See M/M index
location of major machine units 'See MIM index
lock HOA pulley 'See Index of INST
lock PLO to data not active when expected 'See
ECM index
lock/unlock procedure, HOA actuator 'See MIM
index
locking lever, actuator lock/unlock 'See MIM
index
loctite sealant, type E 'See MIM index
log, error INTRO-38
logic board LEOs INTRO-30
logic board pin identification 'See MIM index
logic Enable/Disable Switches INTRO-3S
logic/actuator power supply check, safety
inspection 'See MIM index
lost track following during data check
recovery 'See ECM index
machine safety inspection, external 'See MIM
index
machine serial number labels, spare 'See Index
of INST
machine stability safety inspection 'See MIM
index
machine, front and rear views 'See MIM index
PIN 4519911 39 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
magnet, voice-coil 'See M/M index
main menus, diagram MO-2
mainline power cable disconnect procedure 'See
MIM index
cable connector 'See MIM index
hard wired 'See MIM index
maintenance analysis procedure
See MAP (Maintenance Analysis Procedure)
maintenance device (see MO)
MO description INTRO-37
maintenance device attachment 'See Index of
INST
maintenance device cable MAP 'See MIM index
maintenance device reference 'See MIM index
maintenance, single actuator - power on/off
procedure 'See MIM index
left/right front actuator 'See MIM index
left/right rear actuator 'See MIM index
major units of the machine, location 'See MIM
index
manuals 'See Index of INST
manuals system reference 'See Index of INST
MAP (Maintenance Analysis Procedure)
cable swap analysis 'See MIM index
COP interface errors 'See MIM index
COP wrap error 'See MIM index
clock check analysis procedure
solid 'See MIM index
control interface analysis 'See MIM index
diagnostic CO error at 3880 'See MIM index
routine 77 operating instructions from.::J.
3880 'See MIM index
routine 93 operating instructions from a
3880 'See MIM index
control interface check 1 for 3990 'See MIM
index
control interface check 2 for 3990 'See MIM
index
control interface check-2 'See MIM index
control interface test isolation codes 'See MIM
index
controller A 1 through A4 pin connections 'See
MIM index
CTL-I check 1 for 3990 'See MIM index
CTL-I check 2 for 3990 'See MIM index
CTL-I check-2 'See MIM index
CTL-I test isolation codes 'See MIM index
cylinder seek routine 8A 'See MIM index
data check - drive or device 'See MIM index
diagnostic 03 error stops 'See MIM index
diagnostic error stops 'See MIM index
MASTER INDEX-20
\.
--'
(
(~
(
("
(
(/
('
(
(
(
(
drive motor switch error conditions 'See MIM
index
fault log error isolation 'See MIM index
how t.o use 'See MIM index
maintenance device cable 'See MIM index
MD cable 'See MIM index
no read data 'See MIM index
operator panel LEDs and switches 'See MIM
index
read/write procedures, additional
devices. 'See MIM index
servo low-level and PLO signal cable
assembly 'See MIM index
servo low-level signal analysis 'See MIM index
servo power amplifier cable 'See MIM index
servo power amplifier cable assembly 'See
MIM index
MAP section entry table 'See MIM index
mat, RFI 'See Index of INST
during installation 'See Index of INST
during removal 'See Index of INST
MD
analysis procedures INTRO-38
attachment MD-5
diskette compatibility MD-83
loading MD-5
MD Main option menu MD-52
operation MD-1
power on MD-5
running diagnostics INTRO-38
MD attachment 'See Index of INST
MD cable MAP 'See MIM index
MD connection test 'See MSM index
MD connector 'See MIM index
MD diagnostic error display MD-78
MD diagnostic menu option A MD-37
MD diagnostic menu option a MD-11
MD diagnostic menu option 1 MD-13
MD diagnostic menu option 2 MD-14
MD diagnostic menu option 3 MD-15
MD diagnostic menu option 4 MD-17
MD diagnostic menu option 5 MD-19
MD diagnostic menu option 6 MD-24
MD diagnostic menu option 7 MD-26, MD-73,
MD-75, MD-76, MD-77
MD diagnostic menu option 8 MD-32
MD diagnostic menu option 9 'See Index of INST,
MD-34
MD Main Menu option D MD-82
MD Main Menu option R, review MD-81
MD Main menu option a MD-7
PIN 4519911 39 Sides
© Copyright
(\
EC 476581 01Aug89
IBM Corp, 1987. 1988
MD Main menu option 2 LOG-1
MD Main Menu option 3 MD-79
MD Main option menu MD-52
MD menu MD-2
MD reference
010 select an option MD-52
020 select controller MD-52
030 set/reset CE mode MD-53
035 devices in CE mode/default devices MD-54
050 define the failure boundary MD-55
051 define the failure boundary (EBXX) MD-58
052 define the failure boundary (data
checks) MD-61
090 write inhibit MD-63
092 storage path fenced MD-63
093 fenced devices MD-64
100 diagnostic menu MD-64
200 select controller MD-65
205 invalid controller selected MD-65
210 select device(s) MD-65
215 invalid device MD-65
310 reset CE mode MD-65
320 run connection tests MD-66
330 controller connection test MD-66
340 controller AX connection tests
ready MD-67
350 DPS tests cannot be run MD-67
360 HA scan MD-67
395 scan complete MD-70
400 error log Junctions MD-71
405 select controller (error log) MD-71
410 enter search arguments MD-72
415 enter log number MD-72
420 enter the date MD-72
425 enter logs to be merged MD-72
430 enter the sense formats MD-73
440 error log function stopped MD-73
460 configuration MD-73
462 configuration data MD-75
465 device status MD-76
468 device type MD-77
480 storage control status MD-78
500 MD diagnostic error display MD-78
550 symptom code, enter MD-79
605 cable swap detected MD-79
615 no MD analysis for IC MD-81
700 select review option MD-81
742 analyze or display next error MD-81
800 diagnostic aids MD-82
900 MD disk.not compatible MD-83
MD reference,and entry page 'See MIM index
MASTER INDEX-21
MD use 'See Index of INST
MD, preparing the MO-5
MD, select an option MD-52
media maintenance using ICKDSF 'See MIM
index
merge log LOG-8
merging logs MD-72
MES summary INTRO-2
messages, console IMAP reference
table ISOLATE-34
messages, informational PROG-10
microcode ordering information PRE-2
miscellaneous console messages ISOLATE-53
miscellaneous physical symptoms ISOLATE-6
missing channel end and device end - console
message ISOLATE-46
missing device address bit at selection 'See ECM
index
missing end operation, transfer was
complete 'See ECM index
missing end operation, transfer was
incomplete 'See ECM index
missing interrupt - console message ISOLATE-46
missing safety labels, part numbers for 'See MIM
index
models, 3380 INTRO-1
motor assembly, blower 'See MIM index
motor error conditions, drive 'See MIM index
motor error, drive conditions 'See MIM index
motor on/off switch INTRO-35
motor or soft start thermal or belt guard switch
open 'See ECM index
motor relay active after drop during brake relay
pick test of diagnostic routine 99 'See ECM index
motor relay active after drop during motor test of
diagnostic routine 99 'See ECM index
motor relay active after picking brake relay during
diagnostic routine 99 'See ECM index
motor relay active at the beginning of motor start
routine 'See ECM index
motor relay active with the brake applied during
diagnostic routine 99 'See ECM index
motor relay inactive after pick during diagnostic
