SC19 6202 3_VM_SP_Operators_Guide_Release_4_Dec84 3 VM SP Operators Guide Release 4 Dec84
User Manual: SC19-6202-3_VM_SP_Operators_Guide_Release_4_Dec84
Open the PDF directly: View PDF 
.
Page Count: 406
| Download | |
| Open PDF In Browser | View PDF |
---------_.-------------- - ----...
Virtual Machine /
System Product
Operatorls Guide
Release 4
SC 19-6202-3
Notice: The term VM/SP used in this publication refers to VM/SP when used in conjunction with VM/370 Release 6.
/-
Fourth Edition (December 1984)
This edition, SC19-6202-3, is a major revision of SC19-6202-2, and applies to Release 4
of Virtual Machine/System Product (VM/SP), program number 5664-167, and to all subsequent releases and modifications until otherwise indicated in new editions or Technical
Newsletters. Changes are made periodically to the information herein; before using this
publication in connection with the operation of IBM systems, consult the latest IBM
System/370 and 4300 Processors Bibliography, GC20-0001, for the editions that are applicable and current.
Summary of
Changes
For a list of changes, see the following page.
Changes or additions to text and illustrations are indicated by a vertical line to the left of
the change.
References in this publication to IBM products, programs, or services do not imply that
IBM intends to make these available in all countries in which IBM operates. Any reference to an IBM program product in this publication is not intended to state or imply that
only IBM's program product may be used. Any functionally equivalent program may be
used instead.
Ordering Publications
Publications are not stocked at the address given below. Requests for IBM publications
should be made to your IBM representative or to the IBM branch office serving your
locality.
A form for readers' comments is provided at the back of this publication. If the form has
been removed, comments may be addressed to IBM Corporation, Programming Publications, Dept. G60, P.O. Box 6, Endicott, NY, U.S.A. 13760. IBM may use or distribute
whatever information you supply in any way it believes appropriate without incurring any
obligation to you.
© Copyright International Business Machines Corporation 1980, 1981, 1982, 1983, 1984
Summary of Changes
Summary of Changes
for SC19-6202-3
for VM/SP Release 4
To obtain editions of this publication that pertain to earlier releases of VM/SP, you
must order using the pseudo-number assigned to the respective edition. For:
•
Release 3, order STOO-1351
•
Release 2, order SQ 19-6202.
Hardware Support
Information was added for the support of:
•
•
•
•
•
•
The 3290 Information Panel.
The 4361 Model Group 4 and 5 Processors.
The 4381 Model Group 1 and 2 Processors.
The 3370 Direct Access Storage Models A2 and B2.
The 4248 Printer.
The 3800 Model 3 Printing Subsystem - Existing programs designed to
produce 3800 Modell printer output may produce output for the 3800
Model 3 printer with little or no program change. Use of this support provides improved print quality (240 x 240 pel resolution) and the addition of
a 10 line-per-inch (LPI) vertical space option.
The 3480 Magnetic Tape Subsystem.
Program Support
Saved System 8 M Byte Limit Removal
Changed: SAVESYS, VMSAVE and IPL functions have been modified to
let a page image copy of up to a 16 M byte virtual machine be saved and
restored. Also, the NAMESYS MACRO was changed to enforce the 16 M
byte limit.
Shared/ Nonshared Restriction
Changed: Any attempt to construct a virtual device configuration that
would mix SHARED and NON SHARED device types on the same virtual
control unit is rejected. To permit migration of data on 3420 tapes (a
shared device) to the 3480 (a nonshared device), this restriction will not be
enforced.
Summary of Changes
iii
VM/SP Interactive Problem Control System
Changed: The base VM/IPCS component is enhanced to include the func-
tional equivalent to the VM/IPCS Extension licensed program.
System Initialization
Changed: A new operand was added to the SHUTDOWN command.
Stand-Alone Dump
New: This facility enhances serviceability by letting support personnel dump
up to 16 M of real storage.
Programmable Operator
Changed: This facility can now be used in a mixed environment and the
commands can be used from an NCCF operator station. Slight changes
were added to the LOADTBL and QUERY commands, and a new
command, LGLOPR, was added.
CPTRAP Command
Changed: Two new operands were added to the CPTRAP command:
Groupid and Wrap.
Miscellaneous
Changed Format: The format of the command descriptions in "Section 3. CP
Commands" on page 3-1 has changed to include a "When to Use" section.
Former usage notes have now become a part of the operand description or have
been included in a "Things You Should Know" section. This new organization
is not marked with revision bars since it effects all of Section 3.
Summary of Changes
for SC19-6202-2
for VM/SP Release 3
Program Support
New: A "text" option was added to the SET command to enhance the capa-
bility of the LOGMSG operand.
Information was added for using the MSGNOH command.
Improved functions for the programmable operator facility are explained.
Hardwore Support
New: Information was added for the support of the 3262, 3289-E, and 4245
printers and the 3430 tape drive.
The IBM 3088 Multisystem Communication Unit interconnects multiple
systems using block multiplexer channels. The 3088 uses an unshared subchannel for each unique address and is fully compatible with existing channelto-channel protocol.
iv
VM/SP Operator's Guide
Miscellaneous
Changed: The information for the INDICATE, NETWORK, QUERY, and SET
commands was reorganized to make it easier to determine the operands that
can be issued by each privilege class user and to locate information about the
command operands and system responses.
Various minor technical and editorial changes have been made throughout the
publication.
Summary of Changes
for SC19-6202-1
as Updated by SN24-5735
for VM/SP Release 2
Program Support
New: The missing interrupt handler has been extended so that CP not only
detects missing interrupt conditions, but also attempts to correct them. CP
informs the system operator whether or not the correction action was successful.
To help give you optimum system availability, the missing interrupt handler
allows you to vary the time interval allowed for I/O completion for the supported devices.
Miscellaneous
Changed: Minor technical and editorial changes have been made throughout
this publication.
Summary of Changes
for SC19-6202-1
for VM/SP Release 2
Programmable Operator Facility
New: This facility provides the capability to: log messages, suppress messages,
redirect messages, execute messages, or preprogram message responses. The
capabilities are under control of an editable message routine table in a CMS
file.
Inter-User Communication Vehicle (IUCV) enhancements for message handling are also included.
CMS Nucleus Restructure, and Removal of the CMS Tokenization Eight-Byte
Restriction
New: The restructured nucleus provides a CMS system that is more flexible and
extendable for development, serviceability and maintenance purposes.
The eight-byte tokenize restriction has been removed for parameter passing.
Summary of Changes
V
Trace Table Recording Facility
New: This facility allows service personnel and system programmers to create a
chronological READER spool file of CP trace entries by type, VMBLOK
address, interrupt code, and device address.
Miscellaneous
Changed: Minor technical and editorial changes have been made throughout
this publication.
vi
VM/SP Operator's Guide
Preface
This book is for those persons responsible for the operation and administration of a
Virtual Machine/System Product (VM/SP) system.
This operator's guide describes some of the hardware of the System/370 as well as
some user virtual machine requirements. For general information on how to run
batch jobs in VM/SP virtual machines (other than CMS), refer to the Virtual
Machine/System Product: Running Guest Operation Systems, GC19-6212.
To manage VM/SP resources effectively, a system operator should be familiar with
the operation and hardware requirements of System/370, 4300, and 303X
input/ output devices and should know the broad concepts of computer operations.
In addition, the system operator should be familiar with the use and capability of
the Interactive Problem Control System component of VM/SP. Details of this
major component are in the VM/SP Interactive Problem Control System Guide,
SC24-S260.
Note: For VM/SP users, VM/SP IPCS is more effective than the IPCS Extension
Program Product (S748-SAl).
If your installation uses Remote Spooling Communications Subsystem (RSCS) Net-
working, the system operator is generally the operator of the RSCS virtual machine
as well as of the real machine at the installation. For details on how to operate the
RSCS virtual machine, refer to the VM / SP Program Reference and Operations
Manual, SH24-S00S.
Note: For more effective use of RSCS in VM/SP, the RSCS Networking Program
Product (S748-XPl) is recommended.
To help the system operator gain insight into virtual machine concepts and capabilities, refer to the Virtual Machine/System Product: Introduction, GC19-6200. This
book briefly discusses virtual storage, virtual machine features and facilities of
VM/SP.
This publication describes VM/SP system startup, logon, VM/SP commands,
special service programs, user privilege classes, as well as other data necessary to
manage the resources of the system on a daily basis. Operator commands and
other operator related tasks for the Group Control System (GCS) in VM/SP
Group Control System Guide, SC24-S249.
There· are five sections in this publication. The first section discusses the Operational control of VM/SP under normal and adverse conditions. The remaining four
sections are the "how-to-do-it" sections. Wherever possible, subject matter
adheres to an alphabetic arrangement. The beginning of Section 3, "CP
Preface
vii
Commands" on page 3-1 lists the operator commands and summarizes what each
command does.
Terminology
The user privilege classes referred to in this book are IBM-defined classes. If your
installation restructures the classes, see your installation administrator.
The following terms in this publication refer to the indicated support devices:
•
"2305" -
•
"270x"
IBM 2305 Fixed Head Storage, Models 1 and 2.
IBM 2701, 2702, and 2703 Transmission Control Units
The Integrated Communications Adapter (ICA) on some IBM CPUs.
•
"3330"
IBM 3330 Disk Storage, Models 1, 2, or 11
IBM 3333 Disk Storage and Control, Models 1 or 11
3350 Direct Access Storage operating in 3330/3333 Modell or
3330/3333 Model 11 compatibility mode.
•
"3340"
IBM 3340 Disk Storage, Models A2, B1, and B2.
IBM 3344 Direct Access Storage Model B2.
•
"3350" mode.
IBM 3350 Direct Access Storage Models A2 and B2 in native
•
"FB-512" or "FBA" -
•
"3289" -
IBM 3289 Model 4 Printer.
•
"3262" -
IBM 3262 Printer, Models 1 and 11.
IBM 3310 and 3370 Direct Access Devices.
"3704," "3705," "3725," or "37xx" nications Controllers.
IBM 3704, 3705, and 3725 Commu-
"3705" -
3705 I and the 3705 II, unless otherwise noted.
"2741" -
IBM 2741 and the 3767, unless otherwise specified.
"3800" - IBM 3800 Printing Subsystems, Models 1, 3, and 8. A specific
device type is named only when a distinction is required between device types.
"3800 Model 3" - IBM 3800 Models 3 and 8, unless otherwise stated.
(The IBM 3800 Model 8 is available only in selected world trade countries.)
viii
VM/SP Operator's Guide
•
"3850" -
IBM 3850 Mass Storage System.
•
"3270" - A series of display devices, namely the IBM 3275, 3276, 3277,
3278 and 3279 Display Stations, and the 3290 Information Panel. A specific
device type is named only when a distinction is needed among device types.
"3370" -
IBM 3370 Direct Access Storage Models AI, A2, Bl, and B2.
"3375" -
IBM 3375 Direct Access Storage Device.
"3380" - IBM 3380 Direct Access Storage device. Information on the IBM
3380 Direct Access Storage device is for planning purposes only until the availability of the product.
If the 3380 attached to the 3880 Controller Model 3 with Speed Matching
Buffer (Feature #6550) is part of your installation, CP will permit execution of
extended count-key-data channel programs.
•
"3081" -
IBM 3081 Processor complex.
"3033" - all 3033 processor models including the IBM 3033 Model Group S
processors. Information about the IBM 3033 Model Group S processors is for
planning purposes only until the availability of the product.
•
"3880" -
IBM 3880 Storage Control Unit.
•
"3480" -
IBM 3480 Magnetic Tape Subsystem.
•
"4248" -
IBM 4248 printer.
•
"Attached processors" also refers to multiprocessors unless otherwise indi":
cated.
•
"VSM" refers to the VTAM service machine, a service virtual machine that
lets SNA (Systems Network Architecture) terminals access the VM/SP system.
•
"VSE" refers to the combination of the DOS/VSE system control program
and the VSE/ Advanced Functions program product. Also includes the
VSE/System Product.
•
"DOS," in certain cases, is still used as a generic term. For ex;;t,mple, disk
packs initialized for use with VSE or any predecessor DOS or DOS/VS system
may be referred to as DOS disks.
•
"CMS/DOS" continues to be used when referring to the DOS-like simulation
environment provided under the CMS component of the VM/System Product.
Information about display terminal usage also applies to the IBM 3138, 3148, and
3158 Display Consoles when used in display mode, unless otherwise noted.
Any information pertaining to the IBM 3284 or 3286 also pertains to the IBM
3287, 3288 and the 3289 printers, unless otherwise noted.
Preface
ix
For the relationship of the Virtual Machine/System Product: Operator's Guide to
other VM/SP prerequisite and corequisite publications, refer to "The VM/SP
Library."
Prerequisite Publications
Virtual Machine/System Product:
Introduction, GC19-6200
Terminal Reference, GC19-6206
CMS User's Guide, SC19-6210
If the 3767 Communications Terminal is used as the alternate system console, the
IBM 3767 Operator's Guide, GA18-2000, is also a prerequisite.
The Virtual Machine/System Product: CMS User's Guide is a prerequisite only if
the operator edits files used for system control or accounting purposes.
If the VTAM Communications Network Application (VM/VCNA) product is
used, the following publications are prerequisites:
Communication Network Application General Information Manual, GC27-0501
Communication Network Application Installation, Operation, and Terminal Use,
SC27-0502.
If the ACF/VTAM Release 3 (ACF/VTAM Version 3 for VM/SP with VM SNA
Console Support (VSCS)) product is used, the following publications are prerequisites:
ACF/VTAM General Information (for VM/SP), GC30-3246
ACF/VTAM Network Program Products Planning (MVS Base), SC23-0110
ACF/VTAM Installation and Resource Definition (MVS Base), SC23-0111
ACF/VTAM Customization (MVS Base), SC23-0112
ACF/VTAM Operation (MVS Base), SC23-0113.
Corequisite Publications
Virtual Machine/System Product:
Planning Guide and Reference, SC19-6201
CMS Command and Macro Reference, SC19-6209
x
VM/SP Operator's Guide
CP Command Reference for General Users, SC 19-6211
System Product Editor User's Guide, SC24-5220
Group Control System Guide, SC24-5249
System Product Editor Command and Macro Reference, SC24-5221
Installation Guide, SC24-5237
System Product Interpreter User's Guide, SC24-5238
System Product Interpreter Reference, SC24-5239
Release 4 Guide, SC24-5240
Distributed Data Processing Guide, SC24-5241
System Logic and Problem Determination Guide Volume 1 (CP), LY20-0892
Data Areas and Control Block Logic Volume 1 (CP), LY24-5220
Data Areas and Control Block Logic Volume 2 (CMS), LY24-5221
Device Support Facility User's Guide and Reference, GC35-0033
If your installation has a 3850 Mass Storage System, the following are corequisite
publications:
IBM 3850 Mass Storage System (MSS): Principles of Operation: Theory,
GA32-0035
IBM 3850 Mass Storage System (MSS): Principles of Operation, GA32-0036
IBM 3850 Mass Storage System (MSS): Introduction and Preinstallation Planning, GA32-0038
Operators Library: IBM 3850 Mass Storage System (MSS) Under OS/VS,
GC35-0014
OS/VS Message Library: Mass Storage System (MSS) Messages, GC38-1000.
Information on commands that control the virtual machine and its resources applicable to the G privilege class user are contained in the following publications:
Virtual Machine/System Product:
CP Command Reference for General Users, SC19-6211
Running Guest Operating Systems, GC19-6212
System Programmer's Guide, SC19-6203
Preface
xi
System Messages and Codes, SC 19-6204
OLTSEP and Error Recording Guide, SC19-6205.
The latter publication contains information on the VM/SP error recording process.
It also describes briefly the use of CPEREP, the CMS command that uses EREP
for editing and printing SYSl.LOGREC and VM/SP error recording records.
Detailed information on the use of the operands of the CPEREP command is contained in Environmental Recording Editing and Printing (EREP) Program,
GC28-1178.
Other corequisite publications include:
2821 Control Unit Component Description, GA24-3312
IBM 3211 Printer 3216 Interchangeable Train Cartridge & 3811 Printer Control
Unit Compo Desc. and Operator's Guide, GA24-3543
3262 Printer Models 1 and 11 Component Description, GA24-3733
Virtual Machine/System Product: Library Guide, Glossary, and Master Index,
GC19-6207
Virtual Machine/System Product: Program Reference and Operations Manual,
SH24-5005
/",--
3800 Printing Subsystem Programmer's Guide, GC26-3846
OS/VS and DOS/VSE Analysis Program-l (AP-l) User's Guide, GC26-3855
VM/SP Interactive Problem Control System Guide, SC24-5260
The following are corequisite publications if your installation has a 3704, 3705 or
3725:
Guide to Using the IBM 3704 Communications Controller Control Panel,
GA27-3086
Guide to Using the 3705 Control Panel, GA27 -3087.
ACP/NCP-SSP, V3 Installation and Resource Definition Guide, SC30-3253.
EP / 3725 Installation and Resource Definition Guide and Reference, SC30-3172.
EP / 3705 Generation and Utilities Guide and Reference, GC30-3242.
ACF/NCP V4, ACF/SSP V3 Diagnosis Guide, SC30-3255.
Note: References in the text to prerequisite and corequisite VM/SP publications
will be given in abbreviated form.
xii
VM/SP Operator's Guide
Preface
xiii
The VM/SP Library
Evaluation
Index
GENERAl
INFORMATION
INTRODUCTION
LIBRARY
GUIDE,
GLOSSARY,
AND
MASTER INDEX
GC20-1838
GC19-6200
GC19-6207
Planning
Installation
PLANNING
GUIDE AND
REFERENCE
RUNNING
GUEST
OPERATING
SYSTEMS
DISTRIBUTED
DATA
PROCESSING
GUIDE
RB..EASE 4
GUIDE
INSTALLATION
GUIDE
SYSTEM
DEFINmON
RLES
SC19-6201
GC19-6212
SC204-52o41
SC2-4-5248
SC24-5237
SC24-5256
Administration
Operation
Applications
SYSTEM
PROGRAMMER'S
GUIDE
GCS
GUIDE
GCS
MACRO
REFERENCE
OPERATOR'S
GUIDE
APPLICATION
DEVELOPMENT
GUIDE
SC19-6203
SC2-4-52049
SC24-5250
SC19-6202
SC2-4-5247
TERMINAL
REFERENCE
CMS
PRIMER
CMS PRIMER
FOR
LlNEORIENTED
TERMINALS
CMS
USER'S
GUIDE
GC19-6206
SC204-5236
SC24-5242
SC19-6210
CMS
COMMAND
AND MACRO
REFERENCE
SP EDITOR
USER'S GUIDE
SP EDITOR
COMMAND
AND MACRO
REFERENCE
CP
COMMAND
REFERENCE
SC19-6209
SC2-4-5220
SC24-5221
SC19-6211
SP
INTERPRETER
USER'S GUIDE
SP
INTERPRETER
REFERENCE
EXEC 2
REFERENCE
SC2-4-5238
SC2-4-5239
SC2-4-6219
End Use
Reference Summaries
To order all the Referenc6 Summaries, use order number 380F 3221
------------------------------------------------------II
I
I
II
QUICK
REFERENCE
I
I
I
SX20-4400
I
I
I
I
I
I
I
I
1______ -----------------------------------------------_
xiv
VM/SP Operator's Guide
Diagnosis
SYSTEM
MESSAGES
AND CODES
SYSTEM
MESSAGES
CROSSREFERENCE
OLTSEP
AND ERROR
RECORDING
GUIDE
IPCS
GUIDE
SERVICE
ROUTlNES
PROGRAM
LOGIC
SC19-e2~
SC24-52S-4
SC19-6205
SC24-52S0
LY20-0890
PROBLEM
DETERMINAnON
VOL, 1 (CP)
DATA AREAS
AND CONTROL
BLOCKS
VOL 1 (CP)
PROBLEM
DETERMINAnON
VOL 2 (CMS)
DATA AREAS
AND CONTROL
BLOCKS
VOL 2 (CMS)
LY20-0S92
LY24-5220
LY20-0693
LY2-4-5221
Auxiliary Service Support
DEVICE
SUPPORT
FACILmES
USER'S GUIDE
AND
REFERENCE
Device Suppcrt Facilities
57.48-XX9
GC35-0033
EREP
USER'S GUIDE
AND
REFERENCE
Environmental Recording
r:~ and Printing
GC2B-1378
Auxiliary Communication Support
RSCS Networking
Version 2
566.4-188
RSCS
NETWORKING
VERSION 2
PLANNING
AND
INSTALLAnON
RSCS
NETWORKING
VERSION 2
OPERAnON
AND USE
SH2.4-5057
SH2.4-5058
LY2.4-5228
VTAM
GENERAL
INFORt.AAnON
FOR WISP
VTAM
INSTALLAnON
AND
RESOURCE
DEFlNmON
VTAM
t.AESSAGES
AND
CODES
FOR VWSP
VTAM
DATA AREAS
FOR VWSP
VTAM
DIAGNOSIS
GUIDE
VTAM
DIAGNOSIS
REFERENCE
GC30-3246
SC23-0111
SC30-3276
SC30-3249
SC23-0116
SC23-0117
RSCS
NETWORKING
VERSION 2
GENERAL
INFORMAnON
GH2.4-5055
RSCS
NETWORKING
VERSION 2
DIAGNOSIS
REFERENCE
AdVanced
Communication
Function
For VTIJA
(ACFNTIJA)
666.4-280
VWPasa-
W/PASSTHROUGH
FACILITY
GENERAL
INFORMAnON
W/PASSTHROUGH
FACILITY
GUIDE AND
REFERENCE
W/PASSTHROUGH
FACILITY
LOGIC
GC24-5206
SC24-520B
LY2.4-6208
Through
~~~~~C1
Preface
XV
xvi
VM/SP Operator's Guide
Contents
Section 1. Introduction to Operational Control of the VM/SP System
1-1
System/370 recovery management support ..........................
Machine Check Handler (MCH) ................................
Channel Check Handler (CCH) .................................
Missing Interrupt Handler .....................................
I/O Error Recording and SV C 76 .................................
VM/SP Recovery Features .......................................
Recording Facilities ..........................................
VM/SP Repair Facilities .........................................
VM/SP Restart Facilities ........................................
1-2
1-3
1-4
1-4
1-4
1-5
1-5
1-5
1-6
Section 2. Starting VM/SP Operation .........•..••...••.....••....• 2-1
Using CP and CMS .............................................
Console Procedures ............................................
Operating Procedures (Operator Tasks) .............................
Labeling Disks ..............................................
System Definition ............................................
System Initialization ..........................................
Mass Storage System Initialization ................................
Multiprocessor Initialization .....................................
Attached Processor Initialization .................................
Console Definition ..........................................
Disabling the VM/SP Primary Console ..........................
Selecting a VM/SP Alternate Console during VM/SP Initialization ....
Controlling the 370x Communications Controller ....................
Considerations for Loading All 370x Control Programs .............
Special Considerations for Loading the EP 370x Control Program .....
Messages On Startup and Initialization .............................
Control Messages after Startup ...................................
System Operation Status Messages ..............................
Messages from Virtual Machine Users ...........................
Hardware and Program Status Messages .........................
3480 Magnetic Tape Subsystem Support ...........................
3800 Printing Subsystem Support .................................
Dedicated 3800 Printing Subsystem Support ......................
Performance Options ..........................................
Reserved Page Frames Option .................................
Favored Execution Option ....................................
Virtual=Real Option ........................................
Locked Pages Option ........................................
Priority ...................................................
VM/370 Assistance - Hardware Assist ..........................
Contents
2-1
2-1
2-2
2-2
2-3
2-3
2-14
2-16
2-16
2-17
2-19
2-19
2-20
2-21
2-21
2-22
2-22
2-22
2-22
2-22
2-24
2-24
2-25
2-26
2-27
2-27
2-28
2-31
2-31
2-32
xvii
Affinity ................................................... 2-34
Queue Drop Elimination ...................................... 2-35
System Termination ........................................... 2-35
System Abend Dumps ....................................... 2-37
Section 3. CP Commands ....•.....••.•.•...•••....••••..•...••••• 3-1
Privilege Classes for CP Commands .............................. 3-1
How to Use this Chapter ...................................... 3-3
ACNT ....................................................... 3-6
ATTACH .................................................... 3-8
AUTOLOG ................................................. 3-15
BACKSPAC ................................................. 3-17
CHANGE ................................................... 3-20
CPTRAP .................................................... 3-25
DCP ....................................................... 3-31
DEFINE .................................................... 3-35
DETACH ................................................... 3-37
DISABLE ................................................... 3-41
DMCP ..................................................... 3-43
DRAIN ..................................................... 3-47
ENABLE ................................................... 3-49
FLUSH ..................................................... ' 3-51
FORCE .................................................... 3-54
FREE ...................................................... 3-56
HALT ...................................................... 3-58
HOLD ..................................................... 3-59
INDICATE .................................................. 3-61
LOADBUF .................................................. 3-71
LOCATE ................................................... 3-76
LOCK ...................................................... 3-78
MESSAGE .................................................. '3-82
MIGRATE .................................................. 3-84
MONITOR .................................................. 3-86
MSGNOH .................................................. 3-100
NETWORK ................................................ 3-102
ORDER ................................................... 3-113
PURGE ................................................... 3-115
QUERY .................................................... 3-117
QVM ..................................................... 3-151
REPEAT ................................................... 3-152
SAVESYS .................................................. 3-154
SET. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 3-155
SHUTDOWN ............................................... 3-169
SPACE .................................................... 3-171
SPMODE .................................................. 3-172
SPTAPE ................................................... 3-173
START .................................................... 3-179
STCP ..................................................... 3-184
TRANSFER ................................................ 3-186
UNLOCK ................................................... 3-189
VARy ..................................................... 3-191
WARNING ................................................. 3-195
xviii
VM/SP Operator's Guide
/
Section 4. VM/SP Service Programs ...•.•.........••.•..........•..
Controlling the 3704/3705/3725 Communications Control Program ......
Using CMS To Alter VM/SP Files .................................
Other CMS Commands .......................................
DASD Dump Restore (DDR) Service Program and How to Use It ......
Invoking DDR under CMS .....................................
Invoking DDR as a Standalone Program ..........................
DDR Control Statements ......................................
I/O Definition Statements .....................................
INPUT / OUTPUT Control Statement ............................
SYSPRINT Control Statement ..................................
Function Statements ...........................................
PRINT/TYPE Function Statement .............................
Maintaining the 3800 Image Library ..............................
The Image Library ..........................................
The GENIMAGE Service Routine ..............................
The IMAGELIB Service Routine ...............................
The IMAGEMOD Command ..................................
Device Support Facilities .......................................
Formatting Volumes--General Information .........................
Format/ Allocate Service Program (DMKFMT) ......................
Format/ Allocate Program Card Input ...........................
Format/ Allocate Console Input ................................
How To Print a CMS Dump File .................................
How To Print a CP ABEND Dump from Tape .......................
Network Dump Operations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
NCPDUMP Service Program and How To Use It .....................
ZAP Service Program and How To Use It ..........................
ZAP Input Control Records ...................................
Special Considerations for Using the ZAP Service Program ...........
ZAPTEXT Service Program .....................................
ZAPTEXT Input Control Records ..............................
EXPAND Command ........................................
The Programmable Operator Facility ..............................
Use in a Single System .......................................
Use in a Distributed VM Environment ...........................
Use in a Mixed Environment ..................................
Invoking the Programmable Operator Facility .......................
Invocation of Programmable Operator Facility Commands .............
Issuing Commands in the Local Environment .....................
Issuing Commands In The Distributed VM Environment .............
Issuing Commands in a Mixed Environment ......................
Helpful Hints ..............................................
The Log File .................................................
Ensuring A Complete Log ....................................
Programmable Operator Facility Commands ........................
CMD Command ............................................
FEEDBACK Command ......................................
GET Command ............................................
LGLOPR Command ........................................
LOADTBL Command .......................................
LOG Command ............................................
Contents
4-1
4-1
4-2
4-2
4-2
4-3
4-4
4-4
4-5
4-5
4-7
4-8
4-14
4-19
4-19
4-19
4-21
4-21
4-23
4-24
4-24
4-29
4-35
4-40
4-40
4-41
4-42
4-43
4-45
4-52
4-52
4-53
4-54
4-56
4-57
4-57
4-58
4-59
4-61
4-62
4-63
4-65
4-67
4-67
4-68
4-70
4-71
4-73
4-74
4-75
4-77
4-79
xix
QUERY Command
SET Command
STOP Command
Stopping the Programmable Operator Facility
Stopping the PMX
Problem Determination - DEBUG MODE
Stand-Alone Dump Facility
Overview
Devices that You Can Use to IPL Stand-Alone Dump
Devices to which You Can Send Dump Output
Stand-Alone Dump Program Generation
Using the Stand-Alone Dump Facility
IPL and Dump Formats
Error Handling
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
4-80
4-83
4-86
4-86
4-87
4-87
4-88
4-88
4-89
4-89
4-91
4-92
4-96
4-99
0
000000000000000000000000000000000000000000000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
00000000000000000000000000000000000000000000
000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
000000000000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
00000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
00000000000000
0
0
00000000000000000000
0
000
0
0
000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
000000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Section 5. Operator Spooling Functions .............•...............•
Spooling
Spooling Considerations
Spooling Functions
Spool Buffers
Spool Files
Output Classes
Output Forms
Spooling Commands
Spooled Card Input
Spool Files for Virtual Machines Running Batch Jobs
Using the CMS Batch Facility
Starting the Batch Virtual Machine
Controlling the Batch Facility
Purging, Reordering, and Restarting Batch Jobs
Stopping the Batch Virtual Machine
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0000000000000000000000000.0000000000000000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0000000000000000000000000000
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
5-1
5-1
5-1
5-2
5-5
5-5
5-6
5-7
5-8
5-9
5-12
5-12
5-13
5-14
5-15
5-16
Appendix A. Summary of CP Commands .•..••...•..•....•...•....... A-1
Index ...•.............•••.......••........•.•......•.•......• X-1
xx
VM/SP Operator's Guide
Figures
2-1.
2-2.
3-1.
3-2.
4-1.
4-2.
4-3.
4-4.
4-5.
4-6.
4-7.
4-8.
4-9.
4-10.
4-11.
4-12.
4-13.
4-14.
4-15.
4-16.
4-17.
4-18.
4-19.
4-20.
5-1.
DOS 160K Virtual=Real Storage Assignments ...............
Command Control of VM/370 Hardware Assist .............
CP Privilege Class Descriptions ............................
CP Trace Event Type Codes .............................
Annotated Sample of Output from the TYPE and PRINT Functions
of the DDR Program ...................................
VM-Supplied Character Arrangement Tables for the 3800 Modell
and Model 3 Printers ...................................
VM-Supplied Character Arrangement Tables for the 3800 Model 3
Only. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
TEXT Deck Names Output by the GENIMAGE Command .....
Format of 3330 Cylinders for Use by CP ...................
3330, 3340, or 3350 Cylinder 0 Format ....................
Using the Format Program Label Function ..................
Using the Format Program Allocate Function ................
Using the Format Program Allocate Overlap Function .........
Using the Format Program Label Function for FBA Devices ....
Using the Format Program Allocate Function for FBA Devices ..
Using the Format Program Allocate Overlap Function for FBA
Devices .............................................
Valid Options and Their Output ..........................
Example of a Programmable Operator Facility Configuration. ...
A single system with NCCF ........................... . ..
Example of Communication in the Local Environment .........
Example of Communication in the Distributed Environment .....
Example of Communication in the Mixed Environment ........
Stand-Along Dump Facility Tape Format ...................
Stand-Alone Dump Facility DASD Format ..................
CP Spooling Commands .................................
2-30
2-34
3-2
3-28
Figures
xxi
4-18
4-20
4-20
4-21
4-26
4-27
4-36
4-36
4-37
4-37
4-38
4-39
4-45
4-58
4-59
4-63
4-64
4-66
4-97
4-98
5-8
/
xxii
VM/SP Operator's Guide
O[O>efi'"altnoU1Zl~ COIl1"lcI(O~
•
Section 1. Introduction to Operational Control of the VM/SP
System
System operators, in almost all large System/370 installations, serve in an apprentice capacity until they have acquired the skill and knowledge to maintain the
installation's operating system, be it DOS, OS, or VM/SP.
In an apprentice capacity, an operator acquires a basic proficiency in mounting,
making ready, and loading tape, DASD, line control, and other hardware devices.
Novice operators also learn the address designations of all hardware devices
attached to the system. They become aware of those system resources that can be
switched or patched via alternate channel or telecommunication path to expedite
system operations. The operator becomes aware of system console terminal operation as well as the function of most of the indicators, switches, buttons, and alarms
that are part of the processor control. All of this knowledge is basic and must be
acquired before considering the operational control of any system control program
(SCP).
The second phase of an operator's training is to be knowledgeable about the SCP
that is controlling the data processing operations; he must be aware of its capability
and its operating philosophy. The operator must also be aware of the priorities and
the demands placed upon the system.
Much of this basic knowledge of VM/SP is described in the VM / SP Introduction.
In that document there is discussion on virtual machines; how they are built, identified, and used. Each facet of the virtual machine is described: the virtual
processor, virtual storage, virtual I/O devices, and the virtual system console.
Concurrent virtual machine usage and spooling operations are also described.
With an understanding of VM/SP concepts, the power and versatility of VM/SP is
soon realized. The system operator is not just controlling one process and its
related storage and I/O with control and application programs; the operatorbis
exercising control over the resources of mUltiple systems (virtual systems). Each
virtual system with its own system operation, in turn, is being controlled by a user
via a terminal console. When additional facilities and resources or priorities are
needed by the virtual system, the virtual system operator must request these services from the VM/SP system operator. The system operator then delegates additional system resources to the virtual machine, if feasible.
Some of the operations that are exclusive functions performed by the VM/SP
system operator are as follows:
Reorder, purge, or copy any closed spool files
Section 1. Introduction to Operational Control of the VM/SP System
1-1
•
•
•
Issue warning and high priority messages
Attach and dedicate devices to specified virtual machines
Automatically logon virtual machines
Force users off the system when warranted
Change any virtual machine's dispatch priority and operating characteristics.
For details on how many of these and other functions are accomplished, see
"Section 3. CP Commands" on page 3-1.
Depending on the system installation, the system operator may also be the operator
of a large DOS/VSE, OS/MVS, or OS/VSl batch processing system that is run in
the virtual machine environment. Another duty of the system operator is to control
the data traffic from remote work stations. Information on VM/SP control of such
data transfer is detailed in the VM / SP Program Reference and Operations Manual.
Regardless of the real system resources and the tasks and procedures used, systems
are subject to hardware and program malfunction. The Interactive Problem
Control System of VM/370 provides a method of recording and maintaining a
history of VM/370 and virtual machine program problems. The VM/370 Interactive Problem Control System (IPCS) User's Guide provides the details on how to use
this system.
Problems that are a result of a hardware fault are usually detected by the hardware
itself, and trigger the recording of register contents and sense values (related to the
occurrence of the error) in VM/SP's error recording area.
Regardless of whether the malfunction origin is hardware or software, the system
operator (in most cases) is notified of the seriousness of the situations by means of
console indicators/alarms or by console diagnostic messages. These all serve to
inform the operator that:
•
•
•
The system operation can continue as before.
The system operation can continue with reduced power/resource.
System restart and recovery is commencing.
System operation is terminated.
The type of recovery that is attempted is described in "System/370 Recovery Management Support."
System/370 Recovery Management Support
IBM System/370 attempts correction of most machine errors without program
assistance. CP is notified, via an interruption, of both intermittent and permanent
machine errors to allow error recording and recovery procedures to start.
The following recovery features are implemented in the IBM System/370 hardware:
/'
Retry of the failing processor operations
1-2
VM/SP Operator's Guide
Validity checking on processor and control storage to correct all single-bit
errors
I/O operation retry facilities including an extended channel status word
(ECSW), which provides channel retry data to channel and control unit retry
procedures
Expanded machine check interruption facilities to improve error recording and
recovery procedures.
Machine Check Handler (MCH)
You can set the recording mode to record errors corrected by processor retry (logically termed as processor retry) and Error Correction Code (Eee) with the SET
MODE command. In attached processor or multiprocessor applications, recording
mode can be set for either or both processors. For processor retry, the default
setting is record mode.
Note: The SET MODE MAIN command is invalid for 3031, 3032, 3033, and
308X processors.
When processor retry or ECe succeed in correcting errors, and the processor is in
record mode, the machine check handler records the error. When processor retry
or Eee fail, the machine check handler:
•
Attempts to isolate the failure to one page frame and makes that page frame
invalid or unavailable for paging
•
Attempts to isolate the failure to one virtual machine and logs off or resets that
virtual machine
•
Attempts to isolate the failure to portions of the system and to continue system
operation in degraded mode
•
Abnormally terminates the system when recovery is not possible; or, if VM/SP
is operating in attached processor mode and the malfunction is isolated to the
attached processor and to a particular virtual machine, then, system operation
continues in uniprocessor mode.
Note: If VM/SP is operating in multiprocessor mode and the following conditions
exist:
•
•
The malfunction can be isolated to one processor and one virtual machine.
No system-owned device has its only online path on the failing processor.
The system can continue operation in uniprocessor mode.
Section 1. Introduction to Operational Control of the VM/SP System
1-3
MCH records an error whenever any of the following conditions occur:
•
•
•
•
•
•
Processor retry occurred I
ECC corrected datal
Hardware reported a buffer or DLAT (Data Look Aside Table) error
Multiple-bit storage failure
External damage
Storage protection feature damage
Timer error
System damage
Instruction processor damage.
Channel Check Handler (CCH)
Whenever a channel control check, channel data check, or interface control check
occurs, the channel check handler (CCH) constructs an error record and records
the results in an IOERBLOK. The error recovery procedures use this IOERBLOK
to retry the error. Recovery is not attempted for channel errors associated with
virtual machine I/O events.
Missing Interrupt Handler
The missing interrupt handler monitors system I/O activity on specific device
classes for interruptions that do not occur within a specified period of time. When
a missing interruption is detected, the control program attempts corrective action,
then notifies the system operator of the condition (either cleared or pending), and
writes a record to the error recording area. The operator is notified so that he can
take manual action if the control program's corrective action was unsuccessful.
I/O Error Recording and SVC 76
VM/SP maintains an error recording area that captures I/O, CCH, and MCH
error records. Device and control unit detected unit checks during VM/SP
spooling, paging, and virtual machine I/O errors generate the I/O records.
VM/SP and the virtual machine's SYSl.LOGREC data set contain recorded I/O
errors; this double recording occurs when the virtual machine's operating system
does not invoke SVC 76.
If the virtual machine operating system invokes SVC 76 and passes the correct
parameters to VM/SP, VM/SP records the error in its own error recording area.
VM/SP then passes control back to the virtual machine operating system, thus
bypassing virtual machine error recording facilities.
VM/SP records these errors only under specific conditions. The conditions for
recording these errors are detailed in the VM / SP OLTSEP and Error Recording Guide.
1-4
VM/SP Operator's Guide
r:==:-------.------.--------------------------
(Q) [CD e w~r~ 0l!J) uu a ~
---
C ((i) [('r~ w(Q) ~
VM/SP Recovery Features
The VM/SP recovery features are described more fully in the VM/SP OLTSEP
and Error Recording Guide.
Recording Facilities
The Environmental Recording Editing and Printing (EREP) program is executed
when the CMS CPEREP command is invoked. The output of the CPEREP
command consists of printed reports whose content depends upon the specified (or
defaulted) CPEREP operands and upon the input system error records. The
reports generated by CPEREP have the same format as those generated on an
MVS system. The input system error records may be from the VM/SP error
recording area or from a history tape. The history tape may have been produced
earlier by CPEREP from the VM/SP error recording area data or by an OS/VS
system from SYS l.LOGREC data. Unlabeled tapes produced on OS/VS systems
by EREP and on VM/SP systems by CPEREP are compatible and can be transported between systems. Data from both systems can also be accumulated on the
same tape. For more details on CPEREP, refer to the following publications:
VM / SP OLTSEP and Error Recording Guide and Environmental Recording Editing
and Printing (EREP) Program.
If the facilities of an IBM 3850 Mass Storage System (MSS) are used with VM/SP
virtual machine operations and MSS errors are reflected to VM/SP's error
recording area, CPEREP must be invoked so that MSS-related errors recorded in
the error recording area can be collected on an accumulation (ACC= YES) tape for
further processing by the VS System Data Analyzer Program (SDA). Because MSS
logged-out data is voluminous and the interrelationships of MSS components are
complex, it is imperative that this service program be used to effectively diagnose
and isolate mass storage problems.
VM/SP Repair Facilities
The Online Test Standalone Executive Program (OLTSEP) and associated Online
Tests (OLT) execute in a virtual machine that can run concurrently with normal
system operations. These programs provide online diagnosis of I/O errors for most
devices that connect to the System/370.
The service representative (with a CP command privilege class of F) can execute
online tests from a terminal as a user of the system; CP console functions,
including the ability to display or alter virtual machine storage, are available when
these tests are run. Those tests that violate VM/SP restrictions may not run correctly in a virtual machine environment.
~I
Section 1. Introduction to Operational Control of the VM/SP System
1-5
~
_________________________________________________________________________________J
VM/SP Restart Facilities
VM/SP tries to reload CP when a system failure causes:
1. An abnormal termination that does not result in a disabled WAIT state
2. The Dump to be directed to DASD.
Often, the operator does not need to do anything. The system attempts to execute
a warm start, thus allowing user's terminals to be reconnected (for logon reinitialization by users) and completed spool files as well as open console spool files to be
maintained. In the event of a warm start, device reconfiguration (such as varying a
device off-line) that was performed by the real computing system operator is
remembered by CP for system spooling devices only. Storage reconfiguration data
acquired during the process of recovering from real storage errors is lost. After a
VM/SP system failure, each user must re-access VM/SP (LOGON), and each
virtual machine must be reloaded (IPLed).
If the operator was logged off, running disconnected, or logged on to a machine
other than the primary system console, the operator will be restored as disconnected with the console spooled; CP issues the message:
IIDMKOPE967I DISCONNECT userid - SYSTEM RESTARTED (mmmnnn) AND
SYSTEM CONSOLE NOT VM OPERATOR CONSOLE: II
Otherwise, the operator will be logged on to the system console.
Termination of a virtual machine, whether caused by a real computing system malfunction or a virtual machine program error, normally does not affect the execution
of other virtual machines unless the error involves shared segments.
With virtual storage preservation, the system programmer can tell VM/SP to automatically save the contents of up to a 16M byte virtual machine (main storage and
registers) if:
•
VM/SP terminates the virtual machine.
•
VM/SP itself terminates.
The system programmer must specify at system generation time which virtual
machines are going to be saved. The contents of the virtual machine or virtual
machines are saved on DASD space that the VM/SP system programmer has previously allocated; users can then restore the contents via the IPL command.
Normal recovery procedures for the virtual machines can then be initiated by their
respective owners.
Virtual storage preservation is designed to make the automatically-saved virtual
machine available only to one or both of two previously specified userids. This is
for the privacy and security of the virtual machine. The saved virtual machine can
be loaded into either a V=R or normal non-V=R area. A system generation
macro instruction (NAMESYS) enables the system programmer to assign priorities
1-6
VM/SP Operator's Guide
which indicate the order in which multiple virtual machines should be saved. To
create an environment for saved systems requires both the use of the NAMESYS
macro instruction and enabling the VMSAVE function via the CP SET command
or the directory OPTION statement.
Note: VM/SP restart operations on systems employing virtual storage preserva-
tion can consume more time than is usual because the system must sequentially
page out the contents of the saved virtual storage to the previously allocated DASD
space. If an I/O malfunction prevents the saving of a VMSAVE system, the
system operator is notified.
In VM/SP attached processor operations, the priorities and restart activities are
similar to a uniprocessor mode of operation. However, if system damage assessment indicates an unrecoverable operation on the attached processor and the error
can be isolated to a virtual machine, VM/SP will be continued in uniprocessor
mode on the main processor. All virtual machines with an AFFINITY setting to
the attached processor are automatically reset to AFFINITY OFF for subsequent
virtual machine processing. (Note, the affected virtual machines are placed in
console function mode.) If a virtual machine was running when the malfunction
occurred, its operation will be terminated.
Note: See the VARY command in "Section 3. CP Commands" for instructions
on how to vary the attached processor and multiprocessor back online.
If a similar malfunction occurred on the main processor while VM/SP was oper-
ating in attached processor mode, VM/SP terminates. VM/SP cannot switch to
uniprocessor operation on the attached processor unit because the attached
processor unit has no hardware interface for input/output capabilities.
Note: In certain 303x attached processor environments the channel set switching
facility is present. In the case of a malfunction on the main processor in these environments, VM/SP can continue system operation in uniprocessor mode by
switching the channel set from the failing processor to the remaining processor.
In VM/SP multiprocessor operations, if an unrecoverable error occurs on one
processor and the error can be isolated to one virtual machine, it may be possible
for VM/SP to continue operation in uniprocessor mode on the nonfailing
processor. System operation can continue if all system-owned devices have online
paths from the nonfailing processor. Virtual machines with affinity set to the
failing processor have their affinity set off and are placed in console function mode.
A MESSAGE or WARNING appearing on the screen during restart, puts the
screen in MORE status. Normally, this would put the screen in HOLDING status.
However, if the screen is in MORE status during restart operations, pressing the
ENTER key will put the screen in HOLDING status.
Section 1. Introduction to Operational Control of the VM/SP System
1-7
1-8
VM/SP Operator's Guide
Section 2. Starting VM/SP Operation
Using CP and CMS
CP and CMS commands are never more than eight characters long and can be
truncated to the minimum size indicated in the command format description. The
operands, if any, follow the command on the same input data line. Most CP and
CMS commands may not extend beyond one line except on the 3270. Generally,
the operands are positional, but some commands have keywords to assist in the
translation of the command line. One or more blanks must separate the command
from any operands (except for some EDIT subcommands).
CMS indicates successful command processing with the CMS READY message.
Error messages occur for erroneous operands. CP does not recognize CMS commands. However, a CMS user can issue CP commands without leaving the CMS
environment. Although not required, CP commands entered while in the CMS
environment should be prefixed with "CP" or "#CP" to decrease VM/SP table
and disk search time. CP then handles the command and passes control back to
CMS. If the CMS user wants to enter CP mode, he can do so by keying in CP or
#CP with no operands (that is, no CP commands) or with an attention interruption.
For details, refer to "Attention Handling" in the VM/SP Terminal Reference.
If a user requires communication with CP before responding to specific error mes-
sages issued by some virtual machine operating systems, he should enter CP mode
by keying in #CP with one or more CP command lines separated by a logical line
end character, or by signalling attention. Then perform the necessary console
functions. When returning to the virtual machine environment, the virtual machine
read will have been canceled by a unit exception, and the virtual operating system
normally responds by reissuing the read. The user can then enter the required
virtual machine response.
Console Procedures
You, as the VM/SP system operator, are like other system users, but with some
important differences:
•
You are allowed to perform certain console functions that the normal user is
not.
Section 2. Starting VM/SP Operation
2-1
•
Your terminal normally stays in CP mode and is usually ready to receive messages from the VM/SP system and other users.
Unless you are using a 3270 terminal, before you can type a command, you must
press the REQUEST key (or its equivalent) on the designated VM/SP system
console. VM/SP responds with the time, and sets up to read data from the console
keyboard. You can then enter CP commands into the system.
You can enter all CP commands in lowercase or uppercase, or a mixture of both.
All system responses are prefixed with the clock time at which you entered the
command from the primary console. Full descriptions of CP commands with a
class of A through F are in Section 3, "CP Commands" on page 3-1.
Enter commands into the VM/SP system by pressing the end-of-line function key
on the console to terminate the command line. To cancel a miskeyed or inappropriate VM/SP command that you partially or totally entered on the console, enter
the designated logical line delete character (normally, the ¢ symbol) or press the
CANCEL key (or its equivalent).
Note: The CANCEL key on the 3215 cannot cancel lines of input until you finish
defining the VM/SP system during system generation. Also, the P A2 key or the
CANCEL key on the 3270 and 3066 cannot cancel input lines.
For details on supported VM/SP typewriter-like terminal consoles, display terminal
consoles and the associated keys that manage the display screen, signal attention,
and process line records, consult the VM / SP Terminal Reference.
Operating Procedures (Operator Tasks)
Labeling Disks
CP and CMS, with the exception of the special CP service programs and certain
frequently used CP functions, get all of their system functions from disk. Each of
the disks mayor may not be labeled. If the disks are for CMS file residence, or CP
paging and spooling (temporary use), you must label each pack (count-key-data
pack) with a six-character label in real cylinder 0 track 0 record 3 by using the CP
Format/ Allocate program. For FB-512 devices, this label is placed in block 1.
The VM/SP system is device-independent; each of the volumes that have CP
labels can reside on any available and defined 2314, 2319, 3330, 3333, 3340,
3375, 3380, 3344, 3350, 2305 or FB-512 direct access storage device. This definition occurs at system generation time. The CP system residence (SYSRES)
volume must be mounted on an available IPL device; then use the normal IPL procedure for your processor. If you are using CP-owned volumes, they should be
mounted and ready at this time. All other resident volumes should be mounted and
ready when the system is started, although they can be added later (via the CP
ATTACH command) to the system, while it is in operation.
2-2
VM/SP Operator's Guide
Assignment of Special Performance Options to Logged-On Users:
FAVORED, AFFINITY, SASSIST, CPASSIST, RESERVE, PRIORITY, LOCK,
and UNLOCK commands can be assigned only by an operator with privilege class
A.
Handling of Spooled Input and Output: Card decks for users must be fed into
the real card reader, printed and punched output properly distributed, and the unit
record equipment and spool data files controlled. The control functions for the system's unit record equipment and spool data files can be performed only by a class
D operator. For 3800 virtual printers, the operator must mount the designated
paper stock if directed to do so.
Attaching and Detaching of User and System Volumes: Devices used by
virtual machines in dedicated mode must be attached and detached as appropriate.
Control over the real System/370 computing system's I/O can be performed by an
operator with privilege class B.
Operators with these privilege classes must be logged on to perform these functions. An example of a directory entry that allows an operator to perform all of
these functions is:
USER OPERATOR OPASS 512 1M ABDG
The system and spooling operators do not require virtual devices or options in their
VM/SP directory entries unless CMS is used. Multiple virtual machines for operators may be set up, each with all or some of the associated privilege classes. The
primary system operator must have class A assigned in the directory to properly
initiate VM/SP operation.
System Definition
If you intend to define or attach more than 64 virtual devices for a single virtual
machine, be aware that any single request for free storage in excess of 512
doublewords (a full page) may cause the VM/SP system to issue an appropriate
error message if the extra storage is not available on a contiguous page. Therefore,
two contiguous page frames of free storage must be available in order to log on to a
virtual machine with more than 64 virtual devices (three contiguous page frames
for a virtual machine with more than 128 virtual devices, etc.). Contiguous page
frames of free storage are sure to be available only immediately after IPL, before
other virtual machines have logged on. Therefore, as a system operator you can
inhibit the logon of many virtual machines until the virtual machine or machines
with many I/O devices have accomplished logon and are operating.
System Initialization
Before you invoke any type of start:
The following devices must be powered up and ready:
/-
The processor
Processor storage
Section 2. Starting VM/SP Operation -
2-3
Starting Operation
All necessary devices (and disk volumes) for a minimum VM/SP system
with system residence, paging, and spooling requirements.
The system cannot encounter any unrecoverable hardware errors in the initialization and startup phase of bringing the VM/SP system online.
•
The generated VM/SP SCP and the System/370 hardware and its attached
features must have matching compatibility. If they are not compatible you will
get an error message in addition to VM/SP performance degradation.
To simplify system initialization, be sure that:
1.
All VM/SP resident volumes (specified in the SYSOWN list for paging and
spooling) are mounted and ready at IPL time.
If the volumes specified in SYSOWN are not mounted when you IPL VM/SP,
they are not available to you. If you need the volumes, use the VARY
ONLINE command and the ATTACH command to mount and attach them.
2. Volumes that have user minidisks (such as the CMS system residence volume)
are mounted and ready at IPL time. You can, however, vary online and attach
volumes not required for startup after the VM/SP program is up and running.
Overview: The Steps Involved with System Initialization
1. Initialization Begins: To start the operation of the VM/SP system, load the
CP system residence volume.
Note: When you turn system power on, you may need to run a CS (control
storage) patch routine. This is a System/370 operating procedure and is not discussed further in this book.
CP calculates the real machine's storage size. If a real machine has more storage
than what you specified on SYSCOR, the system does not use the extra storage.
For example, if a real machine's storage size is 1 megabyte and you set SYSCOR to
512K, the system uses 512K of real storage. If less storage is available in the real
machine than what you specified on SYSCOR, the system uses the lesser figure.
The system checks for a valid microcode level. If the microcode level is invalid, the
system disables the microcode assists and stores an error message for display later.
Also, CP searches for the devices defined during system generation and checks to
make sure the volume labels on each DASD are correct. CP varies the devices
online.
Note: If the configuration differs from that specified during system generation
(SYSOWN volumes not mounted or storage size not equal to SYSCOR), the
system stores the message for display later and continues to operate. Label errors
also cause the system to store an error message for later display, and the system
startup may abnormally end.
CP starts the segment, page, core, and swap tables. It also checks to see if VM
assist and the 370E facility are available.
2-4
VM/SP Operator's Guide
S·l!:ar·~Dlng O~erar~DOU1
----------.----------------
2. Locating the Operator's Console: CP locates the operator's console.
During VM/SP system generation, your installation will assign a console or terminal as the primary operator's system console. Your installation can specify more
than one alternate console in the RIOGEN macro instruction during system generation. If the specified primary system console is not operational, CP will use the
first designated alternate console. If the first alternate console is not operational
and your installation specified more than one alternate console, CP will use the
next alternate console.
The primary and alternate consoles must be one of the following devices:
•
A real primary or alternate system console.
A locally attached 3270.
In addition, the alternate console can be any VM/SP supported remote terminal on a leased line connected to a 270x transmission control unit.
The system does not support, as primary or alternate VM/SP consoles, terminals
on switched lines or leased line terminals connected to 3704/3705/3725s.
Next, the following message tells you what release and service level of VM/SP you
are using, and when it was created:
VM/SP RELEASE x, SERVICE LEVEL xxxx, CREATED ON mm/dd/yy AT
hh:mm:ss
3. Time-oF-Day Clock: CP then checks the status of the time of day (TaD)
clock. If the clock is not set, CP requests that you set it. Refer to "Time-of-Day
Clock" on page 2-7 and "Setting the Clock (Uniprocessor Application)" on
page 2-8. If the clock is set, CP lets you change the clock. The following lines are
displayed:
IT IS NOW 08:28:00 EST THURSDAY 01/26/86
CHANGE TOO CLOCK (YESINO):
"YES", CP responds with prompts like those explained in "Setting the
Clock (Uniprocessor Application)" on page 2-8. If you press the ENTER key, CP
assumes the response is "NO".
If you enter
For AP /MP, if the other processor is available, CP begins prefixing and synchronizes the TaD clocks.
Note: If the SYSCOR macro had specified AP= YES or MP= YES and the second
processor is unavailable for startup, the system operates in uniprocessor mode
instead of attached processor or multiprocessor mode.
CP initializes the user directory.
CP verifies that the interval timer is running. If it is not, CP requests that you start
it.
At this point, the system displays any delayed messages. Also, if any CP owned
volumes are malfunctioning or not mounted, the system displays one or more mes-
Section 2. Starting VM/SP Operation
2-5
sages. This lets you select a shutdown start in the next step of the IPL process if
you need to.
4. Selecting the Type of Start: The system then prompts you for the type of
system start you want:
START ((WARMlcKPTIFORCEICOLD) (DRAIN»I(SHUTDOWN) :
You can respond in the following ways:
COLD or COLD DRAIN
WARM or WARM DRAIN
CKPT or CKPT DRAIN
FORCE or FORCE DRAIN
DRAIN
SHUTDOWN
WARM is the default. Specify DRA INif you do not want the system to automatically
start the real unit record devices during initialization. The types of system starts
are explained in:
"Cold Start" on page 2-11
"Warm Start" on page 2-9
"Checkpoint Start" on page 2-10
"Force Start" on page 2-10
"Shutdown Start" on page 2-11.
CP then allocates DASD dump space. If CP cannot allocate the dump space, it
issues the following message:
DMKIDU9531 UNABLE TO ALLOCATE SYSTEM AUTO DUMP
5. Logging on the System Operator: Next, the system automatically logs on
the system operator.
The system operator's userid, password, and command privilege class or classes are
usually stored on the CP system residence volume during system installation. The
primary system operator must have class A command privileges. CP compares the
userid supplied by the SYSOPR macro (used in system generation) with the user
directory. If the userids do not match, CP stops the automatic operator logon
process and gives system operational control to the first privilege class A user who
logs onto the system.
If the primary or alternate devices do not exist or are not ready, automatic logon
stops, and the VM/SP system enters the disabled wait state. If the primary or alter-
nate consoles are operational, but some processing error occurs, you must intervene
to bring up the VM/SP system. CP considers the first user to log on VM/SP with
privilege class A as the primary system operator. For more information see
"Manual Logon" on page 2-12. The system does not accept any users until the
primary system operator logs on.
2-6
VM/SP Operator's Guide
r==----
8'~ a U".~ al111 ~ ([D [CD e U" ~r~ a(D uu
---------------------------
----------===:1
The following message is then displayed:
DMKCPJ9571 STORAGE SIZE = xxxxx K,
NUCLEUS SIZE = xxx K,
DYNAMIC PAGING SIZE = xxxxx K, TRACE TABLE SIZE = xxx K,
FREE STORAGE SIZE = xxx x K,
VIRTUAL=REAL SIZE = xxxxx K
If MONITOR is enabled, the system starts MONITOR. Also, the system logs on
the AUTOLOG 1 virtual machine.
Finally, CP displays the following message:
DMKCPI9661 INITIALIZATION COMPLETE
At this point, the system is ready for normal use. You can now:
Write a message of the day for the system to issue when users log on.
Enable communications lines so users can log on.
Time-ot-Day Clock
The Time-of-Day (TOD) clack provides an accurate measure of time, independent
of system events or activities. It makes accurate measurements available for programming applications. When system power is turned off, the clock value is lost on
some System/370 models. Once CP uses the set clock instructions and you use the
TOD ENABLE SET switch to make the time-of-day (TOD) clock operational, the
system increments the clock at a constant rate. The following do not affect the
timing operation:
•
•
•
•
Any normal activity or event in the system
Wait state
Stopped state
Instruction-step mode
Single-cycle mode
Test model
System reset
Initial program load procedure.
If TOD clock hardware errors occur, the System/370 hardware posts a machine
check and CP enters a disabled WAIT state.
Virtual machines may use the System/370 STORE CLOCK instruction to find the
current clock value. This instruction causes VM/SP to store the current clock
value in the storage location specified in the instruction. Thus, virtual machine
users can use the value of the TOD clock for any purpose. However, virtual
machine users cannot set or change the TOD clock.
VM/SP uses the TOD clock to establish:
Certain accounting records for the virtual machines
The current clock value is lost when the time-of-day micro diagnostic tests are being
run, or when system power is turned off.
Section 2. Starting VM/SP Operation
2-7
•
Queueing and prioritizing tasks for CP and virtual machine operations.
The privileged System/370 instruction, SET CLOCK, sets the clock to a specific
value. If certain operating or program conditions exist, CP executes SET CLOCK
when the system is initialized. SET CLOCK replaces the current clock value with
the value you specified. The SET CLOCK instruction changes the clock value only
when the TaD ENABLE SET switch is enabled.
If your installation has a second processor, you may need to synchronize the TaD
clocks in the initialization procedure. The system may request that you ready the
TaD ENABLE SET key twice: once to set the clock on the main processor and
once to SET and SYNC the clock on the second processor. If the clock value has
previously been established for the main processor, the system will only issue one
prompt message.
Note that the TaD ENABLE SET key is ORed to the other processors' TaD
ENABLE SET key. This lets you set the clock from either processor. If the loworder 32 bits of the TaD clock are not synchronized, a TaD Sync check will be
received. Again, the system will request that you synchronize the TaD clocks by
pressing the TaD ENABLE SET key.
4361 processors are offered with an Auto Start feature. Using a battery operated
clock, this feature maintains the time while the power is off. For these processors,
if you shut down the system using the SHUTDOWN command with the
POWEROFF parameter, during the next IPL, you will not be prompted to set the
time-of-day clock.
Details on the parts of VM/SP initialization that are unique to the attached
processor or multiprocessor applications are explained later in this chapter.
Setting the Clock (Uniprocessor Application)
If the clock is not set, CP issues the following message to prompt you to enter the
date:
SET DATE (MM/DD/YY)
Enter the date.
Note: If the two digit number for the year is less than 50, the system assumes the
century is 2000. If the number is equal to or greater than 50, the system assumes
the century is 1900. Do not use the numbers 41 through 49, because they do not
represent anything.
CP then requests the time:
SET TIME (HH:MM:SS OR HH.MM.SS)
Enter the time. Note that this is a 24 hour clock.
2-8
VM/SP Operator's Guide
r--·--··--··--·-·-··---·-------·------
§'~a r'~D
(Q) LCDer~r~D «J) U1
---------_._._----_._---------------,
uu g
Notes:
1.
Enter data immediately following the colon (:). If you are using a 3270 or a
VM/SP supported processor display console, enter the data in the user input
area. The data then appears in the output display area on the line following the
SET TIME request.