routine 99 'See ECM index
motor relay not active during power on or a search
routine 'See ECM index
motor relay stuck active during either a POR or a
power off 'See ECM index
motor run and soft start relays active after
dropping during diagnostic routine 99 'See ECM
index
PIN 4519911 39 Sides
EC 476581 01Aug89
motor run and soft start relays active at the
beginning of diagnostic routine 99 'See ECM
index
motor run relay active at the beginning of
diagnostic routine 99 'See ECM index
motor run relay active at the end of diagnostic
routine 99 'See ECM index
motor run relay not active after soft start relay
drop;,'\d 'See ECM index
motor sensed off, but still turning at more than 3440
rpm 'See ECM index
motor sequence assembly (MSA) 'See MIM index
motor slowing, AGC active, motor relay not picked
during POR 'See ECM index
motor start routine MD-64
motor switch error conditions, drive 'See MIM
index
motor switch off after the motor is started during
power on routine 'See ECM index
motor switch off at beginning of power on or during
idle 'See ECM index
motor test MD-24
motor turning without relay being picked during
POR 'See ECM index
motor/brake power, drive sequence 'See MIM
index
motor/brake sensing, drive 'See MIM index
motor/brake, drive power 'See MIM index
motor, brake 'See MIM index
motor, drive 'See MIM index
motors, HDA and blower 'See MIM index
move procedure during 4-path 'conversion 'See
Index of INST
MSA 'See MIM index
multiburst ECC 'See Index of INST
multifunction 'See ECM index
multiple MO-56
multiplexer input parity check 'See ECM index
o
(
'
(
'\
\
::._ :7
J
if"-
\l/
next level of support assistance REPAIR-13
no air flow 'See ECM index
no air pressure detected 'See ECM index
no air pressure detected before the motor
start 'See ECM index
no air pressure detected during idle 'See ECM
index
MASTER INDEX-22
© Copyright IBM Corp. 1987, 1988
C\
I;'
-_... "
"
(
(-
f
(-
C
no air pressure detected during power on
routine 'See ECM index
no function 'See ECM index
no interrupt from device during sUb-system
command chain 'See ECM index
no power sequence complete 'See MIM index
no power sequence complete signal MLX-2
no read data 'See MIM index
repair strategy 'See MIM index
no read data check 'See ECM index
no read data found 'See ECM index
no read/write recovery 'See ECM index
no selection response after poll interrupt
(DDC) 'See ECM index
non-repeatable runout MD-37
nondisruptive CDP cables, 4-path
installation, nondisruptive 4-path 'See Index of
INST
nondisruptive installation of a second 4-path
string 'See Index of INST
nondisruptive installations of B units 'See Index of
INST
nondisruptive, string configuration switch,
4-path 'See Index of INST
normal sequence MD-11
normal sequence diagnostics ISOLATE-12
not ready, controller ISOLATE-32
not ready, device ISOLATE-27
not track following after fourth settling delay during
seek routine 'See ECM index
NRRO test MO-37
numbering, device (see PSG Vol. R05, entry
G) 'See MIM index
(
(
(
('
(
offset active cannot be reset 'See ECM index
offset command not saved prior to data check
recovery 'See ECM index
offset routine took too long 'See ECM index
operation, power supply 'See MIM index
operator panel INTRO-7, 'See MIM index, 'See
MSM index, 'See Index of INST
AC power-on LED 'See Index of INST
controller power on/off switches 'See MIM
index
controller ready LEOs 'See MIM index
device enable/disable switches 'See MIM
index
PIN 4519911 39 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
device ready LEOs 'See MIM index
invert, rotate 'See Index of INST
labels when attached to a 3990 'See Index of
INST
labels when attached to 3880 'See Index of
INST
power-on LEO 'See MIM index
switch cables 'See Index of INST
unit emergency switch 'See MIM index
unit not installed label 'See Index of INST
operator panel and +24 Vdc panel 'See MIM
index
operator panel LEDs and switches MAP 'See MIM
index
operator panel switch cable
removal 'See Index of INST
option A, diagnostic menu MD-37
option D, MD Main Menu MD-82
option ~, MO Main Menu MO-81
option 0, diagnostic menu MD-11
option 1, diagnostic menu MO-13
option 2, diagnostic menu MO-14
option 2, MO Main menu LOG-1
option 3, diagnostic menu MO-15
option 3, MD Main Menu MD-79
option 4, diagnostic menu MD-17
option 5, diagnostic menu MO-19
option 6, diagnostic menu MO-24
option 7, diagnostic menu MD-26, MD-73, MD-75,
MO-76, MO-77
option 8, diagnostic menu MO-32
option 9, diagnostic menu 'See Index of INST,
MO-34
ordering manuals 'See Index of INST
oriented not active when tested 'See ECM index
outlet distribution, ac convenience 'See MIM
index
outlet, convenience distribution 'See MIM index
pack and unpack instructions 'See Index of INST
pack scan. See HA scan MO-67
packing 'See Index of INST
pad characters
writing 'See MSM index
padding check 'See ECM index
panel, operator INTRO-20, 'See MIM index
panel, operator and +24 Vdc 'See MIM index
MASTER INDEX-23
o
panel, power sequencing INTRO-34
panel, rear 'See Index of INST
panel, rear UEPO 'See MIM index
panel, 24 Vdc INTRO-32
park failed before starting motor during power
on 'See ECM index
part(s) location, index 'See MIM index
path INTRO-11, INTRO-23, INTRO-25
path and device physical identifier, byte 4 'See
MSM index
permanent equipment check definition 'See MSM
index
permanent failure definition 'See MSM index
permanent field PROG-13
permanent path error 'See ECM inc;lex
phase LEOs INTRO-31
phase rotation rework 'See Index of INST
phase, determine the ac line 'See Index of INST
phasing, ac and primary power 'See MIM index
physical inspection ISOLATE-2
physical parts 'See MSM index
A unit 'See MSM index
B unit 'See MSM index
physical symptoms, miscellaneous ISOLATE-6
pin identification, logic board 'See MIM index
planning the repair action for data
checks START-23
plenum and hose replacement, procedure 2 'See
MIM index
plenum and hose, procedure 1 'See MIM index
port degate check 'See ECM index
port response check 'See ECM index
post-repair checkout REPAIR-8
post-repair controller checkout REPAIR-11
power
controller on/off procedure REPAIR-27
device on/off procedure REPAIR-29
drive on/off procedure REPAIR-28
drive power on, first time 'See Index of INST
LEOs, B1 board INTRO-31
on LED, AC INTRO-8, INTRO-22
on/off switches, controller INTRO-8, INTRO-22,
'See Index of INST
sequence cables INTRO-6, INTRO-19
sequencing panel INTRO-34
string on/off procedure REPAIR-25
unit on/off procedure REPAIR-26
power amplifier card, servo 'See MIM index
power amplifier card, voltage checks - servo 'See
MIM index
power box, primary - B unit 'See MIM index
PIN 4519911 39 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
power box, primary, Aunit(50 hertz) 'See MIM
index
power cable
cable labeling 'See Index of INST
power cable installation, B unit 'See Index of INST
power cable removal 'See Index of INST
power card check 'See MIM index
power card check - sequencer write bus parity
check 'See ECM index
power control summary diagram 'See MIM index
power on/off
controller REPAIR-27
device REPAIR-29
drive REPAIR-28
string REPAIR-25
unit REPAIR-26
power on/off switches, controller 'See MIM index
power on/off, single actuator maintenance (SAM)
procedure 'See MIM index
front left/right 'See MIM index
rear left/right 'See MIM index
power on, CB200 'See Index of INST
power problems on installation 'See Index of INST
power receptacle safety inspection 'See MIM
index
power sequence control 'See MIM index
complete relay (3880) 'See MIM index
control unit device power on 'See MIM index
storage control tailgate connectors 'See MIM
index
power sequence control cables 'See Index of
INST
power sequence did not complete 'See MIM index
power sequencing INTRO-6, INTRO-19
power status SC-FRU-26, SC-FRU-34
power status reporting and collection 'See MIM
index
power supply diagrams 'See MIM index
power supply operation 'See MIM index
bleeder resistor 'See MIM index
capacitor 'See MIM index
power supply diagrams 'See MIM index
rectifier 'See MIM index
transformer 'See MIM index
power supply ripple problem analysis 'See MIM
index
power supply voltage problem analysis 'See MIM
index
all dc voltages are high 'See MIM index
all dc voltages are low 'See MIM index
only one dc voltage is high 'See MIM index
MASTER INDEX-24
C.'\
j
('
"",;'
ff-'
I",.;)
()
If\
Itj
(
(~\
(=
f
(
~,
.., /
(
(
(
only one dc voltage is low 'See MIM index
other voltage problems 'See MIM index
service notes 'See MIM index
two different dc voltages are failing 'See MIM
index
voltage problem analysis 'See MIM index
power supply, controller 'See MIM index
power supply, controller (Ai/A2) 'See MIM index
power supply, drive actuator 'See MIM index
power supply, problem - ripple analysis 'See MIM
index
power supply, problem - voltage analysis 'See
MIM index
power supply, 5/24 Vdc bootstrap 'See MIM index
power switches
A-unit 'See Index of INST
B unit 'See Index of INST
power-on LED, ac 'See MIM index
power-on sequence chart 'See MIM index
power-on/off procedures 'See MIM index
B section of an A unit 'See MIM index
B unit or B section of an A unit 'See MIM
index
complete B unit or B section of an A unit 'See
MIM index
front drive 'See MIM index
rear drive 'See MIM index
power-on/off procedures, 3380-JK 'See MIM index
power-on/off, single actuator (front) 'See MIM
index
power-on/off, single actuator (rear) 'See MIM
index
power, drive motor/brake sequence 'See MIM
index
precomp circuit description 'See MSM index
precompensation check 'See ECM index
precompensation failure, set or reset 'See ECM
index
prepare the MD MD-5
pressure check, HDA 'See MIM index
pressure check, HDA manifold 'See MIM index
pressure checks, air 'See MIM index
pressure table, air 'See MIM index
primary power assembly safety inspection
models AJ4, AK4, and AL4, 60 Hz (50 Hz
Japan) 'See MIM index
models BJ4 and BK4 'See MIM index
50 Hz (except Japan), models AJ4 and
AK4 'See MIM index
PIN 4519911 39 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
primary power box - A unit (50 hertz) 'See MIM
index
primary power box - A unit (60 hertz, 50 hertz
Japan) 'See MIM index
primary power box - B unit 'See MIM index
primary power on 'See Index of INST
primary power, phasing 'See MIM index
primary symptom code definition 'See MSM index
primary, A unit (50 hertz) power box 'See MIM
index
prioritizing symptom codes START-i2, START-i9
probable failing unit, determining ST ART-i6
problem analysis with error log data ISOLATE-24
problem, isolate the ISOLATE-i
problems, power 'See Index of INST
procedure, disconnect mainline power cable 'See
MIM index
procedure, single actuator maintenance
(SAM) 'See MIM index
power on/off 'See MIM index
front, left/right 'See MIM index
rear, left/right 'See MIM index
procedures ~See MIM index
front/rear drive 'See MIM index
power off 'See MIM index
power on 'See MIM index
power off 'See MIM index
power on 'See MIM index
procedures, power on/off REPAIR-23
procedures, 3380-JK power-on/off 'See MIM index
PSG Vol. R05, entry G, (device
numbering/subsystem configuration) 'See MIM
index
publications 'See MIM index
cross references MLX-3
ordering procedures PRE-1, 'See MIM index,
'See MSM index, 'See ECM index
rei: ~ed PRE-1, INTRO-1, 'See MIM index, 'See
MSM index, 'See ECM index
updatil1g PRE-i, 'See MIM index, 'See MSM
index, 'See ECM index
pulley lock, HDA 'See Index of INST
pulley shipping lock, model K 'See Index of INST
pulley unlock, HDA 'See Index of INST
pulley unlock, HDA shipping lock 'See Index of
INST
PWR section entry page 'See MIM index
how to use the PWR section entry page 'See
MIM index
MASTER INDEX-2S
()
quiesce / resume storage path INTRO-19
description INTRO-19
quiesce controller INTRO-19, 'See Index of INST
quiesce storage path 'See Index of INST
R/W data cable failure 'See MIM index
RCC sequence check 'See ECM index
RCC 1 and RCC 2 sequences not successful 'See
ECM index
RCC 1 sequence not successful 'See ECM index
read and write check 'See ECM index
read paths 'See MSM index
read transmit active 'See ECM index
read transmit check 'See ECM index
read/write
cable labeling 'See Index of INST
status 1 SC-FRU-4
status 2 SC-FRU-4
status 3 SC-FRU-5
read/write cable
installation, 2-path 'See Index of INST
installation, 4-path 'See Index of INST
removal 'See Index of INST
read/write channel board 'See MIM index, 'See
MSM index
read/write channel board voltage check 'See MIM
index
read/write channel board, voltage checks 'See
MIM index
read/write channel check 'See ECM index
read/write channel checks
SC = 98XX SC-FRU-22
read/write channel status parity check 'See ECM
index
read/write check 'See MIM index, 'See ECM index
read/write control cable check 'See ECM index
read/write data cable check 'See ECM index
read/write data cable failure 'See MIM index
read/write gate DDC drivers check 'See ECM
index
read/write mode active during safe 'See ECM
index
read/write sequence check 'See ECM index
read/write servo check 'See ECM index
PIN 4519911 39 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1988
read/write status 1
SC = 92XX SC-FRU-4
read/write status 1 byte 12 'See MSM index
read/write status 2
SC
93XX SC-FRU-4
read/write status 3
SC = 94XX SC-FRU-5
read/write status 3, byte 14 'See MSM index
read/write status 4 byte 18 'See MSM index
read/write status-2 byte 13 'See MSM index
ready LEOs 'See MIM index
ac power-on 'See MIM index
controller 'See MIM index
device 'See MIM index
ready lEDs, controller INTRO-8, INTRO-22
ready lEDs, device INTRO-8, INTRO-23
rear drive connectors, HDA 'See MIM index
rear drive power-on/off 'See MIM index
rear drive power-on/off procedures 'See MIM
index
rear panel 'See Index of INST
rear panel lEDs 'See Index of INST
rear UEPO panel 'See MIM index
rear unit emergency power off panel
(3880) INTRO-10
rear unit emergency power off panel
(3990) INTRO-24
rear view 'See MIM index
recable a string 'See Index of INST
recable a string disruptively 'See Index of INST
recable the CTl-1 'See Index of INST
received a command that does not match the HDA
model (command active - busy active) 'See ECM
index
received a command that does not match the HDA
model (command active - busy inactive) 'See