2.
Because CP requests that you press the TOD ENABLE SET key to set the
clock, enter a time value that is 30 to 60 seconds ahead of the current time.
For example, if the date is 01/24/86, and the time will soon be 08:28 (EST),
the exchange on a printing terminal looks like this:
VM/SP RELEASE 4,SERVICE LEVEL OOOO,CREATED ON 01/01/86 AT 11:15:00
SET DATE (MM/DD/YY) :01/24/86
SET TIME (HH:MM:SS OR HH.MM.SS) :08.28.00
PRESS "TOD ENABLE SET" KEY AT DESIGNATED INSTANT
When you press the TOD ENABLE SET key, CP responds with:
IT IS NOW 08:28:00 EST FRIDAY 01/24/86
CHANGE TOD CLOCK (YESINO):
Note: The previous example run on a display terminal would show your
responses on separate lines.
You can now change the clock's value. If you enter "yes," another series of
prompting messages let you enter a new date and time. If you enter "no," or press
the END function key on the console, clock initialization ends.
Programming Note: To VM/SP users, the epoch clock value of all zeros is the
beginning of the day January 1, 1900. To OS and MVS users, the epoch clock
value of all zeros is the beginning of the day January 1, 1980. Thus, OS virtual
machine users will find that stored clock values differ by 80 years. You can,
however, use 1900 in OS as the base epoch clock value.
Warm Start
Warm start means that the previous VM/SP session ended with an orderly shutdown procedure. Use a warm start procedure to recover accounting data and
access previously closed spooled output files. CP tells you when a warm start
cannot be done and requests an alternate recovery method.
Begin a warm start:
1.
If you are not alerted by console messages or visual signs that a checkpoint or
cold start is required.
2.
After VM/SP has abnormally stopped and a system dump operation has completed.
3.
After an orderly shutdown has been performed.
Section 2. Starting VM/SP Operation
2-9
Warm start restores previous spool controls and files. Accounting records remain
valid and are to be kept as a base for continuing operations.
Key in WARM or simply press the END, ENTER, or RETURN key (depending on
the console device) when the following message appears at the console:
START «WARMlcKPTIFORCEICOLD)(DRAIN»I(SHUTDOWN) :
(You can also specify
record devices.)
~/ARM
DRA INto warm start the system and drain all unit
If you try a warm start, but the system has not saved any warm start data, the
system will enter a wait state PSW (code 009). In this case, try a checkpoint start.
In many cases, VM/SP recovers from system failures and restarts itself. When this
occurs, you will see the following message at the console:
VM/SP SYSTEM RESTART DUE TO SYSTEM FAILURE
When you enter the SHUTDOWN RE IPL command, you will see the following
message at the console:
VM/SP SYSTEM RESTART DUE TO SHUTDOWN REIPL
Checkpoint Start
If the system could not do a warm start because of I/O errors or invalid data in the
warm start area, you can request a checkpoint start.
This option tries to start the system using the information that has been dynamically checkpointed during system operation and stored in the checkpoint area.
Checkpoint start reconstructs the spool file chains, but the original order of the
files is lost. Accounting and system message data, recovered under a warm start, is
lost under a checkpoint start. Also, because the system must read each spool
buffer to reconstruct the record allocation blocks, a checkpoint start takes longer
than a warm start.
To begin a checkpoint start, respond with CKPT to the following message:
START «WARMICKPTIFORCEICOLD)(DRAIN»I(SHUTDOWN) :
Force Start
If the system cannot do a checkpoint start because of I/O errors or invalid data in
the checkpoint area, you can request a force start. This option operates like the
checkpoint start, except that any spool file that has unreadable or invalid data is
dropped from the system.
To begin a force start, respond with FORCE to the following message:
START «WARMICKPTIFORCEICOLD)(DRAIN»I(SHUTDOWN) :
2-10
VM/SP Operator's Guide
/
Cold Start
You may want to use cold start procedures if you have tried a warm start, a force
start, and a checkpoint start, and all have failed. This may happen when:
You migrate to a new release of VM/SP.
All closed spool files and accounting records are lost or unreadable because of
a hardware error that prevented valid system recovery and continuation.
VM/SP issues a message telling you to do cold start procedures when it recognizes
that unrecoverable errors have occurred. You may have failures in system component power, building power, hardware logic, or the CP program. Also, some (rare)
kinds of user programs can be classified as system operating procedure errors. All
these types of failures can require that you start the system with a cold start.
VM/SP cold start procedures start after the condition, which caused the
System/370 or the operator to abnormally stop VM/SP, has been corrected.
To start a cold start, respond with COLD to the following message:
START ((WARMICKPTIFORCEICOLD)(DRAIN)) I (SHUTDOWN) :
After the cold start procedure is complete, enter into the LOGMSG the time of the
unplanned shutdown and that a cold start was done. This tells the user that his
spooled files were lost, and that he may need to reconstruct some of his other files.
Shutdown Start
A shutdown start stops the initialization process. You may want to do a shutdown
start if you receive a message that one or more vital CP owned volumes are not
mounted or malfunctioning. For a shutdown start, the system does not process
warm start or checkpoint data. Following a shutdown start, previously saved warm
start and checkpoint data is still valid.
To begin a shutdown start, respond with SHUTDO\m to the following message:
START ((WARMICKPTIFORCEICOLD)(DRAIN)) I (SHUTDOWN) :
If you cannot key in the SHUTDOWN command, press the LOAD button on the
System/370 console without clearing storage. After you see the shutdown
message, press the LOAD button a second time to reload the system with VM/SP.
Automatic IPL
This feature is supported for 4361 processors only. 4361 processors are available
with an Auto Start feature. Using a battery operated clock, this feature maintains
the time while the processor is powered off and sets the system time-of-day clock
after power on.
The 4361 processor with the Auto Start feature can be set to automatically power
on at a specified time, or it can be powered on by telephone from a remote
location.
Section 2. Starting VM/SP Operation
2-11
c=______________________________________________________________________________
~
Once the processor has been powered on, if you previously shut down the system
using the SHUTDOWN command with the POWEROFF parameter, you will not
be prompted to set the time-of-day clock. In fact, you do not need to be present
for the IPL to occur.
Automatic Logon
Automatic logon is normally part of an IPL. When the system operator is logged
on, the following message shows the date and the time of the last LOGMSG
setting:
hh:mm:ss LOGMSG- hh:mm:ss EST FRIDAY mm/dd/yy
The following message is displayed next only if queued, closed spool files exist.
hh:mm:ss FILES: xxx RDR xxx PRT xxx PUN
Notes:
1. A MESSAGE or WARNING on the screen during system IPL puts the screen
in MORE status. Normally, this would put the screen in HOLDING status. If
the screen is in MORE status during system IPL operations, you may press
ENTER to put the screen in HOLDING status.
2.
If the automatic logon fails, the user (privilege class A) will have to explicitly
log onto the system. If the system console is either a 3210 or 3215 (or a 3138,
3148, or 3158 console in printer keyboard mode), the user must protect the
security of his password because there are no print-masking or print-inhibiting
features on these devices.
Manual Logon
If the automatic logon of the primary system operator fails because the VM/SP
directory defines no system operator, the following message occurs:
OPERATOR NOT LOGGED ON; EXPLICIT LOGON REQUIRED
Check your directory entry to confirm your userid and password. A logon prompt
will not be displayed. Log on with your userid:
LOGON userid
In the above response, you enter the CP LOGON command, followed by a blank
and your identifier name (userid), followed by pressing the end function key. CP
checks this entry with the values in the VM/SP directory. If the userid is valid and
you have class A command privileges, the following message on the console
prompts you for the next response:
hh:mm:ss ENTER PASSWORD
You then key in your password, which is checked against the password stored in
your VM/SP directory entry. The system is not accessible to other users until you
log on and enable the necessary teleprocessing lines. Maintaining password secu-
2-12
VM/SP Operator's Guide
§.~ a u--.~ Uru £1 (Q) ~a)(a ~ zr~ ~ «J) rll
r--------------------------------------------------------------------
J
rity on real printer-keyboard system consoles is your responsibility because password inhibiting or masking is not possible on these devices.
If the password is valid, the VM/SP system can be started.
Note: If the system operator as defined in SYSOP in DMKSYS logs off, the next
class A user who logs on will become the system operator. That is, the system will
send operator messages to this class A user, even if the userid originally specified in
SYSOP logs back on. The class A user will continue to receive any operator messages that the system sends until he logs off.
When this class A user logs off, then there is no operator. A user sending a
message to the operator will get the message, DMKMSG045E 'OPERATOR NOT
LOGGED ON'. The next class A user to log on will become the system operator.
Unit Record Spool Files
After starting the VM/SP system, the system responds with the status of the
system spool files:
hh:mm:ss FILES:
031 RDR, 039 PRT, 001 PUN
The response indicates that a total of 31 spooled reader files, 39 spooled print files
(which also includes closed spooled console files), and 1 spooled punch file were
saved at the last system shutdown.
If the unit record devices are drained as a result of the DRAIN option during IPL,
then you should start the unit record devices at this time. For example, the CP
command
START ALL
starts all real unit record equipment. The system then responds with the status of
each unit record device.
37xx Resources
At this point, if the following conditions are met:
3705 control units are part of the VM/SP system support of remote terminals
The automatic load function for the 3705 was not generated into the VM/SP
system
The 370x control program was generated under VM/SP
invoke the NETWORK LOAD command as follows:
NETWORK LOAD raddr ncpname
Raddr is the base address of the 3705 and ncpname is the desired copy of the 3705
Emulator Program. For a description of the NETWORK command, see
"NETWORK" on page 3-102. If you are using 3725 communication control
units, use the ACF /NCP-SSP loader.
Section 2. Starting VM/SP Operation
2-13
You can then enable any or all of the EP teleprocessing lines and 3705/3725
resources for VM/SP users by issuing the CP ENABLE command and
NETWORK ENABLE command, respectively. For example, pressing the
REQUEST function key and entering:
hh:mm:ss ENABLE 50
enables only line 50, while:
hh:mm:ss EN ALL
enables all 3270 and 270x lines. CP responds with:
hh:mm:ss COMMAND COMPLETE
when the requested command has been serviced.
Note: The same message occurs even if no lines connect to the machine. If a
patch panel is in use, make sure that all desired lines connect to the system.
The VM/SP system is now operational and waits for users to log on. Similarly, if:
NETWORK ENABLE ALL
is invoked, the total resources of all VM/SP controlled 3705 control units are
enabled if the system is equipped with the devices.
Mass Storage System Initialization
If an MSS port is attached to a virtual machine running MVS with MSS support
and the MSS communicator program is running in the virtual machine, the control
program can cause automatic 3330V volume mount and demount in response to
any of the following:
A virtual machine logs on with a minidisk defined on a system disk that is not
mounted. The control program attemptd to mount an MSS volume with the
correct volume label.
•
A virtual machine logs on having a dedicate directory statement specifying a
volid for a volume that is not mounted. The control program again attempts to
mount an MSS volume with the specified volume label.
An operator ATTACH command is issued specifying a device address that is
an MSS 3330V address and volid. The control program attempts to mount the
volid on the device address as part of the ATTACH processing.
The operator DEFINE command is used to change the feature (SYSVIRT or
VIRTUAL) of a 3330V containing a mounted volume. The control program
demounts the volume before completing the DEFINE process.
2-14
VM/SP Operator's Guide
/
r----------------
A virtual machine issues a LINK command to link to a virtual device that is all
or part of a 3330V volume. The control program attempts to mount the
volume as part of the LINK process.
The VM/SP control program does not issue orders directly to the mass storage
control (MSC). Rather, it passes requests to an MVS system with MSS support
that is operating in a virtual machine with an MSC port dedicated to it. The MVS
system then issues orders to the MSC and passes response information back to the
control program.
In order for the installation to run with full MSS support, an MVS system must be
IPLed in a virtual machine. An MSC port must be dedicated to this virtual
machine. There are no special requirements for this IPL. The standard installation
procedures for IPLing MVS in a virtual machine should be followed. After the
system is IPLed, the VM/SP communicator program must be started. If the
standard MSS installation procedures as documented in the VM / SP Planning Guide
and Reference have been followed, the communicator program is started in the
virtual machine through the MVS operator command:
START DMKMSS [.pn]
The [.pn] operand is used only in OS/VSl systems to specify the partition in which
the program is to be started.
The absence of error messages received from DMKMSS indicates that MSS
support is initialized. The communicator virtual machine may now be disconnected
if the installation desires.
There are two error messages that the DMKMSS program can produce. If either of
these messages is received, then the VM/SP control program will not be able to .
communicate with the MSS. The first possible message is:
DMKMSS ENDING ERROR - MSS NOT INITIALIZED
Explanation: The MSS did not get initialized as it should during the MVS IPL
process. Either the virtual machine does not have access to the MSC, or there was
an MSS error.
Operator response: Correct the problem, use the MVS VARY command to initialize the MSS, then reissue the START DMKMSS command.
The second possible error message is:
DMKMSS ENDING ERROR - DCB FOR COMM. DEVICE NOT OPENED
Explanation: The DMKMSS program was not able to establish communication
with the VM/SP control program because the MVS DCB control block could not
be opened. The probable cause is the definition of the communicator device for
the virtual machine.
Operator response: Contact the system programmer. After the problem has
been corrected, reissue the START DMKMSS command.
Section 2. Starting VM/SP Operation
2-15
Multiprocessor Initialization
In order for the installation to run with a multiprocessor, it must do the following:
•
Specify MP= YES in the SYSCOR macro during VM/SP system generation
Have the second processor online during system IPL.
If the second processor was generated as part of the system but was not online, and
the multiprocessing feature is installed, the following message is issued:
SECOND PROCESSOR NOT ONLINE
If the second processor is not required, processing can continue. However, if the
second processor is required, the operator should ready the unit and reload the
system.
Assuming MP= YES and the second processor was online the following message is
issued:
PROCESSOR xx IPLED; PROCESSOR yy INITIALIZED
No operator action is required; the system will now utilize the second processor.
If the installation is running with a second or an attached processor, the system
operator will be issued the following message if the clocks are to be.synchronized.
CLOCK SYNCHRONIZATION-PRESS ITOD ENABLE SET I
From this message on, see" Attached Processor Initialization" because the procedure is the same.
Attached Processor Initialization
In order for the installation to run with an attached processor, it must do the following:
•
Specify AP=YES in the SYSCOR macro during VM/SP system generation
Have the attached processor online during system IPL.
If the attached processor was generated as part of the system but was not online,
the following message is issued if the multiprocessing feature is installed:
ATTACHED PROCESSOR NOT ONLINE
If the attached processor is not required, processing can continue. However, if the
attached processor is required, the operator should ready the unit and reload the
system.
/
2-16
VM/SP Operator's Guide
c
The operator will then press the TaD ENABLE SET key on either processor to
synchronize the TaD clocks. Note that once the clocks are synchronized, it will
not be necessary to resynchronize the clocks on subsequent IPL operations unless:
FE diagnostics are run.
Power is turned off on either processor.
The clocks are running but not synchronized.
A clock's error is encountered during initialization.
If the clocks should run out of low-order synchronization, after the clocks are syn-
chronized (during or after system initialization), you will receive the following
message:
TOO CLOCK SYNC CHECK RECEIVED
CLOCK SYNCHRONIZATION-PRESS 'TOD ENABLE SET '
As before, press the TaD Enable SET Key to synchronize the clocks.
Interval Timer
VM/SP uses the interval timer to support time slicing; therefore, during initialization, the system verifies that the interval timer is running before virtual machine
logon operations are performed. The system initialization routine will loop issuing
the following message if the interval timer is found not running when tested:
TURN ON THE INTERVAL TIMER
TURN ON THE INTERVAL TIMER
The messages cease when the operator complies by enabling the interval timer and
the system initialization process continues.
In attached processor application the text of the message indicates the processor
owning the disabled interval timer. This message reads as follows:
TURN ON THE
I
MAIN
ATTACHED
I
PROCESSOR INTERVAL TIMER
To continue, the interval timer must be enabled on the requested processor.
Console Definition
The terms "system console," "primary system console," and "alternate system
console" apply to those console devices that exercise primary control of the
processor.
The VM/SP primary system console can be a real system console, a real alternate
system console, or a local 3270.
The VM/SP alternate system console can be a real system console, a real alternate
system console, a local 3270, or any VM/SP supported remote terminal on a leased
Section 2. Starting VM/SP Operation
2-1 7
Starting Operatooll1l
line connected to a 270x transmission control unit (TCU). The VM!SP alternate
system console cannot be a remote terminal connected to a 37xx TCU.
Note: If the 37xx is loaded with the 270x Emulation Program prior to initializing
VM!SP, any line defined by EP will appear to VM!SP as 270x TCU.
The following examples show the various configurations that can control VM!SP
operations:
•
First Configuration:
Real system console = VM!SP system console
Real alternate system console = VM!SP alternate system console
•
Second Configuration:
Real system console = Batch virtual machine
Real alternate system console = VM!SP system console
Remote terminal = VM!SP alternate system console
Third Configuration:
Real system console = VM!SP system console
Remote terminal = VM!SP alternate system console
Fourth Configuration:
Real system console = Batch virtual machine
Two locally attached 3277 Display Stations = VM!SP system console and
alternate system console
Fifth Configuration:
Enabled remote terminal (not 3270) = VM!SP alternate system console
(No valid VM!SP primary system console is designated in SYSGEN)
Note: The real system consoles must be turned off to prevent VM!SP
from using them.
•
Sixth Configuration:
System console = VM!SP system console (No VM!SP alternate console
designated in SYSGEN)
The first four configurations allow disabling of the VM!SP primary console and
selection of an alternate VM!SP console; or, conversely, reselection of the VM!SP
primary console and disabling of the alternate console.
Configurations five and six indicate that VM!SP system generation specified no
console substitution. Therefore, if the designated VM!SP system console is inoperative at VM!SP startup, the system enters the wait state and cannot continue.
2-18
VM/SP Operator's Guide
,/
After the console becomes operative again, VM/SP must be reloaded into the real
system.
Disabling the VM/SP Primary Console
If the VM/SP primary system console is inoperative, VM/SP automatically selects
the first alternate console (specified in the RIOGEN macro instruction) as the
console for primary system operation. If the first alternate console specified is not
operational, an attempt will be made to start the next alternate console. If an operational console is found, it will be used as the VM/SP system operator's console.
If intermittent errors occur at the VM/SP console, and all alternate consoles are
unavailable (or not specified in the RIOGEN macro instruction), issue the
DIS CONN command and log on again at some other terminal.
If the VM/SP system console becomes inoperative to the point where you cannot
log off, press the System/370 external INTERRUPT button to disconnect you
from the failing system console (or alternate console if that console is currently
controlling the system.) This allows you to log on and reconnect to a terminal with
a communications line that has already been enabled. This terminal can be another
system console (VM/SP supported device) or a remote terminal attached to a 270x
TCU. When the primary console is again operative, control can be regained from
the alternate console by pressing the INTERRUPT button and reconnecting using
the primary console.
When the device specified as the alternate console is a communication line, VM/SP
enables the line. You can then establish a link with the processor through a terminal connected to the line. Once the line is established, VM/SP proceeds with
the normal system initiation. The alternate console must not be a telecommunications line on a real IBM 37xx Communications Controller. If the alternate
console is an IBM 2741 Communication Terminal, it must use the EBCDIC transmission code.
Note: VM/SP 37xx supports multiple alternate console selection only in 270x
emulator mode.
Selecting a VM/SP Alternate Console during VM/SP Initialization
If the VM/SP system console is inoperative, a VM/SP alternate system console
can start VM/SP. To do this, place the primary VM/SP system console in a notready status.
Load VM/SP in the normal way (with the System/370 console switches and the
LOAD key). When the primary system console (assuming it was generated as the
VM/SP system console) indicates "not ready", VM/SP rings the console alarm
bell and enables the alternate system console (such as a 3210 Model 2, 1052
Model 7, or local 3270 if one is specified) or one 2701, 2702, or 2703 transmission
control unit (TCU) line that has a leased line terminal attached, to allow the system
operator to log on. The TCU line address must be defined in the REALIO module
during VM/SP system generation. If the real system consoles are dedicated to
virtual machine use only, the system can be generated with a fictitious VM/SP
Section 2. Starting VM/SP Operation
2-19
primary system console address and the same results occur as the load operation
with the not ready console described above.
You can now establish a link with VM/SP by using the designated device. Once
CP identifies the terminal in the normal procedure, the operator presses the
BREAK, ATTN, RESET, or ENTER key, depending on the terminal type. The
VM/SP system then performs automatic or manual logon procedures, as previously
discussed.
Controlling the 370x Communications Controller
This section only applies to EP gens as defined, created and loaded with VM/SP.
If you have a 3725 communication controller or a 3705 that has been loaded by
ACF /NCP-SSP, refer to the manuals listed in the preface for control information.
Because the 3704 and 3705 Communications Controllers are programmable
devices, in order for them to be controlled by VM/SP, they must be loaded with an
appropriate control program before VM/SP uses them for system telecommunications. This loading is normally an automatic function of VM/SP system initialization, but it can be performed manually.
To load a 370x program into the 370x (either manually or automatically), the 370x
program must previously have been saved on one of the VM/SP system volumes.
Specifications in the VM/SP configuration module, DMKRIO, describe whether
the load process is automatic or manual. For more information on saving or
loading a 370x control program, see the VM/SP Planning Guide and Reference and
the VM/SP Installation Guide.
If manual loading is necessary or desirable for a 370x control program, you should
use the NETWORK LOAD command as follows:
NETWORK LOAD raddr ncpname
where raddr specifies the physical device address of the 370x, and ncpname is the
name of a 370x control program image.
If the specified 370x Communications Controller does not require loading (that is,
it already has an active control program loaded), the following message appears on
your console:
CTLR raddr IPL NOT REQUIRED; ENTER 'YES' TO CONTINUE:
You should verify that you specified the correct 370x, and reply "YES" only if the
load is to be attempted. (Note that execution of the NETWORK LOAD command
causes the 370x to be reset, thus discontinuing any current teleprocessing activity.)
2-20
VM/SP Operator's Guide
§ .~~ W·~ nIn ~ (D ~ e u--~r~ n(1)[11
---,
Other messages relating to the operation of the 370x can appear on your console.
These messages are described in VM / SP System Messages and Codes. For more
information about the facilities of the NETWORK command, refer to "Network
Dump Operations" on page 4-41, Section 3, "CP Commands," and "NCPDUMP
Service Program and How To Use It" on page 4-42.
Considerations for Loading All 370x Control Programs
When using the NETWORK LOAD command to load EP for the 370x, the following events occur:
Active users are disconnected.
Active I/O operations are reset.
Dedicated devices are detached and released.
Dialed lines are released.
Devices enabled but not dedicated are reset.
Binary synchronous communication are reset.
Special Considerations for Loading the EP 370x Control Program
If a 370x Emulation Program (EP) is automatically reloaded after a 370x failure,
the system may loop after the restart. The message
CTLR xxx UNIT CHECK; RESTART IN PROGRESS
and two responses
CTLR xxx DUMP COMPLETE
CTLR xxx ncpname LOAD COMPLETE
indicate that the 370x has been reloaded. If the system loops after the second
response, you must reset all emulator lines from the 370x control panel.
If the automatic dump feature is not enabled, one of the messages
CTLR raddr UNIT CHECK; IPL REQUIRED
CTLR
I
raddr
I
CC=3; DEPRESS 370X "LOAD
II
BUTTON
indicates a 370x abnormal termination. You must reload the 370x Emulation
Program with the NETWORK LOAD command. If the system loops when an
attempt is made to enable the lines, you must reset all emulator lines from the 370x
control panel.
The Guide to Using the IBM 3704 Communications Controller Control Panel and
the Guide to Using the 3705 Control Panel describe the procedure for resetting
emulator lines from the 370x control panel in their "Generating Channel
End/Device End with Emulator Program" section.
Section 2. Starting VM/SP Operation
2-21
Messages On Startup and Initialization
In the course of VM/SP initialization or startup, in addition to the normal expected
messages that are a part of loading the system, you may see other messages to
which you must respond. These messages relate to functions of initialization,
checkpoint, warm start, and automatic checkpoint requirements. These messages
and the recommended course of action are described in VM / SP System Messages
and Codes.
Control Messages after Startup
System console messages are in three categories:
•
•
System operation status messages
Virtual machine user messages
Hardware and program status messages.
The operator action required depends on the type and content of the message.
System Operation Status Messages
These messages inform the system operator of the logging on and logging off of
users, number of spooled files, output printer or punch accounting data, device
attention conditions, and command completion indications. Some of these messages require operator action, while others do not.
Messages from Virtual Machine Users
These messages include requests and inquiries from virtual machine users on the
system; a sampling of these user messages follows:
MSG
MSG
MSG
MSG
FROM
FROM
FROM
FROM
JONES:
SMITH:
SMITH:
BROWN:
CAN YOU GIVE ME A TAPE
CAN YOU PUT UP A SCRATCH 181
PLEASE TAKE RING OUT OF 181 NOW
PLS ATTACH 382
Hardware and Program Status Messages
These messages include VM/SP error recording messages, VM/SP system error
information, and I/O error messages related to users of virtual machines who have
invoked SVC 76. SVC 76 causes VM/SP to send a message to the system operator (for I/O error records only) and to record the error data in VM/SP's error
recording area. The VM / SP System Messages and Codes manual describes these
messages.
For more information about error detection and the recording and editing of errors,
refer to the VM / SP OLTSEP and Error Recording Guide.
2-22
VM/SP Operator's Guide
S'~ar'~a~~ (Q)~Q)er~r~n((»~
-----------------------_._----
c=--
J
The following summarizes the action that the operator must take in response to an
error message:
1.
Determine the severity of the error.
2.
Take a storage dump if the automatic dump routine failed.
3.
Try to preserve spooling and accounting data by issuing the SHUTDOWN
command if it has not been done automatically.
4.
Try to force error recording if it did not occur.
5.
Attempt to analyze the problem by whatever means at his disposal and if necessary, reallocate resources and priorities.
6.
Call for technical assistance if necessary.
Whenever there are indications that I/O errors require further analysis or that basic
integrity of a device needs to be determined, other service programs may be
invoked to help make such an assessment. For example, to test the operational
3344 or 3350 DASD devices, the operator may invoke the OS/VS and DOS/VS
Analysis Program-l (AP-l). This utility program, if resident in the OS/VS or
DOS/VS library of a virtual machine, can be used to check the proper operation of
the device. The use of this program along with error messages and user action is
detailed in the OS/VS and DOS/VSE Analysis Program-I (AP-I) User's Guide,
GC26-3855.
In the event that technical assistance is required, it can be obtained from the following sources:
The customer's own system programmer(s).
The service representative, who has special test equipment for hardware
problem isolation.
The IBM Program Support Representative, who can also assist with VM/SP
system problem diagnosis.
You can assist with problem diagnosis by being alert for the following error
recording messages:
I/O ERROR RECORDING AREA 90 PERCENT FULL; RUN CPEREP
I/O ERROR RECORDING AREA FULL; RUN CPEREP
You should then dump the error recording data to some other device (usually a
tape drive) and clear the error recording area to insure enough space to record
additional error records. Information on how to run CPEREP is contained in
VM/SP OLTSEP and Error Recording Guide. Additional information on EREP,
the program to which CPEREP passes control, is detailed in Environmental
Recording Editing and Printing (EREP) Program.
Section 2. Starting VM/SP Operation
2-23
Other errors may not be severe enough to cause VM/SP system failure, but might
cause a specific virtual machine to abnormally terminate. In most cases, the user
can IPL the virtual machine again.
Should the VM/SP machine enter an unexpected WAIT state or loop, see the
section entitled "How To Use the VM/SP Facilities To Debug" in VM / SP System
Programmer's Guide.
I 3480 Magnetic Tape Subsystem Support
VM/SP supports the 3480 magnetic tape subsystem with the following restrictions:
•
If you IPL a standalone program from a 3480, the 3480 is not assigned to the
processor that IPLed the 3480. Because of this, do not IPL any other
processor that has a path to the 3480, since the other processor might steal the
3480 by assigning it to itself.
•
When a 3480 is assigned to one processor, another processor's attempt to
perform I/O to the device (for example, write, read, forward space file) will
receive an assigned Elsewhere Unit Check. The processor that owns the
device assignment will not be able to perform I/O to the device until the other
processor clears the contingent connection that the Unit Check established (for
example, issuing a SENSE command or Signal System Reset). This problem
can occur if a processor tries to IPL a 3480 device that is assigned to another
system, but fails to clear the resulting unit check.
When running Single Processor Mode, the MVS MP system can perform its
I/O to a 3480 device from only one of its processors, either the V =R
processor or the MVS native processor. Before the MVS native processor can
use a 3480 device, VARY OFF the device from the CP system if the device is
not already offline.
I 3800 Printing Subsystem Support
The 3800 models 1 and 3 are supported as dedicated, spooled, and virtual devices.
The 3800 model 8 is supported as a dedicated and spooled device. The 3800
models 3 and 8 are supported in 3800 model 1 compatibility mode and provide the
following:
An improved pel density of 240 x 240 pels
A 10 line per inch vertical spacing.
Following is a list of features for the 3800 printers:
FORMS CONTROL lets the user select the amount of vertical space between
printed lines. Lines can be vertically spaced at 6, 8, or 12 lines per inch. Users
of the 3800 Model 3 printer have an additional option of 10 lines per inch.
You can also mix different spacings on the same page.
2-24
VM/SP Operator's Guide
§'~a r'~o U1~ (Q) LQ)erCr~O(i)UU
r::=---------------------..----.-.-------------.--.-.-----..- .. -.. ---.---------.---.....-.-.. -... -.-.-----.-.--.. ----J
FORMS OVERLAY permits photographing of predefined data (forms) on one
or more pages. The 3800 printer supports the creation of a form on a page and
the immediate printing of data on that form.
PREDEFINED CHARACTER SETS lets the user select previously designated
character sets that have up to 64 characters each. Different character sets
provide varied character types, sizes, and horizontal spacings between the
characters. The 3800 supports the simultaneous use of two character sets and
optionally supports the use of two additional character sets. The user can mix
characters from the different character sets on a single page. Use the translate
tables to reference the characters within the 3800. You can load up to four
translate tables at anyone time.
CHARACTER SET MODIFICATION lets users modify and extend character
sets.
Note: Due to the change in pel density, customized 3800 Modell character
sets are not interchangeable with the 3800 Model 3 character sets. To convert
3800 Model 1 character sets, you can:
Recode the customized character sets in the 3800 Model 3 pel resolution,
and use the CMS GENIMAGE command to build new modules.
Use the MVS Character Conversion Aid to convert customized character
sets to the 3800 Model 3 pel density. Then use the CMS GENIMAGE
command to build new modules.
•
COpy MODIFICATION permits printing of predefined data (for example,
printing the same header on each page) or the suppression of printing for
selected data.
•
MULTIPLE COPIES permits numerous copies of files and datasets to be
printed without the use of multiple-ply paper. Numerous copies of a single
3800 buffer can also be printed.
The formatting of 3800 output is controlled by character arrangement tables,
library character sets, graphic character modification modules, and forms control
buffers. These are stored in an image library and are loaded into the 3800 before it
prints a spool file. They may also be imbedded directly in the spool file. Use the
START command to control information (including the default character set and
FCB) for the 3800.
Dedicated 3800 Printing Subsystem Support
VM/SP lets virtual machines, including virtual VM/SP, attach the 3800 printer as
a dedicated I/O device. When attached, the full capability of the 3800 is available
to the virtual machine. All of the functions previously described are available when
the 3800 is used as a dedicated device, provided the device is specified in the
RDEVICE macro instruction at system generation. See the VM / SP Planning
Guide and Reference.
Section 2. Starting VM/SP Operation
2-25
Performance Options
VM/SP offers the following performance options to enhance the operating efficiency of one or more virtual machines. This enhancement is normally at the
expense of the performance of other virtual machines. In certain instances,
enhancements to a specific virtual machine (or multiple virtual machines) are detrimental to the operating efficiency of other virtual machines. The performance
options are:
•
Virtual machine assist
Extended Control-Program Support: VM/370
Reserved page frames
Favored execution
Virtual = real
Locked pages
Priority
Affinity in an attached processor environment
Queue drop elimination.
You can dynamically assign or withdraw any of these options except for the
virtual=real option (it can only be dynamically withdrawn) from virtual machine
users during system operations. You can assign to one virtual machine a "mix" of
performance options, or the options may be assigned to separate virtual machines.
Efficient use of these options depends upon the characteristics of the program
applications; the number, size, and complexity of the virtual machines being run;
and the size, I/O configuration, and processor model of the real computer being
used.
In addition to using these performance options, you can also use the SET PRIORITY command to change a virtual machine's execution priority and, thereby,
change its performance.
For example, you could assign the following mix of options.
One virtual machine:
Reserved Pages= 10
Favored Execution=80 %
Virtual Machine Assist
Another virtual machine:
Virtual=Real
Priority = 1
A third virtual machine:
Locked Pages=2
The VIRT=REAL and reserved page options are unique in that both options
cannot be applied to multiple virtual machines at the same time. However, no such
2-26
VM/SP Operator's Guide
restrictions apply to the locked pages, virtual machine assist, Extended ControlProgram Support: VM/370, priority, or favored execution options; they can be
applied to multiple virtual machines.
Reserved Page Frames Option
The CP paging routine uses chained lists of available and page able pages to control
real storage use. CP assigns pages for users from the available list, which it replenishes from the pageable list.
Pages that are temporarily locked in real storage are not available or pageable.
Paging proceeds using demand paging to select the best page for swapping. The
reserved page option gives a particular virtual machine an essentially private set of
pages. The pages are not locked; they can be swapped, but normally only for the
virtual machine for which they are reserved. You specify the number of reserved
pages for the virtual machine as a maximum. When the page selection routine
selects an available page for a reserved user, it marks that page reserved if the
maximum specified for the user has not previously been reached. If CP encounters
an available (idle) reserved page for the reserved user during page selection, it is
used whether or not the maximum has been reached.
Only an operator with privilege class A can specify the number of reserved pages.
Issue the CP SET RESERVE command in the following format:
SET RESERVE userid xx
userid
is the name of the VM/SP user to whose virtual machine you are to assign
the reserved page option.
xx
is the maximum number of pages you assign.
If the CP page selection routine cannot locate an available page for other users
because they are all reserved, the routine uses one of the reserved pages as a last
resort.
Favored Execution Option
The favored execution option guarantees that the specified virtual machine gets up
to a fixed percentage of processor time. Favored in this sense means that CP provides execution time up to the specified percentage to that virtual machine, provided that virtual machine can use that much processor time. At regular time
intervals, the CP dispatcher checks the processor time used by the favored virtual
machine. If the guaranteed percentage is exceeded, the machine gets its normal
priority for the remainder of the time interval. If the percentage used is less than
the guaranteed value, the favored virtual machine has the highest execution priority
until it reaches that threshold. You, as the system operator with privilege class A,
specify the percentage of processor time assured by the favored option in the CP
SET command. The format of the command is:
Section 2. Starting VM/SP Operation
2-27
SET
FAVORED userid
xxx]
[ OFF
userid
is the name of the VM/SP user whose virtual machine is to be assigned the
favored option. -
xxx
is any percentage value from 1 to 100.
OFF
operand stops favored operation.
You can assign multiple virtual machines the favored execution option with, andlor
without a percentage value.
Example:
set favored
set favored
set favored
set favored
set favored
operatns 75
payroll 25
revenue
hot jobs
hot jobs 10
Although the SET FAVORED command will prevent the specifying of more than
100% for a particular virtual machine, nothing is done to prevent more than 100°1<>
being allocated to a number of virtual machines. In situations where more than
100% has been allocated, the individual favored virtual machines will compete for
the available resources on a pro-rata basis: an individual virtual machine's allocation will be roughly proportional to the percentage allocated to it divided by the
total percentage allocated to all virtual machines. The effect that allocating more
than 100% of the system has on interactive (Ql) response is unpredictable.
Virtual=Real Option
The virtual=real option permanently relocates the CP nucleus (except for real page
0) to provide space in real lower storage to contain the largest virtual=real
machine. For the virtual machine, each page from page 1 to its last or nth page is
in its true real storage location; only its page zero is relocated. The virtual machine
still runs in relocate mode, but because each virtual page frame address is the same
as the real page frame address, the virtual machine requires no CCW translation for
the virtual machine. Because no CCW translation is performed, no check is made
of the 110 data addresses. The virtual machine must ensure that no 110 data
transfer occurs into real page zero or beyond the upper boundary of the virtual
machine's storage. Failure to observe these limits can cause damage to the VM/SP
system or to other virtual machines.
Initial program loading of a named or shared system is not allowed in the
virtual=real machine; the hexadecimal device address must be used.
/
2-28
VM/SP Operator's Guide
§'~ar"lcn01~ O~Q)er~r~n(Q)D1
J
Several aspects of the operation of the virtual=real option can affect overall system
operation:
1. The area of contiguous storage built for the virtual=real machine must be large
enough to contain the entire addressing space of the largest virtual=real
machine.
2.
Only virtual machines with the virtual=real option can use the real storage
reserved for virtual=real machines. Only one virtual=real virtual machine can
run at a time. The storage is not available for other users, nor for VM/SP use,
unless the virtual=real machine is not logged on and you issue an UNLOCK
command with the VIRT=REAL option. When you issue this command, all
page frames previously reserved for the virtual=real machine are permanently
relinquished to the available-page pool. Once these virtual=real page frames
are available to the page pool, no virtual=real machine can run until you load
VM/SP again (via IPL). For this reason, it is suggested that the virtual=real
machine should be a machine with high availability and heavy workload
demands. It is not possible to relinquish only some of the virtual=real page
frames (if, for example, a smaller virtual=real machine is to be run). All
virtual=real page frames are locked until all are released to the pool.
3.
The virtual=real machine operates in its allocated storage area with normal
CCW translation in effect until the virtual machine operator issues the CP
command:
SET NOTRANS ON
At this time, all subsequent I/O operations occur from the virtual CCWs in the
virtual=real machine without translation. With NOTRANS on, the virtual
machine must not perform I/O operations into page zero or beyond its
addressable limit. As mentioned previously, violation of this requirement can
cause damage to the VM/SP system or to other virtual machines.
4. If the virtual=real machine performs a reset or IPL, the normal CCW translation routine regains control until the virtual machine operator again issues the
SET NOTRANS ON command. Only the virtual=real machine operator can
use the command. An operator message occurs if the virtual machil1e enters
normal translation mode.
The virtual=real option is used with programs or operating systems that dynamically modify channel programs, or that wish to avoid CP virtual channel program
translation and paging degradation. If virtual=real operation is desired, it must be
specified at VM/SP system generation time. Virtual=real assigns the same real
storage locations to the virtual machine that exist on the real machine (except for
page zero, which is relocated).
For example, in Figure 2-1 on page 2-30, if a DOS system needs 160K bytes to
run virtual=real, CP puts all but the first 4096 bytes of the program at the real
storage address locations. Page 0 of the DOS supervisor is the only part of the
DOS virtual machine that can be relocated, and it follows the last real page frame
of the DOS machine. Figure 2-1 on page 2-30 also shows that the major portion
of the CP nucleus is permanently relocated into upper storage (except for page
Section 2. Starting VM/SP Operation
2-29
zero) whether the virtual=real system is operating or not. The real page frame
zero remains under control of VM/SP. CP dedicates 160K nonpageable bytes to
the DOS machine.
Real
Storage
Addresses
512K
VIRTUAL STORAGE POOL
for other users
260K
RELOCATED CP NUCLEUS
RELOCATED PAGE ZERO
for the DOS machine (locked in real storage)
164K
160K
156K DOS VIRTUAL=REAL MACHINE
CP PAGE 0
Figure
2-1.
4K
DOS 160K Virtual=Real Storage Assignments
However, when the user no longer needs the virtual=real machine performance
option and he logs off, the assigned virtual=real storage can be reallocated as available pageable storage for use by other virtual machines (assuming that no other
users with the VIRT=REAL directory have need of the area). The reallocation of
virtual=real storage as usable and available storage for other virtual machines is
controlled by the UNLOCK command and the VIRT=REAL operand. Once the
pages are released, no virtual=real machine can be run unless VM/SP itself is
reloaded. If more than one virtual machine needs the capabilities of running
virtual=real (at different times), VM/SP system generation parameters must
specify the largest virtual machine that is to run in the virtual=real environment. A
more practical approach may be to generate a different version of VM/SP for each
virtual=real machine, if the storage sizes are substantially different.
,/
2-30
VM/SP Operator's Guide
Locked Pages Option
If sufficient page frames are available for other users, you can choose to lock (fix)
one or more specific pages of a virtual machine or the pageable CP nucleus in real
storage. This performance option differs from the other options in that the LOCK
command can apply to several virtual machines. It is also possible to have the
reserved page frames option apply to one user's virtual machine, and then have the
lock option apply to another user's virtual machine. A locked page is effective
from the time you invoke the lock option until the user logs off the system, or until
you issue the UNLOCK command. If the page or pages that are locked have a
high frequency of reference or change activity, the lock option, like the reserved
option, can cause an improvement in the user's virtual machine performance or
overall system performance because of the reduction in paging activity.
Notes:
1.
If a named system currently has the locked pages option in effect, reloading of
the same named system, loading of another named system, or a system reset
does not affect the locked pages option of a virtual machine. The loading of
named systems over other named systems without using the UNLOCK
command beforehand results in pages being made unavailable for use if the
first named system has locked pages in effect.
2.
In systems generated for AP and MP operation, shared pages cannot be locked.
3.
You should always prefer the reserved page frames option to the locked pages
option, if it is not already in use. The LOCK command locks specific pages
into real storage regardless of activity, while the reserved page frames option
keeps all the currently most active pages in real storage.
Priority
You can assign specific (dispatch) priority values to different virtual machines to
override the assigned or default priority value for each user in the VM/SP directory. In doing so, the virtual machine with a lower nn value is considered for dispatching before a virtual machine with a higher nn value. You set user priorities by
using the following class A console function:
SET PRIORITY userid nn
userid
is the user's identification.
nn
is an integer value from 1 to 99. The higher the nn value, the lower the dispatching priority.
Section 2. Starting VM/SP Operation
2-31
VM/370 Assistance - Hardware Assist
The overhead associated with CP's handling of virtual machines and their collective
I/O resources is extensive and time consuming. To reduce this overhead, CP is
assisted by System/370 processor hardware that duplicates some of the functions
provided by CP routines. This hardware assistance to program function is called
VM/370 hardware assist. Some form of VM/370 hardware assist is available on
most, but not all VM/SP-supported processors. For a complete list of these
processors that support VM/370 hardware assist and ECPS:VM/370 see VM/SP
Planning Guide and Reference. VM/370 hardware assist on certain processors can
consist of virtual machine assist; on other processors VM/370 hardware assistance
is composed of virtual machine assist and VM/370 Extended Control Program
Support (ECPS:VM/370).
VM/370 hardware assist is activated when VM/SP is loaded or by the use of CP
commands, and it is deactivated by CP commands. For the command that
activates/deactivates the VM/370 hardware assist process, refer to the SET
command described in Section 3, "CP Commands." The status of VM/370 hardware assist may be determined by the use of the QUERY command also described
in Section 3, "CP Commands."
Virtual Machine Assist
The virtual machine assist facility can be ordered via feature number, it is available
as an RPQ (Request for Price Quotation), and is standard on some processors. For
a list of VM/SP supported processors and how they support virtual machine assist
and ECPS:VM/370 see VM/SP Planning Guide and Reference. On attached
processor or multiprocessor systems, virtual machine assist may be installed on one
or both processors. Virtual machine assist relieves CP of some of the processing
overhead caused by the simulated execution of certain instructions, privileged operation instructions, and interrupts of virtual machines. On/off system control of
virtual machine assist is provided by the A privilege class command:
SET
SASSIST ION
OFF
I
[[PROC] nn]
Users of virtual machines, in turn, have discrete control over the functions of
virtual machine assist by means of directory entry options or by command. For
control of this function on a virtual machine level, refer to the privilege class G
command SET ASSIST, described in VM/SP CP Command Reference for General
Users.
Extended Control-Program Support:VM/370
Extended Control-Program Support:VM/370 is an expansion of the function provided by virtual machine assist and consists of three major parts:
Expanded virtual machine assist
CP assist
Virtual interval timer assist.
2-32
VM/SP Operator's Guide
/
§'~~~'~nuug OUJ)en-~r~n(Q)nu
[
-------.~---------------.
J
Expanded virtual machine assist is an expansion of function provided by virtual
machine assist in that, via hardware, CP is relieved of processing some conditions
of CP simulated instruction execution that were not covered by the virtual machine
assist process. Expanded virtual machine assist also assists CP in the processing of
other instructions and privileged operations not emulated by virtual machine assist.
Expanded virtual machine assist is activated for all virtual machines when VM/SP
is loaded. It is also activated when both class A commands are invoked as shown:
SET SASSIST ON
SET CPASSIST ON
Expanded virtual machine assist is deactivated when either SET SASSIST OFF or
SET CPASSIST OFF is invoked.
The CP assist portion of the Extended Control-Program:VM/370 Support duplicates via hardware the function provided by various high-usage significant overhead portions of CP routines. CP assist is activated upon system initialization or
by invoking the privilege class A command:
SET
CPASSIST (ON
OFF
I
[[PROC] nn]
When the OFF operand of CP ASSIST is invoked, CP assist is deactivated (as well
as expanded virtual machine assist portion of ECPS, if previously active).
For attached processor applications, specify PROC nn to activate or deactivate CP
assist on the desired main or attached processor. If you desire the same setting for
both processors, omit the PROC nn option and use the ON or OFF options to activate or deactivate CP assist for both the main and attached processor at the same
time.
The third function provided by ECPS:VM/370 is virtual interval timer assist. This
hardware assist provides virtual machines with a more accurate method of updating
the virtual machine's interval timer (location X'50'). Overall system control of this
function is by the class A commands SET SASSIST ON and SET SASSIST OFF.
Control of this facility on the individual virtual machine level is by the class G
command SET ASSIST TMR and SET ASSIST NOTMR. For more details on this,
refer to the VM / SP CP Command Reference for General Users.
To summarize the use of the SET CP ASSIST and SET SASSIST in relationship to
virtual machine assist and Extended Control-Program Support:VM/370, see
Figure 2-2 on page 2-34.
~I
Section 2. Starting VM/SP Operation
2-33
VM/370:
Hardware Assist
SET
SET
SASSIST ON
SASSIST ON
SET
SET
CPASSIST OFF CPASSIST ON
Virtual Machine
Assist
X
X
Virtual Interval
Timer Assist
X
X
Extended Virtual
~1achine Assist
X
CP Assist
X
legend:
Figure
2-2.
SET
SASSIST OFF
SET
CPASSIST ON
SET
SASSIST OFF
SET
CPASSIST OFF
X
X = function activated
Command Control of VM/370 Hardware Assist
Note: When you install the CPFRET Trap, CP disables CP Assists DSP1, DSP2,
UNTFR, FREE, and FRET. This causes performance to lower for systems that
use CP Assists. However, this lowering in performance is not expected to be a
problem when CP traps suspected free storage problems. When the trap is not
installed, system performance remains the same.
VM/370 Hardware Assist Restrictions
A virtual machine running in certain virtual machine modes or running certain
operations cannot use certain facilities provided by VM/370 hardware assist.
When these modes or operations are encountered, the action taken is dependent
upon the processor facilities of either virtual machine assist or Extended ControlProgram Support:VM/370.
For all VM/SP supported processors that have VM/370 hardware assist active
when the CP ADSTOP or certain CP TRACE command functions are encountered, the virtual machine assist is deactivated during the interval of their use.
For System/370 Models 135-3, 138, 145-3, and 148, the use of the DOS emulator
or PER (Program Event Recording) curtails some of the function provided by
Extended Control-Program Support:VM/370. For more details on VM/370 hardware assist, see the VM / SP System Programmer's Guide.
Affinity
In attached processor or multiprocessor systems a virtual machine may be designated to execute only on one specified processor. This may be advantageous for
performance or functional requirements. Performance gains might be realized with
affinity because of the nature of the workload or the availability of performance
assists on one processor. Functional requirements could be the existence of
required operation codes on one processor. In either case, CP code executed on
behalf of the virtual machine may execute on either processor.
2-34
VM/SP Operator's Guide
§.~~ w·~o uu~ ([D [Q)G[(Er~O(Q)O"il
r:=-------------:.-_--_--_--_-_----_--==_----_--_-_-_________.:::.:::.:::.::-.:::.::-_-__-_---_'--_--_---_--_--_-_-_--_-_--_-_-_-_'-_---_---_-'-_---_--_-----1--]
Queue Drop Elimination
VM/SP attempts to optimize system throughput by monitoring the execution status
of virtual machines. When a virtual machine becomes idle, VM/SP drops it from
the active queue. The virtual machine's page and segment tables are scanned, and
resident pages are invalidated and put on the flush list.
In certain cases, where cycle of queue dropping and reactivation is repeatedly executed, there is a significant increase in the overhead involved in invalidating and
revalidating pages. One example of this is the SNA service virtual machine processing IUCV messages.
The SET QDROP userid ON/OFF class A command allows an installation to
control this situation. If SET QDROP OFF is in effect for a virtual machine, that
virtual machine's pages are not scanned or flushed when the machine becomes idle.
Specifying SET QDROP OFF for a service virtual machine can improve system
performance and throughput when queue dropping would otherwise occur frequently. Specifying SET QDROP OFF USERS allows the QDROP OFF status to
be extended to any virtual machine communicating via VMCF or IUCV to a
(server) virtual machine for which SET QDROP OFF has been specified. Thus,
performance gains can be realized in systems with heavy usage of products such as
IFS or PVM (invoked via the CMS PASSTHRU command). Note, however, that
indiscriminate use of the SET QDROP OFF command could degrade system
throughput by forcing page stealing to take place.
System Termination
The VM/SP system may be shut down in one of two ways:
1.
2.
By an operator initiated shutdown
By an abnormal termination of operation.
The normal shutdown is initiated by a class A operator entering the SHUTDOWN
command from the operator's console. The format of the SHUTDOWN command
is:
SHUTDOWN [REIPL
]
POWEROFF
If the SHUTDOWN command is ineffective, press the LOAD button on the system
console once to shut down the system (a second pressing of the load button reloads
VM/SP).
Note: When the LOAD button on the system console is used, spooling and
accounting files can be lost because the files are not closed properly.
The POWEROFF option is valid only for 4361 processors. When you specify this
option for a 4361 processor, the system shuts down normally, and the processor is
powered off. If the 4361 processor has the Auto Start feature, you will not be
prompted to set the time-of-day clock when the next IPL takes place.
Section 2. Starting VM/SP Operation
2-35
S11: OJ ~1c DD1 9)
(()) [0 e ~~arrd 0) IT1
c
The shutdown operation checkpoints spooling and accounting functions; in addition, user activity is terminated by automatically logging off all virtual machines
and disabling communication lines. Open and active files on the spool unit record
devices are purged and, if applicable the SUSPEND order is issued to all 3851
device addresses (suspends MSC interface activity to the host virtual machine).
Note that the SUSPEND order issued to a Mass Storage Control interface may
cause the SHUTDOWN command to run longer than usual.
Note: Before issuing the SHUTDOWN command, a message or warning should be
sent to all logged-on users so that they can quickly bring their virtual machines to
an orderly halt. Time should also be allotted to the class D (spooling) operator so
that he too can terminate in an orderly manner the active files being processed on
the unit record devices.
When you specify the REIPL option with the SHUTDOWN command, the system
shuts down and does an automatic warm start. For this type of automatic restart,
you do not need to set the time-of-day clock or choose the type of start you want
the system to do. In fact, you do not need to be present for the automatic warm
start to occur.
The dump unit for VM/SP system failures is specified in the SYSDUMP operand
of the SYSOPER macro during VM/SP system generation, but can be changed by
a privilege class A or class B system operator.
If the dump unit is set to disk (by default or via the SET DUMP AUTO command)
at the time of system failure, the system dumps all or parts of real storage to the
specified disk and automatically restarts the VM/SP system.
When automatically restarted, the system preserves all accounting information and
spool file data on disk, performs an automatic logon of the primary system operator, restores the system LOGMSG, and continues system operation. It is not necessary to re-enable the lines, since dump and re-IPL automatically re-enable them.
The VM/SP online message is sent to the terminal users, indicating that they must
log on again.
When the system is automatically restarted after a system failure, the system operator is automatically logged on only if he was logged on the primary system console
at system failure time. If logged off, disconnected or logged on elsewhere, the
operator must explicitly log on.
If, at the time of the system failure, the dump unit is set to a printer or tape, the
VM/SP system writes the dump on the specified unit, preserves the spooling and
accounting data, and stops. The operator must then re-IPL the VM/SP system as
for normal system start up, specifying a WARM START to preserve the accounting
and spooling file data. The operator must re-enable the communication lines to
permit users to log on again.
2~36
VM/SP Operator's Guide
./
s·~ a u-.~ II [ii} ~ (Q) Lu)(2H-- ~r~ II CO> [ii}
--------::J
System Abend Dumps
Conditions can occur within the CP program that may force an abnormal ending
condition (ABEND) and cause the dumping of system registers and storage. The
device that receives these records can be a tape, printer, or disk device.
Dumping operations are caused by any program interruptions or system restart
condition. These interruptions cause routines to gather data from registers and
storage and place this data on a previously defined device. The system CP
command SET DUMP defines the quantity of data to be dumped. For example:
SET DUMP AUTO
SET DUMP raddr
SET DUMP raddr ALL
SET DUMP AUTO ALL
SET DUMP AUTO places the VM/SP system dump on a preselected file device.
(The disk dump area is automatically selected at system initialization time if sufficient contiguous space is available.) You can verify the device type and address by
entering the QUERY DUMP command.
Use SET DUMP raddr when the dump device is to be a high speed printer or tape
device. Substitute the real hexadecimal address for raddr.
The ALL operand used with SET DUMP AUTO or SET DUMP raddr dumps all of
storage onto to the DUMP device. If the ALL operand is not specified, the system
defaults to dumping only those areas that pertain to CP, and not those areas that
pertain to virtual machine operations.
If you dump to a disk file, an additional operation is necessary to transform these
records into readable output for programmers or system analysts. You can do this
with the CMS IPCSDUMP program.
Note that expanded function is available to the IPCSDUMP command as generated
for the IPCS virtual machine.
If the records are dumped onto a tape drive, other CMS command options must be
invoked for printout.
Only abend dumps that are a result of using the SET DUMP AUTO command are
spooled as a special virtual card reader file. This card reader file is assigned during
system generation to a specific virtual machine user via the SYSOPR macro. The
CMS IPCSDUMP command formats and prints these CP abend dumps.
The IPCSDUMP command creates a CMS file from the CP disk dump data and
prints the dump from the CMS file.
Section 2. Starting VM/SP Operation
2-37
VM/SP Interactive Problem Control System (VM/SP IPCS) provides installations
with expanded facilities for reporting and diagnosing software failure. The
VMDUMP command allows the user to dump a virtual machine's storage area.
The VM/SP IPCS component or a user written program must process the file
created by the VMDUMP command. For additional information, see the VM / SP
Interactive Problem Control System Guide.
I"
2-38
VM/SP Operator's Guide
J
[
Section 3. CP Commands
The CP commands are interactive console functions that you can use to control the
VM/SP system. CP commands let you control your virtual machine. The commands that you can issue depend on your assigned privilege class(es), as described
in the following section. If appropriate, CP shows its processing results with
responses (for example: COMMAND COMPLETE, MISSING ARGUMENT, or
INVALID OPTION).
Privilege Classes for CP Commands
Each CP command has one or more of the following function types:
1. Operations
2. Resource
3. Programmer
4. Spooling
5. Analyst
6. CE (Customer Engineer - Service)
7. General.
The IBM -defined class structure is based on these seven function types. A
command keeps its function type even if your installation establishes its own class
structure. In most cases, each command class (A-G) has a corresponding function
type (O,R,P,S,A,C or G). Some commands fall into more than one class.
Figure 3-1 on page 3-2 shows the function of each privilege class and function
type. Refer to the appendix for a list of each CP command, its IBM-defined privilege class, function type, and a description of what each command does.
Your installation will assign each user, as part of the user's entry in the directory,
one or more privilege classes. The exceptions are users with a password of
NOLOG. These users have no privilege class and can only:
Send messages
Receive spooled output as punched cards or printed forms.
The NOLOG password identifies them to receive spooled output when a virtual
machine user spools output for them.
If a user tries to issue a command that does not have his command class, CP does
not execute the command and issues an error message. The CP command
Section 3. CP Commands
3-1
descriptions for classes A-F are in this book. For descriptions of the class G commands refer to the VM/SP CP Command Reference for General Users.
Note: If your installation adds or removes any commands from the general user
class (IBM-defined class G), your installation should update the HELP files to
show these changes. See VM / SP eMS User's Guide for information on "Tailoring
the HELP Facility."
IBM-Defined
Class
Al
Function
0
Function, Primary User, and Use
1. operat;ons -
Primary system operator
The system assigns class A to the user at the
VM/SP console during IPL. The class A user is
responsible for VM/SP's availability and its
communication lines and resources. These
commands control system accounting, broadcast
messages, run virtual machine performance options
and affect VM/SP performance.
Note: The Class A system operator who is automatically logged on during CP initialization is
designated as the primary system operator.
R
Any2
2. Resource - System Resource Operator
These commands control allocation and deallocation of real resources of the VM/SP system,
except those that the primary system operator and
the spooling operator control.
P
3. Programm;ng - System programmer
These commands update functions of the VM/SP
system and change real storage in the real
machine.
S
4. Spool;ng - Spooling operator
These commands control spool data files and
specific functions of the system's unit record
equipment.
A
5. Analyz;ng - System analyst
These commands examine and save certain data in
the VM/SP storage area.
C
6. CE -Service Representative (Customer Engineer)
These commands get and examine data about input
and output devices connected to the VM/SP system.
G
7. General - General User
These commands control functions to run users'
virtual machines.
None
These CP commands are available to any user.
These are to gain and take away access to the
Vr1/SP system.
IThis book describes classes A, B, C, D, E, and F.
2Classes G and Any are described in VM/SP CP Command Reference for
General Users.
Figure
3-1.
CP Privilege Class Descriptions
/
3-2
VM/SP Operator's Guide
How to Use this Chapter
This chapter describes CP commands available to users with the privilege classes of
A, B, C, D, E, and F. These privilege classes are assigned to users who manage the
VM/SP system and its spooling and hardware resources.
Note: The privilege class G commands control the functions of the general user's
virtual machine. Refer to the VM / SP CP Command Reference for General Users
for information on all G privilege class commands and for information on commands with the class of "Any." The Group Control System (GCS) operator commands are described in the VM / SP Group Control System Guide.
Command Description Formats
The descriptions of each command in this chapter have the following sections:
1. When to Use the Command and when to use it.
2.
Describes what you can do with the command
Command Fonnat and Operands - Lists the command format and command
operands. Refer to "Commands, Subcommands and Variables" for information about abbreviations, etc.
3. Things You Should Know about the Command - Lists notes about the
command that pertain to more than one subcommand, or that pertain to specific hardware, etc. Some command descriptions may not require this section.
4.
System Response - Lists the standard system responses from invoking the
various commands and subcommands. Some command descriptions may not
include this section. In those cases, there is either no system response, or the
system response is self-explanatory.
Note: For the QUERY command description, the system responses for each
parameter are included with the parameter description, instead of a separate section
titled "System Response".
Commands, Subcommands and Variables
When you issue a command you may not need to enter the entire command name.
In many cases, you may either truncate or abbreviate the name.
For truncating, in the command format, uppercase letters represent the shortest
version of the command name that you can enter. (Remember, however, that
you can enter VM/SP commands with any combination of uppercase and lowercase letters.)
Example:
INDicate
means that you can enter IND INDI, INDIC, INDICA, INDICAT,
and INDICATE to execute this command.
Section 3. CP Commands
3-3
If there is no minimum truncation listed, enter enter the entire word (repres-
ented by all capital letters ).
•
You can also abbreviate the name of the command. Abbreviations for
command names appear below the full name in the format box. Abbreviations
.for operands are with the operand descriptions that follow the format box.
Example:
Reader
RDR
means that the abbreviation is RDR. Thus, you enter READER as
R, RE, REA, READ, READE, READER, or RDR.
Lowercase letters, words, and symbols that appear in the command format box in
italics represent variables that you will substitute with specific information. For
example, hex/oc indicates that you should enter a hexadecimal storage location
address.
Symbols Used in Command Descriptions
The following symbols help define the command format. Do not type them when
you enter the command.
Braces
{}
/
When the operands are in braces, you must select one of the operands.
Example:
{~}
Brackets
means that you must specify either A, B, or C. If a list of choices
is not enclosed by brackets or braces, treat it as if it is enclosed by
braces.
[]
When the operands are in brackets, you may select one of the operands, but it
is not a requirement.
Example:
[~]
means that you can enter A, B, or C, but you may omit the field.
Ellipsis
When you see an ellipsis following an operand, you may use the operand or
group of operands more than once in succession.
/~
I
3-4
VM/SP Operator's Guide
[
----_._----------_._-
Example:
{spoo/idl spoo/id2 •.• }
means that you can code more than the amount of spoolids shown.
•
Underscore
Defaults are represented with underscores.
Example:
A
!!
c
means that!! is the default. You can choose A, ll., or C.
However, if you do not enter any choice the system will assume B.
Use the following symbols as specified in the format box.
asterisk
colon
comma
equal sign
hyphen
parentheses
period
*
=
()
When there are nested braces or brackets on the text lines, the nested operands
that you choose depend on the higher level operand that you choose. The highest
level of nesting is the operand that is enclosed in only one pair of brackets, and the
lowest level of nesting is the operand that is enclosed by the maximum number of
brackets.
Example:
[level 1
fJilename
[level 2
fJiletype
[level 3] ] ]
fJilemode] ] ]
Lets you select a file by:
•
•
•
Filename only
Filename and filetype only
Filename, filetype, and filemode.
You cannot select file type alone because file type is nested within filename. The
same is true if you want to select filemode; you must also select filename and
filetype.
Section 3. CP Commands
3-5
AleNT
ACNT
Privilege Class: A
When to Use ACNT
Use the ACNT command to do the following:
1.
Process accounting data
2.
Create accounting records for logged-on users
3.
Reset the accumulated accounting data
4.
Close the spool file that is accumulating accounting records.
CP sends a message to each user that indicates the accumulated accounting data,
and resets the accumulated accounting data for each user to zero. (Enter the CP
command, SET ACNT OFF, to suppress the accounting message.)
ACNT Format and Operands
ACNT
useridl [userid2 ..• ] }
ALL [CLOSE]
{
CLOSE
userid
specifies the user(s) for which you want accounting records created. You
can specify more than one user on the command input line.
For each user that you specify, ACNT creates an accounting record. This
record gives the time since the user logged on or the time since you last
issued the ACNT command for the user. The record that ACNT produces is
identical to the record that the system gives when the user logs off of
VM/SP.
ALL
creates accounting records for all logged-on users.
CLOSE
immediately closes the spool file that ACNT used to accumulate accounting
records. CP sends the records to the virtual punch or reader (whichever you
specified as the output device on the SYSACNT macro instruction).
3 -6
VM/SP Operator's Guide
ACNl"
c
If you specify CLOSE with ALL, CP will perform the ALL function first.
Then, CP will do the CLOSE as described.
System Response
CONNECT=hh:mm:ss
VIRTCPU=mmm:ss.hh
TOTCPU=mmm:ss.hh
CONNECT hh:mm:ss
is one of the following:
•
The actual clock time spent in the user's current terminal session
The time since you issued the last ACNT command for the user.
VIRTCPU mmm:ss.hh
is the amount of processor time to execute instructions within the virtual
machine.
TOTCPU mmm:ss.hh
is the total of:
1. The amount of processor time to execute instructions within the virtual
machine, and
2.
The amount of processor time that CP used to maintain the virtual
machine (paging, I/O).
Section 3. CP Commands
3 -7
ATTACH
Privilege Class: B
When to Use ATTACH
Use the ATTACH command when the user requests that you do any of the following:
Logically connect a real device to a virtual machine for that virtual machine's
exclusive use
•
Logically connect a DASD device for CP access and control
Dedicate all devices on a certain channel to the specified user.
ATTACH Format and Operands
AITach
TO
raddr
Jraddr...
1raddr-raddr
I
J userid
1SYSTEM
[AS] vaddr [RIIOJ]
[AS] volid
I
[3330V] [volid]
[TO] userid [R[/O]] [3330V]
CHANnel c [PROC
nn) [TO) ["!.erid]
raddr
raddr•..
raddr-raddr
is the real address of the device you want to attach. On the command line,
you can specify:
One real device address
More than one real device address
A range of real device addresses.
With the range form of raddr (raddr-raddr), enter the values cuu-cuu with no
imbedded blanks. For the range, you must supply contiguous addresses, and
the second address in the range must be greater than the first.
When you specify more than one device or a range of devices, ATTACH
processes all real device addresses, even if ATTACH meets errors while it is
trying to attach the devices.
3-8
VM/SP Operator's Guide
AuuAC~-~
You can specify up to 48 multiple addresses on the ATTACH command line.
If you try to attach more than 48 addresses, the system will:
•
Only process the first 48 addresses
Ignore the remaining addresses and will not issue an error message for
these device addresses.
You may specify more than 48 addresses, however, by specifying the
addresses in a range.
Example:
attach 108-l0P used r
if real devices lOA and 10D were not ready and online, this
command would attach devices 108, 109, lOB, 10C, 10E, 10F.
[TO] userid
identifies the user of the virtual machine that will receive the device(s). The
user must be logged on to receive the device(s).
If you want to attach the device to your own virtual machine, specify userid
as an asterisk (*). You must use TO when:
You specify a userid with 3 numbers or less in a range of addresses or
multiple device addresses.
•
You specify a userid that could also be a valid hexadecimal device
address (for example, virtual machine id CE).
You specify a userid as "to" or "t".
[AS] vaddr
is the virtual address (cuu) you will assign to the device. If you omit the
optional keyword AS, vaddr cannot be A.
Be careful to avoid a sub channel protocol conflict in the virtual control unit
(VCU) when you specify vaddr. Sub channel protocols for all devices that
VM/SP supports are listed in Appendix B under "Virtual Device Characteristics" in the VM/SP Planning Guide and Reference. CP checks for a con·flict in the VCU. If CP finds a conflict, CP does not do the operation and
sends an error message to you. The first device attached to the VCU sets
the sub channel protocol. For the ATTACH command with multiple devices,
CP creates the devices in the order that you specify.
The following is an ATTACH statement that, when trying to attach real I/O
devices to a virtual machine, is restricted by the sub channel protocol.
USER Wilson:
cp define graf as 101 3270
OPERATOR:
Section 3. CP Commands
3 -9
CP ATTACH ZERO TO WILSON AS 10E
CP SENDS THE OPERATOR THE FOLLOWING MESSAGE:
DMKVDE331E DEVICE 10E NOT ATTACHED - USE A NONSHARED VCU
INSTEAD
In the above example WILSON already had a device in the VCU range
(100-10F) that needed SHARED support. In this case, you should contact
WILSON and do one of the following:
•
ATTACH the device to a different virtual address.
•
Ask WILSON to DETACH or move (with DEFINE) all devices in the
100-10F range (Le. device 101).
Note: Such conflicts can also arise when the ATTACH command creates an
address on a VCU, which was created during LOGON by a device defined in
the user's directory entry.
If a device that has minidisks is mounted and ready after the system loads
VM/SP, attach the device to the system. After you have done a DASD
ATTACH, users who had been logged on to VM/SP and who have a directory entry for the minidisk can issue the LINK command to access the
device.
R/O
(or R) lets you apply read-only protection to a user's disk by using it for
attached DASD devices.
If you specify R/O with a range of addresses or with more than one address,
it will only apply to DASD addresses. CP ignores the R/O operand for all
other device addresses.
[TO] SYSTEM
tells ATTACH, along with volid, to make the DASD device available to CPo
CP can then access users' disks.
[AS] volid
is the volume label on the device that CP is attaching. When you use volid to
attach:
•
A raddr that is a 3330V to a userid, CP mounts the MSS volume with
the label volid on the raddr during the attach.
A raddr that is a 3330V to the system, CP mounts the MSS volume with
the label volid on the raddr during the attach.
A raddr that is not a 3330V to a userid, the volid operand is invalid.
./
3-10
VM/SP Operator's Guide
A raddr that is not a 3330V to the system, CP verifies the volume label
of the volume on raddr.
3330V
tells CP to treat the vaddr as a 3330V. CP assumes the virtual machine has
MSS device support and does not invoke MSS cylinder fault support for the
vaddr.
Do not attach virtual 3330 volumes (3330V) to the system; this applies only
to 3330V devices that were generated (at system generation) with
FEATURE = VIRTUAL on their RDEVICE macro. In this case, you cannot
use 3330V for minidisks, CP residence, paging, or spooling functions. On
the other hand, if the system programmer generated 3330V devices with
FEATURE=SYSVIRT on their RDEVICE macro, you can and should
attach the 3330V devices to the system.
CHANnel c
is the real address of the channel you will attach to the user. c can be any
hexadecimal character from 1 through 9 or A through F . You can only use
CHANNEL for real channels whose devices do not have defined alternate
paths.
The device addresses that the virtual machine user specifies must be the
same as the real addresses (he cannot use minidisks) . You do not have to
attach the devices one at a time on the attached channel to the user; he can
use any or all of the devices. However, even if the devices are idle, you
cannot attach the devices on the channel to another user.
When you issue CHANNEL, CP checks if the SHARED and NON SHARED
device types are defined to the same virtual control unit (VCU). If CP
detects a conflict, it does the operation for all devices except those that
would cause a subchannel protocol conflict. CP sends an error message for
the devices not attached. However, to permit migration of data on 3420
tapes (a shared device) to the 3480 (a nonshared device), VM/SP will not
enforce this restriction.
The first device applied to the VCU sets the sub channel protocol. CP
attaches the devices in ascending order by device address. See Appendix B
in the VM / SP Planning Guide and Reference for a complete listing of the
. Sub channel Protocols.
Notes:
1. If you issue an ATTACH CHANNEL command right after a DETACH
CHANNEL command, you may get a message that the specified channel
is in use. DETACH CHANNEL processing schedules I/O at different
times for each dedicated device; but, DETACH CHANNEL processing
does not wait for all I/O to finish before it returns control to th~ terminal operator. Because of the timing, the ATTACH CHANNEL
command may fail.
I
~'
Section 3. CP Commands
3 -11
AuuAC~-~
r:::===
1
2.
A virtual machine should not issue a Clear Channel to any dedicated
channel; to do so would cause unpredictable results.
PROC nn
is the address of the processor that owns the channel. CP does not
require this operand in uniprocessor or attached processor mode. If
you do specify PROC, CP will check its validity.
[TO]
~serid]
identifies the user of the virtual machine that will use the dedicated
channel. The user must be logged on to get control of the channel.
If you want to attach the channel to your virtual machine, specify the
userid as an asterisk (*). If you omit the optional keyword TO, userid
cannot be TO or T.
Things You Should Know about the ATTACH Command
1.
You can attach a disk to "SYSTEM". If no one accessed the disk when it was
on the system, you can attach this same disk to a user without first detaching
the disk.
2.
If a user finds he is using certain devices many times, he may want to attach
these devices to his virtual machine with the DIRECTORY DEDICATE statement. Thus, when the user logs on, these devices (if online) automatically
attach to his virtual machine.
When virtual DASD volumes are associated with the 3850 Mass Storage
System, the ATTACH command can attach 3330V devices to a virtual
machine, even if the volume is not mounted. This is because CP does not stage
cylinder 0 on a virtual volume, until you issue an MSC mount order that associates the virtual unit address with a volume serial. Even if you issue a mount
order before the ATTACH, the host system may issue another MSC mount
order to change the data staged on the virtual volume.
.
3. If CP attaches a device with advanced features before the device is on, the
advanced features will not work.
4. If the device you want to attach is:
•
A 370x, VM/SP cannot use it as a 370x unless the system programmer
loaded the 370x by VM/SP, DOS/VS, OS/VS, or ACF/SSP with the
270x EP program.
The ATTACH and DETACH commands reset the loaded status.
A 370x in PEP mode, first disable all emulator lines with the DISABLE
command. If you reload a 370x after an ATTACH, all EP lines will
become disabled.
3-12
VM/SP Operator's Guide
A 2305, specify both the real device address and the virtual device address
as the first exposure on the 2305 (that is, device address 0 or 8). When
CP attaches or detaches a dedicated 2305 to a user, CP processes all eight
exposures.
A device that has the ECMODE option, any address up to X'FFF' is valid
as vaddr; otherwise, any address up to X' 5FF' is valid.
System Response
CP sends messages to the following to tell them the ATTACH was successful.
The virtual machine user
The operator who issued the command (if different from the user)
The primary system operator (if different from the operator who issued the
command).
type raddr ATTACHED
raddr ... ATTACHED
raddr-raddr ATTACHED
Depending on the form that the user uses on the command line, the
response indicates that the real device or devices are attached to the
virtual machine. If you specify more than one address or a range of
addresses, CP does not display the device type.
type raddr ATTACHED TO userid vaddr
type raddr ATTACHED TO SYSTEM volid
raddr ... ATTACHED TO userid
raddr-raddr ATTACHED TO userid
Depending on the form that you use on the command line, the
response shows that the real device(s) are attached to CP or a specified user's virtual machine (and, if appropriate, the virtual address of
the device). If you specify more than one address or a range of
addresses, CP does not display the device type.
type raddr ATTACH TO SYSTEM volid BY userid
type raddr ATTACH TO userid vaddr BY userid
CP sends a response to the primary system operator, if he did not issue
the command, to show that CP attached the real device to either:
The specified user's machine at virtual address vaddr
The CP system by the specified class Buser.
CHANNEL c [PROC nn] ATTACHED TO userid
CHANNEL c [PROC nn] ATTACH TO userid
CHANNEL c [PROC nn] ATTACH TO userid BY operator
CP sends responses to the following to tell them the ATTACH
CHANNEL was successful:
The user whose virtual machine the channel is attached to
The operator who issued the command
• The primary system operator (if someone else issued the
command).
Section 3. CP Commands
3 -13
ATTACH
In the preceding responses, type is one of the following:
Type
Meaning
DASD
TAPE
LINE
Direct access storage device
Magnetic tape
Communication line
Card reader
Line printer
Card punch
Graphics device
Console
Channel-to-channel device
3704, 3705, or 3725 Communications Controller
Any other device
RDR
PRT
PUN
GRAF
CONS
CTC
CTLR
DEV
/'
3-14
VM/SP Operator's Guide
AUJJ70l0G
AUTOLOG
Privilege Class: A or B
When to Use AUTO lOG
Use the AUTOLOG command when you want to log on any virtual machine that is
defined in the VM/SP directory. The virtual machine that you log on with
AUTOLOG:
1.
Must have an automatic IPL defined in its directory
2.
Can issue one read to its virtual console
3.
Operates in disconnected mode.
AUTO lOG Format and Operands
AUTO LOg
userid password [variable data]
userid
identifies the virtual machine user that you want to log on.
password
is the password for the virtual machine that you want to log on.
If your installation has generated password suppression, CP assumes all data
following userid is variable data. CP will prompt you to enter the password
at the terminal. The only time CP lets you issue AUTOLOG with the password on the command line is when you are issuing AUTOLOG from an exec
or module running under CMS.
If LOGON journaling is:
OFF, and you issue AUTOLOG with a password from a virtual machine,
CP checks the password and keeps track of incorrect passwords. If you
enter 4 incorrect passwords, CP issues an error message and does not let
you issue the AUTOLOG command from the virtual machine. You can
still issue AUTOLOG directly from the terminal (that is, in CP console
mode).
ON, CP keeps track of incorrect passwords without regard to the origin
of the command. If you enter the installation defined limit of incorrect
Section 3. CP Commands
3 -15
AUTOLOG passwords (specified in the SYSJRL macro of DMKSYS),
CP does not let you issue AUTOLOG, regardless of how you entered it.
variable data
is the information that CP passes to the virtual machine. CP uses this data to
satisfy reads to your terminal. You may place up to 108 bytes of data in this
field. If you do not enter variable data, CP pads the field with blanks.
If you enter the AUTOLOG command to log on a 'virtual machine that has
an IPL for CMS in its directory, the absence of variable data satisfies the first
read to the virtual console. To prevent CP from forcing off the user because
another read is being issued to the virtual console, include 'SET
AUTOREAD OFF' in either:
•
The variable data field, or
The AUTOLOGed machine's PROFILE EXEC. (CP treats embedded
line-edit characters in this profile exec as data.)
Things You Should Know about the AUTOLOG Command
1. Do not use the AUTOLOG command to log on the Primary System Operator's
virtual machine. If you AUTOLOG the primary system operator's virtual
machine, it cannot receive any messages (including CP messages) that are
intended for the primary system operator.
2. If the VM/SP directory has a userid of AUTOLOG1, CP issues the
AUTOLOG command for the AUTOLOG1 virtual machine. You can set up
the PROFILE EXEC of the AUTOLOG 1 virtual machine to have more
AUTOLOG commands to log on other virtual machines (for example, the
batch virtual machine).
System Response
hh:mm:ss AUTO LOGON *** userid USERS=nnn BY
!
SYSTEM
userid2
I
***
indicates there is no assigned line address for the virtual console.
userid
identifies the virtual machine that AUTOLOG started.
USERS=nnn
is the number of logged-on users.
SYSTEM
indicates that the AUTOLOG process is part of VM/SP system initialization.
userid2
identifies the virtual machine that invoked the AUTOLOG command.
3-16
VM/SP Operator's Guide
r=----
BACKSPAC
Privilege Class: D
When to Use BACKSPAC
Use the BACKSP AC command to restart or reposition the current output on a real
punch or printer. If you issue BACKSPAC when an intervention required condition is pending on the printer or punch, CP terminates the interrupted channel
program and does the backspace.
BACKSPAC Format and Operands
Printer BACKSPAC Format
BAckspac
i
raddr [ File
~
ages [EOFI
J
Punch BACKSPAC Format
IBAckspac
I
raddr [FileI
raddr
is the address (cuu) of the real unit record device that you want CP to backspace. raddr must be a punch or printer address.
File
restarts the output spool printer or punch file from the beginning.
pages
specifies the number of pages that you want CP to backspace on the printer.
If you do not specify a field, the printer will backspace to the beginning of
the page that it was printing when it received the command.
CP counts pages in BACKSP AC operations by looking for skip to channell
specifications (page ejects). If a file does not have skip to channell specifications, CP restarts processing at the beginning of the file, regardless of the
number of pages indicated. Counting begins at the end of the current buffer.
Do not use this operand for a virtual 3800 spool file.
Section 3. CP Commands
3-17
EOF
starts printing from the end of the spool file. Backspacing will start the
requested number of pages from the end of the spool file. Printing restarts at
this location.
Do not use this operand for a virtual 3800 spool file.
Things You Should Know about the BACKSPAC
1.
For buffered printers, such as the 3800 printer, the BACKSPAC command
does not affect files that have already been sent to the printer.
2.
CP counts pages for backspacing, starting from the end of the current spooling
buffer. Output that has a small number of lines per page may require a count
of more than one to backspace one page.
An average print file requires approximately two spooling buffers for each
printed page. However, if you issued a skip-to-channel-l (carriage control
" 1 ") every five lines, the print file might have six printed pages in each
spooling buffer.
3.
Some spool files may have imbedded forms control buffers (FCBs). CP loads
the imbedded FCB, which controls line skipping and spacing, into the real
printer. If CP backspaces over an imbedded FCB within a spool file, printed
results may be incorrect. This is because the FCB that CP backspaced over is
still located in the real printer. If the file had been printed normally, that FCB
would not have been loaded at this point in the spool file.
To avoid backspacing over an imbedded FCB within a spool file, use
FILEFCB (START option) only for selected printers and for print files on the
printers that allow FCBs. Backspace the file to insure that CP handles the
FCBs properly.
System Response
PRT} raddr BACKSPACE USERID FILE CLASS RECORDS RECLEFT CPY DIST
{ PUN
userid file a typ norecs noleft nnn distcode
raddr
is the real device address.
userid
identifies the user whose file was active and CP backspaced.
file
is the unique, system-assigned spoolid number by which VM/SP refers to
the file.
3-18
VM/SP Operator's Guide
SEQ
seq
a
is the spool file class.
typ
is the originating device type, PRT, CON, or PUN.
norees
is the total number of logical records in the file.
noleft
is the number of records remaining for CP to process.
nnn
is the remaining number of copies of the file to be created, where 001 is the
last copy.
disteode
is the distribution code of the file.
seq
is the sequence of the file (printed on the separator page).
~
I
Section 3. CP Commands
3 -19
CHANGE
Privilege Class: D
When to Use CHANGE
Use the CHANGE command to alter one or more of the external attributes of a
closed spool file. (To find the current attributes of a file, issue QUERY Printer,
QUERY PUnch, or QUERY Reader).
Before changing an output file, it must be closed but not yet selected for printing or
punching. You can change an input (READER) file at any time before CP opens
it, that is, before CP issues the first read for the file.
CHANGE Format and Operands
CHange
CHars nameO
~(CHarS] namel]
1
(CHars] name2
(CHars] name3
CLass c2
COpy[*]nnn
DIst distcode
FCB name l
r~~~;EMl {::~:;}I~~~~ormll :~:;;:,~~nn
L~
J
~HOLD ~
ALL
PUnch
[ft]l~
rNAmeflfn
dsnameU
L
spoo/id
NOHOLD
MOdify name[n]1
fsys 1
LNOSYSJ
lYou must choose one of these options; however, you may specify more than
one in any order.
2The CHars, FCB, and MOdify options are valid for only the 3800 printer.
userid
identifies the user whose spool files are to be changed.
3-20
VM/SP Operator's Guide
(
I
SYSTEM
lets you change all spool files in the system regardless of userid.
*
lets you change your own spool files.
Reader
RDR
changes reader files.
Printer
PRT
changes printer files.
PUnch
PCH
changes punch files.
CLass cl
selects the files to change by class. cl is a one-character letter or number.
FORM form 1
selects the files to change by form number. If you use the ~ option, the form
number is the user form. If you specify the userid/SYSTEM option, the
form number is the operator form. forml is one to eight characters.
spoo/id
is the spoolid number of the file that you want to change. Each file has a
unique spoolid.
ALL
changes files regardless of class or form.
CHars nameO [ [CHars] namel [ [CHars] name2[ [CHars] name3] ] ]
names the character arrangement table that CP uses to print the file. The
parameters in brackets, [ ], are optional. name may be one to four characters
in length with a maximum of four names per command.
Note,' You only need to repeat CHars when one of the names that you
specify is the same as one of the option names (such as CLASS, DIST,
FLASH, etc.) If you do not specify a character arrangement table for a file,
CP uses the default table used for the separator page on the selected printer
to print the file. 1
You can specify this option for any file, but the change will only take affect
if you send the file to a real 3800 printer for printing.
A specification of NULL will nullify any previous setting.
Section 3. CP Commands
3 -21
CLass c2
indicates the new class of the file.
COpy [*] nnn
alters the number of copies for the file. The value of nnn (number of copies)
must be a number from 1 through 255.
When you include * with the COPY parameter, CP transmits the file to the
3800 printer only once. The 3800 performs the duplication internally, one
page at a time. (For example, if you specify COPY *3, CP copies the page
numbers as 111 222 333 ... ). Otherwise, CP transmits the file once for each
copy that you want, with VM/SP indicating the copy number to the 3800.
DIst distcode
changes the distribution code of the file.
distcode is a one to eight character alphameric identification. It appears on
the output separator of printer or punch output. CP changes the distribution
code for this file only; CP does not affect other files or change the normal
user code.
FeB name
names the forms control buffer that CP uses to print the file.
/-
name may be one to four characters, or the unique numbers 6, 8, or 12. For
the 3800 Model 3 device, this unique number may also be 10. These
numbers tell VM/SP to print the entire spool file at 6, 8, 10 or 12 lines per
inch, regardless of the size of the paper that is currently loaded. If you use
the unique numbers, you will not cause an FCB-forms mismatch. If you do
not specify this operand, CP uses the value that you specified for the separator page in the START command. 2
You can specify this option for any file, but the change will only affect the
file if you send it to a real 3800 printer for printing.
FLash name nnn
specifies the flashing status of the file.
name identifies a forms overlay sheet in the 3800 printer and may be one to
four characters. The number nnn specifies the number of copies you want
CP to flash. This flash count may range from 00 to 255. The flash count
defaults to zero and indicates that you do not want flashing.
You can specify this option for any file, but the change will only affect the
file if you send it to a real 3800 printer for printing.
Note: The FLASH name must match the FLASH name that you specify in
the START command. The exception is if the flash count is zero.
2
3-22
VM/SP Operator's Guide
A specification of NULL will nullify any previous setting.
c
FORMlorm2
changes the spool form number of the file to lorm2. If you use the * option,
CP changes your form number. If you specify the userid/SYSTEM option,
CP changes the operator form number.
HOLD
prevents CP from printing, punching, or reading a file. Instead, CP puts the
file in a user HOLD status, until the NOHOLD operand releases it.
NOHOLD
releases the specified spool file that the user HOLD status is currently
holding.
MOdify name[n]
is the name of the copy modification module that CP uses to print the file.
name may be one to four characters in length. If you do not specify name,
CP will not use copy modification to print the file. 1 n selects one of the character arrangement tables specified with the CHars keyword to use for the
copy modification text. The values 0, 1, 2, or 3 correspond to the order that
you have specified the table names. If you do not specify n, CP assumes the
first character arrangement table.
You can specify this option for any file, but the change will only affect the
file if you send it to a real 3800 printer for printing.
'-
SYS
prevents CP from printing or punching the file. Instead, CP puts the file in
system hold status until NOSYS releases it.
NOSYS
releases the specified spool file that the system hold status is currently
holding.
Note: A system hold has no effect on the processing of reader files.
NAme
lin I
[ft]
dsname
changes the name of the file. For eMS files, the filename, In, that you
specify (and, optionally, filetype, It) becomes the new file identification. For
a non-CMS file, the dsname that you specify (a 1 to 24-character field)
becomes the new file identification.
Note: If the you select the NAme function and only specify In, CP sets It to
blanks. The dsname specification lets a user specify a file name suitable for
OS, such as SYSl.SYSLIB.MYMAC .
.~.
Section 3. CP Commands
3 -23 .
System Response
In~~nl
FILES CHANGED
is the response to the user who issued the CHANGE command.
If you issue SET IMSG OFF, you will not see the message. If you are the
system operator monitoring the RSCS virtual machine for your installation,
do not specify this option. You need to see information messages to accurately monitor RSCS system activity.
Note: Depending on the options you choose and the combinations you use, CP
may not have actually changed some files because they had the new class before
you issued the CHANGE command. The response indicates how many files you
referenced.
/
3 -24
VM/SP Operator's Guide
CPTRAP
Privilege Class: C
When to Use CPTRAP
Use the CPTRAP command when you want to collect information in a reader file
to use for problem determination. The information can include CP trace table
entries, CP data, and virtual machine data.
CPTRAP Format and Operands
CPTrap
typenum
~vmblOk
DEVaddr
COde
OFF
]
cuu
code-value
[{g~}]
ALL
'-. ..
address
/
ALLOWid
userid
GRoupid
group-name
STArt
[TO {~rid}
WRAP wrap-size]
CLOSE
STOP
Note: Enter the CPTRAP subcommands on separate command lines. You can,
however, put more than one typenum subcommand on the same command line.
typenum
[vrnblOCk address]
DEVaddr cuu
COde code-value
OFF
Use typenum to identify the type of entry that you want CP to collect in the
CPTRAP reader file. You can specify any hexadecimal number from X'O 1'
up to X'3F':
X'Ol'-X'lB' - Corresponds to the CP trace table entry
X'3D' - Group virtual machine entry
X'3E' - General virtual machine entry
Section 3. CP Commands
3-25
X'3P' - CP entry.
See VM / SP System Programmer's Guide for an explanation of these entries
and how to create them.
You can specify typenum before or after CPTRAP START. If you do not
use the typenum subcommand or ALL ON (to specify all typenums) before
you start CPTRAP, CPTRAP does not collect any entries in the reader file.
If ALL ON is in effect when you enter a typenum subcommand, CP sets all
other typenums off. You cannot specify typenum on the same command line
with other subcommands.
Example:
CPTRAP 56
will collect trace table types X'05' and X'06' in the CPTRAP
file along with any previous definitions.
There are three options for typenum that give you additional selectivity.
Figure 3-2 on page 3-28 shows which options you can use for each CP
trace table entry. All the options are not valid for each typenum.
Vmblok address
lets you select CP trace table entries based on a VMBLOK address.
This option is only valid for trace table types that have a VMBLOK
field. The address of the VMBLOK can be up to six digits.
Examples:
CPTRAP 6 collects all CP trace table entries for FREE STORAGE
(X'06').
CPTRAP 6 VMBLOK 41D80
collects only CP trace table entries for FREE STORAGE
(X'06') that the system did for a user with VMBLOK
address 41D80.
DEVaddr cuu
lets you select CP trace table entries based on device address. This
option is valid for CP trace table entries that have a real or virtual
device address field. cuu is the device address and can be up to 4
digits.
Examples:
CPTRAP OB
collects all CP trace table entries for START I/O
(X'OB').
CPTRAP OB DEV 250
collects only CP trace table entries for START I/O
(X'OB') that the system did for device 250.
3-26
VM/SP Operator's Guide
,/
CODE code-value
lets you select CP trace table entries based on a specific code field in
the trace entry. This option is valid for CP trace table entries that
have one of the following types of code fields:
•
•
•
•
Interrupt code
Condition code
Order code
Function code
Transaction type.
The code- value can have up to four digits, except CP trace type 4,
which can have up to 8 digits.
Examples:
CPTRAP 2 collects all CP trace table entries for SVC interrupts
(X'02').
CPTRAP 2 CODE 8
collects only CP trace table entries for SVC 8.
CPTRAP 2 CODE 8 2 CODE C
collects only CP trace table entries for SVC 8 and SVC
12.
OFF
deletes selectivity for a specified typenum.
Examples:
CPTRAP 2 OFF
does not collect X'02' CP trace table entries.
CPTRAP 2 OFF 3E OFF
does not collect X'02' CP trace table entries or X'3E'
general virtual machine entries
To use more than one typenum option, assign typenum with each option you
choose.
Example:
CPTRAP 5 DEV 190 18 DEV 190 18 DEV 192 8 VMBLOK 2D4900
collects
Type
Type
Type
Type
5 with device 190
18 with device 190
18 with device 192
8 with VMBLOK address 2D4900.
~/
Section 3. CP Commands
3 -2 7
TYPENUM OPTIONS
TRACE TADLE
EVENT TYPE
(typenum)
EVENT CODE
Vmblok
DEVaddr
External Interrupt
X'OI'
Xl
SVC Interrupt
X'02'
Xl
Program Interrupt
X'03'
Xl
Machine Check Interrupt
X'04'
I/O Interrupt
X'OS'
Free Storage (FREE)
X'06'
X
Return Storage (FRET)
X' 07'
X
Enter Scheduler
X'08'
X
Queue Drop
X'09'
X
Run User
X'OA'
Xl
start I/O
X'OB'
Unstack I/O Interrupt
X'OC'
X
X
Virtual CSW Store
X' OD'
X
X
Test I/O
X' 0 E'
X
X4
Halt Device
X' 0 F'
X
X4
Unstack IOBlOK/TRQBlOK
X'IO'
NCP BTU
X'II'
Spinning on lock
X'12'
SIGP Issued
X' 13'
X2
X5
Clear Channel Issued
X'I4'
IUCV Communication
X'IS'
X6
SNA CCS
X'16'
X7
Diagnose X'80'
X'I7'
X4
Start I/O Fast Release
X'18'
X
Simulated I/O Interrupt
X'19'
X
Clear I/O
X'IB'
X
Xl
X
X
X
refers
refers
l refers
4 refers
S refers
6 refers
7 refers
I
2
Figure
VM/SP Operator's Guide
3-2.
to
to
to
to
to
to
to
RUNUSER Value.
Real Processor Address.
Interrupt Code.
Condition Code.
Order Code.
Function Code.
Transaction Type.
CP Trace Event Type Codes
X4
X
X
X
X
The following superscripts describe the fields in the CP
trace table entry:
3-28
COde
X4
X4
----.-----.---J
ALL [ON ]
OFF
turns on the selection of all typenums. This is the default.
OFF
turns off the selection of all typenums.
ALLOWid userid
enables an individual virtual machine to send data to add to the CPTRAP
file. The virtual machine must currently be logged on or disconnected. The
virtual machine can enter data into the CPTRAP file until:
•
•
The virtual machine logs off, or
You use CPTRAP STOP to stop CPTRAP.
GRoupid group-name
enables all the virtual machines in the machine group that you specify, which
are logged on or disconnected, to send virtual machine data to the CPTRAP
file. Also, any virtual machine that enters the group after you have issued
CPTRAP GROUPID is also enabled to enter data into the CPTRAP file.
The virtual machine can enter data into the CPTRAP file until:
•
The virtual machine logs off, or
You use CPTRAP STOP to stop CPTRAP.
If all virtual machines in a group leave the group, issue a new GROUPID
subcommand when the group re-establishes itself. This will cause CP to
automatically enable any virtual machine when it enters the group.
STArt
~TO]
{use;id} [WRAP wrap-SiZe]]
starts CPTRAP and defines how many CPTRAP records you want the
SPOOL file to maintain. The system will not collect data in the CPTRAP
file until it receives the START subcommand.
[TO]
userid
*
identifies the owner of the CPTRAP file. If you use the TO operand,
follow it with a userid or *. If you omit the TO operand you cannot
have a userid "to" or "wrap". If you do not specify userid at start
time, the command will default to *. This means that the virtual
machine that issued CPTRAP START will own the CPTRAP file.
WRAP wrap-size
starts a wrap type SPOOL file. A wrap file reuses SPOOL space.
wrap-size, a decimal number from 16 to 20000, is the number of
CPTRAP records (4K blocks of data) that you want the SPOOL file
to maintain. After the SPOOL file collects the number of records you
specify with wrap-size, the newer CPTRAP records overlay the old
records.
Section 3. CP Commands
3-29
You cannot change the size of the wrap file during the CPTRAP
session. You can issue CPTRAP CLOSE to close the current wrap
file and open a new wrap file. However, the original wrap-size that
you specified on the CPTRAP START command will be in effect for
the new file also. If you want to change the size of the wrap file, end
the current CPTRAP session with CPTRAP STOP and start a new
session.
If you do not specify the WRAP option, the system will start a non-
wrap SPOOL file. A non-wrap SPOOL file does not reuse SPOOL
space; so, the space for this file is limited by the SPOOL space available on the system. The newer CPTRAP records do not overlay the
old records in this file; when the file reaches 3480 4K blocks of data,
the system closes the file and opens a new file.
CLOSE
closes the current file and creates a reader file that includes the data collected up to this point. CPTRAP opens a new file so you can continue to
collect data. If you issue CLOSE and there is no data in the file, CPTRAP
will not create a file.
STOP
stops CPTRAP and creates a reader file if there is data. When you issue
CPTRAP STOP, the system stops collecting data, discards all selectivity definitions, and completely stops the CPTRAP operation. If you issue STOP
and there is no data in the file, CPTRAP will not create a file. STOP is valid
when the CPTRAP facility is active or inactive.
Things You Should Know about the CPTRAP Command
3-30
1.
Only one user can invoke CPTRAP at anyone time. If you try to invoke
CPTRAP when another user has invoked it, you will get a message saying
CPTRAP is already active.
2.
You can display or print the data that CPTRAP collects with the TRAP RED
command. Refer to VM / SP System Programmer's Guide to learn how to use
this command.
VM/SP Operator's Guide
c
DCP
Privilege Classes: C or E
When to Use DCP
Use the DCP command to display the contents of real storage locations at your terminal. In a system running with an attached processor, you can also disclose the
values of PSA for the main and the attached processor.
DCP Format and Operands
~
DCP
-MLhexlocl - NLhexlocl
MThexlocl
NThexlocl
Mhexlocl
Nhexlocl
Lhexlocl
Thexlocl
hexlocl
~
"-----/
- ""
{~l ~;:;c~
f.} ~ bytecounj
>
END
L
T
0
"
-
--
- ,.,
MLhexlocl
NLhexlocl
MThexlocl
NThexlocl
Mhexlocl
Nhexlocl
Lhexlocl
Thexlocl
hexlocl
L
T
o
specifies the first storage location that CP will display. If you specify
hexlocl, CP displays, in hexadecimal, only this storage location. When you
. specify Lor T (the designated type character), the display will start with
storage location O.
Section 3. CP Commands
3 -31
•
T includes an EBCDIC translation with the hexadecimal display.
L displays in hexadecimal only.
If hexloel is followed by a period and is not on a fullword boundary, CP
rounds it down to the next lower fullword. Note that hexloel and Lhexloel
get the same results.
In attached processor (AP) systems, M causes CP to interpret addresses like
the main processor generated them. Thus, MO refers to location 0 of the
PSA of the main processor. In multiprocessor (MP) systems, M corresponds
to addresses on the IPL processor.
In attached processor (AP) systems, N causes CP to interpret addresses like
the attached processor generated them. In multiprocessor (MP) systems, N
causes CP to interpret addresses as if they were on the non-IPL processor.
N is valid only if the attached processor is operational.
Notes:
1. When you do not specify M or N, CP interprets the address as an absolute address on both AP and MP systems.
2.
The M and N prefixes are not meaningful for uniprocessor systems.
3. In AP applications, the PSA values for the main and attached processors
are not located at absolute page zero but are displaced from it. Therefore, if you prefix the hexadecimal location with an M or N, the system
will add the hexadecimal value to the displacement.
Example:
dcp m40.8
Displays the CSW of the main processor.
dcp n78-7f
Displays the I/O new PSW of the attached processor.
{~ }[~~~C2J
displays a range of locations. Use the colon (:) or dash (-) to display the
contents of one or more storage locations by addresses that you specify.
The hexloe2 operand must be 1 to 6 hexadecimal digits and greater than or
equal to hexloel. hexloe2 should not exceed the size of real storage. If you
specify END, and a colon (:) or dash (-) follows the first operand, CP will
display real storage from hexloel through the end of real storage. In other
words, the following commands are identical:
•
•
•
3-32
VM/SP Operator's Guide
DCP O:END
DCP 0:
DCP O-END
/
DCPO-
{. }
byteeount]
[ END
byteeount is a hexadecimal number that represents the number of bytes of
real storage (starting with the byte at hexloel) that you want CP to display.
The byteeount operand must be 1 or greater and may not exceed six
hexadecimal digits.
The sum of hexloel and the byteeount must be an address that does not
exceed the size of real storage. If this address is not on a fullword boundary,
CP rounds it up to the next higher fullword.
Things You Should Know about the DCP Command
1.
Normally, you'll define the beginning and ending locations of storage as
follows:
dcp
dcp
dcp
dcp
dcp
Lhexlocl-hexloc2
Thexlocl-hexloc2
hexlocl:hexloc2
hexlocl.bytecount
hexlocl:hexloc2 hexlocl.bytecount
Do not enter blanks between the limit or range symbols (: or - or .), or
between any of the operands. The exception to this is the blank(s) between
the command name and the first operand. Also, you'll need to enter a blank
between each set of operands when you enter more than one set of operands
on one command line.
If you put a blank immediately after the designated type character (T or L),
DCP displays all of real storage. If the next operand is a colon (:), a dash (-),
or a period (.) followed by a blank, CP displays all storage locations. This is
because CP assumes a second set of operands. Do not put a blank on the right
or left of range or length symbols, unless you want the missing operand to
default.
2.
All of the following produce full storage displays:
dcp 1
dcp t
dcp dcp:
dcp.
dcp
dcp
dcp
dcp
dcp
11:
t:
1.
t.
dcp
dcp
dcp
dcp
dcp
00:
I-end
t-end
O-end
dcp
dcp
dcp
dcp
dcp
t:end
t:end
O:end
I.end
O.end
Each of the following produce three full dumps because of the imbedded
blanks:
dcp I . t
If you want CP to apply default values (that is, the beginning and ending of
storage), specify one operand.
Section 3. CP Commands
3-33
System Response
If you enter an invalid operand, the DCP command stops. However, CP processes
any previous valid operands before it terminates the command.
xxx xxx
= wordl
word2 word3 word4 [key]
*EBCDIC translation*
CP displays locations that you request.
xxxxxx
is the real storage location of word 1.
wordl
is displayed (word-aligned) for a single hexadecimal specification.
CP displays up to four words on a line. If required, CP will display
more than one line.
EBCDIC translation
is displayed aligned to the next lower 16-byte boundary if you
specify Thexloc. Note that CP displays nonprintable characters as a
period (.).
If the location is at a 2K page boundary, CP also displays the key for that page.
Press the ATTN key (or its equivalent) to stop the output and terminate the
command.
3-34
VM/SP Operator's Guide
DEFINE
Privilege Class: A or B
When to Use DEFINE
Use the DEFINE command to specify the status of a 3330V volume.
For example, a 3330V may be defined as VIRTUAL so you can dedicate it to a
specific virtual machine. If you want the 3330V volume to be available to the
control program for mounting system volumes, you can define it as SYSVIRT.
DEFI N E Format and Operands
DEFine
Sysvirt }
{ Virtual
raddr 1 [-raddr2]
Sysvirt
reserves the 3330V for CPo You can mount VM/SP system volumes that
are MSS 3330V volumes on SYSVIRT 3330V devices. You cannot dedicate
these volumes to a virtual machine by address, and you cannot attach them
to anything other than the system.
Virtual
allows the 3330V device to be dedicated or attached to a virtual machine.
raddrl [-raddr2]
specifies the device address or range of addr,esses you want CP to define. If
you specify a range, CP sets all 3330V devices in the range to the status that
you specify.
Things You Should Know about the DEFINE Command
CP demounts an existing MSS volume before completing the DEFINE command
when:
An MSS volume is already mounted.
The MSS volume is not being used on the 3330V that is being defined.
~I
Section 3. CP Commands
3-35
c==________ .
System Response
CP sends responses to you when you issue the DEFINE command. The responses
show the device status after CP executes the command. Possible responses are:
raddr NOT KNOWN
The r add r is not a valid device address.
raddr NOT 3330V
The raddr is not a 3330V. CP does nothing for the raddr.
raddr ALREADY SYSVIRT
raddr ALREADY VIRTUAL
The raddr already has the status that you tried to define.
DEVICE raddr IS NOT AVAILABLE
The raddr is currently dedicated to a virtual machine. CP does not change
the status from VIRTUAL to SYSVIRT.
raddr IS OFFLINE
The raddr is currently offline. When you vary it online, the raddr will
receive the status that you specified in the command.
raddr IN USE BY SYSTEM
The raddr is currently attached to the system and there is at least one user
minidisk allocated on the volume. Do not define the raddr as VIRTUAL.
raddr HAS A MOUNTED VOL.
MSS 10 NOT AVAILABLE TO DEMOUNT IT
There is a 3330V volume mounted on the raddr. The MSS communicator is
not active and the CP cannot demount the volume. CP does not change the
status of the raddr.
raddr - ERROR DEMOUNTING VOLUME
There is a 3330V volume mounted on the raddr. CP tried to demount the
volume, but MSS encountered an error. CP does not demount the volume
and does not change the status of the r add r .
INVALID RANGE - raddr
You specified a range of address, but either raddrl or raddr2 does not exist
or is not a 3330V. CP does not change the status.
raddr DEFINED AS VIRTUAL
raddr DEFINED AS SYSVIRT
CP has given the status that you specified to the raddr. If a volume was
previously mounted on the raddr, CP has now demounted it.
3-36
VM/SP Operator's Guide
[------.------------..-=====-=~~::.:.:..:...:.:~~~:=_=_=~=__-_=~~:_-~-~~-=. ==.:..~-~.-~~
[jJ) [:&ru"~~ C ~-J
. _
__
:~.:~~ ~_~.-=:~::.~::=_~:::~_==~ ::::..::..-~_:_~~=:_.:..:==_~-.:::-=:=~.=._-~:.-_==~=
_.=-:.::::-=:=_=:J
DETACH
Privilege Class: B
When to Use DETACH
Use the DETACH command to remove a real device from the CP system. You can
detach a previously attached device from a user, even if the device is still in use.
When you detach a tape device, CP automatically rewinds and unloads it.
Use DETACH CHAnnel to detach a dedicated channel from a user when the
virtual machine operations do not need the channel path any longer.
DETACH Format and Operands
DETach
{
raddr
}
raddr...
raddr-raddr
(FROM] {USerid
}
SYSTEM
*
CHAnnel c [PROC nn) [FROM)
{
{ us:rid
I
raddr
}
raddr...
raddr-raddr
specifies the real address (cuu), multiple addresses, or range of addresses of
the device or devices that you want to detach.
You can specify up to 48 multiple addresses on the DETACH command line.
If you try to detach more than 48 addresses, the system will:
.•
•
Only process the first 48 addresses
Ignore the remaining addresses and will not issue an error message for
these device addresses.
You may specify more than 48 addresses, however, by specifying the
addresses in a range.
FROM {userid
}
SYSTEM
*
identifies the virtual machine user to which the the real device [raddr] is currently attached.
Section 3. CP Commands
3-37
[D)lElrACH
If you omit the optional keyword FROM, do not specify userid as from, fro,
fr or f. When you specify a userid that could also be a valid hexadecimal
device address (for example, virtual machine id CE), use the FROM operand
to distinguish it as userid.
SYSTEM shows that the device is a DASD device, which CP is using for
allocation and control. If you want CP to detach a device from your own
virtual machine (class B operator), use an asterisk [*].
CHANnel c
is the real address of the channel that you want to detach. c may be any
hexadecimal character from 1 through 9 or A through F. This operand
works only with real channels whose devices have no explicitly defined alternate paths.
If you issue DETACH CHANnel before normal I/O activities on the dedicated channel path are complete, CP terminates those activities.
Note: If you issue an ATTACH CHANNEL command right after a
DETACH CHANNEL command, you may get a message that the specified
channel is in use. DETACH CHANNEL processing schedules I/O at different times for each dedicated device; but, DETACH CHANNEL processing does not wait for all I/O to finish before it returns control to the
terminal operator. Because of the timing, the DETACH CHANNEL
command may fail.
PROC nn
is the address of the processor that owns the channel. CP does not
require this operand in uniprocessor or attached processor mode. If
you do specify PROC, CP will check its validity.
FROM
{US:,idl
identifies the virtual machine user from whom CP will detach the
channel. If you omit the optional keyword FROM, do not specify
userid as from, fro, fr, or f. If you want CP to detach a channel from
your own virtual machine (class B operator), use the asterisk (*).
Things You Should Know about the DETACH Command
1.
If the device you want to detach is:
A 370x, VM/SP cannot use it as a 370x unless the system programmer
loaded the 370x by VM/SP, DOS/VS, OS/VS, or ACF/SSP with the
270x EP program.
The DETACH and ATTACH commands reset the loaded status.
A 2305, specify both the real device address and the virtual device address
as the first exposure on the 2305 (that is, device address 0 or 8). When
3-38
VM/SP Operator's Guide
/
c=:------.-.-----..-.--------..-----.---.--------------
CP attaches or detaches a dedicated 2305 to a user, CP processes all eight
exposures.
•
An active shared device from the system, you must inform affected users
to detach the device from their virtual systems. You can use the QUERY
SYSTEM raddr command to determine when all users have detached the
device from their virtual machines.
•
A device that has minidisks in use or a device with volumes in the
SYSOWN list, CP cannot detach them.
2. If your installation is using DOS or OS virtual machine systems, vary the device
offline before invoking the DETACH command. Similarly, if you are a CMS
user, issue the RELEASE command before invoking DETACH.
3. If you use DETACH while the virtual machine is tracing I/O, it will cause
abend DSP004.
System Response
CP sends messages to the following to tell them the DETACH was successful.
•
The virtual machine user
The operator who issued the command (if different from the user)
The primary system operator (if different from the operator who issued the
command).
vaddr. . .
} DETACH BY operator
vaddr-vaddr
{
type vaddr
is the response sent to the user if you detach one or more of the user's
devices.
{
raddr. . .
} DETACHED
raddr-raddr
type raddr
[user i d]
SYSTEM
is the response if the user detaches a previously attached device(s) or if you
detach the device(s) from a user or the system.
raddr. . .
} DETACHED
raddr-raddr
{
type raddr
[user i d] BY operator
SYSTEM
is the response issued to the primary system operator if a class B operator
other than the primary system operator issued the DETACH command, and
the device or devices had been previously attached.
Section 3. CP Commands
3-39
DETACH
Note: In the preceding responses, the term "type" is one of the following:
Type
Meaning
DASD
TAPE
LINE
RDR
PRT
PUN
GRAF
CONS
CTC
CTLR
DEV
MSC
Direct access storage device/3330V (MSS 3330 virtual volume)
Magnetic tape
Communication line
Card reader
Line printer
Card punch
Graphics device
Console
Channel-to-channel device
3704, 3705 or 3725 communications controller
Any other device
3151 port address
CTC vaddr DROP FROM userid vaddr
is the response if the device detached was a virtual CTC connected (by the
COUPLE command) to another CTC on the virtual machine with the userid
you specified. This response is always followed by the response:
CTC vaddr DETACHED
CHANNEL c [PROC nn] DETACHED BY operator
is sent to the user from whose virtual machine the channel is being detached.
CHANNEL c [PROC nn] DETACHED userid BY operator
is sent to the primary system operator if he did not issue the command.
3 -40
VM/SP Operator's Guide
DISABLE
Privilege Class: A or B
When to Use DISABLE
Use the DISABLE command to prevent the following from accessing the VM/SP
system:
Low-speed communication lines, including 2701, 2702, and 2703 lines (collectively referred to as 270x lines)
•
EP (Emulator Program) controlled lines in the 370x
Locally attached 3270 lines.
If the line that you select is not active (not dialed into or logged on), CP disables
the line immediately. If the line that you select is active, CP disables it when the
user logs off or when the user issues a DISCONN command without specifying the
HOLD operand.
,-''
DISABLE Format and Operands
DISAble
raddr...
}
SNA [userid]
{ ALL
raddr...
are the addresses (cuu) of the communication lines that you want to disable.
The address that you give may represent up to 2700 line addresses. CP
. checks these three-digit hexadecimal addresses against the line addresses
stored in the system table. If CP does not find a match, CP sends an error
message to the system console.
SNA [userid]
disables the SNA logical units for the VTAM Service Machine (VSM) that
you specify. If you do not specify userid, CP disables all SNA units for all
VSMs. See the VM/VCNA publications and ACF /VT AM publications
listed in the Preface for information on using SNA terminals.
Section 3. CP Commands
3-41
ALL
disables the following:
•
•
•
•
270x communication lines
EP (Emulator Program) controlled lines in the 370x
All SNA (Systems Network Architecture) terminals
Locally attached 3270 lines, including the operator's line.
Note: To examine the status of the communication lines, issue a QUERY LINES
and NETWORK QUERY command.
Things You Should Know about the DISABLE Command
1. Use the NETWORK DISABLE command instead of DISABLE when:
You want to disable a remote 3270 configuration
•
You want to disable remote 370x resources.
2. If you want to negate the DISABLE command, issue an ENABLE command to
the line before you physically disable the line.
System Response
COMMAND COMPLETE
3 -42
VM/SP Operator's Guide
DMCP
Privilege Classes: C or E
When to Use DMCP
Use the DMCP command to print the contents of real storage locations on the
user's virtual spooled printer. The output format is eight words per line with
EBCDIC translation.
In a system running with an attached processor, you can also dump values that are
displacements from the assigned PSA (Prefix Storage Area) locations of the main
or attached processor. If you want the output to print on the real printer, use the
CLOSE command to stop the virtual spooled printer.
DMCP Format and Operands
DCP
,--------/
r-
-
MLhexlocl
NLhexlocl
MThexlocl
NThexlocl
Mhexlocl
Nhexlocl
Lhexlocl
Thexlocl
hexlocl
0
-
-
r-
-.
l~} [:;;C2J
[*dumpid]
1·\
[bYlecounJ
END
-
Note: If you want CP to apply default operand values (the beginning and ending
of storage), specify at least one operand.
Section 3. CP Commands
3-43
L -_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _._ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
~
MLhexlocl
NLhexlocl
MThexlocl
NThexlocl
Mhexlocl
Nhexlocl
Lhexlocl
Thexlocl
hexlocl
o
specifies the first storage location that CP will dump. If you specify only
hexlocl, CP dumps the specified storage location. If you follow hexlocl with
a period and it is not on a full word boundary, CP rounds it down When you
specify Lor T, the dump will start with storage location 0 and end at the
location that you specify.
•
T includes an EBCDIC translation with the hexadecimal display.
•
L displays in hexadecimal only.
In attached processor (AP) systems, M causes CP to interpret addresses like
the main processor generated them. Thus, MO refers to location 0 of the
PSA of the main processor. In multiprocessor (MP) systems, M corresponds
to addresses on the IPL processor. M is valid if the system has been generated for an attached processor.
In attached processor (AP) systems, N causes CP to interpret addresses like
the attached processor generated them. In multiprocessor (MP) systems, N
causes CP to interpret addresses as if they were on the non-IPL processor.
N is valid only if the attached processor is operational.
Notes:
1. When you do not specify M or N, CP interprets the address as an absolute address on both AP and MP systems.
2.
The M and N prefixes are not meaningful for uniprocessor systems.
3. In AP applications, the PSA values for the main and attached processors
are not located at absolute page zero but are displaced from it. Therefore, if you prefix the hexadecimal location with an M or N, the system
will add the hexadecimal value to the displacement.
Example:
dmcp m40.8
Dumps the CSW of the main processor
dmcp n78-7f
Dumps the I/O new PSW of the attached processor.
3 -44
VM/SP Operator's Guide
,----------_.-------_._._---_._---_..._-._--._--------_.---------
l~} [~~~C2J
dumps a range of real storage locations. To dump to the end of real storage
either:
Specify hexloe2 as END, or
•
Do not specify hexloe2 at all, in which case CP assumes END by default.
When you specify more than one storage location on one command line and
the line has errors, CP processes all correct operands up to the error. CP
rejects the error and the rest of the command line.
{ .\
byteeountJ
[ END
byteeount is a hexadecimal number that represents the number of bytes of
real storage (starting with the byte at hexloel) that you want the printer to
print. The sum of hexloe 1 and the byteeount must be an address that does not
exceed the size of real storage. If this address is not on a fullword boundary,
CP rounds it up to the next higher fullword.
*dumpid
helps you identify dumps. CP prints the dumpid right before the dump data.
You can specify up to 100 characters with or without blanks after the
asterisk prefix. You must specify hexloe2 or byteeount when you specify this
operand. You need the asterisk (*) to identify the dumpid.
Things You Should Know about the DMCP Command
1.
Normally, you'll define the beginning and ending dump locations as follows:
dmcp Lhexloc-hexloc
or
dmcp hexloc. bytecount
Do not enter blanks between the limit or range symbols (: or - or .), or
between any of the operands. The exception to this is the blank(s) between
the command name and the first operand.
If you put a blank immediately after the designated type character (T or L), CP
defaults to the beginning and end of virtual storage. Similarly, if you follow the
range or length symbol with a blank or END, CP dumps all real storage.
2.
If you enter more than one operand on the same command line, separate operands or sets of operands with blanks. Do not put a blank on the right or left of
range or length symbols, unless you want the missing operand to default.
Section 3. CP Commands
3-45
All of the following produce full storage dumps:
dmcp 1
dmcp t
dmcp dmcp:
dmcp.
dmep
dmep
dmep
dmep
dmep
1t1:
t:
1.
dmep
dmep
dmep
dmep
dmep
t.
00:
O.
I-end
dmep
dmep
dmep
dmep
t-end
O:end
l.end
O.end
Each of the following produce three full dumps because of the imbedded
blanks:
dmep 1. t
dmep -: .
System Response
DUMPING LOC hexloc
CP sends this as the dump proceeds. This message indicates that the
dump is continuing from the next 64K boundary. If the user signals attention on the terminal while CP is displaying this message, the dump ends.
hex loc is the segment (64K) address for the dump continuation, such as
020000,030000,040000.
COMMAND COMPLETE
indicates normal completion of the dump.
,/
3-46
VM/SP Operator's Guide
DRAIN
Privilege Class: D
When to Use DRAIN
Use the DRAIN command to stop spooling operations on a real unit record device.
With DRAIN, you can:
•
Bring the spooling system or a specified device to a controlled halt.
Halt the activities on a device whose spooling status you want to change.
For example, you usually drain all unit record devices before system shutdown; you
must drain a printer before you change the contents of the UCS printer buffer.
CP immediately drains an inactive device when you issue DRAIN. If the device is
active, CP drains the device when it finishes processing the current file.
DRAIN Format and Operands
DRain
-
-
Reader
Printer
PUnch
raddr•.•
ALL
-
-
Reader
RDR
drains all of the system readers.
Printer
PRT
drains all of the system printers.
PUnch
PC"
drains all of the system punches.
raddr
drains the real spooled devices at the addresses (cuu) that you specify. You
can specify more than one address.
~)
Section 3. CP Commands
3 -47
c====__________________________________________________________________________
~
drains every spooled unit record device in the system.
Things You Should Know about the DRAIN Command
1. For buffered printers, such as the 3800 printer, the BACKSPAC command
does not affect files that have already been sent to the printer.
2. Restart a drained device with the START command. If you issue START to a
device before draining is complete, the device does not enter the drained status
but continues processing.
System Response
The following is an example of a response when the DRAIN is successful. For
more information about the variables, see the QUERY Unit Record description.
PRT 001 DRAINED
001 MANUAL SEP
3-48
VM/SP Operator's Guide
CLASS A
N03800
FORM STANDARD
FILEFCB
ENABLE
ENABLE
Privilege Class: A or B
When to Use ENABLE
Use the ENABLE command to enable the following previously disabled or
nonenabled devices so users may access the VM/SP system:
3270 (locally attached only) and the 3138, 3148, and 3158 consoles. (Use the
NETWORK ENABLE command for 3270 remote devices.)
2701, 2702, and 2703 (270x)
EP (Emulator Program) controlled lines in the 370x
Systems Network Architecture consoles.
ENABLE does not affect previously enabled lines.
ENABLE Format and Operands
ENable
raddr...
}
SNA luserid]
{
ALL
raddr...
are the addresses [cuu] of the lines that you want CP to enable. If you
specify more than one address, separate each address by one or more blanks.
SNA luserid]
enables the SNA logical units for the VT AM Service Machine [VSAMI that
you specify with userid. If you do not specify userid, CP enables all SNA
units for all VSMs. See the VM/VCNA publications listed in the Preface for
information on using SNA terminals.
ALL
enables the following previously disabled or non-enabled devices:
•
Locally attached 3270 display terminals and display copy printers
(3284/3286 etc.)
All SNA (Systems Network Architecture) terminals
Section 3. CP Commands
3-49
ENABLE
•
EP (Emulator Program) controlled lines in the 370x
•
270x communication lines.
Note: Do not try to enable terminal devices that CP does not support as VM/SP
virtual machine system consoles. You can only use these types of devices as dedicated devices attached to multiple-access system virtual machines.
System Response
COMMAND COMPLETE
Note: CP does not indicate an error if there are no lines available for it to enable.
Issue QUERY LINES to see how many lines are enabled.
,/
3-50
VM/SP Operator's Guide
----==:=-1
FLUSH
Privilege Class: D
When to Use FLUSH
Use the FLUSH command to halt and purge or, optionally, hold the current output
on a specified real unit record device. The device resumes activity with the next
scheduled spool file.
You will usually issue FLUSH when:
•
You see that a mistake in output queuing has occurred, or
You must start a high priority job before CP finishes the current job.
A mistake in output queuing might be that a user directed a file to a printer that is
equipped with the wrong printer train to process that file. Or a user may misdirect
a file to the real punch instead of his virtual punch unit. For example:
FLUSHE HO
FLUSH Format and Operands
Flush
raddr [ALL) [HOld)
raddr
is the address (cuu) of the real printer or punch whose activity you want to
terminate.
ALL
deletes all copies of the current output file. If you specify ALL, and CP is
printing or punching more than one copy of the current output file, CP
deletes all copies. If you do not specify ALL, CP deletes only the current
copy and prints or punches the next copy.
HOld
places the spool file in system hold status. CP does not purge the terminated
spool file. If you want to reset the status, use the CHANGE command.
If you do not specify HOLD, CP requeues the file and starts it on another
device. You must issue both the FLUSH and DRAIN commands to recover
the file.
Section 3. CP Commands
3 -51
Things You Should Know about the FLUSH Command
1. For buffered printers, such as the 3800 printer, the FLUSH command does not
affect files that have already been sent to the printer.
2. If CP starts a spool file on a real output device that it cannot make ready, CP
issues an error message to you. Recover the file and restart it by entering the
following commands:
FLUSH raddr [HOLD]
DRAIN raddr
After you enter these commands, CP issues you the FATAL I/O ERROR and
logically places the device offline. Use the VARY command to put the device
back online.
System Response
PRT} raddr FLUSHED USERID
{PUN
userid
FILE
fi Ie
CLASS RECORDS
class norecs
REC LEFT
recsleft
Cpy DIST
amt
If you specify the HOLD option, CP gives the FILE HELD response in
addition to the FLUSHED response.
raddr
is the real address of the printer or punch that you want to affect.
userid
identifies the user who issued the command. With DIST, user i d identifies
the user who will receive the output.
fi Ie
is the unique spoolid number that identifies the file to the VM/SP system.
class
was the class of the printer or punch when it started and the class that the
file was spooled as.
norecs
is the total number of logical records in the file.
recsleft
is the number of records left in the file after the system executes the
command.
3-52
VM/SP Operator's Guide
SEQ
userid seq
[
•. n _
.--
--.--. --
- - - - .... -
•. - - - - - - - - ... - - -
___n
_ _ _ _ _ _ ..: : - _ _ _ _ _ _ _ _ _ _ - - - - - - - .. - -
- . - . . - - - - - - - - - - - - - - . - - - -• • - - . - -
----=-...::~:::------
. -_____
n
_ _ _ _ . . ______
r-llUJ s =-:J
~GJ
• - - - - - - . - - .--------------- ------.
amt
is the number of copies you request to be printed or punched.
seq
is a sequence number that the system resets when you IPL. For example,
seq=5 means that this is the fifth file on this printer since IPL.
Section 3. CP Commands
3-53
FORCE
Privilege Class: A
When to Use FORCE
Use the FORCE command to force a logoff of any user on the system.