ECM index
received a command that does not match the HDA
model (command signal inactive) 'See ECM
index
reconfiguration, 3990 vital product data 'See Index
of INST
record 'See MSM index
record ID, bytes 8 through 12 'See MSM index
record ready interrupt is active too long during
search routine 'See ECM index
record updating 'See Index of INST
rectifier and capacitor service checks 'See MIM
index
redundant line / selection check 'See ECM index
reference item, safety inspection. 'See MIM index
(', J
'\
=
MASTER INDEX-26
'\
i
,/
(fl' ' \
i
,",-y'
c
(;
"\
C
"~I
,
-..:-"'
( "\
./
C
()
c
c
c
(
(
{
reference manuals system 'See Index of INST
reference page 'See MIM index
register 'OF' error codes
SC = E801 through E813 SC-FRU-42
SC = E814 through E853 SC-FRU-42
SC= E855 through E8FF SC-FRU-43
register 'OF' errors SC-FRU-42
register 10 'See MSM index
register 1F 'See MSM index
regulator card LEOs INTRO-32
regulator card, controller voltage 'See MIM index
relay, complete 'See MIM index
relocate a string 'See Index of INST
relocate a string disruptively 'See Index of INST
relocation procedure, start 'See Index of INST
removal procedure 'See Index of INST
removal procedure, complete string 'See Index of
INST
removal procedure, start 'See Index of INST
removal, HOA 'See MIM index
repair procedures (see also service
procedures) REPAIR-1
repair procedures entry tab;e REPAIR-1
replacement procedures, FRU REPAIR-2
replacement, FRU INTRO-38
reporting and collection, checkpoint error 'See
MIM index
reporting and collection, power status 'See MIM
index
reset allegiance, invalid device address received
from storage control 'See ECM index
reset allegiance, start conditions not correct 'See
ECM index
reset CE mode MO-?, MO-65
resistance check, drive belt 'See MIM index
resistance range, drive belt 'See MIM index
resume controller INTRO-19, 'See Index of INST
resume storage path 'See Index of INST
review
CE mode MO-49, MO-81
diagnostics MO-81
error log MO-81
ha scan MO-51
last repair displayed MO-50
last 4 diagnostic errors MO-50
pack scan MO-?O
repair MO-81
review option (MO Main Menu option R) MO-81
review, select option MO-81
rework, phase rotation 'See Index of INST
RFI mat 'See Index of INST
PIN 4519911 39 Sides
EC 476581 01Aug89
II') Copyright IBM Corp. 1987, 1988
during installation 'See Index of INST
during removal 'See Index of INST
ripple analysis, power supply problem 'See MIM
index
ripple problem, power supply analysis 'See MIM
index
rotational position densing (RPS) check 'See ECM
index
routine
CO failed - routine 03 correct MLX-2
03 failed (CTRL error IC = 031 B) M LX-2
03 failed - routine 77 correct MLX-2
RPS check active during search 'See ECM index
RPS failed to lock detected during the
search 'See ECM index
RPS functions 'See MSM index
record ready interrupt 'See MSM index
RPS lock inactive during data check
recovery 'See ECM index
rrack physical address miscompare while
oriented 'See ECM index
run connection tests MO-66
run the device status routine 'See Index of INST
running diagnostics ISOLATE-8
running with 2 MOs 'See Index of INST
runtime, diagnostic 'See Index of INST
safety checks 'See Index of INST
safety inspection
controller dc power check
models AJ4and AK4. 'See MIM index
drive ac power check
50 Hz (except Japan) 'See MIM index
60 Hz (50 Hz Japan) 'See MIM index
external machine check 'See MIM index
guidelines 'See MIM index
internal machine check 'See MIM index
front 'See MIM index
rear 'See MIM index
logic/actuator power supply check 'See MIM
index
machine stability 'See MIM index
models BJ4 and BK4 'See MIM index
power receptacle safety check procedure 'See
MIM index
preparation 'See MIM index
primary power assembly check
MASTER INDEX-27
o
50 Hz (except Japan), models AJ4 and
AK4 'See MIM index
60 Hz (50 Hz Japan), models AJ4, AK4, and
AL4 'See MIM index
purpose 'See MIM index
reference items 'See MIM index
safety ECs 'See MIM index
safety label 'See MIM index
language group part numbers 'See MIM
index
special tools needed 'See MIM index
unit emergency (UEPO) switch check 'See MIM
index
when to perform 'See MIM index
safety label
locations 'See MIM index
safety label language group part numbers 'See
MIM index
SAM (single actuator maintenance), control
panel 'See MIM index
SAM power on/off procedures 'See MIM index
SAM, power on/off procedure 'See MIM index
left/right front actuator 'See MIM index
left/right rear actuator 'See MIM index
SAM, single actuator maintenance control
panel 'See MIM index
SC (see symptom code)
SC 2004 (sync in check) MLX-2
SC 2020 (DDC interface check) MLX-2
SC-FRU charts SC-FRU-1
SC-FRU charts, how to use SC-FRU-1
SC = DEXX (XX = format 7 byte 11 < 0 SC-FRU-34
SC = DFXX (format 7) SC-FRU-37
SC = DFXY SC-FRU-35
SC= D1XX (XX= format 7 byte 11) SC-FRU-26
SC = D2XX (XX = format 7 byte 12) SC-FRU-27
SC = D3XX (XX = format 7 byte 13) SC-FRU-28
SC = D4XX (XX = format 7 byte 14) SC-FRU-29
SC = D5XX (XX = format 7 byte 15) SC-FRU-30
SC = D6XX (XX = format 7 byte 16) SC-FRU-31
SC = D8XX (XX = format 7 byte 18) SC-FRU-33
SC=EBXX SC-FRU-45,
SC=ECXX SC-FRU-46
SC = EDXX (XX = format 8 byte 09) SC-FRU-46
SC=EEXX (XX= format 8 byte 14) SC-FRU-48
SC = EFXX (format 8) SC-FRU-49
SC = EFXX (XX = format 8 byte 07) SC-FRU-49
SC =EOXX (XX = format 8 byte 10) SC-FRU-37
SC=E1XX (XX= format 8 byte 11) SC-FRU-38
SC =E2XX (XX = format 8 byte 12) SC-FRU-39
SC=E3XX (XX= format 8 byte 13) SC-FRU-39
PIN 4519911 39 Sides
EC 476581 01Aug89
(\:J Copyright IBM Corp, 1987, 1988
SC=E5XX (XX= format 8 byte 15) SC-FRU-40
SC=E6XX (XX= format 8 byte 16) SC-FRU-41
SC = E7XX (XX = format 8 byte 17) SC-FRU-41
SC=E8XX (XX= format 8 byte 18) SC-FRU-42
SC=4XYY (4X= format 4 byte 07, YY = format 4
byte 23) SC-FRU-2
SC = 9CXX (XX = format 1 or 9 bytes
11-15) SC-FRU-22
SC=9DXX (XX= format 1 byte 15) SC-FRU-23
SC = 9E10 SC-FRU-23
SC=9E10 (format 1) SC-FRU-23
SC = 9FXX (format 1 and 9) SC-FRU-25
SC = 9FXX (XX = format 1 or 9 byte 07) SC-FRU-24
SC=90XX (XX= format 1 byte 10) SC-FRU-3
SC=91XX (XX= format 1 or 9 byte 11) SC-FRU-3
SC=92XX (XX= format 9 byte 12) SC-FRU-4
SC = 93XX (XX = format 9 byte 13) SC-FRU-4
SC=94XX (XX= format 9 byte 14) SC-FRU-~
SC=95XX (XX= format 1 byte 15) SC-FRU-6
SC=97XX (XX= format 1 byte 17) SC-FRU-19
SC=98XX (XX= format 9 byte 18) SC-FRU-22
scan complete MD-70
scheduling the hardware START-35
searched cell, cell pulse off 'See ECM index
sector compare not detected during a sector
search 'See ECM index
sector compare not found during data check
recovery 'See ECM index
sector compare stuck on during data check
recovery 'See ECM index
sector number, byte 13 'See MSM index
sectors 'See MSM index
seek head and cylinder address bytes 'See MSM
index
seek incomplete 'See ECM index
seek process 'See MSM index
seek routine took too long 'See ECM index
select a diagnostic option MD-10
select controller error log MD-71
select review option MD-81
select, controller MD-52, MD-65
select, device(s) MD-65
selected device check 1 'See ECM index
selection 'See MSM index
selective feature codes INTRO-2
sense bytes SC-FRU-2
sense data introduction 'See MSM index