Note: The FORCE command is like issuing the LOGOFF command for another
user in that the system does no special processing to terminate outstanding virtual
machine I/O simulation. Thus, if an I/O device (such as a disk or tape drive)
drops ready while it is processing virtual machine I/O activity, the virtual machine
user performing the I/O on that device will not be able to continue processing. If
you issue the FORCE command at this time, it will not be effective because it
cannot complete until all waiting I/O is finished. Determine which I/O device is
involved and make that device ready so that the user can continue operations or be
forced from the system. Usethe QUERY command to determine the status of all
or selected I/O devices.
FORCE Format and Operands
FORCE
userid
userid
identifies the user who you want to log off.
In some cases, such as when system throughput is low due to thrashing, you
may need to force one or more users from the system. In this case, when it is
unclear as to which users you should force off, base your selection on which
users have VMSAVE specified in the CP directory.
Force a VMSAVE user, if possible, because the system will restore the
current state of the user's virtual machine when the user is logged back on.
If you force a non-VMS AVE user, the system destroys its current virtual
state and will not restore it when the user is logged back on. In this case, the
user will have to re-IPL the virtual machine at LOGON.
3-54
VM/SP Operator's Guide
r
System Response
LOGOFF AT hh:mm:ss zone weekday mm/dd/yy BY SYSTEM
The user receives the normal accounting message produced at logoff in addition to this logoff message.
GRAF raddr}
{ LINE raddr
DEV rid
LOGOFF AS userid USERS
DSCONNECT
nnn FORCED
raddr
is the real line address.
rid
identifies the real resource.
userid
identifies the user who CP logged off.
nnn
is the number of users still on the system.
Note: If an unrecoverable I/O error occurs on the virtual console, the
system may force a virtual machine user into a disconnected state. CP
then issues the preceding message with DSCONNECT instead of
LOGOFF. The user may log on at any time while his disconnected
machine is still running. See the System Logic and Problem Determination Guide Volume 1 (ep) for more information.
userid NOT LOGGED ON
You will get this response if the user is not logged on at the time when you
issue FORCE, or if the user has already started the LOGOFF/FORCE
process but has virtual I/O pending. (An interrupt pending condition on a
virtual I/O device could cause the virtual I/O pending condition.)
Section 3. CP Commands
3 -55
FREE
Privilege Class: D
When to Use FREE
Use the FREE command to take a set of spool files out of system hold status. Only
a spooling operator can free a spool file held by a spooling operator. The user or a
spooling operator can free a spool file held by the user. If a spool file is in double
hold status, CP can only process it after you remove both holds (system and user).
You can use the FREE command with the HOLD command to free output files
that are in a hold status. Use HOLD to detain a file or a group of files temporarily
to correct or change a spooling condition. Ybu will severely impact the spooling
area, if you block files from spooled output.
The following occurences may hold a file from output:
A device restriction
An 110 error on the printer or punch
A device modification.
For example, a printer may not be available,or it may have an incorrect print train
mounted. When you correct the problem, CP releases the files for output.
FREE Format and Operands
FRee
userid
Printer]
PUnch
[ ALL
userid
is the user whose spool files you want CP to release from a system hold
status. The user does not need to be logged on, but HOLD may have previously held his files.
Printer
PRT
frees the printer files for the user that you specify.
PUnch
pcn
frees the punch files for the user that you specify.
3-56
VM/SP Operator's Guide
/
frees all printer and punch files for the user that you specify.
System Response
None.
Section 3. CP Commands
3 -57
HALT
Privilege Class: A
When to Use HALT
Use the HALT command only in extreme cases to end any active channel program
on a real device that you specify. VM/SP issues a RESET IMMEDIATE
command to reset the status.
Note: Only use HALT if every other method you have tried fails. If you use
HALT indiscriminately, you may cause unpredictable results.
HALT Format and Operands
I
raddr
HALT
raddr
is the address (cuu) of the real device that you want to stop.
Some device addresses have multiple I/O paths defined by the VM/SP
system generation process. In these cases, HALT raddr causes CP to issue a
HIO on only that specified path.
The HALT command does not affect 3704 or 3705 Communications Controllers running in NCP or PEP mode. See the class A NETWORK
command.
System Response
DEVICE HALTED
indicates that VM/SP reset the status and halted the device.
3-58
VM/SP Operator's Guide
r::::=----
HOLD
Privilege Class: D
When to Use HOLD
Use the HOLD command to place user spool files in a system hold status. You can
hold files of any user on the system. You can reset both the user's files and his
hold status with the FREE command.
Only a class D operator can free files that a class D user is holding. A spool file
that a spooling operator or a user is holding can only be freed by the same spooling
operator or user. You must remove both holds (system and user) before CP can
process a spool file in double hold status.
HOLD Format and Operands
HOLD
Printer]
userid [ ~~ch
userid
identifies the user whose spool files you want CP to place in a system hold
status. The user does not need to be logged on when you issue this
command.
Printer
PRT
holds printer files for the user that you specify.
PUnch
pcn
holds punch files for the user that you specify.
Section 3. CP Commands
3 -59
L: _________________________________________________________________________
~
ALL
holds all the printer and punch files for the user that you specify.
System Response
None.
/
3 -60 VM/SP Operator's Guide
INDICATE
Privilege Class: A or E
When to Use INDICATE
Use INDICATE to help you monitor system performance. INDICATE displays at
the console:
•
The use of the major system resources of processor and storage
The contention of the major system resources of processor and storage.
System analysts can be aware of heavy load conditions or low performance situations. Analysts may need to resolve these situations by using more sophisticated
data collectiC?n, reduction, and analysis techniques.
Use INDICATE to identify:
•
•
The users in queue 1 and queue2
The 110 devices that the users are queued up on
The paging devices that may have been filled
The execution characteristics of any user
The total amount of resources that a user is using.
•
•
INDICATE Formats
Class A INDICATE Format
INDicate
FAVORed
Section 3. CP Commands
3-61
Class E INDICATE Format
INDicate
-
FAVORed
I/O
LOAD
PAGing
WAIT]
[ ALL
Queues
-
USER
INDICATE Operands (Class A, E)
FAVORed
provides a list and SET FAVORED statistics for all users who have the
favored execution option.
I/O
provides the userids of all the users in I/O wait state at that instant in time.
This helps you foresee conditions that may lead to possible I/O contention
within the system. CP also gives you the address of the real device to which
the most recent virtual SIO was mapped. CP does not show the queue of
users who issue SIOF to busy devices.
Because INDICATE I/O indicates a sample according to an instant in time,
use I/O several times before you assume a condition is persistent. If the
condition does persist, use the MONITOR SEEKS command to more thoroughly investigate the condition.
LOAD
provides information about VM/SP's operating load by displaying:
•
•
The number of users in queue 1 and queue 2
How real storage is being used
The ratio of active users to users being serviced.
PAGing [WAIT]
ALL
PAGing WAIT
displays the following:
The userids of the users currently in page wait
The number of page frames allocated on drum storage and on disk
storage.
3-62
VM/SP Operator's Guide
,/
·-..
_._----._.._-_...._-._---.-.
~.---~~----.~--~~-.---.-
..-- .. -•. -
.
__ ........ _-..
__
_.
.-. " - ' - ' - ' ... -- ............. __ ..... _]
If your installation has 2305s as primary paging devices and other
direct access devices as secondary paging devices, you may want to
use this operand. If the primary device is full and the system is allocating paging space on the slower device, system performance may be
lowered. Use PAGing WAIT when the Queues operand shows that a
number of users in queue 1 and queue 2 are persistently in page wait.
PAGing ALL
displays the page residency data of all users of the system (including
the system nucleus and pageable routines). The format of the reply
message is the same as that of the P AGing WAIT operand.
Queues
displays the following:
•
•
•
The active users
The queues that they are in
The storage that they are occupying
The status that they are in.
This display shows the users that are currently dominating main storage. The
display also includes the users waiting in eligible lists, because they are contending for main storage.
USER
[*userid]
USER~
lets you determine the activity of your own virtual machine.
USER userid
lets you determine the resources that the specified virtual machine is
using and the events that have taken place. If you are a class E user,
you can access data from the VMBLOK of any user currently logged
on to the system. This helps you understand an overload or poor performance situation.
You can use INDICATE USER, before and after you execute a program, to
find the execution characteristics of a program and the resources that is uses.
Do not, however, use INDICATE USER during the execution of a program
when it is issuing spooling 110 requests, because CP updates some fields
once and other fields dynamically. Do not halt or abort any spooling operation before it finishes, if you want to use the INDICATE USER command.
Section 3. CP Commands
3-63
System Response
INDICATE FAVORED
CP repeats the following response three times per line. There may be as many
lines as necessary to indicate the SET FAVORED options of all virtual machines
logged on.
userid
[Q]
[P xxx yyy]
userid
identifies the user of the virtual machine that has one or both of the
favored execution options.
Q
indicates that you specified SET FAVORED without a percentage for
this userid.
P
indicates that you specified SET FAVORED with a percentage for this
userid; both Q and P may appear in this response.
xxx
is the percentage you specified with SET FAVORED.
yyy
is the approximate percentage of the processor that the system has
given the indicated virtual machine over the past eight minutes.
INDICATE I/O
useridl xxx
userid2 yyy
useridl
userid2
identifies the users.
xxx
indicates the real device address.
yyy
indicates that two users are waiting for I/O to complete on the indicated device.
Note: If a virtual machine issued mUltiple SIOs, the response shows the real device
address corresponding to the most recent one issued.
CP issues the following response for the I/O operand when appropriate:
NO USERS IN
3-64
VM/SP Operator's Guide
1/0
WAIT
INDICATE LOAD
CPU-nnn% APU-nnn%
}
{ PROC xx-nn% PROC yy-nn%
Ql-nn Q2-nn STORAGE-nnn% EXPAN-nnn
PAGING-nnnn/sec, STEAL-nn%, LOAD-nn%
n
is always a decimal number.
CPU-nnn%
indicates the percentage of time that the main processor is running;
CP derives the percentage from the smoothed wait value that the
scheduler maintains.
APU-nnn%
is the percentage of time that the attached processor is running.
PROC xx-nn%
is the percentage of time that the system is running on the IPL
processor.
PROC yy-nn%
is the percentage of time that the system is running on the non-IPL
processor.
Ql-nn Q2-nn
represents the contention for the processor by the average numbers of
users in queue1 and queue2, maintained by the scheduler.
STORAGE-nnn%
is an approximate percentage that measures the usage of real storage.
It shows the relationship between the number of pages in storage for
in-queue virtual machines and the number of pageable pages in the
system.
Due to the algorithm that the scheduler uses to approximate the
number of pages in storage, the value of STORAGE can exceed
100%. The value may be less than 100 0/0, even when the sum of the
estimated working set for in-queue virtual machines is greater than the
number of pageable pages.
EXPAN-nnn
is the total delay in response time that a virtual machine experiences,
because it contends for both real storage and the processor. If the
virtual machine does not contend for either resource, EXP AN is 1.
The scheduler contention value, EXP AN, is the measure of time it
takes a virtual machine to receive a given amount of processor time.
Its derivation follows:
EXPAN
QT
Q2 51 ice
Section 3. CP Commands
3 -6 5
______::1
L
QT
is the average elapsed time between queue drops that a Q2 virtual
machine spends in the Q2 dispatch list.
Q2 51 ice
is the amount of processor time that the system allows a Q2 virtual
machine in the dispatch list.
PAGING-nnnn/sec
is the average number of page I/O operations (page reads and writes)
that the system performs per second.
STEAL-nn%
is the percentage of page read operations that required CP to steal a
real storage page from another in-queue virtual machine.
LOAD-nn%
is an artificial value that attempts to measure (in elapsed time) the percentage of the system devoted to paging because of real storage contention. The following contribute to the value of LOAD:
•
Estimated processor time involved in paging
The amount of time spent in pagewait
The percentage of steals.
INDICATE PAGING
useridl xxx:yyy
userid2
xxx:yyy
useridl
userid2
identifies the user.
xxx
are the number of pages, in hexadecimal, that CP allocates on drum
storage for these users.
yyy
are the number of pages, in hexadecimal, that CP allocates on disk
storage for these users.
Note: This response may indicate one or more users. This sample response indicates two users. If the two users shown in the response executed similar programs,
userid 1 would experience more page wait than userid2.
You should know which users are occupying most of the primary paging device
space, and whether or not they are still active. CP may have allocated large
amounts of primary paging device space at IPL time to a virtual machine that is
executing a large operating system. If this paging space becomes inactive, the
3-66
VM/SP Operator's Guide
/
-------------_.. _-----_._------------_._-_._---_._--------,
c
machine is occupying a critical resource but is not using it. If you invoke the ALL
operand, CP gives xxx and yyy values for all users on the system.
The system issues the following response for PAGING WAIT, when appropriate:
NO USERS IN PAGEWAIT
INDICATE QUEUES
userid aa bb sss/ttt
userid
userid
identifies the user. (This response may display more than one user.)
aa
represents the eligible lists or queues that are occupied. The value is
E1, E2, E3, 01, 02, or 03.
bb
is one of the following status indicators:
runnable: the user is runnable and waiting for service by the CPU.
RU
is the current RUNUSER in uniprocessor applications. In attached
processor applications, RU is the current RUNUSER on the main
processor. In multiprocessor applications, RU is the RUNUSER
value on the IPL processor.
RA
is the current RUNUSER on the attached processor. In
uniprocessor applications, RA does not appear in the response
line. In multiprocessor applications, RA is the RUNUSER value
on the non-IPL processor.
PG
in page wait-the user's program is not executing because CP is
trying to bring in a page from a paging device.
10
in 110 wait-the user is in 110 wait because access to the device is
not available at the moment.
EX
in instruction simulation wait-the user is waiting for the system to
complete the instruction simulation.
PS
in PSW wait-the user is in an enabled wait state for high-speed
110 devices.
Section 3. CP Commands
3-67
OF
in attached processor configurations, means that the system
deferred the processing of a synchronous (program or SVC) interrupt for this user until the system lock is available. DF is not
applicable for uniprocessor operations.
Note: If a virtual machine is in more than one of the above states,
CP displays only one state. The state that CP displays is the first
one it encounters in the order of priority indicated above.
sss
is a hexadecimal number that indicates the number of pages resident in
real storage.
ttt
is a hexadecimal number that the dispatcher estimates to be the
number of pages in the working set of the specified user.
Note: CP orders the lines of the above response as follows:
1. Eligible list El users in scheduling priority order
2. Eligible list E2 users in scheduling priority order
3. Eligible list E3 users in scheduling priority order
4. Ql, Q2, and Q3 users in runlist priority order (dispatching priority).
These values help you analyze system performance. For details see VM / SP System
Programmer's Guide.
CP issues the following response for the INDICATE QUEUES command when
appropriate:
NO USERS IN QUEUE
INDICATE USER
PAGES: RES-nnnn WS-nnnn REAOS=nnnnnn WRITES=nnnnnn MH-nnnn FH-nnnn
lists the data from the user's VMBLOK that relates to his or her virtual
machine's paging activity and resource occupancy.
VTIME=nnn:nn TTIME=nnn:nn SIO=nnnnnn ROR-nnnnnn PRT-nnnnnn
PCH-nnnnnn
lists the following:
•
The user's CPU usage
•
The user's accumulated I/O activity counts since he logged on, or since
you last issued ACNT for his virtual machine.
n
is a decimal number.
/
3-68
VM/SP Operator's Guide
[-------------------------------_.
__ .._------._---------------"--
RES-nnnn
is the amount of the user's virtual storage page able pages that are resident in real storage when the class E user issues the command.
WS-nnnn
is the most recent system estimate of the user's working set size.
READS
is the total number of page reads for this user since he logged on or
since the class E user last issued the ACNT command for his virtual
machine.
WRITES-nnnnnn
is the total number of page writes for this user since he logged on or
since the class E user last issued the ACNT command for his virtual
machine.
MH-nnnn
is the number, taken when the class E user issues the command, of
virtual pages allocated on system DASD (moveable head) paging
space for the particular user.
FH-nnnn
is the number, taken when the class E user issues the command, of
virtual pages allocated on system DASD (or fixed head cylinder)
paging space for the particular user.
VTIME=nnn:nn
is total virtual time for the user.
TTIME=nnn
is virtual and simulation time for the user.
SIO=nnnnnn
is the total number of nonspooled 110 requests that the user issued.
RDR-nnnnnn
is the total number of virtual cards that the system reads. CP gets this
value from spool file control information and increments the total
number of cards to be read once to the RDR=value. CP increments
this value at the start of each file read operation.
PRT-nnnnnn
is the total number of virtual cards that the system prints. CP gets this
value from spool file control information and increments the total
number of lines to be printed once to the PRT=value. CP increments
this value at the end of each file print operation.
Section 3. CP Commands
3-69
PCH-nnnnnn
is the total number of virtual cards that the system punches. CP gets
this value from spool file control information and increments the total
number of cards to be punched once to the PCH=value. CP increments this value at the end of each file punch operation.
3 -70
VM/SP Operator's Guide
l (Q) !~\ [J) [3 UJ r-
c:----------=--===--=-----_--==_--_--___
------..-.----------------------::=:J
LOADBUF
Privilege Class: D
When to Use LOADBUF
Use the LOADBUF command for the following printers:
The IBM 1403 - use LOADBUF to load the Universal Character Set (UCS)
buffer with a print chain/train image that you specify.
•
The IBM 3203,3211, or 3262 - use LOADBUF to load the UCS buffer or
the Forms Control Buffer (FCB) with an image that you specify.
•
The 3289 Model 4 - use LOADBUF to load the Font Offset Buffer (FOB)
with the image print belt and FCB.
Note: Do not load the 4245 and 4248 printers with the LOADBUF UCS
command. Load the UCS image by the printer.
'---~'
Drain the device before you issue the LOADBUF command. Use this command to
load the FCB/UCS/UCSB/FOB buffer of a printer under the following conditions:
•
•
•
Any time you want to change the print belt/ chain/ train
If the buffer has been causing an excessive amount of parity errors
When the printer or control unit is powered up since the last buffer load
If the printer was previously attached to a user.
Load the FCB buffer:
•
Anytime you want to change the FCB image
If the printer was previously attached to a user.
LOADBUF Format and Operands
LOADBUF
raddr
( raddr
UCS name
FCB name
[Fold]
[Index
(Ver] }
[nn] ]
raddr
is the address of the printer whose buffer you want the system to load.
Section 3. CP Commands
3 -71
UCS name
name is a one to four character name of the UCS, UCSB, or FOB image that
you want CP to load into the printer at "raddr." For the 4245 and 4248
printers, LOADBUF only verifies that the image name you specify here is
really the image that the 4245 or 4248 printer loaded internally.
FCB name
name is a one to four character name of the FCB image that you want CP to
load.
Note: When you want to load both a print chain image (UCS) and a'forms control
buffer (FCB), you must use two distinct LOADBUF commands for the specified
printer.
The supplied UCS and FCB names are as follows:
UCS--1403
AN
HN
PCAN
PCHN
QN
QNC
RN
YN
TN
PN
SN
or 3203
normal AN arrangement
normal HN arrangement
preferred character set, AN
preferred character set, HN
PL/I - 60 graphics
PL/I - 60 graphics
FORTRAN, COBOL commercial
high speed alphameric
text printing - 120 graphics
PL/I printing - 60 graphics
text printing - 84 graphics
UCS--3211
All
H11
G11
P11
T11
Standard Commercial
Standard Scientific
ASCII
PL/I
Text Printing
UCS--3262
P48
P52
P63
P64
P96
Pl16
P128
48-character belt
52-character belt (Austria/Germany)
63-character belt, optimized
64-character belt
96-character belt
116-character belt (French - Canadian)
128-character belt (Katakana)
UCS--3289
F48
F64
F94
F127
- Model 4 (FOB)
48-character belt
64-character belt
94-character belt
127-character belt
FCB--3203, 3211, 3262, 3289E, 4245, or 4248
3-72
VM/SP Operator's Guide
CP provides two names for an FCB image: FCB land FCB8. See VM / SP System
Programmer's Guide, for information on how to add more FCB images. Look in
this section under "Things You Should Know about the LOADBUF Command"
for information about FCBl (6 lines/inch) and FCB8 (8 lines/inch).
Fold
loads the UCS or FOB buffer with the folding operation code, so the printer
can print uppercase for lowercase bit configurations. Use LOADBUF UCS
again without the FOLD option to turn FOLD off. Note that the
LOADBUF UCS command does not load a buffer for the 4245 and 4248
printers. FOLD is valid for UCS but not for FCB.
Notes:
1.
When the FOLD option compares character codes from the UCS buffer and
the print line buffer, the control unit ignores bit positions 0 and 1 of the
EBCDIC code. Thus, the printer only prints uppercase characters from either
uppercase or lowercase data codes.
2.
Use the FOLD option with discretion, since the printer may print more than
one character for the same EBCDIC bit configuration, depending on the image
used.
3.
Depending on the image used, you may experience incorrect printing for a dash
(-) and a backward slash ( \ ), lowercase and uppercase characters, superscript
numbers and standard numbers, and special character combinations.
Ver
prints the contents of the UCS/UCSB/FOB buffer on the printer that you
specify to verify the buffer loading function. Ver is not valid for FCB.
You can use Ver for the 4245 and 4248 printers, but Ver performs no functions for these printers. The LOADBUF UCS command automatically verifies the band loaded in the 4245 and 4248 printers. The system displays
messages for the 4245 and 4248, even if you don't specify the Ver option.
LOADBUF automatically sets the Block Data Check latch for the
UCS/FOB buffer load. If you specify Ver after the system has executed the
command, the printer that you specify prints the image of the UCS/FOB
load. The printed output must correspond to the description of the specified
buffer load in the following publications:
2821 Control Unit Component Description
IBM 3211 Printer 3216 Interchangeable Train Cartridge & 3811 Printer
Control Unit Compo Desc. and Operator's Guide
3262 Printer Models 1 and 11 Component Description.
Use LOADBUF UCS with this option to send a block data check command
code.
Section 3. CP Commands
3-73
L __.___.. _...._.._.. _..
Index [nn]
starts printing the output in the designated (nn) print position. nn must be a
number from 1 to 31. If you specify INDEX without a value, the index value
in the FCB macro becomes the index value. For a description of the FCB
macro and forms control images, see the VM / SP System Programmer's
Guide.
Index is not valid for a 3203 Model 4 or 5 printer or 3262, 3289E, 4245, or
4248 printers.
Things You Should Know about the LOADBUF Command
3-74
1.
Spool files that were created on a virtual spooled 3203, 3211, 3262, 3289E,
4245, or 4248 may have virtual FCB images imbedded in them. When you
print these spool files on a real 3211, 3203, 3262, 3289E, 4245, or 4248
printer, the system loads the imbedded FCB image into the real printer. The
FILEFCB, CFILEFCB, or DEFFCB option that you specified on the START
command dictates how the system wi11load the FCB image. For the FILEFCB
and CFILEFCB options, the system does not use the FCB image that you
loaded with the LOADBUF command.
2.
The system prints spool files, created on a virtual spooled 1403 printer, with
the FCB image loaded by the last LOADBUF command you issued.
3.
For FCBl, when the length of page is 66 lines and there are 6lines/inch, the
following is true:
LSkip
Represented
Channel
Specification
1
1
3
2
5
3
7
4
9
5
11
6
13
7
15
8
19
10
21
11
23
12
64
9
VM/SP Operator's Guide
lOADSlLDlF
For FeB8, when the length of the page is 68 lines and there are 8 lines/inch,
the following is true:
LSkip
Represented
Channel
Specification
1
1
4
2
8
3
12
4
16
5
20
6
24
7
28
8
32
10
36
11
63
12
66
9
System Response
For all Printers Except 4245 and 4248: If you specify VER, the system
prints the contents of the ues and/or FOB buffer on the specified printer.
For 4245 and 4248 Printers:
UCS IMAGE VERIFIED
The system compares the ues name with the ues that is loaded in the
printer, even if you don't specify VER. You get this message if the ues
name in the LOADBUF command matches the ues that is loaded in the
printer.
DMKCSB1581 UCS IMAGE NAME MISMATCH.
IMAGE NAME LOADED IS xxxx
indicates that the the ues name in the LOADBUF command does not
match the ues that is loaded in the printer.
Section 3. CP Commands
3-75
lOCATE
LOCATE
Privilege Classes: C or E
When to Use LOCATE
Use the LOCATE command to find the
with:
•
addres~es
of CP control blocks associated
A particular user
A user's virtual device
A real system device.
•
The control blocks and their use are described in the Data Areas and Control Block
Logic Volume 1 (CP).
LOCATE Format and Operands
LOCate
userid [vaddr]
{ raddr
I
userid
identifies the logged-on user. CP displays the address of this user's virtual
machine block (VMBLOK).
vaddr
displays the following addresses with the VMBLOK address. These
addresses are associated with the virtual device address that you specify:
•
The virtual channel block (VCHBLOK)
•
The virtual control unit block (VCUBLOK)
The virtual device block (VDEVBLOK).
raddr
displays the following addresses asssociated with the real device address that
you specify:
The real channel block (RCHBLOK)
•
The real control unit block (RCUBLOK)
,/
•
3-76
VM/SP Operator's Guide
The real device block (RDEVBLOK).
------------
Note: The raddr option is not valid for CP control blocks associated with
SNA terminals that are connected with the VT AM Network Communications Application (VM/VCNA) product and with VTAM SNA Console
Support (VSCS).
System Response
VMBLOK
xxxxxx
is the response for the LOCATE userid command.
VMBLOK
VCHBLOK
xxxxxx
xxxxxx
xxxxxx
xxxxxx
is the response for the LOCATE userid vaddr command.
VCUBLOK
RCHBLOK
RCUBLOK
xxxxxx
xxxxxx
xxxxxx
is the response for the LOCATE raddr command.
VDEVBLOK
RDEVBLOK
Section 3. CP Commands
3-77
LOCK
Privilege Class: A
When to Use LOCK
Use the LOCK command to permanently lock in real storage selected pages of:
•
•
The pageable CP nucleus
A user's virtual storage.
LOCK excludes the selected pages from future paging activity.
The LOCK command can make a virtual machine more efficient; LOCK keeps
high activity pages, such as virtual page zero, in real storage if some other virtual
machine is already using the reserved page frames option. If the amount of page
frames available for paging is limited, do not issue LOCK without the approval of
the system programmer.
Issue LOCK as many times as you need for one virtual machine to lock noncontiguous pages of storage. The remaining virtual machine storage blocks will stay
pageable.
Note: If you lock too many pages of real storage, the remaining virtual machines
may not have enough available page frames left to operate efficiently. Since the
virtual machines will contend for the remaining available page frames, throughput
may be severely degraded.
Other ways to unlock a locked page are:
•
Use the SYSTEM CLEAR command, which unlocks the user's locked pages
Issue DIAGNOSE 14,30, 34, or 38 against the locked page.
LOCK Format and Operands
LOCK
I
userid
SYSTEM
I
firstpage lastpage [MAP]
userid
identifies the logged-on user.
SYSTEM
locks one or more of the pageable CP pages. Once a page is locked, it
remains locked until the user either logs off the system, or issues the
3 -78
VM/SP Operator's Guide
UNLOCK command for that page. If you have the locked pages option you
can unlock pages by doing any of the following:
Re-IPL the system by device address, and specify the clear option. The
system unlocks the pages, and they become available to the system being
loaded.
Re-IPL the system by device address, and do not specify the clear
option. All locked pages stay locked, except the page given to
DMKVMI for IPL.
Re-IPL the system by name (shared system). The system unlocks the
pages only if:
The locked pages are not in the shared segment, or
The page is in the shared segment, and the user who is re-IPLing is
the last user of the shared segment.
Example:
LOCK SYSTEM 2A 2A MAP
VIRTPG REALPG (response to the MAP operand)
02AOOO 04 FOOO
DCP 4F038-4F048
etc.
UNLOCK SYSTEM 2A 2A
Locks a CP page for display purposes and unlocks it when finished.
firstpage
is the hexadecimal value of the first user page that you want the system to
bring into storage and lock.
lastpage
is the hexadecimal value of the last user page that you want the system to
bring into storage and lock. If you only want the system to bring one page
into storage, lastpage must be the same as firstpage.
For firstpage and lastpage, specify only the page numbers.
Examples:
LOCK USERA 12 2C
locks USERA's virtual storage locations X'12000' to X'2COOO'
in real storage.
LOCK USERA 0 2
locks the first three pages of USERA's storage.
Section 3. CP Commands
3-79
LOCK USERA BA BA
locks one page of USERA's storage.
MAP
prints a map of the virtual storage pages locked and the real page that is
assigned for each page locked.
Note: If you omit the MAP operand from a LOCK command and you later
want to find the real page, reissue LOCK with the MAP operand. You do
not need to unlock the page first.
The SYSTEM and MAP operands help you look at (with DCP) or change
(with STep) a page that is normally not resident, such as console function
routines. The MAP operand locates the real page for display purposes. Page
numbers are defined as follows:
Storage Locations
(in hexadecimal)
Page
Number
OOO-OFFF
1000-1FFF
2000-2FFF
o
12000-12FFF
12
1
2
Things You Should Know about the LOCK Command
3-80
1.
You cannot lock shared pages in a system generated for AP or MP operation.
However, in uniprocessor mode, if the system was not generated for AP or MP
operation, you can lock a shared page. If the shared page becomes nonshared,
the system unlocks it. In addition, if you lock a protected shared page and then
change the page, the system unlocks it.
2.
The virtual pages locked in processor storage are blocks of 4K (4096) bytes.
This block of storage does not need to represent all of the user's virtual
storage. You can issue the LOCK command as many times as needed for one
virtual machine to lock noncontiguous pages of storage. The remaining virtual
machine storage blocks may remain pageable.
VM/SP Operator's Guide
System Response
COMMAND COMPLETE
is the response when LOCK finishes processing.
VIRTPG
virt-firstpage
:j
REALPG
real-firstpage
virt-lastpage
real-lastpage
is the response to the MAP operand.
Section 3. CP Commands
3-81
MESSAGE
~--------------------------------------------------------------------------------~
MESSAGE
Privilege Class: A or B
When to Use MESSAGE
Use the MESSAGE command to:
•
Transmit message text to a specified userid
•
Transmit message text to the primary system operator
•
As the primary system operator, send message text to one or all logged-on
users.
If you want the users to receive messages automatically when they log on, see the
class B SET command.
MESSAGE Format and Operands
Message
Msg
I
AL~
userzd
*
I
msgtext
OPerator
ALL
is for the operator's use only. It lets you broadcast a message to all
logged-on users.
userid
identifies the user to whom you want to send the message.
*
sends a message to yourself.
OPerator
sends a message to the primary system operator, without having to know his
or her userid.
msgtext
is the message you want to transmit. You can enter as many characters as
will fit on the rest of the input line.
3-82
VM/SP Operator's Guide
,/
Things You Should Know about the MESSAGE Command
1. A user will not receive a message if he or she is not logged on or has suppressed the receiving of messages. You, the sender, will receive a message that
tells you so. If you send a message to ALL, you get a message for every user
that has message typing suppressed.
The system does not save a message that a user does not receive. You must
send the message at a later time when the user is receiving messages.
2.
A typewriter terminal or display terminal that has AUTOREAD set ON, will
receive the message when the user hits a carriage return.
System Response
hh:mm:ss MSG FROM OPERATOR: msgtext
is the message format that the user receives from the system operator.
hh:mm:ss MSG FROM userid: msgtext
is the message format that a user receives from another user (specified with
userid).
hh:mm:ss MSG FROM LOGONxxx: msgtext
is the message format sent from a user on the VM/SP system to another
user. The message shows that the user has not achieved logon status. xxx
is the real line address from which the MESSAGE command was issued.
For all of the above responses:
•
hh: mm: ss is the time in hours:minutes:seconds when the message was sent to
the user.
If the user receiving the message is the primary system operator, the alarm bell
at the central computer console rings.
•
If you are using a 3270, the message puts the screen in a HOLD status and
rings the audible alarm, if present. An exception to this is during system IPL or
restart operations, in which case the screen is put in MORE status. Press the
cancel key to return to the RUNNING status.
Section 3. CP Commands
3 -83
MIGRATE
Privilege Class: A
When to Use MIGRATE
Use the MIGRATE command to:
•
Activate the normal page/swap table migration routines
•
Force the pages of the particular user to the secondary device, even if that user
is currently active.
You will rarely use this command for a general user.
MIGRATE Format and Operands
MIGrate
[useridJ
userid
identifies the virtual machine user for whom you want to invoke page/swap
table migration.
When you issue MIGRATE with the userid operand, the system invokes
pagel swap table migration only for the specified virtual machine. In addition, the normal time-limit criteria for determining infrequently used segments is not active. The system migrates all pages that belong to the
specified virtual machine if:
•
•
The pages are not currently in real storage.
The pages are not in the process of being brought in or written out.
The system migrates all swap tables, as long as the segment they belong to
does not have any pages that are not migrated.
Note: If you issue the MIGRATE command without userid, the system will
immediately invoke the page/swap table migration, regardless of when it was
last invoked.
3-84
VM/SP Operator's Guide
[ - - - - - - - - - - - _ . _ - - - _..----
.-----"]
System Response
PAGE MIG DONE
The system migrates out the virtual machine's pages until none of its
"working set" of pages is resident. If the user becomes active, he will build
up his "working set" on his own.
~I
Section 3. CP Commands
3-85
MONITOR
Privilege Class: A or E
When to Use MONITOR
Use MONITOR to Trace and Record Events
Use MONITOR to override the system generated monitor function, and to initiate
or end the recording of events that occur in the real machine. This recording is
always active after a VM/SP IPL (manual or automatic). The CP internal trace
table records the following events:
•
•
•
•
•
•
•
•
•
External interruptions
SVC interruptions
Program interruptions
Machine check interruptions
I/O interruptions
Free storage requests
Release of free storage
Entry into scheduler
Queue drop
Run user requests
StartI/O
Unstack I/O interruptions
Storing a virtual CSW
Test I/O
Halt device
Unstack IOBLOK or TRQBLOK
NCP BTU (Network Control Program Basic Transmission Unit) .
Clear Channel operation (CLCH)
Simulated I/O interrupts.
Use the trace table to determine what happened before a CP system failure. For
information on finding and using the internal trace table see the VM / SP System
Programmer's Guide.
Use MONITOR to Measure Performance
To measure performance, MONITOR:
Displays the status of the following:
Internal trace table
Each implemented class of data collection
Specifications for automatic monitoring using spool files.
3-86
VM/SP Operator's Guide
/'
MONruo~
Enables various classes of MONITOR CALL
Specifies the time intervals for timer-driven data collection
Starts and stops data collection by MONITOR using tape or spool files
Overrides the automatic monitoring assigned at system generation time
Closes the monitor spool files
Specifies device addresses to be included in or excluded from a selection list for
DASD seeks analysis.
For more information about the MONITOR command, see the VM / SP System
Programmer's Guide.
Section 3. CP Commands
3-87
MONITOR Format and Operands
r
MONitor
AUTO disk ION
OFF
CLOSE
I
Display rSPOOLj
TAPE
ALL
ENable
PERForm
RESPonse
SCHedule
USER
INSTsim
DAStap
SEEKs
SYSprof
1
INTerval nnnnn [SEC1mm
MIl'{j
>
1-<
LIMIT n tNOSTOPj
STOP
SAMPLE
I
SEEKS INclude
EXclude
DELete
DISplay
/
raddr raddr···l
raddr raddr•..
I"'"
STArt SPOOL [TO userid] [BUFFS n]
CPTRACE
TAPE raddr
[BUFFS n]
~
STOP [SPOOL ~
CPTRACE
TAPE
TIME/FROM
FOR
ALL
\..
NONE
hl:ml TO h2:m21
hh:mm
lSelect one or more of the classes, subject to the restrictions listed
with the EN able operand.
2See the operand description for the default value.
3 -8 8
VM/SP Operator's Guide
-
AUTO disk {ON \
OFF
overrides the specification in the SYSMON macro for automatic startup of
monitoring. If automatic monitoring is already active, only a MONitor STOP
command will stop it manually.
Note: In general, when you try to override the definitions of the SYSMON
macro the results are temporary. Because the system does not do monitor
checkpointing, an IPL or system crash restores the automatic monitoring
definitions from SYSMON.
CLOSE
closes the current spool file, so it is available to the reader of the recipient
virtual machine. The system continues monitoring with a new spool file.
You can use CLOSE when the VM/SP monitor is spooling the collected performance data and you want to reduce the data thus far.
Display [SPOOL
TAPE
ALL
J
displays the status of VM/SP monitor variables and the internal trace table.
For any of the operands, the system lists the class of monitor call and its
current enabled/disabled state.
SPOOL
displays the automatic monitoring specifications. These include:
If automatic monitoring has been requested, and its start and stop
times
The number of monitor buffers that the system will use
The userid of the virtual machine to receive the spool file
The spool file class and record limit
The classes of monitoring that the system will enable.
If automatic monitoring is already in progress, the system lists the
spool file number and the number of monitor buffer records that the
system already wrote to it.
TAPE
displays the status of the monitor classes and CPTRACE table.
ALL
displays both SPOOL and TAPE responses.
I
~
Section 3. CP Commands
3-89
ENable
PERForm
RESPonse
SCHedule
USER
INSTsim
DAStap
SEEKs
SYSprof
enables the monitor call classes that you specify. Each time this command
completes successfu~ly the system creates a new value for control register 8.
The results from using the the MONitor ENable command differ, depending
on if data collection is active or inactive when you issue the command. If
data collection is active (MONitor STArt has been issued), the system moves
the new mask directly into control register 8, replacing the previous mask.
The new mask takes effect immediately. Collection then continues with the
classes just entered.
If data collection is not active when you issue the command, the system saves
the new mask and does not start collecting data until you issue MONitor
STArt. If you issue MONitor STArt without first issuing MONitor ENable,
the system uses the SYSMON class specifications. Any mask stays in effect
until the next MONitor STOP command.
PERForm
periodically saves system counters, which provide you with system
resource usage data (for example, information about how busy the
CPU is and how much storage is being used).
RESPonse
collects data on terminal I/O. This data helps simplify your analysis
of command usage, user response time, and system response time.
The data can relate user activity to system performance. In order for
RESPonse to be valid, the system programmer must set the TRACE
bit (1) in the LOCAL COPY file to a 1 and reassemble DMKMCC.
SCHedule
does the following:
•
Collects data about scheduler queue manipulation
Monitors the flow of work through the system
Indicates the resource allocation strategies of the scheduler, which
is one of the key functions of the system.
USER
periodically scans the chain of VMBLOKs in the system. The result is
user resource and status data.
INSTsim
records every virtual machine privileged instruction that the control
program (CP) standard simulation routines (DMKPRV and
DMKPRW) handle. Because INSTsim records the instructions, it may
3-90
VM/SP Operator's Guide
/
help your installation improve performance. This command does not
record privileged instructions that DMKFPS simulates.
If the VMA feature is active, CP handles less privileged instructions
for the virtual machines that are running with the feature activated.
DAStap
periodically samples device I/O activity counts (SIO) for online tape
and DASD devices.
Note: Specify DAStap before the MONitor STArt TAPE command.
You can disable DAStap at any time by re-entering MONitor ENable
without DAStap.
SEEKs
collects data for every I/O request to DASD devices. Displays
channel, control unit, or device contention and arm movement interference problems.
Notes:
1. If SET NOTRANS is in effect, CP does not extract information
for V=R regions.
2.
The MN700CYL field of the output record is in cylinders for FBA
devices and CKD devices.
SYSprof
adds data to the DAStap and SCHedule classes. This data helps build
a more detailed profile of system performance, so you can study
DASD utilization.
Note: The system cannot activate the SYSPROF class unless both the
DAStap and SCHedule classes are also active.
Restrictions:
•
Every MONitor ENable command yields a new mask. Thus, for example, if
the system is currently collecting PERForm and USER classes and you enter
MONitor ENable INSTsim, the system stops the PERForm and USER classes
and starts the INSTsim.
If the system is collecting data when you issue a MONitor ENable command
and CP detects an error in the command line, CP does not change the monitoring status. Conflicting keywords, keywords that the system does not recognize, or missing operands cause error messages.
Section 3. CP Commands
3-91
INTerval nnnnn[SEC] mm
MIN
Use this operand if:
•
The system has enabled or will enable a class that involves collecting
data at periodic intervals, and
•
If you want to override the default intervals of 60 seconds and 2
seconds.
Consider the 60 second interval as the primary sampling interval. This
interval sets how often the system collects PERForm, USER and DAStap
classes of monitor data. The two second interval applies to the highfrequency data sampling routine. This routine collects statistics about system
use from the I/O subsystem (channels, control units, and devices) when the
DAStap class is enabled.
You can issue MONitor INTerval at any time, if data collection is already in
progress; the new interval does not take effect until the current interval has
elapsed. The system resets MONitor INTerval to the defaults when:
•
•
•
•
•
You issue MONitor STOP.
The monitor stops automatically.
The system stops the monitor because of an unrecoverable I/O error.
The end-of-tape is reached.
The spool record limit is reached.
nnnn
is the interval between data collections. You can follow nnnn with
SEC or MIN to represent seconds or minutes. If you give an interval
without SEC or MIN, the system defaults to SEC. Do not specify
more than nine hours (540 minutes or 32,400 seconds) or less than
five seconds. If you do not specify an interval, the system sends an
error message.
mm
is the collection interval for the high-frequency sampler. This operand
applies to the DAStap class of data collection only. The default value
for mm is two seconds. You can specify from 1 to 99 seconds, but
mm must be less than the value that you specify for nnnnn.
LIMIT n [NOSTOP]
STOP
SAMPLE
overrides the LIMIT options of the SYSMON macro.
n
is a decimal number from 10 to 50000 that represents the maximum
buffer count in each spool file. If you want to change the automatic
monitoring specification without changing the limit number, specify an
asterisk (*) instead of a decimal number for n.
3-92
VM/SP Operator's Guide
i-
-_-::::::_--:----------------------------------J
L..-_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
NOSTOP
changes the automatic monitoring specification after the system
reaches its limit and closes the spool file. This option lets MONITOR
start again, after you change the limit.
STOP
changes the automatic monitoring specification after the system
reaches its limit and closes the spool file. This option stops
MONITOR, after you change the limit.
SAMPLE
n becomes the number of samples of PERForm, USER, or DAStap
data (or any combination of the three) that you want the system to
collect before closing the spool file.
After the system has collected n samples of data, the system closes the
monitor spool file and moves it to the virtual reader of the data
reduction virtual machine. Monitoring continues uninterrupted using a
new spool file. If you want to specify automatic monitoring, use the
SYSMON macro instruction in DMKSYS at system generation.
I
INclude raddr raddr... \
EXclude raddr raddr.. .
DELete
DISplay
provides seeks trace data on non-2305 DASD devices that you specify. You
can specify a range of addresses; for example, 270-275. This command lets
you form, display, or delete a list of devices for data collection. With a list of
collected data, you can decrease both the overhead to collect the data and
the processor time to reduce the data.
SEEKS
INclude
creates a list of non-2305 DASD device addresses for which the
system collects seeks data. If a list already exists, the system erases
the old list. Issue the appropriate MONitor ENable and STArt commands to invoke this function.
EXclude
excludes non-2305 devices from the existing list. The system creates a
new list and erases the old list.
DELete
cancels the list, and frees the occupied real storage.
DISplay
lets you review the contents of the list.
Notes:
1.
The system keeps the device list that you specify with the INclude or
EXclude option across separate monitoring sessions. Use the DISplay
Section 3. CP Commands
3-93
and DElete options to revise the list when SEEKs analysis is completed.
The system does not clear a list when you enter MONitor STOP.
2.
You can enter as many device addresses as you can fit on a single line.
However, for performance reasons, keep the list as short as possible.
STArt SPOOL3 [TO use rid] [BUFFS 0]
CPTRACE
TAPE
raddr
[MODE I~~i~ll
[BUFFS n]
Starts VM/SP m01'litor data collection to the spool file or tape
Starts the CP internal trace table.
SPOOL [TO userid] [BUFFS n]
starts the VM/SP monitor data collection using a spool file to store
the data.
Close the monitor spool file by a MONitor STOP or MONitor CLOSE
command when:
The spool file reaches the record count limit (specified in the
SYSMON macro).
•
The system restarts or shuts down.
The system generates the filename and file type of the monitor spool
file. The filetype identifies the date and time of starting. The class of
spool file is specified in the SYSMON macro and defaults to M. If you
do not specify the classes of data collection with an ENABLE
command, the system uses those specified in the SYSMON macro.
userid
identifies the virtual machine user whose virtual reader will
receive the reader files. These reader files result from the
system when it stops collecting data, closes the spool file, and
adds the spool file to a the reader files.
If you are issuing the START command and want to be the
recipient virtual machine, specify an asterisk (*) for userid. If
you omit the userid option, the system uses the userid in the
SYSMON macro instruction. The option that you specify stays
in effect until the system is re-IPLed or until someone issues a
new command that uses the TO userid option.
The default value is the active trace facility that is SPOOL or TAPE.
3-94
VM/SP Operator's Guide
BUFFS n
overrides the number of monitor buffers that the system uses, as
specified in the SYSMON macro. The BUFFS option overrides
SYSMON for the length of the data collection session. Then
the system returns to the SYSMON specification. If the number
of buffers specified in the SYSMON macro has been defaulted,
the system adopts the defaults described in the MONitor STArt
TAPE command.
CPfRACE
starts the tracing of events that occur on the real machine. The system
records the events on the CP internal trace table in chronological
order. When the table fills, recording continues by overlaying data
from the beginning of the table. You can tell what data was overlaid
by reading the time stamp for each entry.
TAPE raddr MODE
starts data collection onto a tape mounted on a 9-track tape drive or a
3480 tape drive. The system starts collecting data for the classes of
monitor call that you specified in a previous MONitor ENable
command. The system moves the mask that the MONitor ENable
command saved into control register 8. The system collects the data
into buffer pages in real storage. These pages are separate from the
internal trace table pages. As each data page is filled, the system
writes it on the tape.
raddr
is the real hexadecimal address of the tape drive that you want
to use.
BUFFS n
specifies the number of 4096-byte buffers that the system will
use to monitor operations. n is a value from 1 to 99. If you
don't specify a value for n, the default is based on the real
storage sizes shown below.
Storage Size
Less than 1 Meg
Between 1 and 2 Meg
Between 2 and 4 Meg
Between 4 and 8 Meg
Between 8 and 16 Meg
Over 16 Meg
Default
2 buffers
3 buffers
4 buffers
5 buffers
6 buffers
7 buffers
BUFFS 1 is valid if the command entry is limited to ENable
PERForm for data collection. Once monitoring is in progress
with one buffer specified, you cannot invoke MONITOR with
more ENable operands. Single buffer operation is useful for
basic performance analysis in minimum main storage configurations.
Section 3. CP Commands
3-95
MODE
requests a mode setting for the reset value of tape density when
the system starts to monitor. You can specify mode values of
800, 1600, 6250, or 38K. If you specify a density mode that
the tape drive cannot handle, the control unit does not return an
error condition. In this case, the system ignores the mode
setting and uses the default control unit setting.
A MODE option of 38K is valid only for a 3480 tape. Since the
3480 records only at a density of 38K, the default for the 3480
is 38K.
STOP [SPOOL
J
CPTRACE
TAPE
SPOOL
stops data collection on a spool file if automatic VM/SP monitor data
collection is active. Monitoring will not start again (even if the current
time is within the bounds of the TIME operand of the SYSMON
macro), unless the system fails or shuts down and your installation
reloads it.
CPTRACE
stops the internal trace table event tracing. Event recording stops, but
the system does not release the pages of storage that have the CP
internal trace table. You can restart tracing at any time by issuing the
MONitor STArt CPTRACE command.
TAPE
stops data collection on tape. The system stores a zero mask in
control register 8, and, thus, disables monitor call interruptions. The
system writes out the last partially filled page, writes two tape marks,
and rewinds and unloads the tape. The system. then releases the two
buffer pages that it obtained at the time it processed the MONitor
STArt TAPE command.
Note: If you do not specify an option and the VM/SP monitor is active,
data collection stops whether or not a spool file or tape is in use. You must
use the CPTRACE option to stop internal tracing. Note that the operands
are separate functions, and commands that affect the status of one function
have no effect on the other.
1
TIMEIFROM hl:ml TO h2:m2
FOR hh:mm
ALL
NONE
overrides the automatic monitoring start and stop times defined by the
SYSMON macro instruction.
3-96
VM/SP Operator's Guide
[-------.---------_-===-:::-_-_-._-_--:::--_-_-_______________- - - - - - - - - - _ - 1 " ]
FROM hl:ml TO h2:m2
specifies a start and stop time in hours and minutes using a 24-hour
clock. The default start time is 09:00 and end time is 17:00.
FOR
specifies the length of time that you want the system to collect data.
Use FOR to collect performance data over short periods of time for
testing or benchmarking. Automatic monitoring must be in effect
before you specify this parameter (AUTO= YES in the SYSMON
macro of DMKSYS or MONitor AUTOdisk ON command). When
you specify this option, the new time overrides the time defined in the
SYSMON macro.
ALL
specifies 24-hour monitoring.
NONE
tells the system not to monitor performance.
Things You Should Know about the MONITOR Command
The MONitor command's response to unusual conditions are:
During Automatic Monitoring Using Spool Files
Shutdown: The system stops collecting data with a MONitor STOP SPOOL
command. The system closes and moves the spool file for data reduction to the
reader chain of the designated virtual machine.
System Failure: The system abandons the current monitor data collection
buffer and closes the spool file that has all previous output buffers. The system
then makes the spool file available to the virtual reader of the designated virtual
machine for further data reduction.
Initial Program Load: The system automatically starts monitor data collection
if:
•
The SYSMON macro of DMKSYS specifies automatic monitoring. A system
programmer at your installation is responsible for setting up DMKSYS.
•
The IPL occurs within the time range that the macro specifies.
CP will inform you, but you do not need to do anything. You can, however, use
the MONitor Display SPOOL command to check the specifications for automatic
monitoring.
Section 3. CP Commands
3-97
Spool Space: If automatic monitor data collection is regularly active and the
system is issuing messages that show a serious loss of spool space, do the following
checks:
Be sure that the system reads and reduces the monitor spool files on a
day-to-day basis. This is so that the system gets back the occupied spool
space.
Use MONitor DISPLAY to ensure that the data collection classes do not
include trace classes, which produce large volumes of data (e.g. INSTsim or
SEEKs). Keep in mind that the specifications in the DMKSYS macro
SYSMON may be temporarily overridden by you or someone else using the
MONitor command.
If you have little spool space, the system can monitor to tape or use the LIMIT
parameter. You can use the LIMIT parameter of the SYSMON macro or the
LIMIT option of the MONITOR START SPOOL command to restrict the size
of the monitor spool file.
Suspension: Data collection suspension may occur because of insufficient
buffer space. You may need to increase the buffer allocations. Use the BUFFS n
option of the MONitor STArt command until you can revise the SYSMON macro
inDMKSYS.
During Manual Monitoring Using Tape
Suspension: Data collection suspension may occur because of insufficient
buffer space. This usually occurs when:
Classes USER and/or DAStap are enabled.
Large numbers of users are logged on.
The system I/O configuration is extensive.
Avoid losing data by specifying more buffers with the BUFFS option of the
MONitor STArt TAPE command.
Unrecoverable Tape Error: Unrecoverable tape errors are errors that the
I/O error handling routines cannot recover from by standard recovery procedures.
When an unrecoverable error occurs, monitor tries to write two tape marks and
rewind/unload the tape. The system abandons the use of the tape and data collection stops. The system informs the operator or takes action. Whether or not the
write-tape-marks and rewind/unload is successful, the system releases the tape
drive.
End-of-Tape Condition: When end-of-tape occurs, the system writes a tape
mark on the tape and rewinds and unloads it. The system stops monitoring, and
informs you of the action taken.
Initial Program Load: MONitor STArt CPTRACE is active after real system
IPL (manual or automatic). The monitor tape data collection is off after IPL.
3-98
VM/SP Operator's Guide
c
Shutdown: System shutdown implies a MONitor STOP TAPE command. The
system performs normal command processing for the STOP TAPE function.
System Failure: If the VM/SP system fails and data collection is active, the
system tries to write two tape marks and rewind/unload the tape. If the tape drive
fails to rewind and unload, write a tape mark before you rewind and unload the
tape. The system failure stops monitor data collection.
I/O Devices: You must dedicate a supported tape drive to the system for the
duration of the monitoring.
System Response
COMMAND COMPLETE
for MONITOR commands, other than MONITOR DISPLAY, that run successfully.
CLS
KEYWORD
nn
nn
keyword
keyword
STATUS
ENABLED
DISABLED
CPTRACE
ENABLED
if you issue the MONITOR DISPLAY command and the data collection is.
on tape.
If monitor is using spool files, the automatic monitoring specifications and status
are listed as described in the MONITOR DISPLAY section.
MONITOR HAS BEEN MANUALLY STARTED
when you issue the MONitor STArt CPTRACE command. Only the
primary system operator gets this message.
SELECTIVE SEEKS NOT CURRENTLY ACTIVE
occurs if SEEKS is not active at the time you issued the DISPLAY option.
SEEK, STOP, OR CLOSE CMD IS IN PROCESS, RETRY
if you issue a MONITOR command when the monitor is suspended.
MONITOR HAS BEEN MANUALLY STOPPED
when you issue the MONITOR STOP command. Only the primary system
operator gets this message.
Section 3. CP Commands
3-99
MSGNOH
Privilege Class: A or B
When to Use MSGNOH
Use the MSGNOH command to:
Transmit message text to a specified userid
Transmit message text to the primary system operator
As the primary system operator, send message text to one or all logged-on
users
without the standard header associated with the MESSAGE command.
MSGNOH Format and Operands
/'
MSGNOH
userid
ALL
msgtext
f
1*
userid
identifies the virtual machine to whom you are sending the message.
msgtext
is the message text that you want to transmit. You can enter as many characters as will fit on the rest of the input line.
ALL
sends the message to all users.
*
sends a message to yourself.
Note: MSGNOH also accepts a keyword of OPERATOR.
3 -1 00
VM/SP Operator's Guide
Things You Should Know about the MSGNOH Command
You can have your TELL EXEC use MSGNOH to send messages to users on the
same system as you. Change the default set up in the GLOBALV file by issuing:
DEFAULTS set tell msgcmd msgnoh
When you set the default, it remains that way until you reset it. To change default
back to MSG, issue:
DEFAULTS set tell msgcmd msg
System Response
msgtext
the specified virtual machine receives the message text only.
Section 3. CP Commands
3 -1 0 1
~----------------------------------------------------------------------------~
NETWORK
Privilege Class: A or B
When to Use NETWORK
Use the NETWORK command to:
Control the operation of a 3704, 3705, or 3725 (collectively referred to as
37xx) control program operating in 270x emulation mode (EP).
Load or dump the contents of a 3704 or 3705 communications controller.
Control remote 3270 devices attached to VM/SP with binary synchronous
lines.
Be aware of the different classes of resources that are defined at generation time
for the 37xx control program and the 3270 remote support.
Examine the output from the first stage of the 370x control program generation to
tell which resources are lines and which are terminals. The installation system programmer (or whoever performs the 370x control program generation) should
prepare a cross-reference list of resource IDs and their characteristics, such as line
or terminal, type of line, and location. You will use most of the NETWORK commands for terminal resources.
NETWORK subcommands for terminals only are:
NETWORK ENable
NETWORK DISAble
NETWORK Query ACTive
NETWORK Query FREe
NETWORK Query OFFline
NETWORK Query ALL
NETWORK sub commands for either lines or terminals are:
NETWORK Query resid
NETWORK VARY ONline
NETWORK VARY OFFline
3-102
VM/SP Operator's Guide
-----0-::1
Using the NETWORK Command
For 37xx Functions
The NETWORK command does the following:
•
•
•
Causes 3704/3705 dump operations
Starts 3704/3705 load operations
Enables or disables terminal resources
Varies resources online or offline
Stops all 37xx operations
Queries 37xx resource status.
When operating with a 270x or an Emulation Program there is only one address for
each logon device, and that is the physical sub channel address for the telecommunications line. For a simple leased line configuration, there is one resource ID for
each line and one resource ID for each terminal (one terminal per line). These
resource IDs alternate in numeric value.
For 3270 Devices
Use the NETWORK command to:
Control remote 3270 resources
Vary the polling delay on the binary synchronous lines that are connected to
these resources.
You can use the NETWORK command if the remote 3270 resources are connected
to a 2701, 2703, or to a 37xx in Emulation Program (EP) mode.
The operands that specifically affect remote 3270s are:
•
•
•
SHUTDOWN (Class A)
POLLdlay (Class A,B)
ENable (Class A,B)
VARY (Class A,B)
DISAble (Class A,B)
Query (Class A,B)
Referring to a Remote 3270 Resource: You will specify a four-digit
hexadecimal number to refer to a remote 3270 resource (display station, printer, or
control unit (lines) in the 3270 remote system). For example, in the number 0001,
the two low order characters, 01, identify the resource for a display station or
printer. This value indicates the relative position of the TERMINAL macro from
the first TERMINAL statement within the CLUSTER/TERMINAL grouping.
This value begins at 01. The two high-order characters, 00, are a relative line code
associated with a physical Binary Synchronous Communication (BSC) line. This
value indicates the relative position of the CLUSTER macro from the first
CLUSTER statement within the DMKRIO module. This value begins at 00.
Section 3. CP Commands
3-103
NETWORK Formats
Class A NETWORK Format
NETWORK
ATTach resid [To] userid [As] euu
DETach resid [From] userid
J
DISAble fALL
l:esid [resid•.• ]
-
DISPlay raddr hex/oel
hex/oe2] [ END
{ .} [bytecountl
END
...
DUMP raddr
~MMEDJ
J-
OFF
AUTO
1
ENable [ALL
~esid [resid... ]J
LOAD raddr nepname
POLLdlay nnnn [ALL
J
raddr
Query [ACTive
OFFline
]
FREe
ALL
resid [resid••• ]
SHUTDOWN [raddr]
ALL
I
VARY (ONline resid [resid.•• ]
OFFline
/
3-104
VM/SP Operator's Guide
Class B NETWORK Format
NETWORK
ATTach resid [To) userid [As) vaddr
DETach resid [From) userid
DISAble [ALL
resid [resid...
1
U
-
DISPlay raddr hex/oel ( ;
-
I [::~Oc2]
{. } [bytecountl
END
...
raddr~MMEDJ
DUMP
J-
OFF
AUTO
1
ENable fALL
tesid [resid..• ~
LOAD raddr ncpname
POLLdlay nnnn [ALL]
raddr
r-
-
Query ACTive
OFFline
FREe
ALL
resid Iresid•.• )
...
I
-
VARY (ONline resid [resid... )
OFFline
NETWORK Operands (Class A, B)
ATTach resid ITO) userid [AS) vaddr
attaches the remote printer device (resid) to a virtual machine (vaddr).
Before you issue this command, be sure the device is powered on and
enabled; otherwise, the advanced features of the device will not be operational.
Section 3. CP Commands
3-105
resid
identifies the device. resid is the four-digit resource ID as specified in
the real I/O configuration module (DMKRIO). For a complete
description of resid, read about the four-digit hexadecimal number
under "Referring to a Remote 3270 Resource" on page 3-103.
userid
identifies the user to whom you want to attach the device.
vaddr
is the virtual device address of the virtual machine where you want to
attach the printer. A network resource is represented as a graphic
device in the user's virtual machine configuration. Therefore, you
should specify an address on a SHARED virtual control unit (VCU).
When you issue NETWORK ATTACH, be aware that VM/SP does not
support a mixture of SHARED and NONSHARED device types on the same
virtual control unit. If you do not specify vaddr correctly, the system does
not attach the device. Refer to Appendix B in the VM / SP Planning Guide
and Reference, SC19-6201 for a complete list of the virtual device characteristics.
Note: NETWORK treats the remote printer as a shared device in the virtual
machine configuration. When you are attaching a remote printer to a virtual
machine, the virtual machine may only access the 3270 Information Display
printer with an SIO instruction.
DETach resid [From] userid
detaches a remote printer device that you request from the virtual machine.
resid
identifies the device. resid is the four-digit resource ID as specified in
the real I/O configuration module (DMKRIO). For a complete
description of resid, read about the four-digit hexadecimal number
under "Referring to a Remote 3270 Resource" on page 3-103.
userid
identifies the user from whom you want to detach the device.
DISAble [ALL
]
resid [resid•.. ]
disables 37xx resources (terminals only) and remote 3270 resources.
If you specify a resource that the system is currently using, CP does not
immediately disable it. CP disables the resource when it becomes free
(usually after the user logs off).
disables all 37xx terminals.
3 -1 06
VM/SP Operator's Guide
- -.----------..- - - -
resid
disables resources that you select. resid is the four-digit resource ID of
each terminal resource that you want to disable. For a complete
description of resid, read about the four-digit hexadecimal number
under "Referring to a Remote 3270 Resource" on page 3-103.
DISPlay raddr hexlocl [
{~ } [:~~C2J
1
{ . } [bytecounil
END
J
displays the contents of 37xx storage in fullwords. This operand does not
provide EBCDIC translation.
raddr
is the real address of the 37xx whose storage you want CP to display.
hexlocl
specifies the hexadecimal address of the start of the display. This
operand is required. Specify a colon (:), dash (-) or period (.) to
display more than one fullword.
hexloc2
specifies the hexadecimal location of the end of the display.
DUMP raddr [IMMEDJ
OFF
AUTO
dumps the contents of 370x storage for EP 370x control programs.
raddr
is the real address of the 370x that you want CP to dump.
IMMED
dumps the 370x immediately. See "Network Dump Operations" on
page 4-41 for additional information.
OFF
automatically dumps the 370x if the 370x control program abnormally
ends.
AUTO
automatically dumps and reloads the 370x if the 370x control program
abnormally ends (ABENDS).
ENable [ALL
]
resid [resid•.. }
activates 37xx resources (terminals only) and remote 3270 resources for
VM/SP to use. The resource that you specify must be a terminal device.
The NETWORK ENABLE command:
1.
Ensures that the associated line resource is activated
Section 3. CP Commands
3-107
2.
Enables the terminal device or, in the case of remote 3270s, formats the
screen.
Note: If you enable a 3275 remote display unit, and the unit is equipped
with a printer to handle the display's hard copy output (via PFnn COPY),
you must also enable the printer.
enables all 370X resources.
resid
enables resources that you select. resid is the four-digit resource ID of
each terminal resource that you want CP to enable. For a complete
description of resid, read about the four-digit hexadecimal number
under "Referring to a Remote 3270 Resource" on page 3-103.
LOAD raddr ncpname
loads an EP 3704/3705 control program.
raddr
is the real address of the 3704/3705 that you want CP to load.
ncpname
is the name of the 3704/3705 control program image that you want
CP to load into the specified 3704/3705. This name is previously
defined by a NAMENCP macro and saved on a CP volume.
POLLdlay nnnn [ALL
J
raddr
changes the duration of the polling delay interval for the binary synchronous
line to nnnn.
The polling delay interval, defined at system generation, is a half second.
The polling delay interval minimizes unproductive polling and processor
meter time. In general, if no data or other communications is being received
from the stations on the binary synchronous line, the system starts the
polling delay interval and gives control to the dispatcher.
Note: The polling delay interval is the amount of time from when a binary
synchronous line receives a negative response from a general polling
sequence until the polling delay interval expires, or CP sends a message to
the station on the binary synchronous line.
nnnn
is the polling delay interval. nnnn is a decimal number in tenths of a
second (not to exceed 9999).
raddr
is the address of the binary synchronous line.
I
3-108
VM/SP Operator's Guide
r
sets the polling delay interval for all the 3270 remote lines.
Query [ACTive
OFFline
1
FREe
ALL
resid [resid•.. ]
displays the status of 37xx resources (lines or terminals) and remote 3270
resources. You can use NETWORK QUERY to display the status of a line
resource, but only when you use the NETWORK QUERY resource
command format.
ACTive
displays the resources (terminals, remote 3270 display and printer
stations) that are being used by VM/SP users.
OFFline
displays the resources that are not available to VM/SP users.
FREe
displays the resources that are not offline and also not currently in use.
ALL
displays the active, offline, or free status of each resource attached to
a1137xx and 3270 remote control units on the VM/SP system.
resid
displays only the resources that you specify with the four-digit
hexadecimal identifier. For a complete description of resid, read about
the four-digit hexadecimal number under "Referring to a Remote
3270 Resource" on page 3-103.
SHUTDOWN [raddr]
ALL
controls 3270 remote device applications.
Use NETWORK SHUTDOWN to detach a binary synchronous line that is
coupled to a 3270 remote console unit cluster or work station. CP places
either of the following in a disconnected state:
A virtual machine dependent on the binary synchronous line(s)
A virtual machine dependent on the 37xx for which you issued
NETWORK SHUTDOWN.
raddr
identifies the 3270 remote binary synchronous line on which you want
all telecommunications to stop.
~!
Section 3. CP Commands
3-109
stops telecommunications on all binary synchronous lines in the 3270
remote system configuration.
CP does not attempt to preserve line status or messages in the 37xx.
I
VARY !ONline
OFFline
resid [resid... ]
varies the status of the 37xx resource that you specify. You can use VARY
ONLINE and VARY OFFLINE for line resources. However, use these
commands primarily for terminal resources, because the state of the line
changes automatically if you enable or disable the terminal.
Examine the output from the first stage of the 37xx control program generation to tell which resources are lines and which are terminals.
ONline
places a resource (line or terminal) online.
OFFline
places a resource (line or terminal) offline.
Note: Only the ONLINE and OFFLINE operands are valid for
remote 3270 resources.
resid
identifies the resource. resid is the four-digit resource ID of each terminal resource that you want CP to vary. For a complete description
of resid, read about the four-digit hexadecimal number under
"Referring to a Remote 3270 Resource" on page 3-103.
System Response
NETWORK ATTACH
resid ATTACHED AS cuu
CP successfully attached the resource.
NETWORK DETACH
resid DETACHED
CP successfully detached the resource.
NETWORK DISABLE
COMMAND COMPLETE
3-110
VM/SP Operator's Guide
-.-.-..- - . - . - - - - . - - - - - - - - - . - - -...-------.----------........... -.-.. --.-----=:=1
NETWORK DUMP
CTLR raddr DUMP COMPLETE
CP successfully dumped the 370x "raddr."
NETWORK ENABLE
COMMAND COMPLETE
NETWORK LOAD
CTLR raddr ncpname LOAD COMPLETE
CP successfully loaded the 370x "raddr" with the control program ncpname.
NETWORK QUERY
DEV
DEV
DEV
DEV
resid
resid
resid
resid
LOGON AS userid
DISABLE
ENABLED
OFFLINE
LINE resid ACTIVE
LINE resid OFFLINE
DEV residl ENABLED, DEV resid2 ENABLED, .. .
DEV residl DISABLE, DEV resid2 DISABLE, .. .
DEV residl OFFLINE, DEV resid2 OFFLINE, .. .
ACTIVE resid NOT FOUND
FREE resid NOT FOUND
OFFLINE resid NOT FOUND
resid NOT FOUND
Note: For any of the" ..... NOT FOUND" responses, a problem exists with the
Network Control Program and/or the 37xx.
ACTIVE
indicates that the line resource is online and has been activated by the
system. Users mayor may not be using the terminals on the line.
DISABLE
indicates that the resource is online, but VM/SP users cannot access it.
ENABLED
indicates that VM/SP users can access the resource.
LOGON
indicates that "userid" is using the resource as a virtual machine operator
console.
OFFLINE
indicates that the resource is inactive and unavailable for use.
Section 3. CP Commands
3 -111
~--------------------------------------------------------------------------~
resid
identifies the real resource.
userid
identifies the user.
NETWORK VARY
COMMAND COMPLETE
3-112
VM/SP Operator's Guide
ORDER
Privilege Class: D
When to Use ORDER
Use the ORDER command to reorder closed spool files of a device type that you
specify. To determine the filename, filetype, originating userid, and other attributes of all of your files, use the QUERY command.
ORDER Format and Operands
ORDer
[~YSTE'
userid
Reader I RDR}{CLaSS cl CLass c2...
}
Printer I PRT spoo/idl spoo/id2
{
PUnch I PCH FORMform1 FORMform2 .•.
userid
identifies the user whose spool files the system is to purge.
SYSTEM
orders all spool files in the system.
*
orders your own spool files.
Reader
RDR
orders the reader files.
Printer
PRT
orders the printer files.
PUnch
PCH
orders the punch files.
CLass c
orders the files by their class. c is a one-character field (0 through 9 or A
through Z) that specifies classes for input and output spool files. This lets
you place files in the order of classes that you specify with CLass c 1, CLass
c2, etc.
Section 3. CP Commands
3 -113
spoolid
identifies the spool files and lets you order the files for processing.
FORMforml FORMform2
orders the files by their form. forml and form2 are one to eight character
form numbers. The form number is the user form if you specify the ~ option
or let it default. The form number is the operator form if you specify the
userid/SYSTEM option.
You can do sequencing with the ORDER command using a combination of
CLASS c, FORM, and spoo/id specifications.
Example:
ORDER PRINTER FORM XYZ CLASS A 1963 CLASS C
queues the printer files in the following order: all files with the
form XYZ, all Class A files, the file with the spoolid 1963, all
Class C files.
System Response
{~~nnl FILES ORDERED
is the response to the user who issues the ORDER command.
3-114
VM/SP Operator's Guide
---.------..-.-.--.-----....- - - - - -...--.-.---.------.-.-.------.-=:1
PURGE
Privilege Class: D
When to Use PURGE
Use the PURGE command to remove closed spool files from the system either:
•
Before the spooling device prints or punches the files, or
•
Before the user reads the files.
You can purge any file in'the system, regardless of its status, as long as it has not
been selected for processing.
PURGE Format and Operands
PURge
Reader I RDR
~userid
Printer I PRT
~YSTEJ PUnch I pcn
J
ALL
~
ALL
CLass cl CLass c2 .•.
spoolidl spoolid2 ..•
l
FORM/orml FORM/orm2
userid
identifies the user whose spool files the system is to purge.
SYSTEM
purges all files on the system, regardless of the userid.
*
lets you manipulate your spool files.
Reader
RDR
purges reader files.
Printer
PRT
purges printer files.
Section 3. CP Commands
3 -115
c====. ____________________________________________________________________________
~
PUnch
PCH
purges punch files.
ALL
purges all the files. The system ignores any operands that follow ALL. If
you specify ALL after Reader, Printer, or PUnch, the system only purges the
files for that device type.
CLass c
purges the files for the class that you specify and for the device type that you
specify. c is a one-character field (0 through 9 or A through Z) that specifies
classes for input and output spool files. This lets you purge files in the order
of classes that you specify.
spoo/id
purges the files of the spool identification that you specify and for the device
type that you specify.
FORMform1 FORMform2
tells the system to purge all files that have the form numbers that you
specify. form1 and form2 are one to eight characters. The form number is
the user form if you specify the ~ option or let it default. The form number
is the operator form if you specify the userid/SYSTEM option.
You can use PURGE using a combination of CLASS c, FORM and spoo/id
specifications.
Example:
PURGE PRINTER CLASS A FORM XYZ CLASS D 619
purges the files in the following order: all class A files, all files
with form XYZ, all class D files, and the file with spoolid 619.
System Response
{~~nnl FILES PURGED
shows the number of files that the system purged.
Note: You will not get this response if you have issued the SET IMSG OFF
command. Therefore, if you are monitoring the RSCS virtual machine for your
installation, do not specify this option. You need to see informational messages to
accurately monitor the RSCS system activity.
3-116
VM/SP Operator's Guide
QUERY
Privilege Classes: A, B, C, D, E, or F
When to Use QUERY
Use the QUERY command to request the following types of information:
•
System status
Machine configuration
Paging and scheduling information.
(For 3704 or 3705 Communication Controllers and remote 3270 resources, see the
Class A and B NETWORK command.)
Section 3. CP Commands
3-11 7
L
QUERY Formats
Class A QUERY Formats
Query
AFFinity [userid]
CPAssist 1
CPLEVEL'
JOurnal 2
LOGmsg
Names
PAGing
PRIORity userid
PROCessr
QDROP
S370E
SASsist 1
-< SPMODE
SRM APAGes
DSPSIice
IB
MAXDrum
MAXWss
MHFULL
PB
PCI
PGMStat
PGMTIim
userid
" Users [userid]
...
I'
>
/
.,I
lUse both QUERY CPASSIST and QUERY SASSIST to determine the
current status of the expanded virtual machine assist portion
of the Extended Control-Program Support:VM/370.
2The JOURNAL operand is valid only if STQUERY=YES is specified in the
SYSJRL macro instruction in DMKSYS.
3-118
VM/SP Operator's Guide
-
[
------------]
Class B QUERY Format
,.
J
Query
DAsd [Sysvirt
Virtual
GRaf
LINES
UR
TApes
ALL
""
[ACTive]
ATTach
[PATHS]
FREe
OFFline
ALL
CPLEVEL
DAsd volid
DUMP
Lnnn
LOGmsg
MITime
Names
PROCessr
raddr1 (-raddr2)
STATUS raddr
STORage
SYStem raddr
TDsk
userid
Users [userid]
i""<
'-
~
~
Class C QUERY Format
Query
CPLEVEL
LOGmsg
Names
PROCessr
userid
Users [userid]
Section 3. CP Commands
3 -119
Class 0 QUERY Format
Query
CPLEVEL
Files [CLass c[[FORMform] ruserid 21
L*
..
Hold
J
~HOld 1
]
NOHold
SYSllold
USERhold
l
Printer [CLass c[[FORM form] ruserid 2]~HOld
PUnch
L*
NOHold
SYShold
Reader
USERhold
spoo/id
LOGmsg
Names
userid
Users [userid]
~null
ALL
TBL
lSpecify the options in any order.
2Using a 1-4 digit, all numeric userid will cause unpredictable results
for the QUERY command, which also has a 1-4 digit all numeric spoolid
parameter.
3-120
VM/SP Operator's Guide
J
Class E QUERY Format
,..
Query
AFFinity [userid]
CPAssist 1
CPLEVEL
JOurnal 2
LOGmsg
Names
PAGing
PRIORity userid
PROCessr
QDROP
SASsist 1
I~ SRM APAGes
DSPSlice
IB
MAXDrum
MAXWss
MHFULL
PB
PCI
PGMStat
PGMTlim
""-
)-
userid
,-Users [userid]
lUse both QUERY CPASSIST and QUERY SASSIST to determine the
current status of the expanded virtual machine assist portion of the
Extended Control-Program Support:VM/370.
2The JOURNAL operand is valid only if STQUERY=YES is specified in the
SYSJRL macro instruction in DMKSYS.
Class F QUERY Format
Query
CPLEVEL
LOGmsg
Names
userid
Users [userid]
Section 3. CP Commands
3 -121
QUERY Operands (Class A,8,C,O,E,F)
AFFinity [userid]
(For attached processor applications only) displays the affinity setting of a
virtual machine that you specify or all virtual machines that are logged onto
the system.
Set the affinity in the VM/SP directory OPTION statement or with the CP
SET command. The affinity setting of a virtual machine indicates the
processor (main or attached) that the system will use to run the programs of
a virtual machine. If you do not set an affinity for a virtual machine, the
system dispatches the virtual machine for execution on the first available
processor.
Response:
USER
userid
PROCESSOR AFFINITY
affinity
userid
identifies the virtual machine user.
affinity
is the address of the processor (main or attached) that will execute
the virtual machine instructions. a f fin i t Ycan be any
hexadecimal value from 0 to 3P.
If NONE is in the response, the system did not set an affinity, and
the virtual machine is dispatched for execution on the first available processor.
Note: In System/370 Model 168 installations, you can reverse
the role of the attached processor and main processor with a
console switch setting.
CPAssist
displays the current status of the CP assist function for the VM/SP system.
Response:
CPASSIST ION
OFF
I
ON
shows that the CP Assist portion of the Extended ControlProgram Support:VM/370 is active.
3 -122
VM/SP Operator's Guide
OFF
shows that the CP Assist portion of the Extended ControlProgram Support:VM/370 is inactive or not available on the
System/370 processor.
For Query CPAssi,st (attached processor applications):
I
CPASSIST ION) PROC nn, ION
OFF
OFF
PROC nn
ON
shows that the CP assist portion of the ECPS:VM/370 is active
for the processor, PROC nn.
OFF
shows that the CP assist portion of the ECPS:VM/370 is inactive
or not available for· the processor, PROC nn.
nn
is the processor address assigned to the main processor or the
attached processor by the installation when the system hardware is
installed. nn can be any hexadecimal value from 00 to 3F.
Note: To determine the active status of the Expanded Virtual Machine
Assist portion of ECPS you need a positive response to QUERY CPASSIST
and QUERY SASSIST.
CPLEVEL
displays the current level of the system, most recent generation, and the most
recent IPL date.
Response:
VM/SP RELEASE xx, SERVICE LEVEL xxxx
GENERATED AT date time zone
IPL AT date time zone
DAsd
SYSvirt ]
[ Virtual
ACT~Vej [PATHS]
[
ATTach
FREe
OFFline
GRaf
LINES
ALL
UR
TApes
ALL
ACTive
displays the status of the active devices in the group that you specify.
An active device is one that a user or the system is using. Active
devices do not include devices that are free or offline.
Section 3. CP Commands
3-123
ALL
as the first operand, displays all devices and the size of real storage.
Response:
The response is the same as if you issued the following commands:
QUERY STORAGE
QUERYUR
QUERY LINES
QUERYDASD
QUERY TAPES
QUERYGRAF
ALL
as the second operand, displays the status of all devices within the
group that you specify.
Response
The response is the same as if you issued the following commands:
QUERY type
QUERY type
QUERY type
QUERY type
ACTIVE
ATTACH
FREE
OFFLINE
ATIach
displays all the devices that are dedicated to any user on the system.
An attached device is also an active device.
DAsd
displays the real addresses of disk or drum devices.
Response:
For Query DAsd:
DASD raddr ATTACH TO userid vaddr
PATH raddr\OFFLINE
14
ONLINE [PROC nn[,PROC nn]]
indicates that the real device specified by raddr is attached to a
user's (user i d) virtual machine at virtual address vaddr.
4
3-124
Additional lines (PATH ... ) will only appear if you specify the PATHS operand and the
device has alternate paths. Note that in systems generated for MP mode, all devices
have at least a second logical channel path.
VM/SP Operator's Guide
DASD raddr CP SYSTEM volid nnn
PATH raddr\OFFLINE
14
ONLINE [PROC nn[,PROC nn]]
,
indicates that the real device specified by raddr is allocated to
the system for use as user's minidisks. nnn is the number of
active user's minidisks on the physical disk. vo lid is the
volume serial number of the real disk.
DASD raddr CP OWNED volid nnn
PATH raddr\OFFLINE
14
ONLINE [PROC nn[,PROC nn]]
indicates that the real device specified by raddr is used by the
system for paging and spooling activity. nnn is the number of
active user's minidisks and T-disks (if any) on the physical disk.
vo lid is the volume serial number of the real disk.
FREe
displays all the devices that neither a user nor the system is currently
using. Free devices do not include offline devices.
Response
The system produces a response for each device that is not active or
offline in the following format:
type raddr FREE
type raddr [PROC nl[,PROC n2]] DRAINED
is the response for unit record devices.
5
Note: This response implies that no spool files are queued for
this device.
ENABLED
I
type raddr [PROC nl[,PROC n2]] DISABLED
is the response for communication devices.
type raddr \FREE
vol id
1
5
I
is the response for DASD devices with mounted volumes.
GRaf
displays the real address of each locally and remotely attached display
device and graphics devices dedicated to a virtual machine.
For VTAM Service Machine (VSM) supported terminals, GRaf displays:
..-
-
5
The system will display the processor address(es) if the system is generated for MP
mode.
Section 3. -CP Commands
3 -125
QUErRV
The userid of the VSM that controls the terminals from which
users have DIALed a guest virtual machine
The userid and luname (logical unit name) associated with each
terminal dialed to the VSM
The userid of each terminal's owner.
Response:
GRAF raddr [PROC nl[,PROC n2]] LOGON AS userid
6
shows that userid is currently logged on the terminal located at
real address, raddr.
GRAF raddr [PROC nl[,PROC n2]] ATTACH TO userid vaddr 6
shows that the display device at real address, raddr, is attached
to the virtual machine of user i d at virtual address, vaddr.
GRAF raddr [PROC nl[,PROC n2]] DIALED TO userid vaddr 6
shows that the display device at real address, raddr, is dialed to
the virtual machine of user i d at virtual address, vaddr.
PRT raddr ATTACHED TO userid
shows that the remote printer device at real address raddr is
dedicated to the virtual machine of user i d.
GRAF
I
raddrloFFLINE
ENABLED
DISABLED
DISABLE PND
shows that the remote graphics device at real address, raddr, is
either offline, enabled, disabled or has a disable pending.
vvvvvvvv CONTROLS THE FOLLOWING LUiS
GRAF luname LOGON AS userid
GRAF luname DIALED TO userid
shows that the VTAM service machine (VSM) whose userid is
vvvvvvvv is controlling the graphics devices identified by the
SNA logical unit names (1 uname). This response also identifies
the use rid of each user logged on through one of these devices,
or the userid of the virtual machine that each device is DIALed
to.
LINES
displays the real address of each communication line.
raddr, and the
system is generated for MP mode, the system will add the processor addresses to the
response line as shown.
If you specified the PATHS operand or if the command was QUERY
3 -126
VM/SP Operator's Guide
/'
r:::
Response:
LINEl raddr [PROC nl[,PROC n2]] LOGON AS userid 7
( CONS
shows that user i d is currently logged on at the terminal located
at real address, raddr.
LINE
raddr [PROC nl[,PROC n2]] ATTACH TO userid vaddr
7
shows that the communication line at raddr is attached to the
virtual machine represented by use rid at virtual address,
vaddr.
LINE
raddr [PROC nl[,PROC n2]] DIALED TO userid vaddr
7
shows that the communication line at raddr is dialed to the
virtual machine represented by use rid at virtual address,
vaddr.
OFFline
displays the devices in offline status in the group that you specify. An
offline device is not available for access by any user or the system.
Response
This command produces a response for each offline device in the following format:
type raddr OFFLINE [PATH raddr, ... ]
8
Multiple responses are displayed in the following format:
type raddr OFFLINE,
Note: In the preceding responses the term type refers to one or more
of the following device types:
7
Type
Meaning
DASD
TAPE
LINE
RDR
PRT
Direct access device
Magnetic tape units
Communication line
Card reader
Line printer
If you specified the PATHS operand or used the QUERY raddr command, and the
system is generated for MP mode, the system will add the processor addresses to the
response line as indicated.
The second line will only appear if the device is an alternate path device and if you
specify the PATHs operand or the QUERY raddr command.
Section 3. CP Commands
3 -127
CQ.I!JIE~V
PUN
GRAF
CONS
DEV
Card punch
Graphics device
Console
Any other device
PATHS
displays the path status for all alternate path devices. In multiprocessor mode, the system will also display the processor address for
each device.
Sysvirt
displays the addresses and status of 3330V devices that have the
SYSVIRT feature.
TApes
displays the real addresses of magnetic tape units.
Response:
TAPE raddr CP SYSTEM
PATH raddr (OFFLINE
19
ONLINE [PROC nl[,PROC n2]]
indicates that the real tape device (raddr) is attached to CP for
its exclusive use.
TAPE raddr ATTACH TO userid vaddr
PATH raddr (OFFLINE
19
ONLINE [PROC nl[,PROC n2]]
The system displays this response if the real tape device (raddr)
is attached to a user's (user i d) virtual machine at virtual
address vaddr.
TAPE raddr status
the system displays this response for conditions that test the
status of the tape device.
raddr
is the real device address (cuu).
status
is any available status for the tape device. The status conditions can include:
INT REO
ATTACH
FREE
OFFLINE
9
3-128
Additional lines (PATH ... ) will only appear if you specify the PATHS operand and the
device has alternate paths. Note that in systems generated for MP mode, all devices
have at least a second logical channel path.
VM/SP Operator's Guide
/
----=:==I
UR
displays the real addresses of unit record devices (card reader, card
punches, printers).
Response:
The response that you get depends on how the UR device is used.
The UR device can be owned by CP or attached to a virtual machine.
UR Devices Attached to a Virtual Machine
{ :~~}
raddr [PROC nn] ATTACH TO userid vaddr
RDR
raddr
is the real device address (cuu).
userid
is the name of the spool file owner.
vaddr
is the user's virtual device address.
UR Devices that CP Owns
PRT raddr [PROC n 1[,PROC n2]] 10
raddr status CLASS a ... FORM form
raddr xsetup xxsep xxx3800 xxxxxFCB
is the response for a 3211-type printer.
PRT raddr [PROC nl[,PROC n2]] 10
raddr status CLASS a ... FORM form
raddr xsetup xxsep xxx3800
raddr FLASH ffff IMAGE imagelib CHARS char FCB nn
xhold
is the response for a 3800 printer.
PRT raddr [PROC nl [,PROC n2]] 10
raddr status CLASS a ... FORM form
raddr xsetup xxsep xxx3800
is the response for all other printers.
PUN raddr [PROC nl[,PROC n2]] 10
raddr status CLASS a ... FORM form
raddr xsetup xxsep
is the response for a punch device.
10
If you specified the PATHS operand or used the QUERY raddr command, and the
system is generated for MP mode, the system will add the processor address(es) to the
response line as indicated.
Section 3. CP Commands
3-129
~-------------------------------------------------------------------------------:::=J
RDR raddr [PROC nl [,PROC n2]]
10
status
is the response for a reader device.
The operands are:
a ...
specifies the classes that the output device services. An
output class can service up to four classes. Do not put
blanks or commas between classes.
char
is the name of the character arrangement table used for the
separator page.
ffff
is the forms overlay sheet that the device contains.
form
is the operator form that the output device is processing.
image 1 i b
is the named system image library that the system uses for
tables.
nn
is the value of the forms control buffer that the system uses
for the file.
raddr
is the real device address (cuu).
status
is the status of the device and can be one of the following:
DRAINED
the device is not currently available for processing.
Issue a START command to activate the device.
INT REQ
the device is not ready. You must manually make it
ready before spool processing continues.
MOUNT REQ
the output device is waiting for" a forms mount.
PRINTING
the printer is currently printing a file.
PUNCHING
the punch is currently punching a file.
3-130
VM/SP Operator's Guide
SETUP REQ
the printer is waiting for a forms mount and alignment
(setup).
START REQ
tells you to start the punch if you gave the START
command with the NOSep option. This lets you
remove the last punch deck before starting to punch the
new job.
STARTED
is the response to the START command.
WAITING
the device is available for spooling activity.
xhold
can be HOLD or PURGE.
HOLD
indicates a save status for spool files that have load
errors.
PURGE
indicates a deleted status for spool files that have load
errors.
xsetup
is one of the following:
MANUAL
the system started the device with the MANUAL
option.
AUTO
the system started the device with the AUTO option.
SETUP
the system started the device with the SETUP option.
xxsep
is one of the following:
SEP
the system started the device without the NOSEP
option.
NOSEP
the system started the device with the NOSEP option.
xxx3800
is N03800, BEG3800, or ANY3800.
Section 3. CP Commands
3-131
xxxxxFCB
is one of the following:
DEFFCB
the system is using the default FCB that was loaded in
the printer.
FILEFCB
tells you that you can send File FCBs to the printer.
File FCBs are LOAD FCB CCWs imbedded in the
spool file.
CFILEFCB
tells you that you can send File FCBs to the printer.
File FCBs are LOAD FCB CCWs imbedded in the
spool file. After the system has finished printing each
copy of the spool file containing the file FCB, you will
get this message:
DMKRSP239I PRT cuu DRAINED.
FCB-FORMS MISMATCH.
POSSIBLE
You will also get one of the following responses if the printer, punch,
or reader is currently active with a spool file.
PRT} raddr {PRINTING} USERID FILE CLASS RECORDS RECLEFT CPY DIST SEQ
{PUN
PUNCHING userid file a typ norecs noleft nnn code seq
RDR
raddr {READING} USERID FILE CLASS RECORDS
HAS READ userid file a RDR norecs
userid
is the name of the spool file owner.
file
is the spool file spoolid number.
a
is the spool file class.
typ
is the originating device type (PRT, PUN, CON).
norecs
is the total file logical record count.
noleft
is the number of records left for the system to process.
3-132
VM/SP Operator's Guide
nnn
is the number of copies remaining for output, where 001 indicates
the last copy.
code
is the distribution code of the file.
seq
is the sequence number of the file (printed on the separator page).
Virtual
displays the addresses and status of 3330V devices that have the VIRTUAL
feature.
DAsd volid
displays the active or free status of the DASD volume that you specify with
volid.
Response:
The command response is in either the active or free format depending on
the device status.
Note: If you query any real device or group of devices (such as QUERY
DASD), the system will display the following message for all devices in a
not-ready status:
type raddr INT REQ
DUMP
displays at your terminal the type of device and device address. If the device
is DASD, DUMP displays the type of space on which is the abnormal termination dump. Allocate DASD space for dumps as either 'DUMP' or 'TEMP'
space.
Response:
type ~addr DUMP UNIT \CP
I
ALL
[TEMP]
DUMP
The system response shows that the device of type located at raddr is
the system dump unit. If the system dump is allocated on DASD,
DUMP, or TEMP, the message will indicate space allocation.
Files CLass c FORM/orm
{
[US~i1~~g~Old ]}
SYShoid
USERhold
displays the number of spooled input and output files. The system does not
include files that it is currently processing.
CLass c
selects spool files of class c for counting.
Section 3. CP Commands
3 -13 3
~_ _ _ _ _
!!..........e,...'_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
FORM form
selects spool files of the form that you specify for counting. form is
one to eight characters.
The form number is the user form if you specify the
•
* option.
The form number is the operator form if you specify the userid
option, or neither userid nor *.
HOld
selects spool files that are in hold status for counting. These are the
files held by system hold or user hold.
NOHold
selects spool files that are not in hold status for counting. These are
the files not held by system hold or user hold.
SYShold
selects spool files that are in system hold status for counting.
USERhold
selects spool files that are in user hold status for counting.
userid
displays the spool files of the user that you specify. If you don't
specify a userid, CP examines the spool files that all users own.
*
displays only your (or the one who issued QUERY) spool files.
Response:
FILES: INO
nnnn
I
RDR, INO
I
nnnn
PRT, INO
nnnn
I
PUN
displays the total number of spool files in the system, based on the
options you select.
Hold
displays a list of users whose output is being held as a result of issuing the
HOLD command.
Response:
HOLD: INO
nnn
I
RDR, INO
nnn
I
PRT, INO
nnn
I
PUN
displays the total number of files within the system that are in system
hold status.
3-134
VM/SP Operator's Guide
-----------
~------------------------------------
userid -
I
ALLj , ...
RDR
PRT
PUN
shows the type of hold (if any) for any user in the system for which
HOLD is in effect. When you issue QUERY HOLD you may receive
either or both of these responses, depending on the status of your
spooled files.
JOurnal
indicates the status of the logon/link journaling facility as active or inactive.
Response:
JOURNAL
LOGON {ON}
OFF
LINK {ON}
OFF
ON
shows that the relevant journaling function is active.
OFF
shows that the relevant journaling function is inactive.
Lnnn
displays the status of the logical device with address nnn.
Response:
LDEV nnn LOGON AS userid
shows that the logical device with address nnn is the virtual console for
the virtual machine of user i d.
LDEV nnn ATTACHED TO userid vaddr
shows that the logical device with address nnn is attached to the virtual
machine of user i d at virtual address vaddr.
LDEV nnn ENABLED
shows that the system created a logical device nnn, but no one has
entered the CP DIAL or CP LOGON command through it.
Lnnn NOT LOGGED on
shows that the logical device n n n does not exist. The system interpreted Lnnn entered on the QUERY command line as a userid.
LOGmsg
displays the system log messages of the day.
Section 3. CP Commands
3 -135
Response:
*
logmsg text 1 i ne
*
logmsg text 1ine n
logmsg additional text 1i nes
The system displays all lines (both those with and without an asterisk) in the
log message file.
MITime
displays the current time interval settings for missing interruption monitoring.
Response:
DASD=mm:ss
GRAF=mm:ss
TAPE=mm:ss
UR=mm:ss
MISC=mm:ss
shows the current time interval settings for DASD (FBA and direct
access), GRAF (graphics), TAPE (magnetic tape units), and MIse
(miscellaneous) devices.
MITIME OFF
shows that no monitoring for missing interruptions is taking place.
MISSING INTERRUPTION MONITORING NOT AVAILABLE
shows that DMKDID is not in the system.
INVALID OPTION - option
shows that you have specified an invalid parameter on the QUERY
MITIME command.
Names
displays a list of all the users logged on and the real address of the line to
which each is connected. The system displays the DSe, instead of the line
address, of any user that is disconnected.
Response:
userid -
DSC }'
{raddr
user i d - {DSC }, ...
raddr.
VSM userid
userid-LUNAME, userid-LUNAME, ...
lists all logged-on users. If the user is currently connected, the system
displays the real address to which he is connected (raddr). If the user
is not connected, the system displays DSe.
3-136
VM/SP Operator's Guide
1""
OJ!JE~V
If the user is logged on to the system with a SNA virtual console, CP is not
aware of the real address of the LU (logical unit) device. In this case, CP
displays the LUNAME for the user i d instead of the real address. The
system lists the userid of the VTAM Service Machine (VSM) that supports
the SNA console before the console listing. The VSM and LUNAME displays follow all userid and real address listings.
Note: The logical unit is the port through which an end user can:
•
Access the SNA network to communicate with another end user
Access the facilities of the system services control point.
PAGing
displays the current system paging activity. In Basic System Extensions, this
operand displays the current SET PAGING value.
Response:
PAGING SET: nn
nn
is the number the system will use in the working set size estimate
control algorithm.
Printer {[CLass c] [FORM form]
PUnch
Reader
{
ruseridJ
~HOld
L
*
NOHold
]}[nUllJ}
ALL
SYShold
TBL
USERhold
spoo/id
displays the following types of information about printer, punch, or reader
spool files:
Userid of the spool file owner. If you are looking at files for a specific
user (userid option), the userid indicates the originator of the spool file.
Spool file spoo/id number.
Class and originating device type.
Number of logical records in the file.
You can enter Printer as PRT, PUnch as PCR, and Reader as RDR.
You can display more information using the null, ALL, or TBL operand as
follows:
null
displays the following information for the requested device:
•
Number of copies the user specified for the file
File hold status
Form name that the user assigned
Operator form that corresponds to the user form.
Section 3. CP Commands
3-13 7
ALL
displays the following information for all spool files examined:
•
Number of copies the user specified for the file
File hold status
Date and time that the system created the file
Filename and file type of the file (if any)
Distribution code of the file.
TBL
displays the following table information for 3800 spool files:
Flash name
Forms control buffer (FCB)
N arne of the copy modification module
Number of copies to be flashed (flash count)
Where the 3800 LOAD commands are within the spool file
Name of the character arrangement table.
CLass c
selects spool files of class c for displaying.
FORM form
selects spool files of the form that you specify for displaying. form is
one to eight characters.
•
If you specify the
* option, the form number is the user form.
If you specify the userid or specify neither the userid or
* option,
the form number is the operator form.
HOld
selects spool files that are held (either by system hold or user hold) for
displaying.
NOHold
selects spool files that are not held (either by system hold or user hold)
for displaying.
SYShoid
selects spool files that are in system hold for displaying.
USERhold
displays spool files that are in user hold status.
userid
displays only the spool files of userid. If you don't specify userid, the
system examines the spool files owned by all users.
*
3-13 8
VM/SP Operator's Guide
displays your (or the person who issued QUERY) spool files.
,/
[ ' - - - - - - - - - - - - - - - - - - - - - -..-----
Response:
If you specify the spoolid operand, the system only lists one file for Query
Reader, Query Printer, or Query PUnch.
•
If you enter Query Reader null, Query Printer null, or Query PUnch null,
the system displays the following information: ownerid, file, class,
records, cpy, hold, userform, and operform.
If you enter Query Reader ALL, Query Printer ALL, or Query PUnch
ALL, the system displays the following information: ownerid, file, class,
records, cpy, hold, date, time, name, type, and dist.
•
OWNERID
userid
11
FILE
file
CLASS
a typ
If you enter Query Reader TBL, Query Printer TBL, or Query PUnch
TBL the system displays the following information: ownerid, file, class,
records, flash, feb, mdfy, flshc, load, and chars.
RECORDS
norecs
CPY
HOLD
.I.
"nnn
stat
CPY
HOLD
.I.
"nnn
FLASH
ovly
stat
FCB
plpi
USERFORM OPERFORM
user form
DATE
mm/dd
MDFY
cmod
operfor".l
TIME
hh:mm:ss
FLSHC
ccc
NAME
name
LOAD
{~~G}
ANY
TYPE
type
DIST
code
CHARS
ctab ....
userid
identifies the user who owns the file.
fi le
is a unique, system-assigned number that VM/SP uses to identify the
file.
a
is the spool file class.
typ
is the originating device type (PRT, PUN, CON, or RDR).
norecs
is the number of logical records in the file .
.I.
"nnn
is the number of copies for the file that you specify. (Has no effeet
for reader files.)
11
The system alters the OWNERID heading in the title line for the spool file data to
ORIGINID when you use the userid operand. In that event, ORIGINID represents the
originator of the file.
Section 3. CP Commands
3 -13 9
QlJJrE~V
stat
is the file hold status and can be any of the following:
NONE - no hold
USER - user hold
SYS - system hold
USYS - user and system hold.
user form
is the 1-8 character form name that the user assigns.
operform
is the 1-8 character operator form that corresponds to the user form.
mm/dd
is the date that the system creates the file in month/day.
hh:mm:ss
is the actual time when the system creates the file in
hours: minutes: seconds.
name
is the filename assigned to the file (if any). If the file has a
24-character data set name (dsname), the system only displays 18
characters. These characters extend from the name field through the
type field.
type
is the file type assigned to the file (if any).
code
is the distribution code of the file.
ovly
is the 1-4 character name of the flash overlay sheet loaded into the
3800 printer.
plpi
is the print lines/inch value that the system will use for the separator
page.
cmod
is the 1-4 character name of the copy modification module that the
system will use to print the file.
ccc
is the flash count for the file.
3 ~ 140
VM/SP Operator's Guide
-----------..--
~~G
{ ANY
}
is where the 3800 LOAD commands are positioned within the spool
file:
NO - not at all
BEG - at the beginning
ANY - interspersed throughout the spool file.
ctab
is the 1-4 character name of the character arrangement table that the
system will use for the separator page.
PRIORity userid
displays the current priority of userid. The VM/SP directory establishes this
priority. However, you can override it with a SET PRIORITY nn command.
Respome:
userid PRIORITY
nn
nn
is the the assigned priority of the user that you specify. The lower
the value, the higher the priority.
PROCessr
displays the online processors in the system.
Respome:
PROCESSOR nn ONLINE, [PROCESSOR nn ONLINE, [SYSTEM IN AP/MP
MODE ]]
is a display of the processors that are online. In an attached processor
environment, the first processor is the main processor. In an MP
system, the first online processor that the response specifies is the IPL
processor.
QDROP
displays the userids that have QDROP set off. See the class A SET
command for a description of the QDROP performance option.
Respome:
QDROP SET OFF FOR: useridl [+(USERS)], userid2 [+(USERS)],
The values, user i d 1 and user i d2, are userids that have the QDROP
OFF option set for them. You may get more than one line of output.
+ (USERS) shows that the system is extending the option to the virtual
machines communicating with the displayed userid.
QDROP SET OFF FOR: NO USERS
This is the response if no userids have QDROP set off.
Section 3. CP Commands
3 -141
QUERY
'fP,
raddrl [-raddr2]
displays the device status at the address you specify. If you specify a range
of addresses, the system displays the status of all devices in the range. If
raddrl or any address in the range raddrl-raddr2 is the address of a 3800
printer, the system displays the table information unique to the 3800, in
addition to the current status of the device.
Response:
This response depends on the type of device located at the raddr. See the
QUERY DASD, TAPES, UR, GRAF, and LINES responses.
When you issue QUERY raddr and raddr is a 37xx, the response is in the
form:
CTLR raddr DEV xxoo PGM ={nnnnnnnn} DUMP ={OFF }
EP NAME
AUTO
CTLR
is the keyword for the Communications Controller.
raddr
is the real address of the controller in the cuu form.
DEV
is the keyword DEVICE.
xxoo
tells which device raddr is relative to the CLUSTER macro.
PGM=
is the program loaded into the 37xx (nnnnnnnn) by VM. This
area will be blank if VM didn't load the 37xx. If VM loads the
37xx, this area will contain the name of the EP that VM loaded.
DUMP=
tells the user if an automatic dump (AUTO) was requested or if
no dump will be performed (OFF) in case of a failure.
S370E
displays the status of MVS/System Extensions support. In attached
processor mode, the system reports the state (enabled or disabled) of
MVS/System Extensions support for each processor.
Response:
S370E {ON}
OFF
ON or OFF
shows whether MVS/System Extensions support is active or inacti~e (not available) on the system.
3 -142
VM!SP Operator's Guide
S370E {ON } PROC nn, {ON } PROC nn
OFF
OFF
ON or OFF
shows whether MVS/System Extensions support is enabled or disabled (not available) on the specified processor, PROC nn.
nn
can be any value from 0-63.
SASsist
displays the following:
The current status of the virtual machine assist for the VM/SP system
The current status of the virtual interval timer assist portion of
ECPS: VM/3 70.
Response:
SASSIST
shows if the virtual machine assist (and the virtual interval timer
assist portion of ECPS:VM/370 - if the processor is equipped this
way) is active or inactive (or not available on the system).
For Query SASsist (attached processor and multiprocessor applications)
SASSIST {ON} PROC nn, {ON} PROC nn
OFF
OFF
ON or OFF
shows if the virtual machine assist (and the virtual interval timer
assist portion of ECPS:VM/370 - if the main processor or the
attached processor is equipped this way) is active or inactive for
the designated processor, PROC nn.
nn
can be any hexadecimal value from 00 to 3F. Your installation
assigns values to the processors when the system hardware is
installed.
Note: If you get a positive response to both the QUERY CPASSIST and
QUERY SASSIST commands, the expanded virtual machine assist portion of
the Extended Control-Program Support:VM/370 is active.
SPMODE
displays the status of the single processor mode of operation (on or off).
~I
Section 3. CP Commands
3 -143
Response:
SPMODE {ON }
OFF
ON or OFF
shows if the single processor mode of operation is on or off.
SRM APAGES
DSPSLICE
IB
MAXDRUM
MAXWSS
MHFULL
PB
PCI
PGMSTAT
PGMTLIM
displays the current status of the system's internal parameters.
APAGES
displays the number of available pageable pages.
Response:
AVAIL. PAGES
nnnn
nnnn
is the number of available paging pages.
DSPSLICE
displays the duration of the dispatching time slice.
Response:
TIME SLICE
nnn MILS
nnn
is the size of the dispatching time slice in milliseconds.
IB
displays the current interactive shift bias value. The system uses this
value for Ql virtual machines that are using less than their allocated
share of the processor.
Response:
INTACT. SHIFT
nnnn
nnnn
is the current interactive shift bias value.
3-144
VM/SP Operator's Guide
----------::::=1
c
MAXDRUM
tells if the maximum drum page allocation limit has been set and what
the limit value is.
Response:
DRUM MAX
nnnn x
nnnn
is the maximum drum page allocation.
x
is D if the system set it dynamically and C if you set it with
a command.
DRUM MAX NOT SET
shows that the system has not set a value for the maximum
drum page allocation.
MAXWSS
tells if the maximum page allocation set estimate has been set and
what the limit value is.
Response
WSS MAX
nnnn x
nnnn
is the maximum working set estimate.
x
is D if the system set it dynamically and C if you set it with
a command.
WSS MAX NOT SET
shows that the system has not set a value for the working set
estimate maximum.
MHFULL
displays either the moveable head page migration limits (nnn %) or
that the function is shut off (OFF).
Response:
MHFULL {nnn}
OFF
nnn
is a value from 001 to 100 that shows the percent of available MH page space to be used before migration takes place
for pages in the preferred paging area.
Section 3. CP Commands
3-145
QUJE~V
c==
__________________________________
OFF
shows that the system has not set a percentage or if set, is
set to OFF.
PH
displays the maximum page bias value. The system calculates the
actual page bias dynamically. It is proportional to the paging contention.
Response:
MAX. P. PRIOR
nnnn%
nnnn%
is the maximum page bias value.
PCI
displays the current PCI flag setting mode for 2305 page requests.
Response:
2305 OR FH PCI FLAG {DRUM}
DISK
DRUM
shows that the PCI flag setting mode is X'OF'.
DISK
shows that the PCI flag setting mode is X'OO'.
PGMSTAT
displays the page migration activity counters.
Response:
CALLS
= nnnnnn, USERS
nnnn
nnnnn, SEGS
nnnnnn, PAGES
CALLS = nnnnnn
is the number of calls to the page migration module.
3 -146
VM/SP Operator's Guide
USERS
= nnnnn
is the number of users that had some page migration
activity.
SEGS
=
PAGES
= nnnn
is the number of pages that the system migrated.
nnnnnn
is the number of segments where there was some page
migration activity.
-----_-===============================-'J
c=::----mr----
PGMTLIM
displays the unused segment elapsed time criteria for the page
migration routine to use.
Response:
TIME LIMIT
nnnn
nnnn
is the unused segment of elapsed time criteria for the page
migration routine to use.
STATUS raddr
displays the status of the DASD that has the specified real address.
Note: When you issue the Query STATUS command, be sure the Missing
Interrupt Handler (MIH) is enabled and monitoring for missing interrupts. If
the MIH is disabled, the system will not detect the missing interrupt condition if it occurs upon executing the Query STATUS command. Refer to the
VM/SP System Programmer's Guide under "Interruption Handling" for
information about the Missing Interrupt Handler.
Response:
DEVICE raddr OPERATIONAL
device is online and ready.
DASD raddr NOT OPERATIONAL
shows one of the following:
•
The DASD at address raddr is powered off.
The control unit that the device is attached to is powered off.
The control unit is logically removed from the channel.
DEVICE raddr OFFLINE
when the system checks the device status and finds that the device is
disabled or offline.
UNABLE TO COMMUNICATE WITH THE DEVICE raddr
reports that the specified device or control unit is busy for a prolonged
period.
DEVICE raddr EQUIPMENT CHECK, SENSE=sense data
the system has done an equipment check and shows the 24 or 32 bytes
of sense data. Although this is an "informational" type of message,
you may need to intervene.
Section 3. CP Commands
3-147
QUERY
raddr CHANNEL ERROR
is sent to you (the class B user) when one of the follqwing occurs:
•
A missing interrupt
A channel error (channel data check, channel control check, or
interface control check).
DEVICE raddr INTERVENTION REQUIRED
when the device is not in a "ready" condition. For example, one of
the following could occur:
The device is not mounted properly.
•
The device is logically removed from an operational control unit.
DASD raddr NOT READY
when the system checks the device status and finds that the device is
not ready due to a previous hardware check that has not been fixed.
STORage
displays the size of real storage.
Response:
STORAGE
= xxxxxK
displays the size of real storage (xxxxx) in multiples of 1024 bytes.
SYStem raddr
displays the following:
•
•
The userid, virtual address, and access mode of virtual disks that reside
on the specified channel for the particular device at raddr
The control unit address raddr of logged-on users.
Response:
You use this command to request the number of user minidisks residing on
the physical disk located at raddr. The system gives the response for each
minidisk in the following format.
userid vaddr mode,
userid
identifies the users who have links to minidisks on raddr.
vaddr
is the virtual address by which the user refers to the minidisk.
3-148
VM/SP Operator's Guide
QUJlE~V
c-·-------··
J
mode
is the type of access the user has:
R/O
•
•
R/W
nnn for the number of cyls/blks of TDSK space defined by
users with the DEFINE command. nnn is also the number of
cylinders of TDSK space defined on a count-key-data device.
nnnnnn for the number of blocks of TDSK space defined on
an FBA device.
TDsk
displays the currently defined temporary disk space (TDSK) from all available system-owned volumes assigned to virtual machine users.
Response:
This command displays all the currently defined user TDSK space from all
available system-owned volumes. The system produces an entry of the following format for each TDSK:
userid vaddr {nnnlnnnnnn}
userid
identifies the user.
vaddr
is the user's virtual device address.
nnn
is the number of cylinders that the user has defined. This operand
is also the number of cylinders allocated for count-key-data
devices.
nnnnnn
is the number of blocks defined for FBA devices.
Note: If no user TDSK space is currently defined, you will get this message:
DASD TDSK NOT FOUND
userid
if the user is logged on, userid displays the userid and device address of the
user's terminal. If the user is not logged on, you will get a message.
Response:
userid - raddr
displays the real address (raddr) to which the specified user is connected.
Section 3. CP Commands
3 -149
~----------------------------------~------------------------------~2~'--------~
userid - luname
when you specify QUERY USERS userid and the userid is logged on
with a SNA virtual console. Since CP is not aware of the real device
address (raddr) of the LU device, it supplies the LUNAME in place
of the real address.
Users
displays the number of logged on users and the number of users dialed to
other virtual machines. If you specify userid and the user is logged on, the
system will display the userid and device address of his or her terminal. If
the user that you sp'ecify is not logged on, you will get a message.
Use the USERS operand if the userid you wish to specify is the same as an
operand (or its minimum truncation) of the QUERY command.
Note: The number of users logged on as shown by the NAMES operand
may differ from the number logged on as shown by the USERS operand.
The number of users in the process of logging on and logging off accounts
for this difference.
Response:
nnn USERS, mmm DIALED, xxx NET
nnn
is the total number of logged on users.
mmm
is the total number of users logically attached with the DIAL
command to virtual machines.
xxx
is the number of users currently logged onto the SNA network.
Note: DIALED means that the line is not available to CP because it is logically attached to a logged-on user and is a part of that user's virtual machine
operation.
3-150
VM/SP Operator's Guide
c-----·-----------------·------------··-------·-·--·-_.
QVM
Privilege Class: A
When to Use QVM
Use this command to change a V =R machine to run in native mode, instead of
under VM/SP. Native mode refers to running an operating system stand alone on
the real machine.
QVM Format and Operands
QVM
userid (NORETURN]
userid
identifies the V =R machine that you want to run in native mode.
When you transfer control of the operating system from under VM/SP to
native mode, and VM/SP is running in uniprocessor mode, you can issue this
operand to give the V =R machine control of the processor. Refer to
Running Guest Operating Systems for more information.
NORETURN
indicates that you do not want to return from native mode to the VM/SP
environment. If you do not specify NORETURN, the system will let you
change from native mode back to VM/SP.
Things You Should Know about the QVM Command
1.
Before you pass control to the V =R guest machine to run in native mode, vary
offline all 3480 tape drives that the guest machine will use. After the Control
Program gains control again, vary back online the tape drives that you want to
use.
2.
When an SCP transfers control to the V =R virtual machine in native mode,
error records for the SCP are in two locations. For details on finding all the
error records that pertain to the SCP, see VM/SP OLTSEP and Error
Recording Guide.
Section 3. CP Commands
3 -151
REPEAT
Privilege Class: D
When to Use REPEAT
The REPEAT command applies to files that the system is currently writing to a real
punch or printer. Use this command to:
1.
Request more copies of an output file than that requested by the file's creator.
2.
Place the current output file in a HOLD status, with or without increasing the
number of copies that you want the system to create.
REPEAT Format and Operands
REPeat
raddr
is the address of the output device (printer or punch) whose current active
spool file you want the system to repeat or hold.
nnn
is the number of extra copies of the current file that you want the printer or
punch to make. If you do not specify nnn or the HOld operand, the number
of extra copies defaults to 1. The maximum value that you can specify for
nnn is 255.
HOld
holds the file that the system is currently printing or punching. When the
system finishes the the last copy, it places the file in system hold status. The
file will remain held until you explicitly release it.
Note: Before the HOLD operand takes effect, the system prints all copies
that a SPOOL command or a previous REPeat command indicated.
3-152
VM/SP Operator's Guide
System Response
PRT} raddr {REPEATED} USERID FILE CLASS RECORDS RECLEFT CPY 'DIST SEQ
FILE HELD userid file a typ norecs noleft nnn dist-seq
{ PUN
code
raddr
is the real address of the device whose current output the system is repeating.
userid
identifies the user who will receive the output.
file
is the unique, spoolid number that the system assigns to refer to the file.
a
is the spool file class.
typ
is the originating device type (PRT, CON, or PUN).
norecs
is the total number of logical records in the file.
noleft
is the number of records that the system has left to process.
nnn
is the number of copies of the file that the system has left to create. 001 is
the last copy.
distcode
is the distribution code of the file.
seq
is the sequence number of the file (printed on the separator page).
Section 3. CP Commands
3 -15 3
SAVlESVS
SAVESYS
Privilege Class: E
When to Use SAVESYS
Use this command to save named systems that your installation has defined in the
system name table (DMKSNT). Invoke SAVESYS immediately after loading the
system that you want to save (while the system to be saved is in a stopped state).
This will cause CP to save the virtual machine storage with register contents, PSW,
and storage keys as they currently exist.
SAVESYS Format and Operands
SAVESYS
systemname
systemname
is a predefined name in the system name table (DMKSNT) that defines the
installation requirements for the named system. The definition indicates:
•
The number of pages to be saved
The DASD volume on which the system is to be saved
•
The shared segments (if any).
For more information about saved systems, refer to the discussion of named
systems in the VM / SP Planning Guide and Reference.
System Response
SYSTEM SAVED
3-154
VM/SP Operator's Guide
/'
~c_------·----------~---------==~----------~-a=~------------------------------------,~,~-------~~
SET
Privilege Classes: A, B, E, or F
When to Use SET
Use the SET command to establish system parameters. The SET command performs functions to control the CP system and virtual machine options.
SET Formats
Class A SET Format
r
AFFinity
SET
userid
CPAssist {ON}
OFF
FAVORed
[?:F]
[PROC [nn] ]
J
userid ~nn
OFF
JOurnal {LOgOn}
[ONJ
LInk
OFF
1
"'<
PRIORity
userid
nn
QDROP
userid
{g~F}
REServe
userid
{nnn}
OFF
~
[USERS]
SASsist
{ON}
IPROC Inn]]
OFF
S370E
{ON}
IIPROC] addr]
OFF
"-
oJ
IThe JOURNAL operand is valid only if STQUERY=YES is specified in the
SYSJRL macro instruction DMKSYS.
Section 3. CP Commands
3-15 5
~------------------------------------------------------------------------~
Class B SET Format
SET
DUMP /AUTOj
raddr
[cp J
ALL
LOGmsg rnn
LNULL
text]
Class E Command Format
,..
SET
PAGing
""I
nn
'"
SRM r APAGES nnnn
DSPSlice nnn
IB n
MAXDrum
/nnnnj
OFF
MAXWss
~
MHFULL
/nnnnl
OFF
/
nnn
OFF
~
>-
j
PB nn
PCI
.... PGMTlim
/DRUMj
DISK
.J
~
3 -15 6
VM/SP Operator's Guide
Class F Command Format
SET
RECord
MODE
Ig~TaddT LIMIT nn BYTE nn BIT n
{~~y} {~:!:~d} [cpui~
SET Operands (Class A,B,E,F)
AFFinity userid
~ON ~
OFF
nn
sets affinity on or off for the userid that you specify (attached processor and
multiprocessor systems only). AFFINITY lets the virtual machine execute
only on the processor that you specify; however, AFFINITY does let CP
code execute for a virtual machine on any processor.
userid
is the user for whom you want the system to set or reset affinity. If
you don't specify a userid, the system sets or resets your affinity (the
user who issued the command).
sets affinity on, using the processor address in the user's directory
entry.
OFF
sets affinity off.
nn
the hexadecimal (OO-3F) processor address of an installed processor of
an attached processor system. If you specify this option, the system
uses this processor address to set affinity on.
1
CPAssist ON ( [PROC [ nn II
OFF \
sets CP assist to the active or inactive state.
If CP assist is active, functions that CP routines normally execute are executed by processor hardware logic. This function made active on VM/SP
initialization, a part of Extended Control-Program Support:VM/370, is
available on most VM/SP-supported processors.
Section 3. CP Commands
3-157
ON
activates CP assist for the processor. If you want the same setting for
main and attached processors, use this option without the PROC nn
option.
This option also controls the expanded virtual machine assist. 12 Use
this option along with SET SASsist ON to enable this function.
OFF
deactivates CP assist for the processor. If you want the same setting
for main and attached processors, use this option without the PROC
nn option.
This option also controls the expanded virtual machine assist. 12 Use
this option or SET SASsist OFF to set this function to the inactive
state.
PROC nn
activates CP assist in 3031 attached processor applications. PROC nn
activates CP assist on the desired main or attached processor. The
value nn is the hexadecimal processor address (00-3F) expressed in
decimal (0-63).
DUMP {AUTO}{ CP }
raddr
ALL
identifies the unit to receive the VM/SP system ABEND dump.
AUTO
specifies that the system dump unit is disk.
raddr
is the address of a real printer, 9-track tape address, or a 3480
address. If the dump unit is a tape drive, the dump data must fit on
one reel because VM/SP does not support dumping to multiple tape
volumes.
ALL
dumps all storage.
dumps only storage that the control program is occupying.
Note: If the dump unit is a tape or a printer, all 3480 tape drives online at
the time of the dump will remain assigned to the system. To unassign the
drives, IPL VM/SP after the dump completes, and vary the devices offline.
I~
12
3 -15 8
The expanded virtual machine assist is another part of ECPS:VM/370. The system
sets this virtual machine assist to the active state when VM/SP is initialized.
VM/SP Operator's Guide
--.-.--.---------.....-.-......... -.-.. . -.. ---..
s~"r
-----.-.-:====.=====~::-=--.-..:=:.:.-~::.~:.=--.:.=::::.:.:~~~~:.J
J
FAVORed userid [nnn
OFF
sets the favored execution performance option.
userid
is the user for whom you specified the option. You can favor more
than one user; however, you can only designate one userid in a single
SET command.
nnn
is a percentage value from 1 to 100. When you specify this option,
VM/SP tries to provide that percentage of processor time to the
virtual machine, if the virtual machine can fully utilize the processor
time.
At regular time intervals, the VM/SP dispatcher checks the processor
time used by the particular virtual machine; if the percentage has been
exceeded, the virtual machine is given his normal priority usage for the
remainder of the interval. If the percentage is low, the virtual machine
is assigned highest priority for the remainder of the interval.
Notes:
1.
If you specify 100% for a virtual machine, the system places the
virtual machine at the top of the dispatch queue and holds it there
until it logs off. Because there can be only one user at the head of
the queue, do not specify 100910 for more than one virtual
machine.
2.
If the total amount of percentage times that you assign to a
number of virtual machines exceeds 100 0/0, the effect it will have
on interactive (Ql) response is unpredictable.
OFF
the system removes the specified virtual machine from favored status.
Note: You can only use this option with percentage xxx for one
virtual machine at any time.
If you don't specify xxx or OFF, the system will give preference to the speci-
fied virtual machine over other users with the same priority anytime the
virtual machine is ready to use processor tIme.
Note: SET FAVORed userid xxx does not designate the userid as always
dispatchable; to do this, issue another SET FAVORed command without a
percentage (xxx) value.
JOurnal
I
LOGON}[ON]
{ LINK
OFF
sets the status of the LOGON and LINK journaling functions.
~/
Section 3. CP Commands
3-15 9
LOGmsg [nn
text]
NULL
nn
specifies the line number of the log message that you want to add or
change. You can supply the text of the log message line on the
command line after this line number.
If you do not supply the text, the system will respond with:
LOGMSG:
Enter the text here. If you do not enter a line, CP will delete log
message line nn. If you do enter text, CP will replace log message line
nn with the new text. If log message line nn does not exist, CP will
add your text as a new log message line after the current last line in
the log message.
text
specifies the contents of the nnth log message line. If you specify text,
CP will not perform a console read.
NULL
deletes the entire existing LOGMSG. If you enter more SET
LOGMSG commands, CP will start to create a new LOGMSG.
Notes:
1. To stop the LOGMSG entry, enter a null line.
2.
The maximum log message length per line is 76 characters, including
spaces.
3.
If you precede the message text with an asterisk, CP will automatically
display the message on the user's terminal as soon as he/she logs on.
MITime
claYS {~;;S}
{OFF
r
lass
{~;;S} ooJ}
alters the current missing interrupt monitoring status in one of the following
ways:
3-160
VM/SP Operator's Guide
•
Changes the missing interrupt monitoring time interval in DMKSYS for
a specific device class (class)
•
Stops missing interrupt monitoring for the device class that you specify
(class OFF)
•
Stops all monitoring of missing interrupts for all device classes (MITIME
OFF).
/
SElf
CP only changes the time intervals for device classes that you specify on the
command line. All other time intervals remain unchanged.
class
refers to one or more of the following:
Class
Meaning
DASD
GRAF
TAPE
UR
MISC
Direct access and FBA devices
Graphics devices (local only)
Magnetic tape devices
Unit record devices
Miscellaneous device
Notes:
1.
Do not specify the same device class more than once on the same
command line.
2.
GRAF refers to devices generated by CLASGRAF, except for
TYP328X and TYP1053 printers.
3.
UR refers to devices generated as CLASURI and CLASURO,
except for TYP3800 and TYP3289E printers
4.
MISC refers to Mass Storage devices and 1053, 328X, 3289E,
and 3800 printers.
mm:ss
is the time interval in minutes and seconds. The maximum value for
mm is 99 and the maximum for ss is 59.
Note: If you set the time interval for a device class below its default
value, you may cause unnecessary missing interrupt handler processing
for devices that are functioning properly. The device class default
time interval values are in the "Interruption Handling" section of the
VM/SP System Programmer's Guide.
OFF
If you issue SET MITime class OFF, the system stops monitoring
missing interrupts for the specified device class for the duration of the
IPL or until you issue another SET MITIME command. SET MITime
OFF stops monitoring missing interrupts for ALL device classes.
MODE
I
I
RETRY} Quiet } [cpuidl
MAIN
Record
sets the error recording mode for soft errors. Soft errors are recoverable
processor storage or instruction errors corrected by hardware or software
. logic.
Section 3. CP Commands
3-161
SElf
61
RETRY QUIET
turns off error recording for soft machine check errors successfully
corrected by instruction retry.
RETRY RECORD
resets the error threshold count to zero and records up to 12 soft
machine check errors, successfully corrected by instruction retry. The
system records the errors on the error recording cylinder.
MAIN QUIET
turns off error recording for single bit main storage errors sucessfully
corrected by Error Correction Code logic.