sense data, procedure to analyze format of 'See
MSM index
sensing, drive motor/brake 'See MIM index
sequence assembly, motor 'See MIM index
MASTER INDEX-28
o
,~
""
1",,_,,/
~~\
\L.,)
,
"
-'
'
, -= ••;,'
C
c
()
sequence chart, controller/drive power-on 'See
MIM index
sequence chart, power-on 'See MIM index
sequence, drive motor/brake power 'See MIM
index
sequence, no power complete 'See MIM index
sequencer failed to detect bad parity during
diagnostic command 84 'See ECM index
sequencer failed to receive required data from the
COP 'See ECM index
sequencers
controller 'See MSM index
device 'See MSM index
sequencing panel, power INTRO-34
SERDES control check 'See ECM index
SERDES path check 'See ECM index
serial number labels 'See Index of INST
service call strategy INTRO-36
service checks, capacitor and rectifier 'See MIM
index
servicenotes 'See MIM index
service notes, power supply ripple problem 'See
MIM index
service procedures SC-FRU-1
service procedures, general instructions INTRO-1
servo cable check 'See ECM index
servo check 2 active prior to issuing rezero
command 'See ECM index
servo check 2 bit active before issuing search
command 'See ECM index
servo control check 'See ECM index
SC = 9700 through 9754 SC-FRU-19
SC = 9755 through 9783 SC-FRU-19
SC = 9784 t~rough 97FF SC-FRU-20
. servo failure during sweep or heat up routines of
power on 'See ECM index
servo inhibit 'See ECM ind~x
servo inhibit active during data check
recovery 'See ECM index
servo low-level and PLO signal cable assembly
MAP 'See MIM index
servo lOW-level signal analysis MAP 'See MIM
index
servo operations 'See MSM index
control 'See MSM index
servo power amplifier cable assembly MAP 'See
MIM index
servo power amplifier cable MAP 'See MIM index
servo power amplifier card 'See MIM index
servo power amplifier card voltage checks 'See
MIM index
PIN 4519911 39 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 1985
connector voltage measurement 'See MIM
index
on-card voltage measurement 'See MIM index
servo status - a bytes 12 and 13 'See MSM index
servo status-1 bytes 14 and 16 'See MSM index
servo surface 'See MSM index
servo system INTRO-30
set CE mode MD-7
set power switches 'See Index of INST
setlreset CE mode MD-?, MD-53
shipping lock 'See MIM index
shipping lock installation 'See MIM index
short busy time-out during device selection 'See
ECM index
single actuator maintenance (SAM) 'See MIM
index
single actuator maintenance (SAM) control
panel 'See MIM index
single actuator maintenance (SAM) power onloff
procedure 'See MIM index
single actuator maintenance control
panel INTRO-34
single actuator mechanism power-on/off
(front) 'See MIM index
single actuator mechanism power-on/off
(rear) 'See MIM index
soft start relay active after dropping it during
diagnostic routine 99 'See ECM index
soft start relay active at the beginning of diagnostic
routine 99 'See ECM index
soft start relay active at the beginning of motor
start 'See ECM index
soft start relay active at the end of diagnostic
routine 99 'See ECM index
soft start relay active when not expected 'See
ECM index
soft start relay inactive after pick during diagnostic
routine 99 'See ECM index
. soft start relay not active at beginning of motor
start 'See ECM index
solid failures INTRO-3'S
spare machine serial number labels 'See Index of
INST
split bus op line 'See MSM index
start procedures entry table START-1
start section entry table START-1
stal1 the installation procedure 'See Index of INST
status
storage control MD-78
status pending LED continuously on ISOLATE-33
MASTER INDEX-29
status, reporting and collection - power 'See MIM
index
storage control
installation verification 'See Index of INST
storage control attachment features 'See Index· of
INST
storage control microcode timed out on
deselection 'See ECM index
storage control status 'See Index of INST, MO-34,
MD-78
storage control unit MD-55
storage director ID 'See Index of INST
storage director ID (SDIO) INTRO-9,INTRO-11
storage path fenced MO-63
storage requirements 'See Index of INST
string address INTRO-9, INTRO-11, INTRO-25
string address switch 'See Index of INST
string addressing
2-path setting range 'See Index of INST
4-path setting range 'See Index of INST
4-path with 2-path setting range 'See Index of
INST
string configuration switch
4-path 'See Index of INST
4-path nondisruptive 'See Index of INST
string configuration switch, 4-path,
nondisruptive 'See Index of INST
string configuration switches
2-path 'See Index of INST
string configuration switches (4-path) REPAIR-22
string features, byte 8 'See MSM index
string 10 INTRO-9, INTRO-11, INTRO-23, INTRO-25,
'See MSM index, 'See Index of INST
string ID switches REPAIR-20
string 10 switches, set 'See Index of INST
string IDs 'See Index of INST
string power on/off REPAIR-25
string removal procedure, complete 'See Index of
INST
string summary PROG-1
string type switches, set 'See Index of INST
subsystem configuration (PSG Vol. R05, entry
G) 'See MIM index
subsystem configuration verification 'See Index of
INST
subsystem diagrams with 3880, mod 3 or mod
23 'See Index of INST
subsystem diagrams with 3990 'See Index of INST
subsystem exception report PROG-1
Subsystem failure determination START-4
subsystem 10 'See Index of INST
PIN 4519911 39 Sides
EC 476581 01Aug89
subsystem ID (SSID) INTRO-23, INTRO-25
subsystem integration test 'See Index of INST
subsystem microcode ordering information PRE-2
successful recovery of one type of uncorrectable
data check 'See ECM index
summary of 4-path conversion 'See Index of INST
summary, power control diagram 'See MIM index
supply, power diagrams 'See MIM index
supply, power operation 'See MIM index
support assistance REPAIR-13
switch
A-t'"'it power 'See Index of INST
device enable/disable switch test 'See Index of
INST
local/remote switch, SW487 'See Index of INST
nondisruptive A2, string configuration switch,
4-path 'See Index of INST
nondisruptive A3, string configuration switch,
4-path 'See Index of INST
nondisruptive A4, string configuration switch,
4-path 'See Index of INST
string address 'See Index of INST
string configuration, 2-path 'See Index of INST
string configuration, 4-path 'See Index of INST
string configuration, 4-path nondisruptive 'See
Index of INST
string ID 'See Index of INST
string type 'See Index of INST
SW487 'See Index of INST
UEPO switch test 'See Index of INST
UEPO switch test 2-path 'See Index of INST
switch adjustment, air flow 'See MIM index
switch error conditions, drive motor 'See MIM
index
switch settings, DDC/OTB card (2-path) REPAIR-18
switch settings, DDC/DTB card (4-path) REPAIR-19
switch, unit emergency INTRO-10
switches
controller power on/off INTRO-8, INTRO-22
device enable/disable INTRO-8, INTRO-22
enable/disable INTRO-7, INTRO-20
logic enable/disable INTRO-35
motor on/off INTRO-35
unit emergency INTRO-7, INTRO-20
24 Vdc panel INTRO-32
switches, device logic 'See MIM index
switches and LEOs INTRO-7, INTRO-20
SW487 'See Index of INST
symptom code
(see individual SC) SC-FRU-2
informational messages PROG-10
tf·· .