MAIN RECORD
resets the error threshold count to zero and records up to 12 successfully corrected main storage errors. The system records these errors
on the error recording cylinder before entering MAIN QUIET mode.
Note: SET MODE MAIN is invalid for 3031,3032, and 3033
processors.
cpuid
in attached processor mode, cpuid is the address of the processor for
which you executed the command. Valid hexadecimal values are 00 to
3F. If you don't specify an address, the system processes the
//
command for both processors.
PAGing nn
is the number that you want the system to use in the working set size estimate control algorithm.
PRIORity userid nn
sets a priority of nn to the user that you specify.
userid
identifies the user for whom you want to set a priority.
nn
sets the user's dispatching priority in relation to other users in the
system. Generally, the lower the value of nn, the more favorable the
user's position is in relation to other users in VM/SP's dispatch
queues.
QDROP userid ION }
OFF [USERS]
controls the QDROP elimination option. QDROP ON is the initial setting
for all users.
A virtual machine may enter an idle wait state with no active high speed I/O
outstanding for it. In this case, VM/SP scans the virtual machine's page and
segment tables, and puts resident pages on the flush list. Pages on the flush
list are the first candidates for page-out or removal from main storage.
3-162
VM/SP Operator's Guide
/
userid
identifies the user for whom you want to set the QDROP option.
ON
sets QDROP elimination on for the user that you specify (userid).
OFF
eliminates the scanning of the virtual machine's page and segment
tables, and the placing of resident pages on flush list. This can
improve performance on some virtual machines.
For example, the VM/VT AM service machine waits for the system to
send IUCV messages to it. Each time the machine processes a
message and enters the wait state, the system must scan the page and
segment tables. If the message rate is moderate to high, the system
does not need to scan, because the VM/VTAM service machine is not
idle for an extended period of time. Setting QDROP OFF for the
VM/VT AM service machine eliminates the overhead. If you set
QDROP OFF and set the redisplay timer for the VT AM service
machine to zero, the system may delay the command output for the
VTAM service machine user until the user starts the next I/O operation.
If you set QDROP OFF for a virtual machine that is idle for extended
periods of time, you may degrade system performance. This is
because the system does not remove the virtual machine's resident
pages from storage as quickly as if you set QDROP ON.
USERS
QDROP OFF USERS provides a temporary extension of the QDROP
OFF status to any virtual machine communicating with VMCF or
IUCV to the (server) virtual machine for which QDROP OFF is being
specified. USERS eliminates overhead with a virtual machine being
queue dropped during communications with a server machine for
which the QDROP OFF specification is in effect. This can occur in
small systems where there is a high degree of inter-virtual machine
communications.
The QDROP status for the served virtual machine remains in effect
only while messages are outstanding between it and the server
machine. Thus, performance gains can be realized in systems with
heavy usage of products such as IFS or PVM (invoked with the CMS
PASSTHRU command). A system where PVM is invoked with CP
DIAL or with the SNA VT AM service machine will not gain in performance, since the machine communicates with CP rather than
another virtual machine.
RECord
I
I
OFF
[
ON raddr LIMIT nn BYTE nn BIT n ~~Dl BYTE
nn BIT n]
I
sets the recording mode for a device.
Section 3. CP Commands
3 -163
SEu
OFF
turns off intensive recording mode. If you do not enter SET
RECORD OFF, the system stops intensive recording after it accumulates ten error records for that device.
ON
turns on intensive recording mode for the device raddr.
LIMIT nn
specifies the number of test conditions that must occur before the
system will record it. (Define the test condition with the BYTE or
BIT operands.)
BYTE nn or BIT nn
defines the test condition in the sense bytes.
AND or OR
lets you test two bits. Note that you can only have one device at a
time specified with intensive recording.
REServe
I I
userid nnn
OFF
sets the reserved page performance option for the user that you specify
(userid).
userid
identifies the user for whom you want set the reserved page performance option. You can only apply this option to one virtual machine at
a time.
nnn
is the number of page frames that you reserve for the virtual machine's
exclusive use. If the number of page frames you assign does not equal
or surpass the most active pages, performance of the virtual machine
will degrade.
The virtual machine can use other available page frames. The page
frames that you reserve are not locked, but no other virtual machine
can use a reserved page frame unless the system has exhausted all
other available page frames. The size of the virtual machine determines the maximum number of page frames that you can reserve.
OFF
returns the reserved pages to the system.
3-164
VM/SP Operator's Guide
SASsist
ION I[[PROC] nn]
OFF
ON
activates the virtual machine assist 13, a facility available on most
VM/SP supported System/370 processors, for the entire system. If
you want the same setting for main and attached processors, use this
option without the PROC nn option.
This option along with CP Assist also helps activate the Extended
Control-Program Support:VM/370. 14 Refer to the VM/SP System
Programmer's Guide for more information.
At system initialization, the default value SET SASsist is ON for
VM/SP processors except the System/370 Model 155 II and the 165
II.
OFF
deactivates the virtual machine assist. 13 If you want the same setting
for main and attached processors, use this option without the PROC
nn option.
This option along with CP Assist also helps deactivate the Extended
Control-Program Support:VM/370. 14
Note: SET SASsist ON or OFF also controls if the virtual interval
timer assist function of ECPS:VM/370. This assist function is a hardware updating facility for the interval timers of virtual machines,
(virtual location X'50') is available to users. Use of this timer hardware update facility by the virtual machine is controlled by the class G
command SET ASSIST TMR or SET ASSIST NOTMR. For details,
see the VM / SP CP Command Reference for General Users.
PROC nn
activates SASsist in the processor that you define with nn (decimal
value 00 to 63) .. If nn is a value that is not the address of either
processor, the system issues an error message.
SRM'
sets the system resource management function.
13
Virtual machine assist increases the efficiency of CP processing for certain instructions,
privileged instructions, and interrupts.
14
Extended Control-Program Support:VM/370 simulates the CP processing of additional privileged instructions and extends the level of handling of other privileged
instructions not covered by virtual machine assist. For details on the virtual machine
assist and the Extended Control-Program Support:VM/370, refer to the VM/SP
System Programmer's Guide.
Section 3. CP Commands
3-165
SEu
APAGES nnnn
the current number of pageable pages available. nnnn can be from
0000 to 9999.
DSPSlice nnn
the duration, in milliseconds (nnn), of the dispatching time slice.
IB n
is the interactive shift bias that the system will use for Ql virtual
machines that are using less than their allocated share of the
processor. n can be from 0 to 9.
I
MAXDrum nnnnl
OFF
is the maximum number of drum pages that you want the system to
allocate to anyone virtual machine (nnnn). If you specify OFF, the
system will shut off the MAXDrum. nnnn can be from 0000 to 9999.
MAXWss Innnnl
OFF
is the maximum value for a virtual machine's working set (nnnn). If
you specify OFF, the system will shut off the MAXWss. nnnn can be
from 0000 to 9999.
I
MHFULL Innn
OFF
is the percentage of the preferred moveable head paging area that the
system must be using before migration from this area takes place. If
you specify OFF, the migration of pages will take place from fixed
head areas only (OFF).
PBnn
indicates the maximum page bias that the system can calculate. nn can
be from 00 to 99.
PCI !DRUMI
DISK
indicates the PCI flag for 2305 and fixed head page requests. DRUM
is the standard default mode; the system only flags requests for sector
1. DISK flags all chained page requests.
PGMTlim
is the unused segment elapsed time criteria that the page migration
routine will use.
S370E
Ig~FI~PROC]
[addrl]
enables or disables MVS/System Extensions support for virtual machines;
thus allowing an MVS system, running in a virtual machine, to use the
MVS/System Extensions Program Product (Program No. 5740-XEl).
3-166
VM/SP Operator's Guide
ON
enables the MVS/System Extensions support for virtual machines. In
attached processor mode, the system enables both processors unless
you specify a processor address.
OFF
disables the MVS/System Extensions support for virtual machines.
When you specify OFF, the system issues a message and enters the
console function mode.
PROC nn
is the 'address of the processor (0-63). In attached processor or
uniprocessor mode, the system issues a message if the processor that
you specify is unavailable or invalid.
SET S370E is invalid if you are running a VM system in a virtual machine.
Things You Should Know about the SET Command
1.
In attached processor mode, if you specify one processor, the change will not
affect MVS/System Extensions support users with affinity established on the
other processor.
2.
If the dump allocation has defaulted to the printer, when dumps are purged
from DUMP DASD allocated space, CP will reallocate the dump to DUMP or
TEMP space.
System Response
When you issue SET, the system displays informational messages relating to the
operands that you specify. These messages are described in the VM/SP System
Messages and Codes.
SYSTEM AUTO DUMP RE-ALLOCATED TO DASDraddr
indicates that the system reallocated the CP dump DASD space described in
"Things You Should Know about the SET Command."
UNABLE TO ALLOCATE SYSTEM AUTO DUMP
. indicates that CP could not find the DASD space that it needs to allocate the
system.
INVALID OPTION - option
or
OPERAND MISSING OR INVALID
means that you specified an invalid parameter on the SET MITIME
command.
Section 3. CP Commands
3-167
SElf
MISSING INTERRUPTION MONITORING NOT AVAILABLE
indicates that DMKDID is not in the system.
Note: Your installation may remove the missing interruption handler module
(DMKDID) from the load list if you do not want missing interruption monitoring.
3-168
VM/SP Operator's Guide
SHUTDOWN
Privilege Class: A
When to Use SHUTDOWN
Use the SHUTDOWN command to:
End all VM/SP functions, disable communication lines, checkpoint the system
for an eventual warm start, and save virtual machines that are enabled to be
saved.
•
Automatically do a warm start, when you specify the REIPL option.
Using SHUTDOWN is the normal method to shut the system down. However, if
you want to stop communications on 370x or remote 3270 binary synchronous
lines, use the NETWORK SHUTDOWN command.
SHUTDOWN Format and Operands
I
SHUTDOWN [REIPL
]
POWEROFF
stops the system, and the CPU enters a wait state of 8.
REIPL
stops the system and does an automatic warm start. All online 3480 tape
devices remain assigned to the system.
When you don't specify the REIPL option, the system rewinds, unloads, and
.unassigns all online 3480 tape devices from the system. The system then
stops and enters a wait state.
POWEROFF
is only allowed on 4361 processors. When you specify this parameter, a
normal shutdown will take place and the processor will power off. If the
processor has no poweroff feature, the system will load a wait state 007 PSW
after it has saved the system warm start data and virtual machines.
Upon power on of the processor, you will not need to intervene during IPL if
the 4361 has the Auto Start feature installed and the warm start data is valid.
During IPL, the system uses the battery operated TOD clock setting and a
warm start will take place. (The TOD clock provides an accurate measure of
Section 3. CP Commands
3 -16 9
L_ . . ___________ .__.____.___._____ ._. _____________
~_.
___._____________
time, independent of system events or activities.) If the 4361 does not have
the Auto Start feature installed, you will have to re-IPL the system using the
steps in "System Initialization" on page 2-3.
Things You Should Know about the SHUTDOWN Command
1.
Because the SHUTDOWN command stops all virtual machine operation, use
the appropriate LOGMSG information, messages and warnings to tell all
virtual machine users of the impending shutdown.
2.
In installations that use the 3850 Mass Storage System, the VM/SP shutdown
process issues a SUSPEND order to all 3851 devices. This suspends the MSC
interface. The SUSPEND order may cause the shutdown process to take
longer than usual, because it causes the destaging of data on 3330V volumes.
3.
During processing, if the device has an outstanding intervention required condition, the system will not log a Miscellaneous Data Record (MDR) for a
device that maintains hardware environmental counters.
System Response
SYSTEM SHUTDOWN COMPLETE
indicates that the system completes the shutdown.
You won't get a response when you issue SHUTDOWN REIPL.
For a normal SHUTDOWN, you will enter a wait state of 8. When you issue
SHUTDOWN POWEROFF for a 4361 that does not have the battery TOD clock,
it will shutdown and poweroff. However, when you IPL the 4361 again, you will
need to use the traditional IPL procedures.
3-170
VM/SP Operator's Guide
c·····--·-··· .....-.. -...... -.-.---.--.-..- ..- - - - - - . - - - - -..--.-.--------.... --....-- -..-.. --.. --.-.-.-.- . - - - - - - - - - - - . - - - - - - - - - - - - - . - - - - - - ]
SPACE
Privilege Class: D
When to Use SPACE
Use the SPACE command to force a printer to single space the output for the
current active spool file. The command will override any carriage control commands that are in the actual file.
SPACE Format and Operands
SPAce
raddr
raddr
is the real address (cuu) of the printer whose output you want to single
space.
Note: For buffered printers, such as the 3800 printer, the SPACE command
does not affect files that have already been sent to the printer.
System Response
None.
Section 3. CP Commands
3 -1 71
SPMODE
Privilege Class: A
When to Use SPMODE
Use SPMODE when you want to restrict VM/SP to run in a uniprocessor environment in a multiprocessor configuration, while letting an MVS V =R virtual machine
control the other processor. SPMODE can establish or reset the single processor
mode environment.
Note: Before entering the SPMODE command, review its operating procedures in
Running Guest Operating Systems.
SPMODE Format and Operands
SPMode
ON
initializes VM/SP in single processor mode if all the proper conditions are
met. SPMODE ON will only execute when VM/SP is running in
uniprocessor environment.
Note: Before issuing SPMODE ON, you may need to VARY OFFLINE
PROCESSOR nn if the system is operating in attached processor mode.
OFF
resets VM/SP to the normal uniprocessor mode of operation.
Things You Should Know about the SPMODE Command
3-172
1.
When in single processor mode, error records for the MVS V =R virtual
machine are in two locations. For details concerning the collection of all error
records see VM / SP OLTSEP and Error Recording Guide.
2.
Before the MVS native processor can use a 3480 tape drive in SPMODE, vary
the drive offline from the CP system. After SPMODE is turned off, vary back
online the tape drives that you want to use.
VM/SP Operator's Guide
c---·---------·--
SPTAPE
Privilege Class: D
When to Use SPTAPE
When the spooling system is overloaded with spool files, you can temporarily store
files on tape to free up those resources. SPTAPE does the following:
•
Stores to tape (DUMP) the unit record files that you want to schedule later on
the VM/SP system
•
Retrieves from tape (LOAD) the unit record files
•
Reads the files back into the system at your selection
Scans the tape for priority files.
Section 3. CP Commands
3 -1 73
SPTAPE Format and Operands
SPTape
....
~
STOP raddr
CANCEL raddr
SCAN raddr SADump option2
LOAD raddr SADump option2
spoolidl
LOAD raddr
PUnch
Reader CLass
{Printer}
[SPOOlid2]
option2
option3
END
cl[c2[c3[c4]]]
FORM/orm
ALL
DUMP raddr
spoolidl
{Printer}
PUnch
Reader
~POOlid2]
option 1 option2 option3 option4
END
CLass
cl[c2[c3[c4]]]
>-
FORM/orm
ALL
OPTIONS
[:~~t~I~ll
~~:~~
REWind
" RUN
option 3
[SYSHOLD ~
USERHOLD
NOHOLD
option 4
[PURGE]
,,;
STOP raddr
stops the operation after CP completes the current file. CP positions the
tape (raddr) at the beginning of the next file. You can continue with a new
command at the current position.
CANCEL raddr
stops the operation immediately. CP then rewinds and unloads the tape
(raddr), without regard to the previous tape positioning options.
SCAN raddr
scans the tape (raddr) from the current point to the first double tape mark it
finds. Use one of the tape handling options to position the tape.
LOAD raddr
specifies the device address of the tape (raddr) from which CP will read the
spool files. Use one of the tape handling options to position the tape.
Note: When you use this operand with the SADUMP operand, do not attach
the tape drive to the virtual machine.
3-174
VM/SP Operator's Guide
/
Lr:=:
___-_-_-_-_-_-__--____________________________________________________________________
,~
DUMP raddr
specifies the device address of the tape (raddr) where CP will write the spool
files. Use one of the tape handling options to position the tape.
SADUMP
indicates that the data on the tape is output from the stand-alone dump
facility.
Note: When you use this operand with the LOAD raddr operand, do not
attach the tape drive to the virtual machine.
Printer
PRT
is a printer-type spool file.
PUnch
PC"
is a punch-type spool file.
Reader
RDR
is a reader-type spool file. This includes MONITOR and CP system dump
spool files.
SPOOlidl [SPOOlid2] }
END
CLass cl[c2[c3[c4111
{ FORM form
ALL
spoolid
identifies the spool file or the start of a range of spool files ending with
spoolid2.
Example:
SPTAPE spoolid 1 50
causes the operation to start at spoolid 1 and operate on
all spoolids from 1 up until 50.
SPT APE spoolid 1 55
operates on spoolids 1, 7, 5, 55, if a reader chain had
spoolids in the following order: 50, 6, 1, 7, 5, 55.
END
dumps or loads all spool files after spoolidl on the output chain.
I
~
CLASS cl[c2[c3[c411 ]
indicates the output classes. You can specify up to four classes. If
you are loading and dumping files with the same class, dump and load
them synchronously to avoid any possible overlap of functions.
Section 3. CP Commands
3 -1 7 5
FORM/orm
dumps or loads all files with the specified operator form.
ALL
dumps or loads all spool file classes.
[MODE 1:1i~ll
is the reset value of tape density for reel-to-reel tape systems. You may
specify mode values of 800, 1600, 6250, or 38K. 1600 is the default, except
for the 3480 tapes. For the 3480 tapes, the default is 38K. MODE also
specifies the recording density for the 3480 magnetic tape subsystem.
Notes:
1. If a user specifies a density mode that the tape cannot handle, the
control unit may not return an error condition. In this case, CP ignores
the mode setting and uses the default control unit setting.
2.
The following combinations of modes and tape drives are invalid:
Mode 800 with 3430 tape drives
Mode 38K with any tape drive other than the 3480
Modes 800, 1600, or 6250 with the 3480 tape drive.
LEAVE]
REWind
[
RUN
LEAVE
does not move the tape after CP performs the tape operation.
REWind
rewinds the tape after CP performs the tape operation.
RUN
rewinds and unloads the tape after CP performs the tape operation.
SYSHOLD ~
USERHOLD
[ NOH OLD
SYSHOLD
includes the system-held files in the tape operation.
3 -1 7 6
VM/SP Operator's Guide
c
USERHOLD
includes the user-held files in the tape operation.
NOHOLD
includes the non-held files in the tape operation.
[PURGE[
tells CP to delete the spool files from the spooling system after successfully
writing them to tape.
Things You Should I(now about the SPTAPE Command
1.
The SPTAPE command does not handle multi-volume tapes.
2.
While SPT APE is processing spool files, other virtual machines cannot access
or process the spool files.
3. If the format of the input spool file is different from the current spool file
format, CP changes (or leaves) the restored spool file in system hold status.
This happens when you move a spool file from:
A VM/SP Release 1 or 2 system to a VM/SP Release 3 system
A VM/SP Release 3 system to a VM/SP Release 1 or 2 system.
4.
When you are using SPT APE, do not attach the tape drive to another virtual
machine.
System Response
The restored files keep the same characteristics as the original file, but CP assigns a
new spoolid to the restored files. The response shows the old and new spoolids. It
also shows that the SPTAPE command is currently processing the designated spool
file on the indicated tape.
DUMPING}
{PRT}
raddr PUN userid spoolidl class form stat [NOW spoolid2]
{ LOADING
SCANNING
RDR
raddr
is the tape drive address where CP is performing the operation (dumping,
loading, or scanning).
userid
identifies the user that originally created the file.
spoo 1i d 1
is the spoolid that CP first gave to the spool file.
Section 3. CP Commands
3 -1 77
class
is the spool class of the file.
form
is the operator form number of the file.
stat
is the status of the files: USER, SYS, USYS (for USER and SYS), or NONE.
spool id2
is the new spoolid that CP gave to the restored file.
If an I/O error occurs:
Or CP detects an end-of tape, while writing files (SPTAPE DUMP) or reading
files from the disk, you will receive a message stating that the tape is backed-up
to the last complete spool file. CP writes a tape mark, rewinds and unloads the
tape, and ends the command.
•
3 -1 78
While trying to read or write files for the SPTAPE LOAD option, you will
receive a message. CP rewinds and unloads the tape, and ends the command.
Spool files that CP has successfully read remain in the spooling system, but the
system purges any partial file.
VM/SP Operator's Guide
-----------_ _-]
..
START
Privilege Class: D
When to Use START
Use the START command to:
Restart a spooling device after it has been drained.
Change the output class that the spooling device may service.
If the device is busy when you issue the command, the changed classes take effect
after the current file is complete.
START Format and Operands
-
~
STArt
ALL
Printer
PUnch
Reader
raddr[ CLass c••. ] [Fonn form
FOnn *
1
J NOSep
J
J-
[AUto
[N03800
SETup
BEG3800
MAnual
ANY3800
-
FLash name
CHars name
FCBpipi
IMage imagelib
PUrge
-
-
DEFfCbJ
FILefcb
[ CFIlefcb
starts all readers, printers, and punches.
Printer
PRT
starts all printers. Printers that are in ready status, with files available for
processing, immediately start to produce output when you issue this
command.
Section 3. CP Commands
3 -1 7 9
PUnch
PCH
starts all punches. Punches that are in ready status, with files available for
processing, immediately start to produce output when you issue this
command.
Reader
RDR
starts all readers. Readers do not start until you supply a card deck and
make the device ready.
raddr
is the address (cuu) of the spooling device that you want CP to start. You
can specify more than one device. If the device is a reader, you cannot use
any options.
This is the only operand with which you can use options.
CLass c...
specifies the output classes that the device can service. You can
specify up to four classes (single characters separated with blanks). If
you specify more than one class, the device services the output files in
the order that you specify the classes.
Example:
CLASS AD 1
specified for a printer, spools all class A files before class
D files, and all class D before class 1.
Specify a class of * for printers or punches if you want the device to
service any spool file class.
FOrm form
specifies the operator form that the printer or punch can process. This
becomes the active form on the printer or punch. form is one to eight
characters.
FOrm *
specifies that the printer or punch processes spool files with any form
number. If you do not specify FOrm on the STArt command, the
same form that you specified on the last STArt command remains in
effect. CP keeps the last form across system warm and checkpoint
starts, but not on cold starts.
If you do not issue the FOrm option with START since the last cold
start, CP uses the default operator form specified in the SYSFORM
macro.
3-180
VM/SP Operator's Guide
NOSep
does not use a file separator for the output files. If you specify NOSep
for the punch, the system does not punch separator cards. Instead,
you will receive the following message:
PUN raddr START REQ CLASS c ... FORM form status
NOSEP
Issue CP STArt raddr or physically activate the START or STOP key
on the device to start the punch for the next file.
If you do not specify this option, CP produces a separator.
AUto
processes files of any form, with automatic prompting messages for
forms changes. CP processes the currently active form first.
SETup
processes files of any form, with automatic prompting messages for
forms changes. You can check and adjust forms alignment when
changing forms.
This option is valid only for printers.
MAnual
tells CP to process only spool files with the currently active form. If
the form is *, CP processes spool files with any form.
N03800
restricts all 3800 LOAD CCWs. The device can print a file from any
virtual device only if it has no 3800 LOAD CCWs. After a cold start,
N03800 is the default value for the 3800 operands (N03800,
BEG3800, ANY3800).
The minimum abbreviation is N03. This option is valid for all real
spooling printers, not just 3800s.
BEG3800
restricts some 3800 LOAD CCWs. The device can print a file from
any virtual device if it has no 3800 LOAD CCWs or if it has 3800
LOAD CCWs only at the beginning of the file.
The minimum abbreviation is BEG. This option is valid for all real
spooling printers, not just 3800s.
ANY3800
does not restrict 3800 LOAD CCWs. The device can print any file.
The minimum abbreviation is ANY. This option is valid for all real
spooling printers, not just 3800s.
Section 3. CP Commands
3 -181
FLash name
is the flash overlay sheet loaded into the device. name may be one to
four characters in length. If you do not specify this operand, the previous value remains in effect.
This option is only valid for the 3800 printer.
CHars name
is the character arrangement table that CP will use for the separator
page. name may be from one to four characters in length. If you do
not specify this operand, the previous value remains in effect.
This option is valid for only the 3800 printer.
FCBplpi
specifies the forms control buffer for the separator page and any spool
file for which you have not specified an FCB (using the SPOOL or
CHANGE command). plpi (print lines per inch) must be a unique
number 6, 8, or 12, and for the 3800 Model 3 device, 10. These
numbers mean that you want the system to print the entire spool file at
6, 8, 10 or 12 lines per inch, regardless of the size of the paper that is
currently loaded. If you do not specify this operand, the previous
value will remain in effect.
This option is only valid for the 3800 printer.
IMage imagelib
specifies the named system image library that CP uses for tables.
imagelib may be from one to eight characters and must be a valid
named system. If you do not specify this option, the previous value
will remain in effect.
This option is only valid for the 3800 printer.
PUrge
purges all files with load errors. If you do not specify this option, CP
will hold all files.
This option is only valid for the 3800 printer.
DEFfcb
uses the default FCB that was loaded in the printer. CP will not send
imbedded LOAD FCB CCWs to the printer. The LOADBUF
command specifies the default FCB.
This option is only valid for 3211-type printers.
FILefcb
sends file FCBs to the real printer. File FCBs are LOAD FCB CCWs
imbedded in the spool file. This is the default when you start the
printer following a non-warm or non-checkpoint start.
3-182
VM/SP Operator's Guide
/'
This option is only valid for 3211-type printers.
Note: When a virtual machine loads a 3211 with a new FCB, the
operator must make sure that the proper form (paper) is loaded and
that the paper is aligned properly.
CFIlefcb
. checks the file FCBs. File FCBs are LOAD FCB CCWs imbedded in
the spool file. You can send file FCBs to the real printer. After CP
prints each copy of the spool file, CP sends a message saying that
forms misalignment may have occurred and that it drained the printer.
Check the forms alignment within the printer.
This option is only valid for 3211-type printers.
Things You Should Know about the START Command
1. If you do not specify either MAnual, AUto or SETup, the previous mode
setting remains in effect. CP keeps the mode across system warm and checkpoint starts. If a cold start occurs, the mode defaults to MAnual.
2.
If you do not specify either N03800, BEG3800 or ANY3800, the previous
setting remains in effect.
3. If you want to print or punch without separator pages, specify NOSEP every
time you enter the START command. Otherwise, START defaults to the SEP
option.
4.
VM/SP will assure that the FCB used for separator pages contains enough
print lines. An FCB with more lines per inch than specified will be loaded, if
necessary, for printing separator pages. For paper:
5 1/2 inches long and smaller, a 12lpi FCB is loaded.
•
7 1/2 inches long and smaller (down to 5 1/2 inches), a 10 lpi FCB is
loaded (121pi for the 3800 Modell), unless 12lpi was specified.
8 1/2 inches long and smaller (down to 7 1/2 inches), an 8 lpi FCB is
loaded, unless 10 or 12 lpi was specified.
Select a character set that matches the FCB so that the tops of characters will
print.
System Response
You may receive more than one response line for this command. The format of the
response depends on the device type. The response is identical to that of the
"QUERY UR" command. In each case the status is given as STARTED. Look in
"QUERY" on page 3-117 for a description of the response to Query UR.
Section 3. CP Commands
3 -18 3
STep
Privilege Class: C
When to Use STCP
Use the STCP command to change the contents of real storage.
Note: You cannot change the real PSW or real registers with this command. Also,
in a system running in attached processor mode, this command cannot modify data
in shared pages.
STCP Format and Operands
STCP
~MLhexloc'
NLhexloc
-<
Mhexloc >
Nhexloc
hexwordl [hexword2 •.. ]
Lhexloc
, hexloc
J
MSheXIOC}
NShexloc
{
Shexloc
hexdata
MLhexloc
NLhexloc
Mhexloc
Nhexloc
Lhexloc
hexloc
stores the data given in hexwordl [hexword2 ... J in successive fullword
locations starting at the address specified by hexloc. The smallest group of
hexadecimal values that you can store using hexloc is one fullword. The
system aligns data to the nearest fullword boundary. If the data that the
system is storing is less than a fullword (eight hexadecimal digits), CP rightadjusts the word, and fills the high order bytes of the word with zeros.
In attached processor systems, hexloc and Lhexloc values are absolute
addresses. The M prefix causes the system to interpret that address through
the prefix register of the main processor. In MP systems, M corresponds to
addresses on the IPL processor.
3-184
VM/SP Operator's Guide
In MP systems, the N prefix causes the system to interpret addresses as if
they were on the non-IPL processor. In attached processor systems, the N
prefix causes the system to interpret the address through the prefix register
for the attached processor.
Example:
STCP N64 20000
stores X'00020000' into the second word of the SVC new PSW
in the PSA (prefix storage area) of the attached processor.
If you don't specify M or N, the system will interpret the address as an absolute address on both AP and MP systems. N is valid while the second
processor is in operation. M is valid if the system was generated for the
attached processor or multiprocessor mode of operation.
MShexloc
NShexloc
Shexloc
stores the data, given in hexdata, in the address that hexloc specifies without
word alignment. The shortest string that the system can store is one byte
(two hexadecimal digits). If the string has an odd number of characters, the
system does not store the last character. An error message occurs and the
function ends.
Attached processor VM/SP applications use MShexloc and NShexloc operands. If you prefix Shexloc with M or N, the system interprets the specified
hexadecimal addresses through the associated prefix registers of the main or
attached processor.
hex word
specifies up to eight hexadecimal digits. If you specify fewer than eight
digits, the system right-justifies the string in a fullword and left-fills it with
zeros. If you specify two or more hexwords, you must separate them with at
least one blank.
hexdata
specifies a string of two or more hexadecimal digits with no embedded
blanks.
System Response
STORE COMPLETE
Section 3. CP Commands
3 -18 5
TRANSFER
Privilege Class: D
When to Use TRANSFER
Use the TRANSFER command to:
Direct one or more spool files to a specified user.
Reclaim reader spool files that CP previously sent to one or more users.
TRANSFER Format and Operands
TRANsfer
,
I
userid
identifies the user to whom CP will transfer or reclaim the spool files.
SYSTEM
lets you manipulate all spool files in the system, regardless of the userid.
*
lets you manipulate your own spool files.
Printer
PRT
is a printer-type queue to which or from which CP will transfer the files.
PUnch
PCH
is a punch-type queue to which or from which CP will transfer the files.
Reader
RDR
is a reader-type queue to which or from which CP will transfer the files.
spoo/id
identifies the file that you are transferring.
3-186
VM/SP Operator's Guide
r
CLass c
transfers all files of the class that you specify.
FORM form
transfers all files of the form that you specify. form is one to eight characters. If you specify the * option or let it default, the form number is the user
form. If you specify the userid/SYSTEM option, the form number is the
operator form.
ALL
transfers all spool files of the queue that you specify, regardless of Class,
form, or spoolid.
To
{*useridI
is the user to whom you are directing the files. If you omit the optional
keyword TO, userid cannot be T or TO. The asterisk (*) means that you
want CP to send the files to you.
From
{AL~
I
userzd
is the user from whom CP will reclaim the spool files. Specify ALL to
reclaim files from all users, starting with the userid that you specified first.
Note: If you use the TRANsfer SYSTEM From command to reclaim files,
CP directs the files back to their originators.
Things You Should Know about the TRANSFER Command
The command will only transfer files to queues of a valid type. If you try to
transfer printer to punch or punch to printer, CP indicates an error. You can
transfer reader files to the printer/punch, but CP will only do valid transfers. For
example, CP will not transfer real reader, punch or Dump/Monitor files to the
printer. The following chart shows the valid queues for each spool file origin.
Spool
Files
Queue
reader printer punch
0
r
i
9
i
n
Real
Reader
*
Printer
*
*
*
Punch
Durnp/
Monitor
*
*
Note: Dump/Monitor includes CP system dumps, user dumps (VMDUMP)
command, NCP dumps, or VM Monitor spool files.
Section 3. CP Commands
3 -18 7
System Response
RDR} FILE spoo 1 i d TRANSFERRED TO user i d { RDR}
PRT .
PRT
{
PUN
PUN
tells what your input was when you issued the
TRANSFER command. This response identifies the spool file and the
user to whom CP transferred the file.
Note: If you used the asterisk (*), the user will be yourself (the person who
issued the command).
{~~nn} FILES TRANSFERRED
is the response to the user who issued the TRANSFER command.
Note: Do not specify the SET IMSG OFF command if you are monitoring
the RSCS virtual machine for your installation; you need to see informational
messages to accurately monitor RSCS system activity.
3-188
VM/SP Operator's Guide
[------------------.-----:--------..-._--_--_-_-_-__-_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _====:1_----1
UNLOCK
Privilege Class: A
When to Use UNLOCK
Use the UNLOCK command to unlock page frames that a previous LOCK
command locked.
UNLOCK Format and Operands
I
,
UNLOCK
~
{
USerid
}!irstpage lastpage
SYSTEM
>
,I
VIRT=REAL\
V=R
userid
identifies the virtual user that you previously specified in a LOCK command.
SYSTEM
unlocks a previously locked pageable CP nucleus page.
!irstpage
is the hexadecimal value of the first page that you want to unlock.
lastpage
is the hexadecimal value of the last page to be unlocked.
Note: For !irstpage and lastpage, specify only the page numbers, not the displacement. For example, to unlock USERA's virtual storage locations
'12000' to '2CFFF', issue the following command:
UNLOCK USERA 12 2C
VIRT=REAL
V=R
releases, for normal paging use, all the page frames normally for the
virtual=real machine. However, the area must not be occupied by a
virtual=real virtual machine.
Note: Once the system releases the virtual=real area, it cannot reactivate
the area to run another virtual=real machine until VM/SP is loaded again.
Section 3. CP Commands·
3 -18 9
System Response
COMMAND COMPLETE
occurs even if some (or all) of the pages unlocked were not previously
locked.
3-190
VM/SP Operator's Guide
r----·-,·-----·--- ----
J
VARY
Privilege Class: B
When to Use VARY
Use the VARY command to mark a device available or unavailable for a user or
CP to use. (For 370x resources running in NCP or PEP mode, or remote 3270
resources, use the NETWORK VARY command.)
VARY Format and Operands
VARY
ONLine} {raddr-raddr}
{OFfline raddr•.•
PROCessr
OFfline PROCessr nn [VPHY] (FORCE]
VLOG
ONLine
makes an offline device(s) available or makes offline paths to an online
device available.
After a WARM IPL or CKPT IPL, the first time you VARY a printer
ONLINE it will have the same status (for example, DRAINED, WAITING)
that was present at the last checkpoint before the system shut down.
OFfline
makes the device(s) unavailable.
VM/SP remembers spooled unit record devices marked offline through any
system failures. A shutdown and subsequent cold start assumes that all
devices are available.
raddr
is the real address of the device whose status you want to change. You can
specify more than one device or a range of device addresses. CP also
changes any alternate paths to the device(s) that you specify online or
offline.
When you enter more than one device or a range of devices, CP continues to
process the VARY command even if it finds an error while trying to vary
one of the devices online/offline. CP issues an error message for every
Section 3. CP Commands
3 -191
VAlPJV
device that causes an error and an informational message for every device
that it could vary online/offline.
Note: If the device is a 2305, the address must be for the first of 8 expo-
sures, that is, address xxO or xx8.
PROCessr nn
is the processor that CP will make available or unavailable. nn is the
processor address (hexadecimal value from 00 through 3F) assigned when
your installation installed the processor.
For VARY ONLINE PROCessr, CP checks to see if the processor is already
online. If it is not, and if the processor is available and operational, CP does
the following:
Establishes prefixing
Initializes the control registers and prefix areas
Sets the clock comparator and CPU timer.
The system can then function in attached processor or multiprocessor mode.
For VARY OFFLINE PROCessr, CP checks to see if the processor is online
and operational. CP also ensures that, with the specified processor offline,
an online path will still remain to all system-owned devices. If the processor
is operating in multiprocessor mode, CP ensures that any device that would
lose its last online path is not currently in use. CP then does the following:
•
•
Stops work flow (quiesces) to the specified processor
Clears the control registers and prefix register
Notifies any users having affinity to the specified processor that the
affinity is set off; CP puts these users in console function mode.
OFfline PROCessr nn [VPHY]
VLOG
VPHY
physically varies the processor offline by the MSSF in a 3081
processor configuration.
VLOG
causes the processor to be logically offline to VM/SP but not physically offline as when MSSF is called. This option is for shared system
environments.
If you want to run a program on a 3081 in single processor mode,
enter the commmand as follows:
VARY OFF PROC nn VLOG
3 -192
VM/SP Operator's Guide
--]
FORCE
abruptly stops the processor that you specify. Use this option only when all
other means of varying the processor offline have been unsuccessful.
Use FORCE when you need to stop the processor immediately or when the
system cannot stop all activity on devices attached to the processor. When
you issue VARY OFFLINE PROC FORCE, CP does not check for device
activity before marking devices offline. This may result in a device being
hung, or it may preclude a device's being varied online again.
Things You Should Know about the VARY Command
1.
If system performance monitoring is taking place when you want to vary a
device, use the MON STOP command to stop monitoring. Restart monitoring
after the system makes the configuration change.
2.
If the system is only enabled for the UP mode, do the following:
a.
Stop the online processor
b.
Enable the MP features for the system
c.
Restart the online processor
d.
Vary the offline processor online.
System Response
I
raddr. . .
raddr-raddr
I
VAR IED /ONL INE
OFFLINE
I
is the response when CP successfully varies the device (r add r) online or
offline.
RADDR MISSING OR INVALID
is the response if you entered the PROCESSR operand without a processor
number or with any invalid number.
PROCESSOR nn DOES NOT EXIST
is the response if the processor that you specified in VARY PROCESSR nn
is not defined to the system.
VARY PROCESSOR COMMAND FAILED
For VARY OFFLINE PROCESSR, this response means one of the following:
The processor that you specified is in uniprocessor mode or not online.
•
The processor that you specified is the main processor in an attached
processor system.
Section 3. CP Commands
3-193
VAPJV
c:::
In a multiprocessor system, the system-owned device would lose its last
online path, or a device that would lose its last online path is currently in
use.
For VARY ONLINE PROCESSR, this response means one of the following:
•
•
•
The system is not generated for AP mode.
The MP function is not installed.
The system is already in AP mode.
PROCESSOR nn ONLINE
the specified processor is now online and operational.
PROCESSOR nn OFFLINE
the specified processor is now offline and not operational.
LAST PATH TO DEVICE raddr ON PROC nn
the system-owned device that you specified by raddr will lose its last online
path if CP varied processor nn offline.
3-194
VM/SP Operator's Guide
WARNING
Privilege Classes: A or B
When to Use WARNING
Use the WARNING command to transmit high-priority messages to a user that you
specify or to all users.
WARNING does not wait for input operations at the terminal to complete.
Instead, WARNING breaks in on the line that the terminal is currently displaying
and sends the message immediately (unless you or someone else previously issued
the SET WNG OFF command for the virtual machine).
WARNING Format and Operands
Warning
Wng
{
userid } msgtext
OPerator
ALL
userid
identifies the user who will receive the message.
OPerator
sends the message to the primary system operator regardless of his userid.
ALL
sends the message to all users able to receive warning messages.
msgtext
is the text of the warning.
Things You Should Know about the WARNING Command
You can have your TELL EXEC use WARNING to send warnings to users on the
same system as you.
DEFAULTS set tell msgcmd warning
changes the default set up in the GLOBALV file.
DEFAULTS set tell warning msg
changes default back to MSG.
',,-
/
You must reset the default again if you want to change it.
Section 3. CP Commands
3 -19 5
WA~N~N(G]
System Response
hh:mm:ss WNG FROM userid: msgtext
is the response to the users who receive warning messages. hh : mm: 55 is the
time in hours, minutes, and seconds when the system sends the message to
the user.
If a user is using a 3270, WARNING puts the screen in a HOLDING status
and rings the alarm, if present. Press the cancel key to return to the
RUNNING status. If you send the WARNING during system IPL or restart
operations, WARNING will put the screen in a MORE status.
3-196
VM/SP Operator's Guide
r-
----------------
§eu"vnce
---
Pu"(l)~U"EJms
'-------------------1
Section 4. VM/SP Service Programs
Controlling the 3704/3705/3725 Communications Control
Program
The EP/3725 Installation and Resource Definition Guide and Reference,
SC30-3172, provides the information necessary to:
Define and generate an Emulator Program for the IBM 3725
Load the control program into the controller
Dump the contents of the controller storage.
VM/SP does not provide loading or dumping facilities for the 3725.
ACF /NCP-SSP provides these facilities for both 3705 and 3725 communications
controllers. Refer to the publications listed in the preface for more information.
Operational control of the 3704/3705 Communications Controller (hereinafter
referred to as 3705) resides with the system operator via the NETWORK
command. In addition, the behavior of the 3705 device is, for a large part,
dependent upon the communications control program with which it was loaded.
The communication control programs are created, assembled, or modified by a set
of CMS commands specifically designed for this purpose. These commands are:
ASM3705
GEN3705
'LKED
SAVENCP
ZAP
The cited commands, described in greater detail in the VM / SP Installation Guide,
are the system programmer's responsibility. The 3705 NETWORK and
NCPDUMP commands, however, are the responsibility of the VM/SP system
operator and are described elsewhere in this publication.
The 3704/3705 Dynamic Trace Facility is supported for virtual machine operation.
However, if the 3704/3705 in emulator mode is not dedicated to the user invoking
the trace, the user is required to have in his directory a privilege class other than (or
in addition to) G.
Section 4. VM/SP Service Programs
4-1
~------------------------------------------------------.--------------------,~
Using eMS To Alter VM/SP Files
The VM/SP system operator is usually assigned a virtual machine with a large
DASD storage allotment. This DASD storage can be assigned to other users as a
temporary work area or for short-term DASD storage. In addition, the operator's
virtual machine can record system statistical data, or create and maintain directory
files and other system files.
For whatever reason the operator uses the system, it is probable that some files will
need to be modified at some time. One of the easiest ways to modify such system
files is to use the VM/SP System Product Editor (XEDIT) facilities.
VM/SP XEDIT facilities provide ways to create and modify all types of eMS files.
For a complete description of VM/SP XEDIT facilities and commands, see
VM / SP System Product Editor User's Guide and VM / SP System Product Editor
Command and Macro Reference.
Other CMS Commands
There may be occasions, caused by user demands, when other facilities of CMS
may be invoked. For example, to create punched output of a file for use by
another system, or to create a history file. The commands likely to be needed by
the system operator are the following:
CMSBATCH
COMPARE
COPYFILE
ERASE
FORMAT
LABELDEF
LISTDS
LISTFILE
MOVEFILE
PRINT
PUNCH
QUERY
READ CARD
SET
STATE
TAPE
TAPEMAC
TAPPDS
TYPE
UPDATE
A complete description of these and other commands and their usage is contained
in VM / SP CMS User's Guide and VM / SP CMS Command and Macro Reference.
DASD Dump Restore (DDR) Service Program and How to
Use It
Use the DASD Dump Restore (DDR) program to dump, restore, copy, or print
VM/SP user minidisks. The DDR program may run as a standalone program, or
under CMS via the DDR command.
The DDR program has five functions:
1.
4-2
VM/SP Operator's Guide
Dumps part or all of the data from a DASD device to tape.
[-_.---------------------------_._---------------
2.
Transfers data from. tapes created by the DDR dump function to a direct access
device. The direct access device must be of the same type as that which originally contained the data.
3.
Copies data from one device to another of the same type. Data may be reordered by cylinder (or block) when copied from disk to disk. In order to copy
one tape to another, the original tape must have been created by the DDR
DUMP function.
4.
Prints selected parts of DASD and tape records in hexadecimal and EBCDIC
on the virtual printer.
5.
Displays selected parts of DASD and tape records in hexadecimal and
EBCDIC on the terminal.
To generate the VM/SP starter system from the distribution tape, the standalone
RESTORE function must be used.
Invoking DDR under eMS
The format of the DDR command is:
[nit [fml
is the identification of the file containing the control statements for the DDR
program. If no file identification is provided, the DDR program attempts to
obtain control statements from the console. The filemode defaults to an
asterisk (*) if a value is not provided.
DDR handles two logical line editing symbols:
1. The logical character delete symbol (@), which allows deletion of one or
more of the previously entered characters.
2.
The logical line delete symbol (¢), which deletes the entire previous
physical line.
Most often, the default values for these two symbols are defined for each
virtual machine at system generation time. When you use the CP TERMINAL command to redefine the CHARDEL and LINEDEL characters,
the re-definitions have no affect on DDR line editing. DDR continues to
recognize "@" as valid CHARDEL and "¢" as valid LINEDEL symbols.
Notes:
1. If you use the CMS DDR command, CMS ignores the SYSPRINT control
statement and directs the output to the CMS printer OOE.
Section 4. VM/SP Service Programs
4-3
2.
Be aware that DDR when run as a standalone program does not have error
recovery support. However, when DDR is invoked in CMS, in a virtual
machine environment, the I/O operation is performed by CP (CP has built-in
error recovery facilities).
3.
When running standalone, DDR will search for a console at address 009 or
OlF. If these consoles are not operational, the program will enter a wait state,
waiting for an interrupt to identify the console. If any nonconsole type device
is physically connected to address 009 or OlF, it must be made nonoperational
or the results will be unpredictable.
Invoking DDR as a Standalone Program
To use DDR as a standalone program, the operator should IPL it from a real or
virtual IPL device as he would any other standalone program. Then indicate where
the DDR program is to obtain its control statements by responding to prompting
messages at the console.
Notes:
1.
Be aware that DDR when run as a standalone program has only the most elementary error recovery support. However, when DDR is invoked in eMS, in a
virtual machine environment, the I/O operation is performed by CP which has
better error recovery facilities.
2.
The standalone DDR utility will not support cylinder faults for MSS virtual
volumes when performing the DUMP or COPY functions.
3.
It may be necessary to disable channel-to-channel devices in order for the
DDR utility to run as a standalone program.
4.
The DDR utility does not assign 3480 tape drives to itself. So, when a 3480 is
in use during standalone DDR processing, do not IPL any other processor that
has a path to the device. This is because the other processor might steal the
3480 drive by assigning it to itself. This would immediately stop DDR's access
to the device. For the same reason, do not vary the 3480 device online to any
other processor until DDR processing completes.
DDR Control Statements
DDR control statements describe the intended processing and the needed I/O
devices. I/O definition statements must be specified first.
All control statements may be entered from either the console or the card reader.
Only columns 1 to 71 are inspected by the program. All data after the last operand
in a statement is ignored. An output tape must have the DASD cylinder header
records in ascending sequence; therefore, the extents must be entered in sequence
by cylinder or by extent. Only one type of function - dump, restore, or copy may be performed in one execution, but up to 20 statements describing DASD cylinder extents may be entered. The function statements are delimited by (preceded
by) an INPUT and OUTPUT statement. If additional functions are to be per-
4-4
VM/SP Operator's Guide
/'
formed, the sequence of control cards must be repeated. If you do not use INPUT
or OUTPUT control statements to separate the functions you specify when the
input is read from a card reader or CMS file, an error message is displayed.
However, the remainder of the input stream will be checked for proper syntax, but
no further DDR operations will be performed. Only those I/O devices defined by
the INPUT statement and the OUTPUT statement must be redefined in subsequent
steps. The SYSPRINT I/O definition remains the same.
To return to CMS, enter a null line (carriage return) in response to the prompting
message (ENTER:). To return directly to CP, key in #CP.
The PRINT and TYPE statements work differently from other DDR control statements in that they operate on only one data extent at a time and it is not necessary
to respecify the INPUT statement ahead of each PRINT or TYPE statement. If the
input is from a tape created by the dump function, the tape must be positioned at
the header record for each step. The PRINT and TYPE statements have an
implied output of either the console (TYPE) or system printer (PRINT), so no
OUTPUT statement is required.
I/O Definition Statements
The I/O definition statements describe the tape, DASD, and printer devices used
while executing the DASD Dump Restore program.
INPUT/OUTPUT Control Statement
An INPUT or OUTPUT statement describes each tape and DASD unit used. The
format of the INPUT/OUTPUT statement is:
INput
OUTput
cuu
type
[VOlser] [(Options ... »)
altape
Options:
[~~;: ~nJ [~g:: ~;~~] r~~:!~dJ
MOde 800
MOde 38K
COmpact
L~Eave
INPUT
indicates that the device described is an input device.
OUTPUT
indicates that the device described is an output device.
Note: If the output device is a DASD device and DDR is running under
CMS, the device is released using the CMS RELEASE command function
and DDR processing continues.
Section 4. VM/SP Service Programs
4-5
cuu
is the unit address of the device.
type
is the device type (2311,2314,2319,3330,3330-11,3340-35,3340-70,
3350,2305-1,2305-2,2400,2401,2415,2420,3375,3380,8809,3410,
3411,3420,3430,3480, or FB-512). There is no 7-track support for any
tape devices. Specify a 3340-70F as a 3340-70, and a 3333 as a 3330.
Specify a 3350 that is in 3330-1 or 3330-11 compatibility mode as a 3330 or
3330-11. Specify a 3344 as a 3340-70, and specify 3350 for a 3350 operating in native mode (as opposed to compatibility mode).
The DASD Dump Restore (DDR) program, executing in a virtual machine,
uses I/O DIAGNOSE 20 to perform I/O operations on tape and DASD
devices. DDR under CMS requires that the device type entered agree with
the device type of the real device as recognized by VM/SP. If there is a
conflict with device types, the following message is issued:
INVALID OPTION
However, if DDR executes standalone in a virtual machine, DDR uses
DIAGNOSE 20 to perform the I/O operation if the device types agree. If
the device types do not agree, an error message is issued.
volser
is the volume serial number of a DASD device. If the keyword
"SCRATCH" is specified instead of the volume serial number, no label verification is performed.
altape
is the address of an alternate tape drive.
If multiple reels of tape are required and "altape" is not specified, DDR
types the following at the end of the reel:
END OF VOLUME CYL xxx HD xxx, MOUNT NEXT TAPE
After the new tape is mounted, DDR continues automatically.
SKIP[~n]
forward spaces nn files on the tape. nn is any number up to 255. The SKIP
option is reset to zero after the tape has been positioned.
MODE
[~;~~]
800
38K
Mode 38K, 6250, 1600, or 800 causes all output tapes that are opened for
the first time and at the load point to be written or read in the specified
density. All subsequent tapes mounted are also set to the specified density.
4-6
VM/SP Operator's Guide
/
If no mode option is specified, then no mode set is performed and the density
setting remains as it previously was unless the tape is positioned at the load
point. When this occurs, the density setting resets to 1600 (the default
value).
MODE also specifies the recording density for the 3480 magnetic tape subsystem. Since the 3480 records only at a density of 38K, the default for the
MODE option of the 3480 is 38K.
REWIND
rewinds the tape at the end of a function.
UNLOAD
rewinds and unloads the tape at the end of a function.
LEAVE
leaves the tape positioned at the end of the file at the end of a function.
COMPACT
causes the output tape to be in a compact format, which uses less tape space
than standard format. DDR stores data in a compact format by compressing
strings of duplicate data into a smaller amount of space and reducing the
amount of space necessary to represent the characters in the data. This
option is valid only on the OUTPUT control statement for the DUMP functions.
You can use tapes in the compact format as input to the RESTORE, COPY,
PRINT, and TYPE functions without any changes. For more information,
refer to the following sections about function statements.
Notes:
1.
When the wrong input tape is mounted, the message DMKDDR 709E is displayed and the tape will rewind and unload regardless of options REWIND,
UNLOAD, or LEAVE being specified.
2.
If DDR is executed from CMS, failure to attach the tape drive or the disk
device (or both) to your virtual machine prior to invoking the input/output
statement causes the following response to be displayed:
INVALID INPUT OR OUTPUT DEFINITION
SYSPRINT Control Statement
The SYSPRINT control statement describes the device that output is to be sent to.
If the SYSPRINT CONS option is specified, the output is directed to the console
for both the CMS environment and the standalone DDR virtual machine.
In the CMS environment, all output is directed (by default) to OOE, unless the
SYSPRINT CONS option is specified. Any SYSPRINT cuu option specification is
ignored.
Section 4. VM/SP Service Programs
4-7
~------------------------------------------------------------------------------~
In the standalone DDR virtual machine, the output is directed to the output device
specified by the SYSPRINT cuu option. If the SYSPRINT CONS option is specified, all output is directed to the console. If no options are specified, the output is
directed (by default) to OOE.
cuu
specifies the unit address of the device.
CONS
specifies the console as the output device.
Function Statements
The function statements tell the DDR program what action to perform. The function commands also describe the extents to be dumped, copied, or restored. The
format of the DUMP/COPY/RESTORE control statement is:
-
roo
DUmp
COpy
REstore
[FIr I)
block]
cylJ
[To)
[To)
[block2
[cyl2
[Reorder) [To) [block3 ))
[Reorder) [To) [cy13 ))
{ChOI
ALL }
...
NUcleus
IThe FTr operand is valid with only the DUMP control statement.
DUMP
requests the program to move data from a direct access volume onto a magnetic tape or tapes. After the system dumps the data, if you specified the
COMPACT option on the OUTPUT control statement, you will receive
these messages:
IBYTES IN
BYTES OUT
ITRACKS NOT COMPACTED ON TAP"""E--IBLOCKS NOT COMPACTED ON TAPE - - - The format of the tape depends on the type of the direct access volume.
Note: If you specify dump with the COmpact option, the system dumps the
data in an FTR format.
4-8
VM/SP Operator's Guide
§eU'"vnce
pU'"«J)~U'"~ms
--,
Ffr
requests an output tape format of variable unblocked records. The size of
the records and the number of records per track written to the tape depend
on the density of the tape.
The option can be used for those devices supporting the full-track-read
(FTR) feature (3330, 3340, 3350, 3375, and 3380) and for FBA devices
(FTR is the default for 3375 and 3380 and, therefore, need not be specified).
If FTr is specified on the DUMP control statement for a count-key-data
DASD but the control unit does not support the feature, a message is written
and the operation proceeds with data written in the old format.
For count-key-data direct access volumes, the data is moved cylinder by cylinder.
The format of the resulting tape is:
Non-FfR format:
Record 1
A volume header record, consisting of data describing the volumes.
Record 2
A track header record, consisting of a list of count fields to restore the
track, and the number of data records written on tape. After the last
count field, the record contains key and data records to fill the 4K
buffer.
Record 3
Track data records, consisting of key and data records packed into 4K
blocks, with the last record truncated.
Record 4
Either the end-of-volume (EOV) or end-of-job (EOJ) trailer label. The
end-of-volume label contains the same information as the next volume
header record, except that the ID field contains EOV. The end-of-job
trailer label contains the same information as record 1 except that the
cylinder number field contains the disk address of the last record on tape
and the ID field contains EOJ.
FfR format:
Record 1
The same as described for the non-FTR format.
Record 2
A track header record, consisting of fields containing the length of the
track, the density of the tape, and the number of count fields in the track
followed by the track contents to fill the record, making it the same
length as Record 3.
Section 4. VM/SP Service Programs
4-9
Service
Programs
________________________________
~
~
________________________________________==:=J
Record 3
Track data records, consisting of count-key-data records in 8K, 12K, or
48K blocks for 800, 1600, or 6250 BPI respectively, or 48K blocks for
the 3480 Tape Subsystem. The last block, in all cases, is a short block.
Record 4
The same as described for the non-FTR format.
For FBA devices, the data is moved in sets of blocks. Each set contains 95
(or less for the last set) blocks of data. Any number of blocks can be moved
with one DUMP statement. The format of the resulting tape depends on
whether the FTR option is used or not, and/or the density of the output
tape.
Record 1
The same as described for the non-FTR format of CKD devices.
Record 2
A data header record. This consists of data that describes the set of
blocks that follow (such as block numbers and the number of tape
records required to hold these FB-512 blocks). Following the control
data are the actual FB-512 blocks filling out the tape record.
Record 3
FB-512 data records. These contain the rest of the blocks making up
the set.
Record 4
The same as described for the non-FTR format of CKD devices.
In non-FTR format the record length of record 2 and record 3 is 4K bytes.
For FTR or compact formatted tapes, record length is 8K, 12K, or 48K
blocks for 800, 1600, or 6250 BPI respectively, or 48K blocks for the 3480
Tape Subsystem. The last block, in all cases, is a short one.
COpy
requests the program to copy data from one device to another device of the
same or equivalent type. Data may be recorded on a cylinder or block basis
from input device to output device. A tape-to-tape copy can be accomplished only with data dumped by this program.
You may use a tape in compact format as input. For a tape-to-tape copy,
the output tape will be in the same format (compact or standard) as the input
tape. The COMPACT option on the OUTPUT control statement is not
valid for the COpy function. If it is specified, the system displays the following message:
'COMPACT OPTION IGNORED FOR COPY OPERATIONS'
Note: You cannot copy between FBA and count-key-data devices.
4-10
VM/SP Operator's Guide
§errVllce ~3r(D~r~lru1lS
c--·---------------------------------------·------------_..-.-._-------------------..---------._--.. -...
RESTORE
requests the program to return data that has been dumped by this program.
Data can be restored only to a DASD volume of the same or equivalent
device type from which it was dumped. It is possible to dump from a real
disk and restore to a mini disk as long as the device types are the same.
You can use a tape in compact format as input. DDR checks if the input is
in compact format, and expands the data back to standard format, if needed.
You do not need to specify anything to the program about the tape format.
After the system restores the data, you will receive the following message:
'BYTES RESTORED
cyll ITOI (cyl2 IREORDERIITOllcyl311
Only those cylinders specified are moved, starting with the first track of the
first cylinder (cyll), and ending with the last track of the second cylinder
(cyI2). The REORDER operand causes the output to be reordered, that is,
moved to different cylinders, starting at the specified cylinder (cyI3) or at the
starting cylinder (cyl1) if cy13 is not specified. The REORDER operand
must not be specified unless specified limits are defined for the operation;
the starting and, if required, ending cylinders (cyll and cyl2) must be specified. If the input device cylinder extents exceed the number of cylinders
specified on the output device, an error message results.
blockliToliblock2 IREORDERIIToliblock311
only the specified blocks are moved, starting with the first block, up to and
including the last block. The REORDER operand causes the data to be
moved to a different DASD location. The REORDER operand must not be
specified unless limits are defined for the operation. If the input block
extents exceed the capacity of the output device, an error message results.
CAUTION:
The REORDER operand is intended to be used to move minidisks to new
locations; it is not intended for re-Iocating non-minidisk cylinders. To
understand the difference, consider a 10 cylinder minidisk. Its cylinders are
numbered 0-9 and the count fields of its records refer to cylinders 0-9. Furthermore, if the minidisk contains location-dependent data, then references
to cylinders 0-9 will be hidden within that data. When REORDER is used to
move the minidisk to a new reallocation, the minidisk cylinders are nevertheless still regarded as being cylinders 0-9 and there is no need to change
the cylinder numbers in the count field of the records. On the other hand,
when moving non-minidisk cylinders, one would generally want the count
fields of the moved records to reflect the new cylinder addresses; but
REORDER keeps the original cylinder numbers in the count fields.
CPVOL
specifies that cylinder 0 for count-key-data devices, or blocks 0-16 for FBA
devices including all active directory, override space, and permanent disk
space are to be copied, dumped, or restored. This indicates that both source
and target disk must be in CP format, that is, the CP Format/Allocate
program must have formatted them.
Section 4. VM/SP Service Programs
4-11
Service PU"ogrrams
ALL
specifies that the operation is to be performed on the entire DASD volume,
either cylinders or blocks. This operation is not valid for alternate track cylinder assignments on some devices. (See "Restrictions".)
Note: The occurrence of message DMKDDR705E (issued upon completion
of the copy restore or dump operation) indicates that an attempt was made
to copy, restore, or dump the contents of cylinders beyond the extents of the
designated minidisk.
NUCLEUS
°
specifies that record 2 on cylinder 0, track and the nucleus cylinders for
count-key-data devices are dumped, copied, or restored. These are blocks
5-12 for FBA devices.
Note: To do a 'DDR RESTORE NUC', the tape must be created by 'DDR
DUMP NUC'. Otherwise, the system would issue message DMKDDR 723E.
Restrictions
1. Each track processed by this utility must have a valid home address on it containing the real cylinder and track location. Even when restoring and copying
data to a track it must have a pre-existing home address on it.
2.
Each track on an input DASD device must have a valid record zero on it, with
no more than eight bytes in the key and data fields of the record. Each track
on an output DASD device must also have a valid record zero on it unless that
device is 2314, 2319, 2305-1, or 2305-2.
3.
Flagged tracks are treated just as any other track for all 2314, 2319, and 2305
devices. That is, no attempt is made to substitute the alternate track data when
a defective primary track is read. In addition, tracks are not inspected to determine whether they were previously flagged when written. Therefore, volumes
containing flagged tracks should be restored to the same cylinders of the
volume from which they were dumped. The message DMKDDR715E occurs
each time a defective track is dumped, copied or restored, and the operation
continues. When ALL is specified for these device types, both the primary cylinders and the high-order cylinders normally reserved for alternate tracks are
dumped, copied, or restored.
4.
Flagged tracks on 3330, 3340, 3350, 3375, and 3380 devices are handled so
that data is transferred to or from the assigned alternate track in place of the
defective track. (For 3330, 3375, 3380 and 3350 this is accomplished automatically by the hardware of the control unit while for the 3340 and 3344 it is
accomplished through software.)
Note: Alternate track recovery for overflow records is not provided by
VM/SP for 3340s and 3344s.