'''-- :
MASTER INDEX-30
r&:J Copyright IBM Corp. 1987. 1988
{/
()
c
c\
c\
c
c
c
C:
(~
c
(
c
C:
C'
c
r
summary PROG-16
symptom code analysis ISOLATE-23
symptom code assignments by sense format 'See
MSM index
symptom code indicated, 3380 MLX-2
symptom code summary PROG-1, PROG-16
symptom code to FRU charts SC-FRU-1
symptom code, enter MO-79
symptom codes
OFOX 'See MSM index
4XYY 'See ECM index
9020 'See ECM index
9101 'See ECM index
9102 'See ECM index
9104 'See ECM index
9108 'See ECM index
9120 'See ECM index
9180 'See ECM index
9201 'See ECM index
9202 'See ECM index
9204 'See ECM index
9210 'See ECM index
9220 'See ECM index
92XX 'See ECM index
9301 'See ECM index
9302 'See ECM index
9304 'See ECM index
9308 'See ECM index
9310 'See ECM index
9320 'See ECM index
9340 'See ECM index
9410 'See ECM index
9420 'See ECM index
9440 'See ECM index
9480 'See ECM index
9510 'See ECM index
9513 'See ECM index
9515 'See ECM index
9516 'See ECM index
9518 'See ECM index
951A 'See ECM index
952F 'See ECM index
9531 'See ECM index
9534 'See ECM index
9539 'See ECM index
953A 'See ECM index
9538 'See ECM index
953F 'See ECM index
9541 'See ECM index
9544 'See ECM index
9545 'See ECM index
PIN 4519911 39 Sides
© Copyright
EC 476581 01AugB9
IBM Corp, 1987, 1988
9546
9547
9548
9549
954C
9540
954F
9550
9552
9553
9554
9555
9556
9557
9558
9559
955A
9558
955C
9550
955E
950,
9560
9562
9563
9564
9565
9566
9567
9568
956A
9568
956C
9560
956E
956F
9574
9575
9576
9577
9578
9579
957 A
9578
957C
9570
957F
9580
9582
9584
9585
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
MASTER INDEX-31
9S86
9587
9588
9S8A
9S8B
9580
958E
9S8F
9592
9SA1
95A2
9SA3
9SA4
9SAS
9SA6
9SA7
9SA8
9SA9
95AA
95AC
9SAO
9SAE
9580
9S81
9S82
9583
9584
9585
9586
9587
9588
9589
958A
9S88
958C
9580
958E
958F
95CO
95C1
95C2
95C3
95C4
95C5
95C6
9SC7
9SC8
95C9
95CA
95CB
95CC
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
9SCO
9SCE
9SCF
9S01
9502
9503
9504
9505
9S06
9507
9508
9509
950A
9508
9S0C
9S00
9S0F
95F1
95F2
9SF3
95F4
95F6
95F8
95F9
95FC
95XX
97XX
9801
9802
9804
980R
9810
9820
9840
9CXX
90XX
9F10
9F11
9F12
9F13
9F14
9F15
9F16
9F18
9F19
9F18
9F1C
9F1F
9F96
9F97
9F9A
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
:See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
,
\.
",--
".~
"'4.. ...:j
IJ'C
'f
\it~
·f
~C~
PIN 4519911 39 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
~~
MASTER INDEX-32
.1
~
'"
--'------~~~-~
._----"------
c
C
c\
C)
C:
C'
C'
Ci
£:
C
C'
()
C
(
(
(~
(
C
(
(
(,'
(.'
9F9E
9FFF
9XXX
011X
012X
0140
0180
0201
0202
0208
0210
0220
0240
0260
0280
02AO
02CO
02EO
0308
0310
0340
0380
0401
0402
0408
0410
0420
0440
0460
0480
04AO
04CO
04EO
0508
0510
0540
0580
0602
0604
0610
0612
0618
061A
0620
0622
0632
0634
0636
0670
0672
0674
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
PIN 4519911 39 Sides
© Copyright
EC 476581 01 Aug89
IBM Corp. 1987. 1988
06CE
06XX
07XX
0802
0804
0810
0812
0818
081A
0820
0822
0832
0834
0836
0870
0872
0874
08CE
08XX
09XX
OF1X
OF2X
OF4X
OF8X
OFFF
OFX1
OFX2
OFX3
OFX4
OFX5
OFX6
OFX7
OFX8
OFX9
OFXA
OFXB
OFXC
OFXO
E01X
E02X
EO·A'~
E102
E104
E108
E110
E120
E140
E180
E220
E240
E280
iSee ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECMindex
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
'See ECM index
MASTER INDEX-33
("
E301
E302
E304
E308
E310
E340
E502
E504
E508
E510
E520
E540
E580
E610
E620
E640
E680
E701
E720
E780
E801
E802
E803
E804
E805
E806
E807
E810
E814
E817
E818
E819
E820
E821
E822
E851
E852
E853
E855
E856
E857
E858
E881
E882
E883
E884
E885
E886
E887
E890
E894
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
'See
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
ECM
PIN 4519911 39 Sides
© Copyright
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
index
EC 476581 01Aug89
IBM Corp. 1987. 1988
E897 'See ECM index
E898 'See ECM index
E899 'See ECM index
E8AO 'See ECM index
E8A1 'See ECM index
E8A2 'See ECM index
E8XX 'See ECM index
EBXX 'See ECM index
ECXX 'See ECM index
ED01 'See ECM index
ED06 'See ECM index
ED07 'See ECM index
ED08 'See ECM index
ED10 'See ECM index
ED11 'See ECM index
ED13 'See ECM index
ED14 'See ECM index
ED16 'See ECM index
ED17 'See ECM index
ED18 'See ECM index
ED19 'See ECM index
ED1B 'See ECM index
ED20 'See ECM index
ED81 'See ECM index
ED87 'See ECMindex
ED90 'See ECM index
ED97 'See ECM index
ED98 'See ECM index
ED99 'See ECM index
EDAO 'See ECM index
EDXX 'See ECM index
EEXX 'See ECM index
EF81 'See ECM index
EF86 'See ECM index
EF87 'See ECM index
EF88 'See ECM index
EF89 'See ECM index
EF8A 'See ECM index
EFFO 'See ECM index
EFFF 'See ECM index
symptom codes, prioritizing START-12, START-19
symptom OFOX 'See MSM index
sync byte missing 'See ECM index
sync bytes
reading and recognizing 'See MSM index
writing 'See MSM index
sync in check 'See ECM index
sync out/in tag counts not equal 'See ECM index
SYSGEN 'See Index of INST
system error summary (part 2) report PROG-2
system error summary part 2 PROG-1
MASTER INDEX-34
r
It. .