The tape created by dumping one of these types of DASD devices will appear
as if it had been dumped from a defect-free device and the tape can be restored
to any device of the same type, even though that device might not have the
4-12
VM/SP Operator's Guide
same tracks flagged defective as the original device had. (The COpy function
works this way also.) If a track is flagged as defective, but has no alternate
assigned, a warning message is issued and the only data transferred is the home
address record and record zero. When ALL is specified for these device types,
only the primary cylinders are processed; the cylinders reserved for alternate
tracks are not processed except that an assigned alternate track is processed
whenever the corresponding defective track is processed. However, by specifying the cylinder range explicitly (cyll to cyl2 format), all cylinders, including
cylinders in the alternate track area, can be dumped or copied from. But these
same cylinders cannot be restored to or copied to explicitly. It is intended that
explicitly dumped cylinders in the alternate track area will be restored to
another area via the REORDER operand. The only reason the explicit
dumping and copying of cylinders from the alternate track area is allowed at all
is to facilitate conversion of 3340 and 3344 disks that were written using early
releases of VM/SP.
5.
For DASD devices other than the 3375 or 3380 (which will default to FTR
mode) that support the full track read (FTR) processing, you must specify the
option. Otherwise, the tape will be produced in the current non-FTR DDR
format of 4096 blocks. The 3330/3340 DASD devices can only take advantage of the full track read feature only when the 3830 or 3880 has microcode
supporting either 3344 or 3350.
6.
The system cannot use tapes created by the DDR DUMP function, which are
in compact format, as input to earlier levels of DDR.
7.
The 8809 tape drive may not operate efficiently in streaming mode while DDR
is processing data in the compact format.
Example:
INPUT 191 3330 SYSRES
OUTPUT 1802400 181 (MODE 800
SYSPRINT OOF
DUMP FTR CPVOL
INPUT 130 3330 MINI01
DUMP 1 TO 50 REORDER 51
6070 101
This example sets the density to 800 BPI, then dumps all pertinent data from the
volume labeled SYSRES onto the tape that is mounted on unit 180. If the program
runs out of space on the first tape, it continues dumping onto the alternate device
(181). A map of the dumped cylinders is printed on unit OOF while the program is
dumping. When the first function is complete, the volume labeled MINIO 1 is
dumped onto a new tape. Its cylinder header records are labeled 51 to 100. A map
of the dumped cylinders is printed on unit OOF. Next, cylinders 60 to 70 are
dumped and labeled 101 to 111. This extent is added to the cylinder map on unit
OOF. When the DDR processing is complete, the tapes are unloaded and the
program stops.
I
~
If cylinder extents are being defined from the console, the user need only enter
DUMP, COpy or RESTORE on the command line. The following is displayed:
Section 4. VM/SP Service Programs
4-13
ENTER CYLINDER EXTENTS
ENTER:
For any extent after the first extent, the message
ENTER NEXT EXTENT OR NULL LINE
ENTER:
is displayed. The user may then enter additional extents to be dumped, restored, or
copied. A null line causes the job step to start.
Notes:
1.
When a cylinder map is printed on the virtual printer (OOF as in the previous
example) a heading precedes the map information. Module DMKDDR controls the disk, time and zone printed in the heading. Your installation must
apply a local modification to DMKDDR to ensure that local time, rather than
GMT (Greenwich Mean Time), is printed in the heading.
2.
Attempts to restore cylinders or blocks beyond the capacity that had been
recorded on the tape produces a successful EO}, but the printout only indicates
the last cylinder or block found on the tape.
PRINT/TYPE Function Statement
Use the PRINT and TYPE function statement to print or type (display) a
hexadecimal and EBCDIC translation of each record specified. The first of a
group of PRINT or TYPE statements must be preceded by an INPUT statement
defining either a direct access device or a tape. The output is directed to the
system console for the TYPE function, or to the SYSPRINT device for the PRINT
function. (This does not cause redefinition of the output unit definition.) PRINT
and TYPE may be used to display the contents of any track including those in the
alternate track cylinders. For 3330, 3340, and 3350 devices, the following is displayed when alternate tracks are involved:
Wh'en displaying a defective track that has a properly assigned alternate, the
home address record displayed is taken from the defective track while record
zero and all other records are taken from the alternate. The "defective" flag,
visible in the displayed home address, is the only hint that this is not a normal
track.
When displaying a flagged defective track which does not have a proper alternate, only the home address record and record zero are displayed, and they are
both taken from the defective track.
When displaying an· alternate track explicitly, all data displayed is from that
track.
You can use a tape in compact format as input to the PRINT or TYPE function.
The input format does not affect the printed or displayed output. You do not need
to specify anything to the program about the format of the input tape.
4-14
VM/SP Operator's Guide
J
The format of the PRINT/TYPE control statement is:
PRint
TYpe
cyll [ hhl [ rr 1 .11 [To cyl2 [ hh2 [ rr2
III
[(Options... [)]]
blockl [ To block2 ]
Options
[Hex] [Graphic] [Count]
cyll
is the starting cylinder.
hhl
is the starting track. If present, it must follow the cyU operand. The default
is track zero.
rrl
is the starting record. If present, it must follow the hhl operand. The
default is home address and record zero.
TO cyl2
is the ending cylinder. If more than one cylinder is to be printed or typed,
"TO cyl2" must be specified.
hh2
is the ending track. If present, it must follow the cyl2 operand. The default
is the last track on the ending cylinder.
rr2
is the record ID of the last record to print. The default is the last record on
the ending track.
blockl
is the starting FB-S12 block number.
To block2
is the ending block number. If more than one block is to be printed or typed,
"To block2" must be specified.
Options:
HEX
prints or displays a hexadecimal representation of each record specified.
GRAPHIC
prints or displays an EBCDIC translation of each record specified.
Section 4. VM/SP Service Programs
4-15
COUNT
prints or displays only the count field for each record specified. The
COUNT option is ignored for FBA data.
Usage Note
If the TYPE statement follows the occurrence of error message DMKDDR705E
and specifies the same cylinder, track, and record extents indicated in the error
message, the contents of the printed record must be interpreted in the context of
the I/O error information given in the initial message.
Examples:
°
PRINT TO 3
Prints all of the records from cylinders 0, 1, 2, and 3.
PRINT 0 1 3
Prints only one record, from cylinder 0, track 1, record 3. Count-key-data devices
print only one record; from cylinder 0, track 1, record 3; FBA devices print blocks
to 3.
°
PRINT 1 10 3 TO 1 15 4
Prints all records starting with cylinder 1, track 10, record 3, and ending with cylinder 1, track 15, record 4.
The example in Figure 4-1 on page 4-18 shows the information displayed at the
console (TYPE function) or system printer (PRINT function) by the DDR
program. The listing is annotated to describe some of the data fields.
The printed output for FBA data is also provided by the DDR program. The
program first prints a heading that lists the block number, and then prints the 512
bytes of data in the block.
Responses
ENTER CYLINDER EXTENTS
ENTER:
-- or --
ENTER BLOCK EXTENTS
ENTER:
These messages are received only if you are entering input from your terminal:
END OF VOLUME CYL xxx HD xx, MOUNT NEXT TAPE
-- or --
END OF VOLUME BLOCK xxxxxxxx, MOUNT NEXT TAPE
DDR continues processing, after the mounting of the next tape reel.
RESTORING volser
vo 1ser
is the volume serial number of the disk dumped. The RESTORE operation
has begun.
4-16
VM/SP Operator's Guide
c
COPYING volser
vo I ser
is the volume serial number described by the input unit. The COPY operation has begun.
DUMPING volser
vo I ser
is the volume serial number described by the input unit. The dumping operation has begun.
PRINTING volser
vo I ser
is the volume serial number described by the input unit. The PRINT operation has begun.
END OF DUMP
The DUMP operation has ended.
END OF RESTORE
The RESTORE operation has ended.
END OF COPY
The COpy operation has ended.
END OF PRINT
The PRINT operation has ended.
END OF JOB
All specified operations have completed.
Enter:
Prompts input from the terminal. A null line (Press the Enter key or equivalent)
causes control to return to CMS, if the virtual machine is in the CMS environment.
In addition to the above responses, other informational messages that call for a
response by the operator to continue, terminate, or reinitiate the current operation
are described in VM/SP System Messages and Codes.
For return codes associated with DDR messages, see the VM / SP CMS Command
and Macro Reference.
Section 4. VM/SP Service Programs
4-1 7
Service Progwams
Home Addrcss-f-_--l
Rc~ord 0
ICYL 019 HD 00 I HOME
I
ADDRESS 0000130000
I RECORD
ZERO 0013000000
I
00
I ODDS I 00000000 FFF F F FF F I
"':'.':.
Cylindcr and hcad
idcntlfi~ation for
Record 0
Homc Addrcss of track
in hcxadc~imal format
Record 0 ID from thc
count field
ro-I-fth~a
I •
I
(hexadccimal)
~
_
040961000 DATA LENGTi-l-..
_---------
Data
(hexadecimal)
I.
Key Data
Len
Length
( hexadecimal)
1;;;11
00 is :;zc;:;-
"
--
l
A heading is printed containing the
data length from the count field first in
decimal. then in hexadecimal
• The data is then printed in hcxadedmal
with graphic interpretation to thc right
~ts~n~. _ _ _
I
I
J
00000 0000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
SUPPRESSED CHARACTERS SAME AS ABOVE ...
1st Halfof-+---_CYL 019 HD 00 REC 002 COUNT 0013000002 00 09AS
Record ~
02472 09AS DATA LENGTH
NIII£'." Data Length field rcpeatcd
in hcading.
00000 0000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
SUPPRESSED CHARACTERS SAME AS ABOVE ...
ABOVE RECORD WRITTEN UStNG RECORD OVERFLOW
0
r:::;--------,
0
I'
I
This statcment indicatcs that this portion
of Record 2 was written using thc Write
Special Count. Kcy, and Data command. Thc
remainder of Record 2 is found on the ncxt
,
track as the first rccord after Rccord O.
L _____
Ilome Address
Rccord (J
__ -1
____ CYL 019 HD 01 HOME ADDRESS 0000130001 RECORD ZERO 0013000100 00 OOOS 00000000 00000000
CYL 019 HD 01 REC 002 COUNT 0013000102 00 0 6 5 S - - - - - - - - - - - - - - - - - - - '
~nd lIalf of
Rccord ~
01624 065S DATA LENGTH
00000 0000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
SUPPRESSED CHARACTERS SAME AS ABOVE ...
e
I
I
~-------
e
RCl'ord3 - - + - - - - C Y L 019 HD 01 REC 003 COUNT 0013000103
II the key length Ileid
•
•
IS
not ICro
A heddlng IS prllllcd containing thc kcy length
first in dccimal. thcn in hexadecimal.
The kcy is thcn printed in hcxadccimal with
graphic interpretation to the right (not shown here).
~ ________
JT
-,
I
I
...J
SO OFSO
0012S OOSO KEY LENGTH - ' - - - - - - - - -
00000 0000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
SUPPRESSED CHARACTERS SAME; AS ABOVE
0396S OFSO DATA LENGTH
00000 0000 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
SUPPRESSED CHARACTERS SAME AS ABOVE ...
Record - I - - i - - - - . l - CYL 019 HD 01 REC 004 COUNT 0013000104 00 0000
END OF FILE RECORD
r::--------,
IICD
L
Figure
4-18
4-1.
Whenever the data length fi.:ld is zero
an end-of-file prints next.
_ _ _ _ _ _ _ ....J
Annotated Sample of Output from the TYPE and PRINT Functions of the DDR Program
VM/SP Operator's Guide
I
I
§ervnce
PnQ)~[j"arullS
J
I
Maintaining the 3800 Image Library
The Image Library
The 3800 image library contains library character sets (LCSs), graphic character
modification modules (GRAPHMODs), character arrangement tables (CATs),
forms control buffers (FCBs), and copy modification modules (COPYMODs).
These modules are selected and loaded into the 3800 printer based on the parameters specified on the CP START command and control information associated with
each spool file. The 3800 modules are created by the GENIMAGE command and
are stored in the 3800 image library by the IMAGE LIB and IMAGEMOD commands.
Both 3800 model! and 3800 model 3 modules may be stored in the same 3800
image library. The 3800 image library must contain the CAT and any
GRAPHMODs referenced by that CAT for each character set that is to be used.
For the 3800 model 3 only, any LCSs referenced by the CAT must also be loaded
in the 3800 image library.
The GENIMAGE Service Routine
The GENIMAGE command invokes the IEBIMAGE program to create the CATs,
LCSs, GRAPHMODs, FCBs, and COPYMODs. These are created on a CMS disk
with a file type of TEXT.
1
1 1
1
J 1
J
GENIMAGE rfn
[it
[fm [sfn
[sft
[sfm
lSYSINj I!ILEJ. ~
SYSPRINTj LISTIN~ At
J
fn
specifies the filename of the input control file. The name may be up to eight
alphameric characters in length. If not specified, the default filename is
SYSIN.
ft
specifies the filetype of the control file. Filetype may be up to eight alphameric characters in length. If not specified, the default file type is FILE.
fm
specifies the filemode of the control file. Filemode may be up to two alphameric characters in length. If not specified, the default filemode is *.
sfn
specifies the filename into which the messages listing is placed. If not specified, the default file name is SYSPRINT.
Section 4. VM/SP Service Programs
4-19
Service Programs
sft
specifies the filetype for the messages listing. If not specified, the default file
type is LISTING.
sfm
specifies the filemode for the messages listing. If not specified, the default
file mode is AI.
Figure 4-2 and Figure 4-3 list the "starter set" of character sets supplied by IBM.
They are supplied with a filename and filetype of XTB1xxxx TEXT.
AN
AOA
AOD
AON
All
BOA
BON
DUMP
FM10
Figure
4-2.
4-3.
GU10
GU12
GU15
G11
HN
H11
KN1
OAA
OAB
ODA
ONA
ONB
PCAN
PCHN
PN
P11
QN
QNC
RN
5N
TN
TU10
T11
XN
YN
2773
2774
VM-Supplied Character Arrangement Tables for the 3800 Modell and Model
3 Printers
AE10
BITR
BRTR
CE10
CE12
COlO
CR10
DOTR
EBTR
Figure
FM12
FM15
GF10
GF12
GF15
GN
G510
G512
G515
EITR
E5TR
GB10
GB12
GF10
GF12
GF15
GFC
G112
GP12
GR10
G5C
GT10
GT12
GT15
GUC
LB12
LR12
OB10
OR10
PB12
P 112
PR10
PR12
RT10
5B12
5110
5112
5012
5R12
5T10
5T12
5T15
VM-Supplied Character Arrangement Tables for the 3800 Model 3 Only
Responses
Completion messages and codes are returned to the user in the file indicated by the
sfn and sft operands of the GENIMAGE control statement. The GENIMAGE
command produces two outputs: the listing file (sft above) and a TEXT deck for
each NAME statement in the input deck. Figure 4-4 on page 4-21 contains the
filemode prefixes of these TEXT decks.
4-20
VM/SP Operator's Guide
[
3800 Model
LCSl
LCS
CAT
GRAPHMOD
FCB
COPYMOD
Figure
4-4.
XTBl
GRAF
FCB3
MODl
3800 Model 3
LCS2
XTBl
GRF2
FCB3
MODl
TEXT Deck Names Output by the GENIMAGE Command
For a description of control statements in SYSIN FILE, messages and codes in
SYSPRINT LISTING, and the character sets listed above, see the following documents:
OS/VS IBM 3800 Printing Subsystem Programmer's Guide for the 3800 Model
1
IBM Printing Subsystem Model 3 Programmer's Guide: Compatability for the
the 3800 Model 3.
The IMAGELIB Service Routine
Before issuing the IMAGELIB command, you must create a control file whose
filename is the same as the IMAGELIB you want to build, and whose filetype is
CNTRL. The format of this file is one statement per TEXT deck you want
included, with the names in columns one through eight.
IMAGE LIB reads the control file containing the list of teAt images. The files are
then loaded into the specified named system. The format of the IMAGELIB
control statement follows.
IIMAGELIB IimageIib
image lib
specifies the named system that is being created or replaced.
The IMAGEMOD Command
The IMAGEMOD command allows an installation to make changes to 3800 named
systems without the need for generating a completely new named system from
scratch. It uses the current contents of the 3800 Named System and adds, deletes,
or replaces only those members that you change. The format of the IMAGEMOD
command is as follows:
Section 4. VM/SP Service Programs
4-21
IMAGEMOD
(GENj
,
ADD imagelib [modname [modname] ••. ]
I~
REP
>
I
J
DEL
\ MAP
imagelib [ (Options) ]
[~i~~Js
PRINT
DISK
GEN
specifies the creation from scratch of the new 3800 named system including
the specified module names.
ADD
specifies the addition of the indicated module to an already existing 3800
named system.
REP
specifies the replacement of the indicated member with a new version of that
member in an already existing 3800 named system.
DEL
specifies the deletion of an existing member from an already existing 3800
named system. This option causes the named system to then compress so
that all members are contiguous with one another.
MAP
specifies the creation of a map of the 3800 named system on the user's terminal, virtual printer, or CMS disk. This map gives the member name, relative byte displacement in both decimal and hexadecimal, and the byte size of
the member in both decimal and hexadecimal.
Note: The MAP function does not require the specification of any module
names.
TERM
specifies that the map of the 3800 named system be written to the
user's terminal.
PRINT
specifies that the map of the 3800 named system be written to the
user's virtual printer.
DISK
specifies that the map of the 3800 named system be written to a CMS
disk file called "libname MAP A5".
4-22
VM/SP Operator's Guide
/
§erVDce Programs
libname
specifies the name of the 3800 named system to be operated upon. Unless
the GEN operand is specified, this named system must have been created
with at least one member by the IMAGELIB command and must be defined
by the installation administrators via the NAME3800 macro in the
DMKSNT module.
modname
specifies the name of a module (CAT, LCS, GRAPHMOD, FCB,
COPYMOD) that is to be added, replaced or deleted. One to 10 modules
may be specified on one IMAGEMOD command.
Device Support Facilities
Device Support Facilities is a program used with IBM operating systems to perform
various operations on direct-access storage devices. It replaces IBCDASDI,
INITDISK, and SURF ANAL in that it can:
Initialize direct-access storage volumes so that they can be used in OS/VS or
DOS/VSE systems
Inspect a volume for defective tracks
Reformat the volume label, and IPL bootstrap and program records
Provide an analysis function for nonremovable DASD, both CKD and FBA.
The CMS FORMAT command initializes mini disks for CMS. The CP format
program formats CP-owned volumes, such as the system residence, paging, and
spooling volumes.
For information on disk initialization, refer to the Device Support Facility User's
Guide and Reference. This manual describes procedures for initializing, formatting,
and analyzing disk space.
Section 4. VM/SP Service Programs
4-23
L __________________________________________.____________________________________
~
Formatting Volumes--General Information
Disk Initialization
Disk initialization and alternate track assignment should be performed by the
Device Support Facility.stand-alone utility which is on the CMS system disk with a
fileid 'IPL DSF S2'.
Disk Formatting
All direct access volumes used by the VM/SP system (for paging, spooling, system
residence, directory, or temporary disk allocation) must be properly labeled, formatted, and allocated. The CP Format/Allocate service program (DMKFMT
module) prepares disks for use by CPo A CMS Format program is also available
and must be used to format CMS and RSCS disks.
All direct access volumes (including both real disks and minidisks) to be used by
VSAM under CMS, as, or DOS, must be formatted by the device support facility
utility.
If certain information in the OS Format 4 label on track 0 cylinder 0 is destroyed,
no additional alternate tracks can be assigned by the device support facility until
the volume is reformatted by the device support facility. When either the device
support facility or the CP Format/Allocate program (IPL FMT) is used to format a
volume, this information in the OS Format 4 label is preserved. But, when the
eMS Format Program formats the volume (or when it formats a minidisk whose
origin is cylinder 0 of the volume) the as Format 4 labelis destroyed and no
further alternate tracks can be assigned until the device support facility is used to
reformat the volume.
Format/Allocate Service Program (DMKFMT)
The Format/Allocate service program formats, allocates, and labels direct access
volumes for paging, spooling and CP file residence. This service program is executed as part of CP system generation procedures and may also be executed as a
stand-alone program to:
Format direct access volumes for CP use
Allocate specific disk areas to particular functions or to CP use
•
Write six-character volume serial number labels.
Note: The Format/Allocate program should be used with care since it destroys
existing data (if any). Also, user minidisks and temporary minidisks must not begin
4-24
VM/SP Operator's Guide
/
Serrvuce Prrogrrams
on real cylinder zero of CP-owned volumes, because information critical to CP is
stored in that cylinder.
An object deck version of the CP Format/Allocate service program is a standalone
program and can be loaded from a virtual or real card reader into a virtual or a real
machine. (If run in a virtual machine, the virtual machine must have write access to
the volume being formatted.) The program accepts control statements from the
operator's system console (commands) or from the IPL device (card reader).
Note: Messages DMKFMT736E and DMKFMT735E may be issued if an avail-
able path to the device cannot be found after an appropriate number of retries.
High activity may cause this situation.
Cylinders used by CP for paging, spooling, and so on, must be preformatted with
fixed length unblocked records of 4096 bytes.
It is important to note the differences in the terminology for fixed-block DASD
devices (for example, 3310) and count-key-data DASD devices (for example,
3330).
The basic unit of DASD space used by the VM/SP system is the page. A page
contains 4096 bytes of data. On count-key-data devices, a page is recorded in a
DASD record data area that is 4096 bytes in size. The count area of this record
indicates the location (by cylinder and track) and size (4096 bytes) of the recorded
page. The preformatting of count-key-data volumes initializes the count areas and
sets the data areas to zero.
On FBA devices, a page is recorded in eight successive blocks of data. Each block
contains 512 bytes of a page. Control data, such as the count area for count-keydata, is not required. Pre formatting FBA volumes sets all the pages to zero.
The unit of formatting or allocation for count-key-data devices is the cylinder.
After you decide how many pages of DASD space are required for system operation, convert the number to a corresponding number of cylinders for communication with DMKFMT.
Device capacity and count-key-data DASD capacities when formatted for CP use
(4096-byte pages) are:
2314/2319
3330
2305
3340
3350
3375
3380
32 pages/cylinder
57 pages/cylinder
24 pages/cylinder
24 pages/cylinder
120 pages/cylinder
96 pages/cylinder
150 pages/cylinder
(8 pages/5 tracks)
(3 pages/track)
(3 pages/track)
(2 pages/track)
(4 pages/track)
(8 pages/track)
(10 pages/track)
The format operation writes 4096-byte blocks on all cylinders being formatted.
The s"ervice program does write-checking to verify that parts of the track are not
defective. A count is maintained of pages with read check errors detected during
Section 4. VM/SP Service Programs
4-25
the format operation. At the completion of the format operation, the count of the
pages with read check errors is printed.
For FBA devices, the unit of formatting or allocation is the page.
The format operation writes 4096-byte blocks on all cylinders being formatted.
Fixed block device capacity when formatted for CP use is:
3310
3370, Al or Bl
3370, A2 or B2
15752 pages per spindle
69750 pages per spindle
89094 pages per spindle
Format for All Cylinders Except Cylinder Zero
For example, the 3330 track format for all formatted cylinders except cylinder 0 is
shown in Figure 4-5.
Track 0:
RO
8 bytes
R1
I I
4096 bytes
R2
I I
4096 bytes
R3
I I
4096 bytes
Track 1 to Track 18:
RO
8 bytes
Figure
4-5.
R1
I I
4096 bytes
R2
I I
4096 bytes
R3
I I
4096 bytes
Format of 3330 Cylinders for Use by CP
Format for Cylinder Zero
All volumes containing space for CP use (paging, spooling, and so on) must have a
properly formatted cylinder o. The only service program that can do this is the
Format/ Allocate program (DMKFMT).
Cylinder 0 is formatted like other cylinders except that the space associated with
the first three 4096-byte blocks is reserved for system use. This area is then formatted as illustrated in Figure 4-6 on page 4-27.
4-26
VM/SP Operator's Guide
§ell"VDCe
~:Dr(!)gU'"ams
r=---------------.-..-----------------.------------.------.-------.-..-.-..- -----.--.----..-.-.. - . . . -..--- ....--.--.--.-..----:J
Track 0
RO
8
R4
Rl
R2
KEY
[ill]
rec
DMKCKP
Module
V
24
4096
KEY
Allocation
Byte map
1024
KEY
R5
Format
4 DSCB
44
96
Format
5 DSCB
44
0
VOLID
4
80
L
1
R6
96
R3
[::J
FR
Track 1 to Track 18 are the same as the nonzero cylinders.
Figure
4-6.
3330, 3340, or 3350 Cylinder 0 Format
The contents of each record in cylinder 0 track 0 are as follows:
RO
Nothing.
Rl
IPL record. Puts the system into wait state if storage volume is loaded
before CP nucleus is built.
R2
Checkpoint record. Used by CP to save and retrieve information for a warm
start.
R3
Volume label. Same as OS VOLllabel. On CP system residence volume,
area in data record marks the beginning of the system directory. A label is
automatically written when cylinder 0 is formatted. The owner field of the
label record contains "CP370" if there is allocation data present in R4.
R4
Allocation Byte Map. Each byte identifies a cylinder and specifies its usage
(paging, spooling, directory, and so on). This map is filled in by the ALLOCATE function of the DMKFMT service program.
RS
Format 4 OS DSCB type label. For compatibility with OS. Also, the device
support facility program uses this label to keep a record of how many alternate tracks remain available for assignment on this disk. The
Format/ Allocate program will preserve this information by first reading it
from any existing Format 4 label, and then writing it back in the new label.
Section 4. VM/SP Service Programs
4-27
§GL"~.fDCe P~"(Ql9J U"8)u-UOS
r-:--
........................... _. __::......_-._._-.._-_-_._-._-..~__ .______._ ..__=._=_--~.-==--====--==-._:_=_ _=.:~=~=~~==:=:.:__=_===~~._=~_
_:_=____==___=::J
R6
Format 5 OS DSCB type label. For compatibility with OS. Label indicates
to OS that no space is available on this volume.
FR
Is one or more filler records.
Format for CP-Owned FBA Devices
Each block is 512 bytes long. The first 16 blocks (pages 0 and 1) are reserved for
system use.
IPL
REC
o
VTOC Allocation
Extent Map
1
2
3-4
DMKCKP RESERVED
5-12
13-15
PAGES
2-X
16-N
Each block contains:
o
IPL record. Places the system into the wait state if IPL occurs before the CP
nucleus is built.
,/
1
Volume label. Same as DOS/VSl VOLllabel. On CP system residence
volumes, a field in this record contains a pointer to the page where the
system directory starts. A label is automatically written when the format
function is used, specifying page 0 as the starting page. The owner field of
the label record contains "CP370" if there is allocation data in blocks 3 and
4.
2
Volume Table of Contents. The Format 4 and Format 5 DSCB's. These
labels are written for compatibility with DOS/VS. They indicate that no
space is available on this volume.
3-4
Allocation Extent Map. Each entry is 12 bytes long and describes a range of
pages on the device as well as the usage of the pages (PERM, TEMP,
DRCT; or TDSK).
5-12
Checkpoint Program. The CP module (DMKCKP).
13-15
Reserved.
16-N
Contains the pages used by the VM/SP system. The area starts at block 16,
which corresponds to page 2 on the volume.
4-28
VM/SP Operator's Guide
§elJ'VDC(9 PlJ'«))~lJ'SJms
c:=-----
- - - - - - - - - - - _ . --_._-----
----.-~
Format/ Allocate program control statements may be supplied in card form via a
card reader, or may be entered at the system console. All error messages regarding
improper specification of control statements are displayed at the console.
Format/Allocate Program Card Input
Punch control statements for card input start in column 1, and each field is separated from the adjacent field by a comma. Two commas in a row cause the
insertion of a default value. Three commas in a row cause the insertion of two
default values.
Note: The only default values permitted are those that define the starting and
ending cylinders or DASD extents. The defaults are the first and last cylinders of
the volume (or pages), respectively.
Comments must be preceded by at least three blanks.
The control card entries for the Format/Allocate program must be in the following
order:
Format function
FORMAT, devadr, devtype, volser, startadr, endadr
Allocate function
ALLOCATE, devadr, devtype, volser
TEMP, startadr, endadr
PERM, startadr, endadr
TDSK, startadr, endadr
DRCT, startadr, endadr
OVRD, startadr, endadr
PAGE, startadr, endadr
DUMP, startadr, endadr
END
•
Label functions
FORMAT, devadr, devtype, volser, LABEL
FORMAT, ALLOCATE, and LABEL are Format/Allocate program control
words and may be abbreviated to one letter.
FORMAT Control Statement
The format of the FORMAT control statement is:
FORMAT, devadr, devtype, volser, startadr, endadr
Section 4. VM/SP Service Programs
4-29
devadr
is a three-digit hexadecimal number that identifies the address of the device
that the Format/Allocate program is to act upon. Valid device addresses
under CMS are '001' to '5FF' for ECMODE OFF and '001' to 'FFF' for
ECMODEON.
Note: To avoid possible I/O contention when formatting two device types
(3375/3380) in two virtual machines, do not use consecutive even/odd pairs
of device addresses (e.g., 290,291); rather, it is advisable to use consecutive
odd/even pair device addresses (e.g., 291,292).
devtype
is a four-to-seven character field that defines an approved device for the
Format/ Allocate program.
2314
2319
3330
3330-11
3340-35
3340-70
3350
2305-1
2305-2
3330
3340-70
3330
3330-11
3375
3380
FB-512
for a 3333 device
for 3340-70F or 3344 devices
for a 3350 device in 3330-1 compatibility mode
for a 3350 in 3330-11 compatibility mode
for a 3310 or 3370 device
volser
is a one-to-six character field that represents the volume serial number of the
volume you are formatting.
startadr
is the starting cylinder address, and in Basic System Extensions for countkey-data devices, this is the starting cylinder address on the DASD on which
the format function is to be performed. For fixed-block devices, this is the
starting page number. The starting address is entered as decimal digits.
endadr
is the last cylinder address, and for count-key-data devices, this is the last
cylinder address on the DASD on which the format function is to be performed. For fixed-block devices, this is the ending page number. The
ending address is entered as decimal digits.
Note: FORMAT is a control word and may be abbreviated to F.
4-30
VM/SP Operator's Guide
Seuv~ce ~:)uOg)u~uu1lS
[------_._---_.__.._-_._-------------_ ..
--_.---.-.-........ -................... --.-.. -.-..-.----.--.--.J
ALLOCATE Control Statements
The formats of the ALLOCATE control statements are:
ALLOCATE, devadr, devtype, valser
TEMP, startadr, endadr
PERM, startadr, endadr
TDSK, startadr, endadr
DRCT, startadr, endadr
OVRD, startadr, endadr
PAGE, startadr, endadr
DUMP, startadr, endadr
END
devadr
is a three-digit hexadecimal number that identifies the address of the device
that the Format/Allocate program is to act upon. Valid device addresses
under CMS are '001' to '5FF' for ECMODE OFF and '001' to 'FFF' for
ECMODEON.
devtype
is a four-to-seven character field that defines an approved device for the
Format/ Allocate program. Approved device types are specified as:
2314
2319
3330
3330-11
3340-35
3340-70
3350
2305-1
2305-2
3330
3340-70
3330
3330-11
3375
3380
FB-512
for a 3333 device
for 3340-70F or 3344 devices
for a 3350 device in 3330-1 compatibility mode
for a 3350 in 3330-11 compatibility mode
for a 3310 or 3370 device
valser
is a one-to-six character field that represents the volume serial number of the
volume you are formatting.
startadr
is the starting cylinder address on the DASD on which the format function is
to be performed. For fixed-block devices, this is the starting page number.
The starting address is entered as decimal digits.
Section 4. VM/SP Service Programs
4-31
c=== ____________________________________________________________________________
~
endadr
is the last cylinder address on the DASD on which the format function is to
be performed. For fixed-block devices, this is the ending page number. The
end address is entered as decimal digits.
TEMP
indicates that the following operands identify temporary storage space
reserved for spooling or paging activity.
PERM
defines an area that can contain the logout area, the CP nucleus, and space
that is not used by the system but is available for use by virtual machine
users (for example, for user minidisks).
TDSK
defines the temporary minidisk space available for virtual machine users
during a single terminal session on the VM/SP system.
Whenever spooling space is needed, DMKPGT allocates TEMP storage
space for spooling.
DRCT
indicates that the following space is reserved for directory files.
Note: If you wish to reallocate the system directory, it may be necessary to
rerun the system directory program (DMKDIR) because the active directory
pointer can be destroyed.
OVRD
indicates the space reserved for an override file. This file assigns classes to
CP commands to override the IBM-defined command structure. For countkey-data devices, the system only allocates the first cylinder that it finds allocated for OVRD. For FBA devices, the system will use the first extent
defined as OVRD. For FBA devices, the minimum allocation is 2 pages.
PAGE
indicates that the following operands identify DASD storage space for preferred paging activity.
Note: Cylinders which are to be used for non-preferred paging space
(spooling, overflow paging operations, and so forth) should be allocated as
"TEMP."
DUMP
indicates that the appropriate DASD space be marked reserved for CP allocation of system dumps. This reserved area will not be used for CP paging
and spooling.
Note: DUMP allocation must be contiguous since CP will look for contig-
uous cylinders or blocks when assigning the system dump spool file.
4-32
VM/SP Operator's Guide
/
c
END
the allocation termination control statement. This statement causes the allocation record to be written when this is the only control statement. The allocation record is not written back, except for FBA devices which have an
allocation record mismatch. For example, a 3370 restored to a 3310 device.
Note: ALLOCATE is a control word that may be abbreviated to its first letter, A.
TEMP, PERM, TDSK, PAGE, DUMP, DRCT, and OVRD are all functions of
ALLOCATE. These cards can follow the ALLOCATE control statement in any
sequence. Each card in turn overlays the cylinder (or allocate) table, and any space
not reallocated remains the same. If an ALLOCATE function overlays the previous allotment, then the previous cylinder space allotment is truncated to the
beginning of the next allotment. The allocation operation is executed after you
re-ipl VM/SP. For example:
Disk Storage
Allocation
First
Cylinder
Last
Cylinder
1st Entry PAGE
2nd Entry TEMP
3rd Entry PERM
4th Entry TDSK
5th Entry DRCT
6th Entry OVRD
7th Entry END
000
003
010
040
003
005
002
202
050
050
004
005
The result of this disk volume allocation is:
Disk Storage
Allocation
First
Cylinder
Last
Cylinder
PAGE
DRCT
OVRD
TEMP
PERM
TDSK
000
003
005
006
010
040
002
004
005
009
039
050
Once-an ALLOCATE control statement is encountered, all cards following it until
an END card is encountered are assumed to be part of a single allocation. The
Format/ Allocate service functions cannot be performed on another disk volume
until the END card is encountered. Any area not allocated will default to "TEMP"
space. (See Figure 4-8 on page 4-36.)
Note: Reallocation of an area containing an active VM/SP directory deallocates
the directory to allow a new directory to be written in the same area.
After reallocation, the directory program (DMKDIR) must be executed to reinitialize the directory. If this is not done, your program will abend with an
ABENDCPI002 when you initialize via IPL.
Section 4. VM/SP Service Programs
4-33
LABEL Control Statement
The format of the LABEL control statement is:
FORMAT, devadr, devtype, volser, LABEL
devadr
is a three-digit hexadecimal number that identifies the address of the device
that the Format/Allocate program is to act upon. Valid device addresses
under CMS are '001' to '5FF' for ECMODE OFF and '001' to 'FFF' for
ECMODEON.
devtype
is a four-to-seven character field that defines an approved device for the
Format/ Allocate program. Approved device types are specified as:
2314
2319
3330
3330-11
3340-35
3340-70
3350
2305-1
2305-2
3330
3340-70
3330
3330-11
FB-512
3375
3380
for a 3333 device
for 3340-70F or 3344 devices
for a 3350 device in 3330-1 compatibility mode
for a 3350 in 3330-11 compatibility mode
for a 3310 or 3370 device
volser
is a one-to-six character field that represents the volume serial number.
LABEL
is a keyword designating the label function of the Format/Allocate program.
Notes:
1.
Format of cylinder 0 is a prerequisite for the label only function.
2.
FORMAT and LABEL are control words and may be abbreviated to FORM
and L, respectively.
If the volume is to be used for paging, spooling, dump, directory, override, or temp
space, 'CP370' is required in record 3 for CKD or block 1 for FB-512. The label
4-34
VM/SP Operator's Guide
Seu·'V'oce
~3[(O~[(aUUUS
__________._. _____________...... _.. ___________ - .. _,, ________________________ .._____________________ .____________ .___ .________________________ .. _______ .________ .___ n]
option does not write it out; the FORMAT function must be used to format cylinder for CKD or page for FB-512 to insure the inclusion of the 'CP370'.
°
°
Examples:
FORMAT
FORMAT,232,3330,MYDISK,OOO,006
FORMAT,232,3330,MYDISK,,,
FORMAT,232,3330,MYDISK,,00
FORMAT,232,3330,MYDISK,OOl"
ALLOCATE
ALLOCATE,232,3330,MYDISK
PAGE,000,020
TEMP,021,150
PERM,055,060
TDSK, 100,108
OYRD,109,109
DRCT,110,120
DUMP,121,150
END
LABEL
F,232,3330,MYDISK,label
Format/Allocate Console Input
The Format/Allocate program can be controlled by control statements entered into
the real or virtual console instead of by a deck of cards containing control statements. If the program finds no control statements at the card reader, it issues a
prompting message to the console. The proper response causes the prompting
message for the next operand to appear until the Format, Allocate, or Label function is completely defined; then the Format/Allocate program is executed.
After execution, the prompting begins again until all DASD allocation requirements
are fu,lfilled.
The sequence for console typewriter processing of the Format/Allocate program
(after making the operator's console ready), is as follows:
1.
Load the card reader with a loader, followed by the Format/Allocate deck
2.
IPL the card reader
3.
Respond to the first message displayed at the system console
4.
Respond to other messages.
Section 4. VM/SP Service Programs
4-35
[
Note: If you are in a virtual machine environment, you must re-IPL CMS to exit
from the Format/Allocate program.
Following are examples of Format/Allocate program execution under CP control.
Figure 4-7 is an example of the label operation. Figure 4-8 is an example of the
allocate operation. Figure 4-9 on page 4-37 is an example of the allocate overlap
operation. All responses are entered after the colon, except for the 3270 device.
After a function is complete, the program returns and issues again the ENTER
"FORMAT" OR "ALLOCATE": statement.
VM/SP FORMAT/ALLOCATE PROGRAM
ENTER "FORMAT" OR "ALLOCATE": f
FORMAT FUNCTION SELECTED
ENTER DEV ICE ADDRESS (CUU): 131
ENTER DEVICE TYPE: 2314
ENTER START CYL INDER (XXX) OR "LABEL ": 1
ENTER DEVICE LABEL: cpdsk2
Figure
4-7.
Using the Format Program Label Function
ENTER "FORMAT" OR "ALLOCATE": a
ALLOCATE FUNCTION SELECTED
ENTER DEV ICE ADDRESS (CUU): 131
ENTER DEV ICE TYPE: 2314
ENTER DEVICE LABEL: cpdsk2
ENTER ALLOCATION DATA FOR VOLUME CPDSK2
TYPE CYL CYL
drct 000 001
ovrd 002 002
perm 004 008
page 009 070
dump 071 099
tdsk 100 150
end
ALLOCATION RESULTS
DRCT 000 001
OVRD 002 002
TEMP 003 003
PERM 004 008
PAGE 009 070
DUMP 071 099
TDSK 100 150
TEMP 151 202
DEVICE 131 VOLUME CPDSK2 ALLOCATION ENDED
Figure
4-36
VM/SP Operator's Guide
4-8.
Using the Format Program Allocate Function
S(3~"\'Jace ~-~l(O~ L"a[{lr~S
l~_.~
__.'_-~~-:.:":~_- ___ ~~ ___ . _______________. __ .__ .__ .__ .
_:::-_:~.:
___~~.. __ .~:~ __ ~_._:____ .__ ~_~ __ ~_~__________________._-___ ._'_'"_-_ . _ . _:1
ENTER "FORMAT" OR "ALLOCATE": a
ALLOCATE FUNCTION SELECTED
ENTER DEV ICE ADDRESS (CUU): 131
ENTER DEV ICE TYPE: 2314
ENTER DEVICE LABEL~ cpdsk2
ENTER ALLOCATION DATA FOR VOLUME CPDSK2
TYPE CYL CYL
perm 004004
temp 000 010
tdsk 000 010
perm 010 202
drct 000 004
ovrd 005 005
end
ALLOCATION RESULTS
DRCT 000 004
OVRD 005 005
TDSK 006 009
PERM 010 202
DEVICE 131 VOLUME CPDSK2 ALLOCATION ENDED
Figure
4-9.
Using the Format Program Allocate Overlap Function
Figure 4-10 shows the label function for FBA devices. Figure 4-11 on page 4-38
shows the allocate function for FBA devices. Figure 4-12 on page 4-39 shows the
allocate overlap function for FBA devices.
VM/SP FORMAT/ALLOCATE PROGRAM
ENTER "FORMAT" OR "ALLOCATE": f
FORMAT FUNCTION SELECTED
ENTER DEV ICE ADDRESS (CUU): 131
ENTER DEV ICE TYPE: FB-512
ENTER START PAGE NUMBER OR "LABEL":
ENTER DEVICE LABEL: cpdsk2
Figure
4-10.
1
Using the Format Program Label Function for FDA Devices
"-_/
Section 4. VM/SP Service Programs
4-37
SGH'"VDce P~"o£1raJ!rll1s
_ _._______._ _ _ _ _ _ _ _ _._ _ _._._..____.___.__J
r:=
ENTER "FORMAT" OR "ALLOCATE": a
ALLOCATE FUNCTION SELECTED
ENTER DEVICE ADDRE~S (CUU): 131
ENTER DEVICE TYPE: FB-512
ENTER DEVICE LABEL: cpdsk2
ENTER ALLOCATION DATA FOR VOLuME CPDSK2
TYPE PAGE PAGE
perm 2 1000
temp 1001 5000
drct 5001 5100
ovrd51015108
tdsk 5109 9000
end
ALLOCATION RESULTS
PERM
2 1000
TEMP 1001 5000
OVRD 5101 5108
DRCT 5001 5100
TDSK 5109 9000
TEMP 9001 15751
DEVICE 131 VOLUME CPDSK2 ALLOCATION ENDED
Figure
4-38
VM/SP Operator's Guide
4-11.
Using the Format Program Allocate Function for FHA Devices
ENTER "FORMAT" OR "ALLOCATE": a
ALLOCATE FUNCTION SELECTED
ENTER DEV ICE ADDRESS (CUU): 131
ENTER DEVICE TYPE: FB-SI2
ENTER DEVICE LABEL: cpdsk2
ENTER ALLOCATION DATA FOR VOLUME CPDSK2
TYPE PAGE PAGE
temp 2 500
page 9500 11000
end
ALLOCATION
TEMP
2
PERM 501
TEMP 1001
DRCT 5001
OVRD 5101
TDSK 5109
TEMP 9001
PAGE 9500
TEMP 11001
DEVICE 131
RESULTS
500
1000
5000
5100
5108
9000
9499
11000
15751
VOLUME CPDSK2 ALLOCATION ENDED
ENTER "FORMAT" OR "ALLOCATE": a
ALLOCATE FUNCTION SELECTED
ENTER DEV ICE ADDRESS (CUU): 131
ENTER DEVICE TYPE: FB-SI2
ENTER DEVICE LABEL: cpdsk2
ENTER ALLOCATION DATA FOR VOLUME CPDSK2
TYPE PAGE PAGE
perm
0 1000
LOWEST ALLOCATABLE PAGE IS PAGE 2 -- RESPECIFY
perm
2 17345
HIGHEST ALLOCATABLE PAGE IS PAGE 15751 -- RESPECIFY
perm
end
2 1000
ALLOCATION
PERM
2
TEMP 1001
DRCT 5001
OVRD 5101
TDSK 5109
TEMP 9001
PAGE 9500
TEMP 11001
DEVICE 131
Figure
4-12.
RESULTS
1000
5000
5100
5108
9000
9499
11000
15751
VOLUME CPDSK2 ALLOCATION ENDED
Using the Format Program Allocate Overlap Function for FHA Devices
Note that before the ALLOCATE function was invoked, cylinder 0 was formatted
and labeled CPDSK2. The area associated with the first three 4096-byte blocks on
cylinder 0 are not used for spooling but contain system information (page allocation map, label, and so on).
These CP-formatted volumes can be made usable by CP in one of two ways:
Section 4. VM/SP Service Programs
4-39
1.
They may be attached to the system by the VM/SP operator.
2.
Their volume serial numbers may appear in the SYSOWN macro in the
DMKSYS module. The CP system residence volume's serial number must
appear in the SYSOWN macro.
How To Print a CMS Dump File
Use the PRTDuMP command to print a previously created dump file under CMS
by entering:
PRTDVMP PRBnnnnn [options}
at the terminal. The message:
PRINTING FILE 'PRBnnnnn Dnn DUMP'
then occurs at the terminal. When the specified file has been formatted and
printed, one of two completion messages is displayed:
DUMP FILE 'PRBnnnnn Dnn DUMP' PRINTED AND KEPT
DUMP FILE 'PRBnnnnn Dnn DUMP' PRINTED AND ERASED
How To Print a CP ABEND Dump from Tape
When the CP abend dump is sent to a tape, the records are 132 characters long,
unblocked, with a blocksize of 132 and carriage control characters.
If the CP dump unit has been specified as a tape drive, and one or more dumps
have been placed on the tape, use the following procedure to print the dumps.
1.
Log on to the VM/SP system with any userid that has the capability of running
CMS. No other special privilege classes or options are required.
2.
Attach a tape drive to the virtual machine as address 181.
3.
Mount the tape that has the CP abend dumps.
4.
IPL the CMS system and perform the usual access requirements.
5. Issue the following CMS commands:
FILEDEF ddnamel PRINTER (RECFM VA LRECL 132)
FILEDEF ddname2 TAPI (optionC RECFM V LRECL 132)
MOVE ddname2 ddnamel
cp close prt
,/
4-40
VM/SP Operator's Guide
--------"----------,
Note: Refer to the FILEDEF command description in VM / SP CP Command
Reference for General Users.
Step 5 can be repeated for as many dumps as are on the tape. Note that the CP
dump routines write two tape marks at the end of each file. Therefore, to process
the next dump, the TAPE FSF command line must be issued to position the tape
for reading the next dump file.
Network Dump Operations
This section only applies to 3704 or 3705 communication controllers that have
been loaded by VM/SP. If you want to dump the contents of a 3725 or a 3705
that has been loaded by ACF/SSP, refer to ACF/NCP V4, ACF/SSP V3 Diagnosis Guide, SC30-3255.
If 3704/3705 operations are erratic, fatal hardware errors occur, or some other
internal error appears, the Communications Controller's storage should be dumped.
The NETWORK DUMP command dumps the contents of 3704/3705 storage for
NCP, PEP, or EP 3704/3705 control programs, if unit check or IPL required conditions are detected.
The format of the NETWORK command with the DUMP operand is:
NETWORK
DUMP raddr [IMMED]
AUTO
OFF
raddr
is the real hexadecimal address of the 3704/3705.
IMMED
is the default operand; it forces an immediate dump. The IMMED operand,
if specified, does not reload the control program. Before 3704/3705
resources can be used again, the control program must be reloaded. To
reload the control program after the NETWORK DUMP raddr IMMED
command has executed, use the NETWORK LOAD raddr ncpname
command.
If the IMMED operand is specified, a check is made to determine whether
the "IPL required" sense status is present. If it is not, the following message
occurs:
CTLR raddr IPL NOT REQUIRED; ENTER YES TO CONTINUE:
This pause in operations allows the operator an opportunity to check the
NETWORK DUMP command line before engaging or terminating the operation.
Section 4. VM/SP Service Programs
4-41
t=:____________
AUTO
causes a dump if VM/SP subsequently detects a unit check condition or
"IPL required" condition. If AUTO is specified, each time a dump is taken,
the Communications Controller is reloaded with the 3704/3705 control
program that was previously active.
OFF
resets a previously set AUTO (automatic dump) status.
Note: The dumps produced by the NETWORK command cannot be processed by
the IPCSDUMP service program. NETWORK-initiated dumps are processed by
the NCPDUMP (Network Control Program DUMP) service program created for
this task.
NCPDUMP Service Program and How To Use It
NCPDUMP applies only to dump files that were dumped with the NETWORK
DUMP command after the 3704/3705 was loaded by VM/SP.
NCPDUMP is a CMS command. It processes CP spool reader files created by
3705 dumping operations, that is, dump files that are produced as a result of the
CP NETWORK command specified with the DUMP operand and either automatic
or immediate mode.
The NCPDUMP file processing operation can include:
Erasing a specific CMS NCPDUMP file after printing it
Formatting the dump
Printing the dump
Assigning an identifier to the CMS NCPDUMP file
Creating the CMS NCPDUMP file from the spool file.
•
•
Although NCPDUMP is a CMS command, its use is restricted to the user identified
by the SYSDUMP operand of the SYSOPER macro in DMKSYS during VM/SP
system generation. The operation of NCPDUMP is similar to IPCSDUMP operations. A general description of the NCPDUMP operation follows the command
description.
The NCPDUMP command has the following format:
NCPDUMP
[DUMPxx] [( [ERASE] [NOFORM] [NCPBUFF] [)]]
DUMPxx
is the filename of a CMS file containing a 3704/3705 Communications Controller program dump. This dump was created by a previously invoked
NCPDUMP command with the ERASE operand not specified.
4-42
VM/SP Operator's Guide
/-
[------~--
Se~vn(Ce P~«Jl~~auuUS
----------- -----------
------------------------------------:1
ERASE
erases the current CP DUMP file or a specified DUMPxx (filename), saved
CMS file.
NOFORM
specifies that a formatted control block is not desired.
NCPBUFF
specifies that a formatted listing of the NCP buffer pool is desired.
The NETWORK command invoked with the DUMP xx operand, as stated previously, produces CP files that contain the contents of a designated 3704/3705
Communications Controller unit buffer. These CP files reside as a spooled reader
input assigned to a system designated user. The CMS NCPDUMP command
invoked by this user formats (if requested) and prints the contents of these files.
The NCPDUMP program creates a CMS file with a filename DUMPxx and a
filetype of NCPDUMP, and erases the original spooled NETWORK initiated dump
reader file. The created CMS file is erased if you specify ERASE; otherwise it is
kept.
A maximum of ten dumped spooled files can be processed and saved, and later
recalled, if necessary, by the system assignment of an xx identifier suffix to the
CMS DUMPxx filename. The "xx" is a decimal number from 00 to 09, depending
on any existing files of a similar name. For example, if the files DUMPOO
NCPDUMP and DUMPOI NCPDUMP already exist, the new file would be called
DUMP02 NCPDUMP. The file thus created is retained for later use unless the
ERASE option is specified, in which case the file is erased immediately after the
dump is printed.
ZAP Service Program and How To Use It
ZAP is a CMS command that modifies or dumps MODULE, LOADLIB, or
TXTLIB files. It may be used to modify either fixed or variable length MODULE
files. It is for use by system support personnel only.
Input control records control ZAP processing. They can be submitted either from
the terminal or from a disk file. Using the VER and REP control records, you can
verify and replace data or instructions in a control section (CSE CT). Using the
DUMP control record, you can dump all or part of a CSECT, or an entire member
of a LOAD LIB or TXTLIB file, or an entire module of a MODULE file.
Section 4. VM/SP Service Programs
4-43
. ___,._._. ________.__._. ____.____....____ ....__._____,_____._____ J
' - - - - - - - - - ._._---,_._----
The format of the ZAP command is:
{~:D~~:}
llibnamel ..• libname3] [(Option....
TXTLIB
ZAP
l)]]
Options
[PRINT ]
TERM
[ INPUT filename NOPRINT
J
MODULE
LOADLIB
TXTLIB
indicates the type of file that is to be modified or dumped.
For LOADLIB files, the maximum blocksize supported by the ZAP service
program is 1024 bytes.
libname
is the library name containing the member to be modified or dumped. You
can specify one to three library names. The libname is valid only for
LOADLIB and TXTLIB files.
Options:
TERM [PRINT ]
NOPRINT
indicates that input to the ZAP service program is submitted through the terminal. If you specify TERM, the prompting message ENTER: is issued, and
you can then enter input control records up to 80 characters long. If you
specify PRINT with TERM, all output prints on the printer, but only error
messages display at the terminal. If you specify NOPRINT with TERM,
nothing prints on the printer. All output except control records displays at
the terminal.
INPUT filename [PRINT ]
NOPRINT
specifies that input is submitted from a disk file, filename. This file must
have a filetype of ZAP, and must be a fixed 80- byte sequential file residing
on any accessible device. If you specify PRINT with INPUT filename, all
output produced by the ZAP service program prints on the printer. In addition, commands and control records in error and error messages display at
the terminal. If you specify NOPRINT with INPUT filename, nothing prints
on the printer. Selected output produced by the ZAP service program prints
on the printer. All output displays at the terminal.
Figure 4-13 shows the resulting output resulting from valid option combinations:
4-44
VM/SP Operator's Guide
Se~VllC(9 P~OgU--~uu1S
:J
Options
PRINT
NOPRINT
INPUT
Commands and control records
in error and error messages on
the terminal. Everything on the
printer.
Everything on the terminal.
Nothing on the printer.
TERM
Only error messages on the terminal. Everything on the
Printer.
Everything except control
records on the terminal.
Nothing on the printer.
Figure
4-13.
Valid Options and Their Output
ZAP Input Control Records
Eight types of ZAP control records exist: DUMP, NAME, BASE, VER or
VERIFY, REP, LOG, COMMENT, and END.
ZAP control records are free-form and need not start in position one of the record
but the ZAP program can accept only 80 characters of data for each control record.
Separate all information by one or more blanks. All address fields including disp
(displacement) fields in VER and REP control records must contain an even
number of hexadecimal digits, to a maximum of six digits (OD, 02C8, 014318).
Data fields in VER and REP control records must also contain an even number of
hexadecimal digits.
If you wish, you may separate the data anywhere by commas (for example,
83256482 or 8325,6482). The commas have no effect on the operation.
Note: Do not use blank spaces as separators within data fields.
The program sets the NO GO switch on if a control record is found to be in error.
A file cannot be modified once the NOGO switch is turned on. The next valid
NAME record turns the NO GO switch off. This means that if the control record is
the NAME record, all succeeding records are ignored until the next NAME,
DUMP, or END record. For any other error, only REP control records that follow
are ignored.
DUMP Control Record
The DUMP control record resets the NOGO switch off. The DUMP control
record must not immediately precede a BASE, VER, or REP control record. A
NAME control record must precede the BASE, VER, and REP control records (if
any) that follow a DUMP control record.
The DUMP control record allows you to dump a portion or all of a specified
control section, or the complete member or module. The format of the output of
the dump is hexadecimal with an EBCDIC translation of the hexadecimal data.
Section 4. VM/SP Service Programs
4-45
The DUMP control record is optional. The format of the DUMP control record is:
DUMP
I
membername} [csectname [startaddress [endaddress11]
modulename
ALL
membername
is the name of the member to be dumped, or the member that contains the
CSECT(s) to be dumped. his member must be found in one of the libraries
specified in the ZAP command line.
For a CMS TXTLIB, the format of the dump control recqrd requires that
you specify both a member name and a CSECT name. Because the library
directory does not contain member names, any word may be used to replace
membername. The program searches for only the second name following the
dump operand; therefore, the second name must be the CSECT name.
modulename
is the name of the module to be dumped, or the module that contains the
CSECT(s) to be dumped. If you specify a module that has no loader table,
the program dumps the entire module.
csectname
is the name of the control section that is to be dumped. If you do not specify
csectname, the program dumps only the first CSECT. The csectname is
required for CMS TXTLIBs, optional for OS TXTLIBs, LOADLIBs, and
MODULE files. (See the discussion of csectname in "NAME Control
Record" on page 4-47). You must not specify csectname for a module
created with the NOMAP option.
ALL
specifies to the program to dump all CSECTs within the specified member or
module. You can specify ALL for MODULE files, LOADLIBs, and OS
TXTLIBs, but not for CMS TXTLIBs. If you wish to dump all the CSECTs
in a member of a CMS TXTLIB, you must issue a separate DUMP control
record for each CSECT.
startaddress
is the location within the specified CSECT where the dump is to begin. This
must be two-, four-, or six-hexadecimal digits. The start address is the displacement from the beginning of the CSECT. For example, if you wish to
start dumping at address 08 in a CSECT that begins at location 400, you
specify start address 08, not 0408.
endaddress
is the last address to be dumped. This must be two, four, or six-hexadecimal
digits. If you specify no address, the program dumps from the start address
of the CSECT to the end of the CSECT. Note that start and end addresses
apply only when you specify a csectname.
4-46
VM/SP Operator's Guide
c----------··------------------··
Notes:
1.
Displacements listed in the dump output for a module file are calculated from
the beginning location of the module. Therefore, the addresses in the output
may differ from the displacements within a CSECT.
2.
If a DUMP control record references a TXT LIB file that contains ORG state-
ments causing more than one occurrence of an address, data found at the first
occurrence will be displayed, but any subsequent redefinition of data for the
same location will be ignored.
NAM E Control Record
The NAME control record specifies the member or module and CSECT that
contain the data to be verified or replaced by the ZAP operation. The NAME
control record must precede the BASE, VER, and REP control records. If it does
not, the program sets the NOGO switch on. The format of the NAME control
record is:
NAME
I
I
membernamel [csectname]
modulename
I
membername
modulename
is the member or module that you want to be searched for the desired
CSECT.
csectname
is the name of the desired control section.
You must specify csectname if the CSECT you wish to modify is in a CMS
TXTLIB (that is, TXTLIB created by the TXTLIB command from CMS
TEXT decks that do not have a NAME card following the END card). The
directory of a CMS TXTLIB contains only CSECT names and no member
names. Select any word to replace membername as the first entry following
the NAME operand in the NAME statement for a CMS TXTLIB. The
CSECT name specified in the NAME record is compared with CSECT
names in the directory. If a CSECT match is found and no member name
match is found, the member selected is the one that contains the CSECT
name.
The csectname is optional if the CSECT you wish to modify is a LOAD LIB
or an OS TXTLIB (that is, a TXTLIB created by the TXTLIB command
from CMS TEXT decks that have a NAME card after the END card). The
dictionaries of the specified libraries are searched for the member name and
the member is then searched for the CSECT name, if you specified one. If
you do not specify csectname for a LOADLIB or an OS TXTLIB, the
program uses the first control section.
Section 4. VM/SP Service Programs
4-47
~-------------------------------------------------------------------------------~
The csectname is optional for a MODULE file. The module named in the
NAME control record is located and, if you specified csectname, the first
record is read to determine the number of records in the module and the
availability of a loader table, which the program can then search for the
csectname. If you do not specify csectname, the program uses the beginning
location of the module.
You are not allowed to specify csectname if the module was created with the
NOMAP option.
Note: The word selected to replace membername for a CMS TXTLIB should be a
meaningful name because the filename of the LOG control record is determined by
the specified membername or modulename in the NAME control record.
BASE Control Record
The BASE control record adjusts displacement values for subsequent VER or REP
control records for a CSECT whose starting address is not location zero in an
assembly listing. The format of the BASE control record is:
I
BASE
address
address
is the starting address of the CSECT. The address must be two-, four-, or
six-hexadecimal digits. For example, for a CSECT starting at location 400,
you would specify the BASE 0400 in the BASE control record. If a subsequent VER card requests verification of location 0408, the BASE of 0400 is
subtracted from 0408, and the program verifies location 08 in the CSECT.
This example applies if you specify TXTLIB, LOADLIB, or MODULE and
the module map is present. However, if no module map is present for a
MODULE file (that is, the module was generated with the NOMAP option),
then all operations are performed as if the BASE address is location O. For
example, if you specify a BASE of 400 and the address you wish to inspect
or modify is 408, then you must specify 08 and not 408 in REP and VER
control records. The address in this case is from the start of the module. If
you do not specify csectname in the NAME control record, you cannot
specify any BASE value other than 00. The BASE control record is
optional. See the discussion under "VER or VERIFY Control Record" on
page 4-49. If specified, the BASE control record must follow the NAME
record, but it need not follow the NAME record immediately. For example,
you could have the following sequence of control records: NAME, VER,
REP, BASE, VER, REP.
4-48
VM/SP Operator's Guide
Se~v~ce p~((j)~[i"~[JU1lS
---.--.---- ..
-----.--.-~---.~-~--~----
-----:J
VER or VERIFY Control Record
The VER control record requests verification of instructions or data within a
CSECT. If the verification fails, the program does not perform a subsequent REP
operation until it encounters another NAME control record.
The VER control record is optional. More than one VER record can follow a
single NAME record.
The format of the VER control record is:
disp
is the displacement from the start of the CSECT containing the data to be
inspected if you did not submit a BASE control record for this CSECT, or
disp is the actual location of the data to be inspected if you did submit a
BASE control record. disp must be two, four, or six hexadecimal digits. This
displacement does not have to be aligned on a fullword boundary. If this
displacement value is outside the limits of the CSECT specified by the preceding NAME control record, the VERIFY control record is rejected.
data
is the data against which the data in the CSECT is to be compared. This
must be an even number of hexadecimal digits.
Note: If the VER control statement references data in a TXTLIB file that is later
redefined by ORG statements, only the first data definition is verified.