I
\"
~./
/4'
~,
C{~
:[/
.
( 'w'\
system reference manuals
'See Index of INST
c
C:
C'
()
c
c\
('
c
(
~:
table, MAP section entry 'See MIM index
tag in check 'See ECM index
tags
CDP or device interface 'See MSM index
in 'See MSM index
out 'See MSM index
DDC or control interface 'See MSM index
in 'See MSM index
out 'See MSM index
tailgate 'See Index of INST
tailgate connectors 'See MIM index
temperature check, HDA 'See MIM index, 'See
Index of INST
temperature conversion chart 'See MIM index
temperature strip, HDA 'See MIM index
temporary equipment check definition 'See MSM
index
temporary offset invk PROG-13
test equipment and tools INTRO-40
the actuator moved too far Juring the safe 'See
ECM index
the DCP either did not take or it did not complete
the safe routine 'See ECM index
threshold logging PROG-13
time between record ready interrupts too long
during a search 'See ECM index
time between record ready interrupts too short
during a search 'See ECM index
time out routine, DDC tag out at selected null with
CDP tag in at valid 'See ECM index
timeout or sense fault log, DDC tag out lines in
invalid state 'See ECM index
timeout routine, 'allow selection' off, with 'device
selected' on 'See ECM index
tools and test equipment INTRO-40, 'See Index of
INST
track 'See MSM index
track follow process 'See MSM index
track following not acquired at the end of
rezero 'See ECM index
track format 'See MSM index
count, key, data 'See MSM index
home address 'See MSM index
track layout
guardband INTRO-28
PIN 4519911 39 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
models AJ4 and BJ4 INTRO-27
models AK4 and BK4 INTRO-28
track repair, CE MD-32
tracking, belt 'See MIM index
transferring ECC correction bytes 'See MSM
index
transformer connections
A-unit. 'See Index of INST
B-unit 'See Index of INST
transformer connections, SO hertz 'See Index of
INST
transformer, voltage settings, A-unit 'See Index of
INST
tripped CB or CP failure analysis 'See MIM index
CB200 trip isolation 'See MIM index
entry point A 'See MIM index
CB201 trip isolation 'See MIM index .
entry point B 'See MIM index
checks 'See MIM index
circuit overload 'See MIM index
controller power CP trip analysis 'See MIM
index
CP20S trip isolation 'See MIM index
entry point C 'See MIM index
CP206 trip isolation 'See MIM index
entry point D 'See MIM index
CP207 trip isolation 'See MIM index
entry point E 'See MIM index
CP208 trip isolation 'See MIM index
entry point F 'See MIM index
CP30S (controller A1/A3) trip isolation 'See
MIM index
entry point G 'See MIM index
CP30S (controller A2/ A4) trip isolation 'See
MIM index
entry point H 'See MIM index
CP30S/CP306 (controller A 1/A3) trip
isolation 'See MIM index
entry point I 'See MIM index
CP30S/CP306 (controller A2/A4) trip
isolation 'See MIM index
entry point J 'See MIM index
CP306 (controller A1/A3) trip isolation 'See
MIM index
entry point K 'See MIM index
CP306 (controller A2/ A4) trip isolation 'See
MIM index
entry point L 'See MIM index
CP306/CP307 both tripped (controller A1/A3) trip
isolation 'See MIM index
entry point U 'See MIM index
MASTER INDEX-3S
(
((
CP306/CP307 both tripped (controller A2/A4) trip
isolation 'See MIM index
entry point V 'See MIM index
CP307 (controller A 1/A3) trip isolation 'See
MIM index
entry point M 'See MIM index
CP307 (controller A2/A4) trip isolation 'See
MIM index
entry point N 'See MIM index
CP405 trip isolation 'See MIM index
entry point 0 'See MIM index
CP406 trip isolation 'See MIM index
entry point P 'See MIM index
CP410 trip isolation 'See MIM index
entry point Q 'See MIM index
CP411 trip isolation 'See MIM index
entry point R 'See MIM index
CP412 trip isolation 'See MIM index
entry point S 'See MIM index
CP602 through CP636 trip isolation 'See MIM
index
entry point T 'See MIM index
general trip analysis information 'See MIM
index
limits 'See MIM index
motor start sequence not normal 'See MIM
index
optional short-circuit analysis procedure 'See
MIM index
thermal indicating CPs 'See MIM index
tripping board voltage CPs 'See MIM index
unequal power load distribution 'See MIM
index
vibration 'See MIM index
tripped, CB or CP failure analysis 'See MIM index
trouble setting CE mode ISOlATE-14
T3880A OlT
OlT 'See Index of INST
UCW 'See Index of INST
UEPO cable 'See Index of INST
UEPO cable removal 'See Index of INST
UEPO control 'See Index of INST
UEPO panel (rear) 'See MIM index
UEPO switch test 'See Index of INST
UEPO switch test 2-path 'See Index of INST
undervoltage
PIN 4519911 39 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987. 1988
SC= EFFO and EFFF SC-FRU-49
undervoltage detection INTRO-36, 'See Index of
INST
undervoltage, checking for ISOlATE-7
undervoltage, checking for with data
checks ISOlATE-15
unexpected end-op response code received 'See
ECM index
unexpected offset active during data check
recovery 'See ECM index
Unexpected R/W mode active 'See ECM index
unexpected sector compare during data check
recovery 'See ECM index
unit. check description, bytes 0,1,2 'See MSM
index
unit emergency (UEPO) switch safety
'
inspection 'See MIM index
unit emergency power off (UEPO) - rear 'See MIM
index
unit emergency switch INTRO-S, INTRO-10,
INTRO-22, INTRO-24, 'See MIM index
unit power on/off REPAIR-26
unlock HOA pulley 'See Index of INST
unlock HOA pulley shipping lock 'See Index of
INST
unpack 'See Index of INST
unselected device check 1 'See ECM index
updating records 'See Index of INST
using the error log to determine failing devices and
controllers ISOlATE-20
VARY system path to off or on 'See Index of INST
views, front and rear 'See MIM index
vital product data, reconfiguration 'See Index of
INST
voice coil interference 'See MIM index
voice coil-to-VCM interference 'See MIM index
voice-coil contamination 'See MIM index
voice-coil magnet 'See MIM index
voltage
and clock lEOs, A board INTRO-30
voltage analysis, power supply problem 'See MIM
index
voltage check, read/write channel board 'See
MIM index
voltage checks 'See MIM index
ac ripple 'See MIM index
MASTER INDEX-36
(
c
c
()
c
c
c
c
«=
c
c
alternate chart - drive 'See MIM index
common 'See MIM index
controller - A 1 board 'See MIM index
drive - B1 board 'See MIM index
general 'See MIM index
R/W channel - drive 'See MIM index
servo power amplifier 'See MIM index
voltage checks, power amplifier card - servo 'See
MIM index
voltage checks, servo power amplifier card 'See
MIM index
voltage problem, powersupply analysis 'See MIM
index
voltage regulator card LEDs INTRO-32
voltage regulator card, controller 'See MIM index
voltage ripple measurement example 'See MIM
index
voltage ripple measurement procedure 'See MIM
index
excess ripple on one dc voltage 'See MIM
index
excessive high-frequency noise 'See MIM
index
excessive ripple on two different dc
voltages 'See MIM index
measure ac ripple 'See MIM index
other dc voltage ripple problems 'See MIM
index
.
ripple problem analysis 'See MIM index
voltage out of tolerance when ripple is
included 'See MIM index
voltage/high frequency noise measurement 'See
MIM index
voltages - ac, dc, common, controller, drive, ripple,
R/W channel 'See MIM index
volume fenced ISOLATE-51
volume fenced from storage path - console
message ISOLATE-51
(
c
c\
~/
weight, HDA 'See MIM index
working with ESD-sensitive ::>arts 'See MIM index
write address mark active when not
expected 'See ECM index
write gap 3 control check 'See ECM index
write gate active 'See ECM index
write inhibit MD-63
PIN 4519911 39 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
write inhibit or DCP servo write inhibit active at the
end of a zero offset routine 'See ECM index
write inhibit signal or DCP servowrite inhibit bit in
wrong state during offset routine 'See ECM index
write inhibit signal or DCP servo write inhibit bit in
wrong state during seek routine 'See ECM index
write inhibited - console message ISOLATE-49
write mode verify 'See ECM index
write overrun check 'See ECM index
write paths 'See MSM index
I
Numerics
I
2-path or 4-path configurations 'See Index of INST
24 Vdc panel INTRO-32
3380 Direct Access Storage
characteristics INTRO-26
description INTRO-1
diskettes PRE-2
models INTRO-1, INTRO-26
symptom code indicated MLX-2
3380-JK
subsystem microcode ordering
information PRE-2
3380-JK and 3380-DE differences 'See MSM index
3380-JK model description INTRO-1
3380-JK power-on/off procedures 'See MIM index
3380-JK procedures 'See MIM index
controller A 1(A3). power-off 'See MIM index
controller A1(A3), power-on 'See MIM index
controller A2(A4). power-off 'See MIM index
controller A2(A4). power-on 'See MIM index
controller power-on/off 'See MIM index
3380-JK string power-off 'See MIM index
3380-JK string power-on 'See MIM index
3380-JK string 'See MSM index
3380-JK string configurations 'See Index of INST
3880 diskette compatibility MD-83
3880 diskettes PRE-2
3880 requirements INTRO-3
3990 requirements INTRO-12
3990 vital product data, reconfiguration 'See Index
of INST
3990, CTL-I check 1 MAP 'See MIM index
3990, CTL-I check 2 MAP 'See MIM index
4-path conversion summary 'See Index of INST
5 volt common circuit protectors INTRO-35
5/24 Vdc bootstrap power supply 'See MIM index
MASTER INDEX-37
--~--------
50 hertz convenience outlet and +24 Vdc
distribution 'See MIM index
50 hertz Japan, ac drive 'See MIM index
50 hertz Japan, primary power box - A unit
MIM index
50 hertz transformer connections, DASD
functions 'See Index of INST
'See
50 hertz, primary power box - A unit 'See MIM
index
60 hertz (50 hertz Japan), primary power box - A
unit 'See MIM index
60 hertz convenience outlet and + 24 Vdc
distribution 'See MIM index
60 hertz, ac drive 'See MIM index
9F10 symptom code 'See ECM index
.~.
c
PIN 4519911 39 Sides
© Copyright
EC 476581 01Aug89
IBM Corp. 1987, 19S8
MASTER INDEX-38
c
c
c
c·
c
c
EC History of Master Index
EC HISTORY OF P/N 4519911
EC Number
Date Of EC
EC Number
475245
14Nov86
475248
21Ju187
476573
475246
475247
llSep87
476581
Date Of EC
25Apr88
30Nov88
01Aug89
Notes:
c
c
(:
C\
~/
C
£
£.'
PIN 4519911 39 Sides
EC 476581 01Aug89
© Copyright IBM Corp. 1987, 1988
MASTER INDEX-39
,f'
'1\._ "
.1
··ct . .
o
READER'S COMMENT FORM
You may use this form to communicate your comments about this publication, its organization, or subject matter, with the
understanding that IBM may use or distribute whatever information you supply in any way it believes appropriate without incurring
any obligation to you.
Your comments will be sent to the author's department for whatever review and action, if any, is deemed appropriate. Comments
may be written in your own language; use of English is not required.
Note: Copies of IBM publications are not stocked at the location to which this form is addressed. Please direct any requests for
copies of publications, or for assistance in using your IBM system, to your IBM representative or to the IBM Branch Office serving
your locality.
Machine Type/Model
c
(
Serial Number
Storage Control
or Processina Unit
Symptom/Error Code
PagelD
Page EC Number
If no symptom or error code, check here
0
Describe machine problem as it was reported to you, or as it appeared to you.
,
/
c
(
\
"
C:
Describe the problem you encountered with the documentation.
()
Want an acknowledgement?
()
For plant use only
Date received _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
Fill in the following information:
Name _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
Sender acknowledged _ _ _ _ _ _ _ _ _ _ _ __
Occupation _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
Research completed _ _ _ _ _ _ _ _ _ _ _ _ __
B/O Number_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
Incorporated by EC _ _ _ _ _ _ _ _ _ _ _ _ __
B/O Address
(streed
()
(city, state, zip code, countryl
Thank you for your cooperation. No postage stamp necessary if mailed in the U.S.A. (Elsewhere, an IBM representative will be
happy to forward your comments.)
PIN 4519912 4 Sides
EC 476581
01Aug89
©Copyright IBM Corporation 1986
Reader's Comment Form RCF-1
()
READER'S COMMENT FORM
,(:.,
;-'
c
(f'
"''4L,
Fold and Tape
Fold and Tape
Please Do Not Staple
------------------------------------------------------------
III
NO POSTAGE
NECESSARY
IF MAILED
IN THE
UNITED STATES
BUSINESS REPLY MAIL
FIRST CLASS
PERMIT NO. 40
ARMONK, N.Y.
POSTAGE WILL BE PAID BY ADDRESSEE:
International Business Machines Corporation
Department G26/61C
5600 Cottle Road
San Jose, California 95193
-----------------------------------------------------~
Fold and Tape
Please Do Not Staple
Fold and Tape ------
-------- ----'--------,
--
-~--®
~
RCF-2 Reader's Comment Form
PIN 4519912 4 Sides
EC 476581 01Aug89
© Copyright IBM Corporation 1986
( :
Source Exif Data:
File Type : PDF File Type Extension : pdf MIME Type : application/pdf PDF Version : 1.3 Linearized : No XMP Toolkit : Adobe XMP Core 4.2.1-c041 52.342996, 2008/05/07-21:37:19 Create Date : 2018:05:24 13:50:42-08:00 Modify Date : 2018:05:24 14:16:14-07:00 Metadata Date : 2018:05:24 14:16:14-07:00 Producer : Adobe Acrobat 9.0 Paper Capture Plug-in Format : application/pdf Document ID : uuid:4e3f547a-1112-4347-8668-70da898b40af Instance ID : uuid:8cbe8858-93ab-6e4a-bdb8-cec83b6d8ed3 Page Layout : SinglePage Page Mode : UseNone Page Count : 404EXIF Metadata provided by EXIF.tools