Example: If the location you wish to verify is 3CC, and the CSECT begins at
location 2BO, you can enter:
BASE 02BO
VER 03CC data
or you can omit the BASE control record, subtract the CSECT start address from
the address of the data, and enter:
VER OIIC data
REP Control Record
The REP control record modifies instructions or data at the specified location
within the CSECT that you specified in a preceding NAME control record. The
data specified in the REP control record replaces the data at the CSECT location
specified by the disp operand. This replacement is on a "one-for-one" basis; that
is, one byte of data defined in the control record replaces one byte of data at the
location that you specified. If the replacement fails, the program does not perform
additional REP operations until it encounters another NAME control record.
Section 4. VM/SP Service Programs
4-49
The REP control record is optional. More than one REP record can follow a single
NAME record.
The format of the REP control record is:
I
REP
disp
data
disp
is the displacement from the start of the CSECT of the data to be replaced if
you did not submit a BASE control record for this CSECT, or disp is the
actual location of the data if you did submit a BASE control record. disp
must be two, four, or six hexadecimal digits. This displacement need not
address a fullword boundary. If this displacement value is outside the limits
of the CSECT being modified, the program does not perform the replacement operation.
data
is the data that is to replace the data in the CSECT. This must be an even
number of hexadecimal digits.
Notes:
/
1. Although you do not have to verify a location before replacing data, you
should do so to make sure that the data being changed is what you expect it to
be.
2. If the REP control statement references data in a TXTLIB file that is later
redefined by ORG statements, the replacement of data will take place at the
first occurrence of the rl:ata address of the TXTLIB member.
Example: If the location you wish to replace is 3CC, and the CSECT begins at
location 2BO, you can enter:
BASE 02BO
REP 03CC data
or you can omit the BASE control record, subtract the CSECT start address from
the address of the data, and enter:
REP 011C data
LOG Control Record
The LOG control record allows the user to specify, after application of a fix, a
unique fix number which is recorded in a log file for the module or member. The
file name of the log file is the same as the membername or modulename in the
NAME control record. The format of the LOG control record is:
4-50
VM/SP Operator's Guide
§eu--vnce
~~u--gu--~ms
.-------.---.----.---.--.--.----------------.---:J
LOG
fixnum
[filetype [user data]]
ZAPLOG
fixnum
specifies the number associated with the fix. Its length may vary from oneto-eight alphameric characters.
filetype
specifies the file type of the log. If not specified, filetype defaults to
ZAPLOG.
user data
specifies any data that the user wishes to enter into the log. If user data is
specified, the filetype operand is mandatory.
Usage Notes
Notes:
1.
The LOG control record is optional and is allowed only if valid NAME and
REP control records are found. The file name is obtained by the log routine
from the module or member named in the NAME control record. However,
whenever a LOG control record is found, a dummy log record will be written
at the completion of the user's valid REPs.
2.
Log multiple names by including a LOG control record after each name. If the
LOG record is not included after each name, error message DMSZAP070E
results. Processing continues after the error messages occur.
3.
The LOG record is 80 bytes in length and has the following information.
Columns 1 - 63: The fixnum and, if specified, the filetype and user data.
Columns 64-80: The date and time of the ZAP.
Comment Control Record
The ZAP program ignores comment control records. If the PRINT option is in
effect, the program prints the comments. The format of a comment record is:
I*
comment
The asterisk must be followed with at least one blank before you enter the text.
Section 4. VM/SP Service Programs
4-51
§ ~ 'tf (G
~"(Q)
L_ ._____________
e . De P
g II'" a uYil S
END Control Record
The END control record ends ZAP processing. The END record is required and
must be the last control record for input from the console. The format of the END
control record is:
END
Special Considerations for Using the ZAP Service Program
Before using the ZAP command against MODULE files, you can use the
MODMAP command to determine whether a module map exists and what it contains.
When a ZAP input file has more than one pair of VER and REP control records
and a VER control record (other than the first) fails, you must remove the records
prior to the failing record and correct the error before you issue the ZAP command
again. Otherwise, the file being modified returns to its original status.
The REP control record cannot be used to place data in an undefined area such as
a Define Storage area. If any part of a data field specified in a pair of VER and
REP control records is an undefined area, warning message DMSZAP248W will be
displayed, and no data replacement will occur. If you do not issue a VER control
record prior to the REP control record, partial modification of data may result.
User-defined data may be inserted in undefined areas of text files by using the REP
statement described under the LOAD command. (See the VM / SP CMS Command
and Macro Reference.)
If the file to be dumped contains undefined areas (such as a DS or ORG statement
in a TXTLIB member), the hexadecimal portion of the dump contains blanks to
indicate that the corresponding positions are undefined.
VER and REP control words can be used to modify TXTLIB members produced
by FORTRAN compilers that store the length of the compiled text in the END
card rather than in the ESD card. However, if a member of this type contains multiple CSECTs, only the first CSECT can be modified by the ZAP program.
I
ZAPTEXT Service Program
ZAPTEXT is a CMS command that modifies or dumps individual text files. Use
ZAPTEXT like the ZAP service program, but only for text files, not MODULEs,
TXTLIBs, or LOADLIBs, which ZAP processes.
The format of the ZAP TEXT command is:
4-52
VMjSP Operator's Guide
[_______ .
SeU"VDce
~:>rro[Jrr~u1ilS
--_.--._-.-_.-...._.-.. ---_. --_--.. ---.-.. .-..------------------.-----.------------------.. -.-.--.------..----.-. ---.....-.--.-------=-:1
_:~._:.._-.-_----:~_=_~_=-_.-.-_.
~
ZAPTEXT
fn [(t [(mIl [(Options .... [) II
Options:
[INPUT filenamel
fPRINT
]
LNOPRINT
fnftfm
is the fileid of the text file that you want to alter. If you omit the filetype or
filemode, the system assumes TEXT for the filetype and At for the filemode.
The filemode must specify a read/write disk.
Options:
INPUT filename
filename is the name of the file that has the ZAP control records. This file
must:
Have a file type of ZAP
Be a fixed 80-byte sequential file that resides on any accessible disk.
If you do not specify a filename, it defaults to fn.
PRINT
prints all output produced by ZAPTEXT on the printer. The system also displays error messages, commands in error, and control records in error.
NOPRINT
does not print anything on the printer, and instead displays all output at the
terminal.
ZAPTEXT Input Control Records
ZAPTEXT uses the same control information as the current ZAP service program,
with the addition of the EXPAND control record. The ZAP service program will
ignore any EXPAND control records. Refer to "ZAP Input Control Records" on
page 4-45 for information about the control records, other than EXPAND, that
ZAPTEXT uses.
Use the ZAP control records with ZAPTEXT according to ZAP's TXTLIB conventions.
Section 4. VM/SP Service Programs
4-53
EXPAND Control Record
The EXPAND control record lets you increase the size of a named control section
contained in the text file. The format of the EXPAND control record is:
EXPAND csect size [ , csect size ••• J
csect
specifies the symbolic name of a control section whose length you want to
increase.
size
specifies the decimal number of bytes for the system to add to the control
section length. The system will initialize the added bytes to binary zeros.
The maximum number of bytes for each control section that you indicate is
4095.
Each control record may have multiple entries, but you must separate them with
commas. Do not spill an entry onto the next line.
Notes:
1.
The system processes all EXPAND control records before any other control
records, regardless of their position in the control file.
2.
The effective length of the expansion, which is the actual number of bytes
added to the control section, may be greater than the length that you specify
for the expansion. This may occur if, after the specified expansion, the system
must add padding bytes to align the next control section or common area.
3.
When you increase a control section's size, it may affect the offset address of
any following control section. This is important when you determine values for
BASE, REP, and VER control records. Use the effective expansion lengths
when you are determining control section offsets.
EXPAND Command
ZAP TEXT calls EXPAND if you specify an EXPAND control record in the ZAP
control file. Use EXPAND to add space to a program in object deck form. The
system creates object decks when you assemble or compile a source program. This
is especially useful when you do not have the source code for a program or the
program does not have a patch area.
Note: EXPAND can only add extra space at the end of named control sections
(csects). EXPAND cannot expand private code (unnamed csect) and common
areas (named or unnamed).
/
4-54
VM/SP Operator's Guide
SeU"VDce PU"ogn"ams
J
Do not increase the length of a program beyond its design limitations. For
example, if you add space to a control section beyond the range of its base register
addressability, that space is unusable.
The format of the EXPAND command is:
EXPAND
fnI [ftl [fmI [fn2 [ft2 [fm2] ] ] ] ] I (Options .... l
)]]
Options
J
[PRINT ]
INPUT filename
[ CSECT csect SIZE size NOPRINT
fnI ftl fmI
identifies the input text file that the system will expand. The file must have
valid object deck information, like that created by an assembler or compiler.
If you don't specify the filetype or filemode, the system assumes TEXT for
the filetype and Al for the filemode.
Note: EXPAND assumes that the input text file follows OS/VS standards
and that the OS/VS Linkage Editor will accept it without error. The system
does a limited check for errors. If the input file is invalid, the system may
not expand the text file correctly.
fn2 ft2 fm2
identifies the output text file that the system will create. You can use an
equal sign (=) for any of the file identifiers to indicate that it is the same
identifier as fnl, ftl or fm!. The default fileid is $fnl = =. The system
truncates the filename (fnl) to 7 characters before appending the $. In any
case, fm2 must be a read/write disk.
Options:
CSECT
specifies the symbolic name of a control section whose length the system will
increase. If you specify CSECT, you must also specify SIZE.
Do not specify CSECT with the INPUT option.
SIZE
specifies the decimal number of bytes that the system will add to the control
section length. The system initializes the added bytes to binary zeros. The
maximum number of bytes for each control section is 4095. If you specify
SIZE, you must also specify CSECT.
Do not specify SIZE with the INPUT option.
Section 4. VM/SP Service Programs
4-55
INPUT
is the name of an EXPAND input file that contains EXPAND control
records. If you do not specify INPUT, the filename will default to the name
of the text file that you are expanding (fnl). The file type must be
EXPAND. The system will search all accessed disks for this file.
Do not specify this option with the CSECT or SIZE options.
PRINT
prints on the printer all output that EXPAND produces. In addition, the
system displays error messages, commands in error, and control records in
error.
NOPRINT
does not print any output on the printer, and instead displays it at the terminal.
System Response
After the system expands each CSECT that you specified, the system will issue
a message indicating the following:
The number of bytes added to the control section.
The number of bytes added may be greater than the length that you
specify for the expansion. This may occur if, after the specified expansion,
the system must add padding bytes to align the next control section or
common area.
The offset, relative to the start of the specified control section where the
expansion began.
If the system encounters an error during processing, it aborts the update and
does not perform the expansions.
The Programmable Operator Facility
The Programmable Operator Facility is designed to increase the efficiency of
system operation and to allow remote operation of systems in a distributed data
processing environment. It does this by comparing incoming messages against a list
of entries in the routing table (a CMS file); when a match occurs, the appropriate
action is performed. The tasks that can be performed by the programmable operator facility include:
•
•
•
Logging messages
Suppressing message display and routing messages to a logical (real) operator
Executing commands
Responding with preprogrammed message responses.
/
4-56
VM/SP Operator's Guide
Informational messages, such as those appearing on the CP system console as a
result of the LOGON and LOGOFF commands, can be filtered out by the programmable operator facility and not clutter the real operator's display.
Messages with requests that cannot be handled by the programmable operator
facility either because they require physical intervention, because the user is not
authorized to issue the programmable operator facility command, or because the
request does not match an entry in the routing table are routed to a real operator
(the logical operator) at another virtual machine for consideration and/or appropriate action. These requests may include tasks such as mounting tapes and loading
paper into the printer or may be messages to the operator from CMS users.
The logical operator mayor may not be located on the same physical system as the
programmable operator facility. When the logical operator and the programmable
operator facility are not on the same VM system, the communications are handled
through RSCS networking. Several different physical systems may be connected
by an RSCS network and controlled by one logical operator from a single console.
When the logical operator is an NCCF (Network Communication Control Facility)
operator, the communications are handled primarily by the PROP /NCCF Message
Exchange (PMX) portion of the programmable operator facility.
Use in a Single System
When the programmable operator facility is operational in a single-system environment, it can:
•
Ease message traffic to the system operator, by:
Filtering (logging) non-essential, information-only messages
Routing messages (for example, I/O intervention requests) to someone
else for specialized action.
Increase productivity, by freeing the system operator from certain routine
responses or tasks. Such responses (whether they consist of one or a series of
commands, whether VM/SP or guest operating system) may be programmed
to execute automatically upon receipt of a given message.
Thus, only essential, non-routine messages (that is, those requiring the skill and
experience of a system operator to handle) are sent on to the operator for
response or action.
Use in a Distributed VM Environment
The capabilities of the programmable operator, outlined above, also allow for the
remote operation of systems in a distributed environment. When the programmable operator facility is operational in a distributed system, it can:
•
Issue responses and perform tasks that do not require an on-site operator.
•
Filter (log) non-essential, information-only messages.
Section 4. VM/SP Service Programs
4-57
[
Route messages requiring on-site (that is, manual) intervention to someone,
not necessarily an operator, at the distributed site for action.
•
Route messages that require the skill and experience of a system operator to
handle to the operator at the host system. The operator at the host site also
has the capability to send commands to the programmable operator facility to
control its operation, as well as commands to execute on the distributed system
to control the system itself.
By running the programmable operator facility on VM/SP systems distributed at
several different locations (network nodes), one operator at a host site can control
a network of systems.
Figure 4-14 shows a simplified configuration of the logical operator controlling a
distributed system through RSCS networking. Communications between the programmable operator facility running in the system operator virtual machine
(OPERATOR) at the remote site and the logical operator virtual machine
(LGLOPR) at the host site are handled through the RSCS network that connects
the two VM systems.
Host site at NODEl
J>---
l
G
l
0
p
R
R
Figure
C
C
S
S
4-14.
0
> P
R
S
---
I
---
S
1
<,
Remote site at NODE2
2
---<
I
E
R
A
T
0
R
Example of a Programmable Operator Facility ConfIguration.
Use in a Mixed Environment
The programmable operator facility also provides for distributed data processing in
an SNA environment with mixed VM, OS/VS, and VSE distributed systems and
host systems, called a "mixed environment". The Programmable Operator/NCCF
Message Exchange (PMX) provides an interface with NCCF so that an operator
on an OS/VS or VSE system can operate a VM distributed system from an NCCF
operator station. This means that an NCCF operator can be the logical operator
for a programmable operator facility. Also, an NCCF operator (not necessarily the
logical operator) can issue programmable operator commands, including the ability
to run any VM command that the programmable operator virtual machine is
authorized to run.
Notes:
1.
4-58
VM/SP Operator's Guide
The programmable operator facility does not let a user logged onto VM issue
NCCF commands.
§elrvnce ProglrSJu{l)S
:J
2.
VM/SP Group Control System (GCS) is a requirement for this support. The
Programmable Operator/NCCF Message Exchange (PMX) uses facilities
unique to GCS, and cannot run on any other supervisor. For more information
on GCS, see VM/SP Group Control System Guide, SC24-5249.
When the logical operator is an NCCF operator, the programmable operator does
not use RSCS to route messages to the logical operator; instead, the programmable
operator passes the messages to NCCF through the NCCF Message Queueing
Service, so NCCF displays the messages at the appropriate NCCF operator
console.
Figure 4-15 shows a single system environment, VM-only, where NCCF has been
added. In this configuration, if you log on to an NCCF operator station you can
operate the VM system. You can extend this idea to multiple unconnected systems
which can be controlled by different NCCF operator stations centralized at the
same location. This, however, requires one operator station for each of the systems
being controlled.
NCCF Op
Station
CP Sys. Ope
PROG
OP
NCCF
<-- IUCV -->
CMS
PMX
--
_______ J
VM/VTAM
GCS
GCS
VM/SP CP
Figure
4-15.
A single system with NCCF
Invoking the Programmable Operator Facility
Before using the Programmable Operator Facility, it must be installed as prescribed
by the steps given in the VM/SP System Programmer's Guide under the section
"Installing the Programmable Operator Facility." After the programmable operator
facility has been installed and if it is not automatically invoked, the PROPST
EXEC can be used to manually invoke the facility.
The PROPST EXEC drops any IBM-supplied programmable operator routines that
are currently loaded as nucleus extensions and loads the programmable operator as
a nucleus extension. It then invokes the programmable operator facility with the
specified RTABLE. If you do not specify a routing table, the RTABLE name
defaults to "PROP." You may specify a disconnect parameter to cause the programmable operator virtual machine to be disconnected before the programmable
operator is invoked. The format of the PROPST EXEC is as follows:
Section 4. VM/SP Service Programs
4-59
PROPST
rtable-name] [DISConn]
[ PROP
Optionally, you can take the following steps each time you invoke the programmable operator facility:
1. Issue a FILEDEF command to assign a CMS filename to the PROPLIB
LOAD LIB file so CMS can read and load from it. This is done as follows:
FILEDEF PROPLIB DISK PROPLIB LOADLIB
2.
*
Load the programmable operator program as a CMS nucleus extension via the
NUCXLOAD command. The command is issued as:
NUCXLOADPROPDMSPOPPROPUB
(The details of the NUCXLOAD command are given in VM/SP CMS
Command and Macro Reference.)
These first two steps may be omitted for subsequent invocations as long as you
do not:
IPL CMS, or
Have a CMS abend from which the programmable operator was not able
to automatically recover.
3. Following its loading as a CMS nucleus extension, invoke the programmable
operator facility in the same way as any CMS command. The format of the
invocation is:
PROP
routing-table-name]
[ PROP
routing-table-name
is the filename of the routing table that is to be used by the programmable operator facility. The routing table provided with the programmable operator facility has a file name of PROP and a file type of
RTABLE.
PROP
is the default file name if no other is specified by the installation.
When the programmable operator facility initialization is completed, the message
PROP RUNNING - ENTER' STOP' TO TERMINATE is displayed on the
console of the programmable operator facility virtual machine. The programmable
operator facility then waits for either an incoming message or a programmable
operator facility console command (either STOP or SET). At this point, the user
may disconnect from the programmable operator facility virtual machine by
4-60
VM/SP Operator's Guide
(
- - - - - ._--_._--_ _.-
§eU"VDce Prr(iJ)~[(a[('ullS
•...
...
_--,
entering CP mode (pressing the PAl key or equivalent) and typing DISCONN.
The logical operator should now log on to the appropriate system or virtual
machine if not already logged on.
If initialization cannot be successfully completed, an error message followed by the
message PROP HAS TERMINATED is displayed on the console of the programmable operator facility virtual machine and programmable operator facility operation is terminated.
Invocation of Programmable Operator Facility Commands
The commands used with the Programmable Operator Facility can be executed by
anyone who is authorized by the active routing table. If an unauthorized user
issues a programmable operator facility command, the command is not executed
but is routed to the logical operator. To send a command to the programmable
operator facility you must send a message to the programmable operator facility
virtual machine. The text of the message is the command to be issued. Unless
otherwise noted, the response from the command is returned to the user who sent
the command to the programmable operator facility virtual machine. Responses
are sent one line at a time via the CP MSGNOH or MESSAGE commands.
The format of the message sent to the programmable operator facility virtual
machine is the same as used with the CP MESSAGE command for local use and
the CP SMSG command for distributed (network) use. In a mixed environment, an
NCCF operator uses PROP, an NCCF command, to send messages to the programmable operator facility virtual machine.
The local format is:
Message
MSG
userid propcmd [ parameters )
The distributed (network) format is:
SMsg
netid Msg nodeid userid propcmd [ parameters)
netid
is the userid of the RSCS network machine at the user's node.
Msg
is the required RSCS message command. It can be entered as M or MSG.
nodeid
I
.~.
is the nodeid of the programmable operator facility virtual machine.
Section 4. VM/SP Service Programs
4-61
userid
is the userid of the programmable operator facility virtual machine.
propcmd
is the command to be executed by the programmable operator facility. See
"Programmable Operator Facility Commands" on page 4-70 for information about the commands.
parameters
are the parameters associated with the command to be executed.
The format for an NCCF operator station is:
IPROP
Ipropcmd [parameters]
propcmd
is the command that the programmable operator facility will execute
parameters
are the parameters associated with the command to be executed.
Notes:
1.
To use this format, you must be authorized by NCCF.
2. If you (NCCF operator) are not attached to the local system, use the NCCF
ROUTE command to specify the domain where you want to execute the
command.
Issuing Commands in the Local Environment
If the Programmable Operator Facility is running in the operator's virtual machine,
a user on the same physical machine might send the messages:
MESSAGE OPERATOR QUERY RTABLE
MESSAGE OPERATOR WHAT TIME IS SHUT DOWN?
where "QUERY" is a programmable operator facility command. See "QUERY
Command" on page 4-80 for information about the programmable operator
facility QUERY command.
Figure 4-16 on page 4-63 shows how messages are transferred between a user,
the programmable operator facility, and the logical operator when all are located on
the same physical system.
4-62
VM/SP Operator's Guide
§eu-vnce ~:Du"!Q)~U"'a[(UllS
c:-----------'----------------------------------------;:-----'--------------------'-----------------------------------_.-------]
Local System
Logical Operator
Vi rtua I ~lach i ne
Userid
A
= LGLOPR
I
(4)(3)
r--~~
Programmable<-(1)
User Virtual
~lachi ne
Operator
~
Virtual
(2)->
Machi ne
Userid =
OPERATOR
(1)
1
<-(3)
I
!-(4)-~
Userid
USER1
=
USER1 issues the message:
MESSAGE OPERATOR QUERY RTABLE
(2)
The programmable operator facility responds by sending
the following message to USER1:
PROP RUNNING WITH ROUTING TABLE 'PROP RTABLE AS'
(3)
USER1 issues the message:
MESSAGE OPERATOR WHAT TIME IS SHUT DOWN?
Because this message cannot be processed by the
programmable operator facility, the message is routed
to the logical operator.
(4)
Figure
The logical operator responds to USER1's message by
sending a message to USERl by way of the programmable
operator facility.
(See "Ensuring A Complete log").
4-16.
Example of Communication in the Local Environment
Issuing Commands In The Distributed VM Environment
A user (USER21) on another VM system (assuming a remote system with the
nodeid of "NODE2," a network machine with the id of "NET2," and that the programmable operator facility is running in the operator's virtual machine with a
userid of "OPERATOR") might send the messages:
MESSAGE OPERATOR QUERY RTABLE
MESSAGE OPERATOR WHAT TIME IS SHUT DOWN?
where "QUERY" is a programmable operator facility command. See "QUERY
Command" on page 4-81 for information about the programmable operator
facility QUERY command.
Section 4. VM/SP Service Programs
4-63
Figure 4-17 shows how a user on one physical system exchanges messages with
the programmable operator facility on a different physical system and how the programmable operator facility responds directly to a user on the same physical system
or routes messages to the logical operator who is located on a different physical
system.
Local System
Nodeld = NODEl
Distributed System
Nodeid = NOOE2
RSCS Virtual ~------+-I(l)->------~ RSCS Virtual
~------~-(2)-+'------~
Machine
Machine
r-------<-(5)-i------~
10 = NETl
r-------r-(6)->-----~
10 = NET2
-1 I I 1
1-1
c:k~
r----------~
User
Virtual
Machine
Userid
USERll
(1)
=
V-V--logical
Operator
Virtual
Machine
Userid
lGlOPR
=
J
TJdcffi
I I
v-
rV
(3»Programmable
User
~
Operator
Virtual «4)
Virtual
Machine ~
Machine
(5»
Useri d =
I
USER2l ~(6)1
Useri d =
OPERATOR
LGlOPR issues the message:
CP SMSG NETl MSG NODE2 OPERATOR QUERY RTABLE
(2)
The programmable operator facility responds by sending the
following message to LGLOPR:
PROP RUNNING WITH ROUTING TABLE 'PROP RTABlE AS'
(3)
USER2l issues the message:
MESSAGE OPERATOR QUERY RTABLE
(4)
The programmable operator facility responds by sending the
following message to USER21:
PROP RUNNING WITH ROUTING TABLE 'PROP RTABlE AS'
(5)
USER21 issues the message:
MESSAGE OPERATOR WHAT TIME IS SHUT DOWN?
Because this message cannot be processed by the programmable
operator facility, the message is routed to the logical
operator.
(6)
Figure
The logical operator responds to USER21's message by sending a
message to USER21 by way of the programmable operator facility.
(See "Ensuring A Complete log").
4-17.
Example of Communication in the Distributed Environment
/
4-64
VM/SP Operator's Guide
[---_.._. __ .•.
_ ... _.•._-----------------_._-._----..•.._-_.
-----
Servoce
ProgU"'~uuUS
J
Issuing Commands in a Mixed Environment
In a mixed environment, message paths are more complex. A user on a VM system
(USER3) might send the messages:
MESSAGE OPERATOR QUERY RTABLE
MESSAGE OPERATOR WHAT TIME IS SHUT DOWN?
where "QUERY" is a programmable operator facility command. See "QUERY
Command" on page 4-80 for information about the programmable operator
QUERY command.
The message sent containing the QUERY command simply follows the paths of a
message sent by a user to the programmable operator facility in the same system.
The message is processed by the programmable operator facility and a response is
returned to the user (USER3).
Figure 4-18 on page 4-66 shows how messages are transferred between a user,
the programmable operator facility, and an NCCF logical operator. It also shows
how messages and commands are transferred from an NCCF operator to the programmable operator facility and its VM system.
Section 4. VM/SP Service Programs
4-65
§e~·vDce ~fi"(Q)91fi"8JmS
L
.. _--------_ .._---------------"
CP Sys Op
User
SNA Servo Mach.
Userid=
USER3
VM/
P --(3)- -> N
VTAM
M <-(4)- -- C
X <-(5)- -- C - ------
PROGRAMMABLE
<-(4)- --- OPERATOR
I
C
M
S
A
I I
I
--(8)- -> F
Userid=
OPERATOR
C
M
S
I
( 1)
V
IUCV *MSG--J
X A AX
AX X A
IIII
IIII
~m~
NCCF
Operator
Stution
Operator id=
lGlOPR
GCS
IUCV
(7)
VM/SP CP
(1)
USER3 issues the message
MESSAGE OPERATOR WHAT TIME IS SHUT DOWN?
that reaches the programmable operator facility through the lUCY
*MSG service. The programmable operator facility determines that
the message is to be routed to the logical operator.
(2)
The programmable operator determines that the logical operator is an
NCCF operator and sends the message though IUCV to the PMX.
(3)
The PMX, upon receiving the messuge from IUCV, queues (through the
NCCF Messuge Queueing Service) the message for display on the
specified NCCF operator's console.
(4)
Upon receiving the message from USER3, the NCCF logical operator
issues the command
PROP CMD MSG USER3 SHUTDOWN IS AT 10:00 PM.
which travels back through the PMX, IUCV, and the programmable
operator facility, before the response reaches USER3's screen.
(S)
An NCCF operator issues a programmable operator command
PROP QUERY RTABlE
which is queued for the PMX.
(6)
The PMX sends the command as an IUCV message to the programmable
operator.
(7)
The programmable operator receives the message from IUCV and processes
it. Upon determining that the requester was an NCCF operator, the
programmable operator sends any responses through IUCV to the PMX.
(8)
The PMX receives the response:
PROP RUNNING WITH ROUTING TABLE 'PROP RTABlE AS'
from IUCV and queues the response for display on the specified
NCCF operator's terminal.
I
Example of Communication in the Mixed Environment
Figure
4-18.
4-66
VM/SP Operator's Guide
*
*
*
**
*
*
*
§errVDce ~rrO~rra[(ll1S
---_.----------_._----_._--------_.._--_._--
Helpful Hints
The typing of large text strings, such as the message command string preceding the
message text, can be minimized by assigning the string, up to the userid, to a PF
key. An EXEC can also be used that prompts for, or accepts as parameters, that
part of the message command that the user must type in.
For example:
SET PFOI IMMED SMSG NETI MSG NODE2 OPERATOR QUERY RTABLE
will allow the logical operator to press PFO I to find out the name of the active
routing table when the logical operator and programmable operator facility are
located on different physical systems.
The CMS TELL and NAMES exec procedures may also be used by the logical
operator to send a message to a user. If the logical operator has a NAMES file
entry assigning the nickname PROPI to OP at node NODEI, the logical operator
could send a message in the following manner:
TELLPROPIQUERYRTABLE
TELL PROPI CMD WNG ALL ... WILL RE-IPL IN 5 MIN - PLS LOGOFF.
These exec procedures are described in VM / SP CMS Command and Macro Reference and VM/SP CMS User's Guide.
The Log File
If LOGGING is not OFF, the programmable operator facility puts every message
that it received into a CMS file called the log file. If LOGGING is set to ALL, the
programmable operator facility puts all generated error messages and command
responses in the log file.
If LOGGING is ON, the programmable operator facility does not log responses
from CP, CMS, and programmable operator facility commands, but does log messages. Each message is identified by the date and time received. The userid and
nodeid are in the message only if the text was sent by a CP MSG, SMSG, WN G, or
sent using SCIF (Single Console Image Facility). For a message sent by CP, the
userid and nodeid are blank. A message from a remote RSCS network virtual
machine has a nodeid, but no userid. A message sent by an NCCF operator
console has '*NCCF' as the nodeid.
Log entries generated and logged by the programmable operator have a userid of
PROP. The log file has the following format:
Section 4. VM/SP Service Programs
4-67
-==oJ
col
co 1 10
~
yy/mm/dd
t
hh:mm:ss
co 1 19
col 28
~
~
[userid node i d]:
col 39
~
text
The log file has variable length records. The maximum record length that can be in
the log file is 132 characters (38 prefix characters plus 94 text characters). Messages with a text length greater than 94 characters will be continued on the next log
record. The continuation record will have the prefix characters (columns 1 through
36), but no colon (:) in column 37.
A separate log file is started for each day. The name of the file is:
LGyymmdd nodeid AS
yy
is the current year
mm
is the current month
dd
is the current day.
nodeid
is the current RSCS nodeid of the system where which the programmable
operator facility is running.
Any non-NCCF user listed in the active routing table can use the GET LOG
command to get the log file from the programmable operator facility as a reader
spool file.
Any user authorized in the active routing table to use the programmable operator
facility CMD command can purge an old log file. Use the CMS ERASE command
to purge the file.
Ensuring A Complete Log
When the programmable operator facility routes a message to the logical operator,
the message has the userid of the sender. To respond to the message, you can send
a message directly to the user without going through the programmable operator
facility. If you do send a message this way, the message will not be logged by the
programmable operator facility.
To ensure that the programmable operator facility logs the messages, use the programmable operator facility CMD command to send the message to the user. See
"CMD Command" on page 4-71 for information about the programmable operator facility CMD command. To be sure that user messages always have the same
id (the programmable operator facility id), send messages from the logical operator
through the programmable operator facility.
4-68
VM/SP Operator's Guide
("
c ____
~==:~-~--·--------··----
------.----.--------------.
To route a message through the programmable operator facility where the operator
and user are not on the same physical VM system, use the SMSG command:
SMSG netl msg nodel operator CMD MSG userl - RESPONDING TO YOUR
REQUEST
netl
is the userid of the network machine at the user's node.
msg
is the RSCS message command.
nodel
is the nodeid of the programmable operator facility virtual machine.
operator
is the userid of the programmable operator facility virtual machine.
CMD
is the programmable operator facility CMD command.
MSG
is the VM command that the programmable operator facility will execute.
userl
is the userid of the user who will receive the message.
RESPONDING TO YOUR REQUEST
is the message text sent to the user.
To route a message through the programmable operator facility where the operator
and user are on the same physical system, use the MSG command:
MSG operator CMD MSG userl - RESPONDING TO YOUR REQUEST
operator
is the userid of the programmable operator facility virtual machine.
CMD
. is the programmable operator facility CMD command.
MSG
is the VM command that the programmable operator facility will execute.
userl
is the userid of the user who will receive the message.
RESPONDING TO YOUR REQUEST
is the message text sent to the user.
Section 4. VM/SP Service Programs
4-69
L __________________________________________________________
To route a message through the programmable operator facility from the logical
operator who is an NCCF operator to the user on a VM system, use PROP, an
NCCF command as follows:
PROP CMD MSG USER1 - RESPONDING TO YOUR REQUEST
PROP
is the NCCF command that sends the message from an NCCF operator to
the programmable operator facility.
CMD
is the programmable operator CMD command.
MSG
is the VM command that the programmable operator facility will execute.
USER 1
is the userid of the user who will receive the message.
RESPONDING TO YOUR REQUEST
is the message text sent to the user.
Note: Refer to "Helpful Hints" on page 4-67 for ways-to reduce typing of long
text strings.
Programmable Operator Facility Commands
In the following command descriptions, an authorized user is any user for which
there is a corresponding routing table entry for the command he wishes to issue. In
other words, authorization for the programmable operator facility is controlled
entirely by the contents of the routing table. All programmable operator facility
commands are sent by the user as messages. (See the programmable operator
section of the VM / SP System Programmer's Guide for more information on the
routing table).
4-70
VM/SP Operator's Guide
CMD Command
Use the CMD command to execute CP or CMS commands in the programmable
operator facility virtual machine. The command is accepted or rejected by CP
based on the CP user class defined for the programmable operator facility virtual
machine. Also, the programmable operator facility may reject or accept the
command based on the authorization granted in the active routing table. The user
class assigned to the programmable operator facility virtual machine is determined
by the installation. If accepted, the response from the command is returned to the
issuer of the command.
The format of the CMD command is:
I CMD
I vmcmd
vmcmd
is the VM command sent to the programmable operator facility virtual
machine for execution according to the CMS IMPCP and IMP EX settings.
Usage Notes
Notes:
1.
Any commands that alter or overlay CMS storage (CP DEFINE STORAGE,
CP IPL CMS, CP SHUTDOWN, etc.) will have an adverse effect on the operation of the programmable operator facility and should not be issued under the
programmable operator facility.
2.
Reissuing the PROP command once the programmable operator facility is
running will cause it to stop operating correctly. The user must then reIPL
CMS and restart the programmable operator facility using the procedure
described under "Invoking the Programmable Operator Facility" on
page 4-59.
3.
Issuing commands that cause a VM READ or a CP READ (such as the DDR
command) will stop the programmable operator facility. It must then be
restarted using the procedure described under "Invoking the Programmable
Operator Facility" on page 4-59, or the read must be answered from the
console of the programmable operator facility virtual machine. Commands of
this type should not be sent to the programmable operator facility.
4.
Line editing characters (CHARDEL, LINEDEL, LINEND, and ESCAPE),
although interpreted by your terminal when you have SET LINED IT ON in
effect, are not interpreted as line editing characters when sent to the programmable operator facility. In other words, they are interpreted as the characters
they are (that is, @, ¢, #, :). For example, the string:
MOP CMD ACCESS 191 A"#RECEIVE
Section 4. VM/SP Service Programs
4-71
will cause an "INVALID MODE' A#RECEIV '" message to be returned to
the issuer. The default line editing characters may be defined by the installation. You may also define your own line editing characters by using the CP
TERMINAL command.
5.
The programmable operator facility does not recognize the CMS immediate
commands (HB, HI, HO, HT, HX, RO, RT, SO, TE, and TS). If you issue any
of these commands, the programmable operator facility issues "UNKNOWN
CP/CMS COMMAND."
Examples
The command issued at a virtual machine console:
MOP CMD INDICATE
M OP CMD QUERY FILES
MOP CMD QUERY PRINTER ALL
\ I
MOP CMD ERASE LG80l0l5 NODEI A5
MOP CMD QUERY SEARCH
M OP CMD LISTFILE
The command issued at an NCCF operator terminal:
PROP CMD INDICATE
PROP CMD QUERY FILES
PROP CMD QUERY PRINTER ALL
PROP CMD ERASE LG80l0l5 NODEI A5
PROP CMD QUERY SEARCH
PROP CMD LISTFILE
Responses
The response returned by CP or CMS is sent to the issuer of the command. After
the response from CP or CMS, the programmable operator facility responds with
"COMMAND COMPLETE".
,/
4-72
VM/SP Operator's Guide
§e[('V'nCG u~u"(Q)~rams
--------------------.-----.-------------.---.------------.- --------.---------------=:=1
FEEDBACK Command
Use the FEEDBACK command to place comments about the operation of the
system and/or the programmable operator facility in the feedback file. These comments are available for review by personnel responsible for maintenance of the
programmable operator facility. The comment (preceded by the date and time it
was received, and the sender's userid and nodeid) is placed in a file named
"FEEDBACK nodeid AS,".
The format of the FEEDBACK command is:
I
::EDBACK
I
text.•.
text..•
is the user's comments placed in the feedback file (FEEDBACK nodeid AS).
Usage Notes
Notes:
1.
The feedback file can be retrieved by authorized non-NCCF users, using the
programmable operator facility GET command.
2.
The length of the message is limited by the maximum length of the command
used to send the message to the programmable operator facility. If the user
desires to send a longer message, the command must be used mUltiple times.
Example
The command issued at a virtual machine console:
M OP FEEDBACK SYSTEM RESPONSE WAS SLOW DURING THE MORNING
SHIFT
The command issued at an NCCF operator terminal:
PROP FEEDBACK SYSTEM RESPONSE WAS SLOW DURING THE MORNING
SHIFT
Response
COMMAND COMPLETE
Section 4. VM/SP Service Programs
4-73
L:.
GET Command
Use the GET command to retrieve one of the programmable operator facility files;
the feedback file (FB or FEEDBACK) or the log file (LOG). The file is sent to
the requesting user if he is authorized in the active routing table to receive it. If
LOG is specified, the user will receive the log file for either the current day or the
specified day.
The format of the GET command is:
FEEDBACK }
GET
{
~~G (yymmdd]
FEEDBACK
FB
indicates that the feedback file is to be retrieved.
LOG (yymmdd]
indicates that the log file for date "yymmdd" is to be retrieved. If no date is
given, the log file for the current day is retrieved.
Usage Notes
1. An NCCF operator cannot use the GET command to get the LOG and FEEDBACK files. Use CMS commands (and the programmable operator CMD
command) to type the file(s) or portions of the files, or send the file(s) to some
userid where you can process them.
2. The file appears in the requesting user's virtual reader in DISK DUMP format.
The user must execute a DISK LOAD or RECEIVE command to read the file.
Examples
MOPGETFB
M OP GET LOG 801030
Response
COMMAND COMPLETE
4-74
VM/SP Operator's Guide
§errvDce Pr(1)~U"cllrtJ1lS
-_._- ------------===,
LGLOPR Command
Use the LGLOPR command to assign or release yourself as the logical operator for
the programmable operator facility under which the command is executed. This
programmable operator command can be used by a VM user or an NCCF operator.
Users are authorized in the active programmable operator routing table.
where:
ASN
assigns the issuer of the command as the logical operator, if a logical operator is not currently assigned (Le. the current LGLOPR is the defaUlt). If a
logical operator is already assigned to the programmable operator facility, an
error response is given.
RLS
releases the issuer from being the logical operator, if he currently is the
logical operator and assigns the default LGLOPR. If the issuer is not the
logical operator, no operation is performed and the system gives the following response:
DMSPOR763E NOT CURRENTLY ASSIGNED AS LGLOPR, CANNOT BE
RELEASED
RPL
replaces the current logical operator with the issuer of the command. The
programmable operator determines if a logical operator is currently assigned.
If there is, an implicit release is done. Then the issuer of the command
becomes the logical operator.
Usage Notes
Notes:
1.
The system keeps the logical operator that is specified on the LGLOPR statement in the routing table as a default. You cannot release this logical operator.
The LGLOPR ASN and LGLOPR RPL commands override the default.
2.
To ensure that messages are not lost when changing logical operators, the new
logical operator should issue a "LGLOPR RPL" command, rather than the
current logical operator issuing a "LGLOPR RLS" command and the new
logical operator issuing a "LGLOPR ASN" command.
Section 4. VM/SP Service Programs
4-75
L_. _______ .___._______________._____ ._.__________.. __._ _ _ _._________ .________--=-_ .___. . _. ___ . ________.______ . __ ._____ _
._. . . __. . ____ .. ____.___::J
3. If the default logical operator is the current logical operator and he issues the
LGLOPR ASN command, CP will send a message stating that he already
assigned as the logical operator
4.
The HOSTCHK function is suspended when an NCCF operator or a local VM
user is assigned as the logical operator. It is resumed when a remote VM user
is assigned as the logical operator.
Examples
The command issued at a virtual machine console:
MSG OP LGLOPR ASN
The command issued at an NCCF operator terminal:
PROP LGLOPR ASN
Responses
Both the new and old logical operators receive the message:
{NCCF/VM} USER 'userid ' ~'nodeid'] IS NOW LGLOPR
FOR PROP ON NODE 'nodeid
to notify them of the change of logical operators. The requester also receives the
response:
COMMAND COMPLETE
/
4-76
VM/SP Operator's Guide
S e u~u n~ (9 f,) [((0 ~ U·C) u"'uU 5
c·--··---------··---------·--··-·-····---·-··-·----.-----.----.---.--------------------- .. ---.----.-- ...... - ... -----.--------.--.------.--J
LOADTBL Command
Use the LOADTBL command to dynamically load a new routing table. You, as a
VM user or an NCCF operator, can specify whether or not the currently assigned
logical operator should be replaced by the logical operator specified in the new
routing table.
ILOADTBL I
[filename) [( RPL [ ) ) )
where:
filename
is the name of the routing table to be loaded. If filename is not given, the
default routing table (filename = PROP) is used.
RPL
replaces the currently assigned logical operator with the logical operator
specified in the new routing table. The logical operator is replaced only after
the new routing table has been successfully loaded.
If RPL is not specified, the logical operator in the new routing table simply
becomes the new default logical operator, and any explicitly assigned logical
operator (Le. a logical operator assigned by the LGLOPR command)
remains the logical operator.
Usage Notes
Notes:
1.
If any action routines named in the specified routing table can not be located
or loaded, an error message is issued, and the programmable operator facility
drops any action routine modules associated with the specified routing table. It
then tries to reload the action routine modules associated with the routing table
that was active before the LOADTBL command was issued. If it cannot load
these modules, the programmable operator facility terminates operation.
Note: If any of the action routines associated with the previous routing table
were modified (that is, replaced in LOADLIB) between the time the programmable operator facility dropped the specified routing table modules and
reloaded the previously active routing table modules, the modified version of
the action routines are used when the previous routing table is reloaded.
2.
Only the filename is used to identify the routing table file to the LOADTBL
command.
3.
Because DMSPOL is the action routine module executing the LOADTBL
command, it is not dropped and reloaded as are the other action routine
modules listed in the routing table. If you want to replace this module, you
must stop the programmable operator facility (using the STOP command),
make the desired modifications or replacement, and reinvoke the programSection 4. VM/SP Service Programs
4-77
mabIe operator facility as described under "Invoking the Programmable Operator Facility" on page 4-59.
If the programmable operator virtual machine is set up so that the program-
mable operator is started automatically when CMS is IPLed in that virtual
machine, it is sufficient to do the replacement and then IPL CMS again.
4.
With the loading done by DMSPOL, it is possible for the other routines in
DMSPOR to be replaced when a LOADTBL occurs. This permits changes to
action routines other than DMSPOL to be made dynamically, without stopping
the programmable operator. It is also possible to specify the name of a table to
be loaded as a parameter to the action routine. The logical operator will be
notified of the loading.
5. If the current logical operator is the default logical operator (not explicitly
assigned), then the current logical operator will be replaced even if the RPL
option is not specified.
6.
When you issue LOADTBL and replace the logical operator, both the ,019 and
new logical operators receive the following message:
(NCCF I VM] USER
nodeid '
I
user i d [node i d]
I
ISNOW LGLOPR FOR PROP ON NODE
Both operators receive this message, even if you have not specified the RPL
option.
Example
The command issued at a virtual machine console:
MSG OP LOADTBL ROUTE3 (RPL
The command issued at an NCCF operator terminal:
PROP LOADTBL ROUTE3 (RPL
Responses
NEW RTABLE NOT LOADED
PROP TERMINATED
PROP RUNNING WITH ROUTING TABLE 'fn ft fm'
/'
4-78
VM/SP Operator's Guide
Se~va«.;e
Pu-ogu"au-u1S
r=:==L.._ _ _ _ _ _._ _ _._ _ _ _ _-_-_-_---_-_-_.-._.-._-_.-.-_-_
..... -_.-_---------------.-----.--.. -----.---.--.---... --------,
LOG Command
Use the LOG command to write a message to the log file. Use of the LOG
command allows messages to be placed in the log file with no action taken by the
programmable operator facility.
The format of the LOG command is:
I
text...
text...
is the message text to be placed in the log file.
Usage Notes
Notes:
1.
All messages are logged whether or not the LOG command is explicitly used.
2.
Authorized non-NCCF users can retrieve the log file using the GET LOG
command. The log file has a fileid of "LGyymmdd nodeid AS" where "yy" is
the year, "mm" is the month, "dd" is the day, and "nodeid" is the nodeid of
the system on which the programmable operator facility is running.
The userid and nodeid of the sender is recorded along with the message.
3.
The length of the message is limited by the maximum length of the command
used to send the message to the programmable operator facility. If you want to
send a longer message, you must use the command multiple times.
Example
The command issued at a virtual machine console:
M OP LOG THIS MESSAGE IS TO BE LOGGED.
The command issued at an NCCF operator terminal:
PROP LOG THIS MESSAGE IS TO BE LOGGED.
Response
COMMAND COMPLETE
Section 4. VM/SP Service Programs
4-79
QUERY Command
Use QUERY RTABLE to find the name of the active routing table.
Use QUERY PROPCHK and QUERY HOSTCHK to query the status of the programmable operator node-checking.
Use QUERY LOGGING to query the status of the logging messages.
Use QUERY LGLOPR to find the name of the currently assigned logical operator.
The format of the QUERY command is:
QUERY
RTABLE
}
PROPCHK [nodeid]
HOSTCHK
{ LOGGING
LGLOPR
RTABLE
displays the name of the active routing table.
PROPCHK
displays a message with node-checking status. The message will state
whether PROPCHK is set ON or OFF. If a nodeid is supplied, only the
node specified is checked.
HOSTCHK
displays a message with node-checking status. The message will state
whether HOSTCHK is set ON or OFF.
LOGGING
displays a message with logging status. The message will state whether messages and responses are being logged (LOGGING ALL), incoming messages
and special programmable operator messages are being logged (LOGGING
ON), or there is no log (LOGGING OFF).
LGLOPR
for a VM logical operator, displays a message with the userid and nodeid of
the operator. For an NCCF logical operator, LGLOPR displays only the
userid (Le. operator-id) of the operator.
4-80
VMjSP Operator's Guide
Seuvnce
~uogu"auu1lS
[---.--.----------.-----------.. --------------.- -_..._. __ ..... __._._----------_._-----._--_._-_. .-._. __.__.._-----J
Example
The command issued at a virtual machine console:
MOP QUERY RTABLE
M OP QUERY PROPCHK NODEl
MOP QUERY HOSTCHK
M OP QUERY LOGGING
MOP QUERY LGLOPR
The command issued at an NCCF operator terminal:
PROP QUERY RTABLE
PROP QUERY PROPCHK VMSYSl
PROP QUERY HOSTCHK
PROP QUERY LOGGING
PROP QUERY LGLOPR
Responses
PROP RUNNING WITH ROUTING TABLE 'fn ft fm'
is received if the programmable operator facility is running.
fn
is the filename of the active routing table.
ft
is the filetype of the active routing table.
fm
is the filemode of the active routing table.
{PROPCHKIHOSTCHK} IS ON
is received if node-checking is in effect.
{PROPCHKIHOSTCHK} IS OFF
is received if node-checking was specified in the RTABLE but is currentlyoff.
PROPCHK IS {oNloFF} FOR NODEID
nodeid
is received if a specified nodeid is being queried.
Section 4. VM/SP Service Programs
4-81
c:,________
o_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
HOST-CHECKING IS SUSPENDED; LGLOPR NOT ON A CHECKABLE NODE
is received if the logical operator is an NCCF operator or a local VM
user.
{PROPCHKIHOSTCHK} NOT SPECIFIED IN RTABLE
is received if node-checking was not specified in the current RTABLE.
PROPCHK NOT SPECIFIED IN RTABLE FOR NODEID nodeid
is received if a nodeid was specified on the QUERY command and
node-checking was not specified in the current RTABLE for that
node.
LOGGING ALL
is received if incoming messages and all programmable operator
responses are being logged.
LOGGING ON
is received if incoming messages and special programmable operator
responses are being logged.
LOGGING OFF
is received if no logging is being done.
[NCCF IVM] USER user i d [node i d]
'nodeid '
I
I
ISNOW LGLOPR FOR PROP ON NODE
is received when the logical operator is being queried.
4-82
VM/SP Operator's Guide
Seu--vnce Programs
--------------------------------.-----J
SET Command
Use SET DEBUG to enter or exit the programmable operator facility DEBUG
mode. DEBUG mode is used to do problem determination on the programmable
operator facility.
Use SET PROPCHK to set the periodic checking of the programmable operator on
the distributed systems ON or OFF. The distributed systems are identified by the
PROPCHK statements in the routing table of the host programmable operator.
The programmable operator facility with the PROPCHK statement (e.g. the host
system) does the checking.
Use SET HOSTCHK to set the periodic checking of the link to the host system ON
or OFF. HOSTCHK must be specified in the routing table of the programmable
operator at the distributed system. The programmable operator facility with the
HOSTCHK statement (e.g. the distributed system) does the checking.
Use SET LOGGING to control messages going to the programmable operator log
file. SET LOGGING allows the message sender control the logging level: no
logging, logging incoming messages and special programmable operator messages,
or incoming messages plus response messages.
The format of the SET command is:
SET
DEBUG
Ig~Fl
PROPCHK
1~~Fl
HOSTCHK
Ig~l
[nodeid]
[nodeid]
LOGGING {'ON}
OFF
ALL
I
DEBUG ION
OFF
SET DEBUG ON stops the programmable operator facility from intercepting
responses to CP commands. SET DEBUG OFF allows the programmable
operator facility to return to its normal function of intercepting messages and
responses from CPo SET DEBUG OFF is the initial setting.
PROPCHK
I I
ON
nodeid
OFF nodeid
SET PROPCHK OFF halts checking of the programmable operators on the
distributed systems, until this command is reissued to set the checking back
Section 4. VM/SP Service Programs
4-83
§e~"vDcce P~1
Predumped Predumped Predumped Predumped Predumped
Page
Page
Page
Page
Page
0
1
3
2
1<-------------------------
4
B
Predumped Predumped Predumped Predumped
Page
Page
Page
Page
5
7
6
B
8
---------------->1
~OOOOOE
*Tape*
*Mark*
******
1<-
C
->1
SFBLOK
DMP1NREC
DMPKYREC
1
DMPKYREC
DUMPED
STORAGE
2
//
//
//
//
1<
B
C
==>
==>
==>
D
==>
A
I
D
Figure
DUMPED
STORAGE
******
*TAPE*
*MARK*
******
******
*TAPE*
*MARK*
******
----------------------------------->1
Written by the Stand-Alone Dump Utility on the 1PL tape at generation time.
Written by BOOTSTRAP on the IPL tape.
Written by BOOTSTRAP if the 1PL tape is the same as the dump tape.
Written by stand-alone dump program if the 1PL tape is not the same
as the dump tape.
Written by the stand-alone dump program on the dump tape.
4-19.
Stand-Along Dump Facility Tape Format
Section 4. VM/SP Service Programs
4-97
I.
1
DASD Format
When you use a DASD device to IPL the stand-alone dump program, the system
uses cylinder 0 to hold the program. Cylinder 0 must be CP formatted and allocated as permanent space. The stand-alone dump facility has the format shown in
Figure 4-20.
COTORl
COTOR2
COTOR3
IPL
Sequence
BOOTSTRAP
Part 1
VO LUr1E
LABEL·
1<---
A
COT2Rl
COT2R2
SAD
Part 2
BOOTSTRAP
Part 2
& SAD
Part 1
1<--COT3R2
------->1<-------
COTOR4
ALLOC.
~lAP
B
--->1
COT3Rl
Predumped
Page
0
A
--->11
COT(tR2
COT SRI
COTSR2
Predumped Predumped Predumped Predumped Predurnped
Page
Page
Page
Page
Page
1
2
3
4
5
1<---------------------COT6RI
COT6R2
D
COT7Rl
Predumped Predurnped Predumped
Page
Page
Page
6
8
7
D
A
B
C
D
==>
==>
==>
==>
------->1
Written by the Stand-Alone Dump Utility on the
IPL DASD at generation time.
Written by FORMAT/ALLOCATE program.
Written by BOOTSTRAP Part 1 on the IPL DASD.
Written by BOOTSTRAP Part 2 on the IPL DASD.
CnTnRn identifies cvlinderl track, and record numbers.
Figure
4-98
VM/SP Operator's Guide
4-20.
Stand-Alone Dump Facility DASD Format
Printer Format
Dumps to printer devices are printed as follows:
CP formats the following data fields for each processor, beginning with the
processor where the stand-alone dump program was IPLed:
CPU address (only if in AP or MP mode)
General purpose registers
Control registers
Floating point registers
Clock comparator
CPU timer values
Stored-status PSW
Prefix value (only if in AP or MP mode)
External interrupt old/new PSWs
SVC old/new PSWs
Program check old/new PSWs
Machine check old/new PSWs
I/O interrupt old/new PSWs.
The following fields are printed for the processor where the stand-alone dump
program was IPLed:
-
TOD clock
Lines of duplicate data will have a suppression message after the first line of
the data is printed.
A half page (2048 bytes) of all zeros has one line of zeros printed with the key,
followed by a line suppressed message.
The dump page is interpreted on the right-hand side of the printout.
Error Handling
Basic error recovery is available for DASD, tape, and printer devices used as IPL or
output devices. In addition, the following information may be of value when the
system detects errors:
The Channel Status Word is at location X'40'.
The I/O address is at location X'BA'.
32 bytes of sense data are at location X'2EO'.
The starting and ending addresses of the CP Trace Table are stored in the PSA
at X'7BO' and X'7B4', respectively, in addition to the low storage locations.
''
....... - - / /
Under certain error conditions, storage areas may be overlaid. This ,could cause
fields in SFBLOK and DMPINREC to be incorrect. (For example, fields containing date and time information.)
Section 4. VM/SP Service Programs
4-99
/
4-100
VM/SP Operator's Guide
r::=----
Section 5. Operator Spooling Functions
Spooling
Input and output files for use and access by virtual machines through unit record
devices are maintained by CP as disk data files using a mechanism called spooling.
Individual files can be identified and manipulated using various console functions.
The disk records are chained to form a logical file from dynamically assigned areas
on specially formatted CP disk areas. Data records from disk are read into available page space obtained through the CP paging mechanism. The data records
contain the actual data to be used and the CCWs to properly control the format.
These CCWs are directly executed to perform the actual unit record operations on
the real hardware. CP can support any number of virtual and real unit record
devices given sufficient system resources.
The data is placed in the spool buffers through the virtual machine unit record simulation routines in CP. Certain spool files have a special data format (system
dumps, for example) and are accessed using a special interface.
Spooling Considerations
VM/SP spooling facilities allow several virtual machines to share one or more unit
record devices. Since virtual machines controlled by CMS ordinarily have modest
requirements for unit record I/O, such device sharing is quite advantageous, and it
is the standard mode of system operation.
Each user has, as a general rule, a virtual reader, a virtual punch, and a virtual
printer as his spooling devices. In addition, the virtual console can also be classified as a spool file generator as all input and output to the console can be logged on
a spool file. This console log, and the files created by the user's virtual spooling
devices, can be processed by the real unit record devices that attach to the system.
CP controls and schedules the operation of the real unit record devices via spooling
techniques. Virtual machine SIO instructions directed to those unit record devices
designated as spool devices in the user directory entry are intercepted and modified
by CP. CP generates another I/O operation, transparent to the virtual machine,
which replaces the one specified. The new operation is directed to a CP spooling
disk area which acts as intermediate storage between the real unit record device
and the virtual machine. The data transfer operation between a spooled unit record
device and the virtual machine is, in reality, between a CP spool file and the virtual
Section 5. Operator Spooling Functions
5-1
machine. Spool file records are page size (that is, 4096 byte blocks), and are transferred between storage media via the CP paging mechanism.
When the system informs you that the spooling space is full or nearly full of spool
files, you can use the SPTAPE command to write those unit record files to tape to
relieve the situation. Then, when spool space is more plentiful you can use the
SPTAPE command to read the spool files back to the system for eventual spool file
output. See the description of the SPTAPE command in Section 3, "CP
Commands. "
Spooling Functions
CP spooling support performs the following five functions for virtual machines:
1.
It simulates, with software routines, the operation of the virtual unit record
devices that are attached to each user's virtual machine. The simulation makes
it appear that the program in the virtual machine controls real unit record
devices. Unit record device simulation involves the interception and interpretation of user Start I/O (SIO) instructions, the movement of data to and from
the user's virtual storage space, and the reflection of interruption codes and
ending conditions in the virtual machine.
2.
It operates the real unit record equipment attached to the real machine.
Spooling moves print-image and punch-card-image files to a real printer and
card punch, and creates spool files from data read in from the real card reader.
3.
It provides an easy-to-use interface between the virtual machine users, the
system operator, and the spooling system to allow flexible and easy switching
of system resources between many users. A set of general user and operator
commands can request these functions.
4.
It allows keyed-in CP commands and responses as well as virtual machine
console input and output to be placed on disk instead of, or in addition to,
being displayed at the terminal. You can initiate or terminate virtual console
spooling at any time during a terminal session.
5.
It provides the ability to spool files across the RSCS teleprocessing network.
Spooling commands can be separated by type into those that affect virtual devices,
those that affect real devices, and those that affect queued spool files within the
system. The commands that affect virtual devices are available to all general users;
a user may affect the status of devices that attach to only his virtual machine. For
a discussion of the Class G spooling commands, see VM / SP CP Command Reference for General Users.
Only the spooling or resource operator can use commands that affect the status of
the real spooling devices. Commands affecting the user's virtual machine closed
spool files that are awaiting processing are available to all users, with some additional capabilities available to the spooling operator. For example, a user can alter
characteristics of only those files that have the same userid as his, but the spooling
5-2
VM/SP Operator's Guide
operator can change the status of any spool file in the system. These commands
are described in Section 3, "CP Commands."
You can create input spool files (that is, data available at a specific user's virtual
card reader), when you place cards in the real card reader that are preceded by a
special VM/SP card that identifies the virtual machine userid of the user requesting
spool files.
When the virtual machine operating system writes to a virtual printer or card
punch, it creates output spool files and stores them on direct access devices. Real
output is scheduled for a real printer or card punch, or for remote output, whenever
a user logs off the system or issues a CP spooling command to close the file.
You can transfer specific files from the spooled printer or card punch of a virtual
machine to the card reader of the same or another virtual machine. (A virtual card
reader is not limited to 80-character records.) Files are not physically printed or
punched when transferred between virtual unit record devices by the spooling routines. With this method, you can make files available to multiple virtual machines,
or to different operating systems executing at different times in the same virtual
machine.
The Remote Spooling Communications Subsystem Networking, with the CP
spooling system, provides support for spooling across a teleprocessing network.
For detailed information about RSCS, see the VM I SP Program Reference and
Operations Manual.
VM/SP spooling includes many options for the virtual machine user and the real
machine operator. These include printing multiple copies of a single spool file,
backspacing any number of printer pages, and defining or reordering spooling
classes and forms for real output scheduling.
Real printers with the Universal Character Set feature must have the block data
check set when using VM/SP spooling. The LOADBUF command automatically
sets the block data check. For more details on spooling, see "Spooling
Considerations" on page 5-1.
-,
Virtual Console Spooling
The terminal user can spool virtual console output and virtual console input keyed
in by him in addition to typing or displaying it at the terminal. Furthermore, if the
virtual console is disconnected and the virtual machine is active, the console spool
file acquires all console output that would normally print or display at the terminal.
Virtual console 110 consists of all initiated CP commands and responses, as well as
the communication to and from the operating system running in the virtual
machine.
You start virtual console spooling when you issue the command:
SPOOL CONSOLE START
Section 5. Operator Spooling Functions
5-3
You close the console spool file at logoff time or when you stop and close the file.
When you close the console spool file, it goes on the printer spool file and then the
operator can manipulate it in the same way as any other printer spool file.
The display output of local or remote 3270s in DISPLAY mode (every 16 lines) is
written automatically on the spool file if the operator is the system operator and the
device was specified as a graphic device supported as a virtual machine operator's
console.
Console spooling for the system operator continues even if the operator disconnects from one terminal and logs onto another. To stop the automatic console
spooling, the operator must issue the SPOOL CONSOLE STOP command. If the
system fails, up to 16 lines of output may be lost from the system operator's spool
file, but the system closes the console spool file when the failure occurs. However,
virtual machine operators may lose up to one page of spooling data in a system
failure because their spooling buffer is one page (4096 bytes) in size. In regard to
DASD I/O, errors occurring on spool and paging devices produce error messages
that appear on the system operator's terminal.
A user can start or stop virtual console spooling any time during a terminal session
with the SPOOL CONSOLE command. For a description of the CP SPOOL and
CLOSE commands, see the VM / SP CP Command Reference for General Users.
Note: The user cannot use console spooling to stack commands for subsequent
execution. The console spool file is for historical purposes only. In addition, the
following types of data will not be placed in the console spool file for the indicated
conditions:
•
CP command output - if this is being received in a buffer via DIAGNOSE 8
•
Messages and Warnings - if they are being trapped via the IUCV and MSG
System Service.
I/O Error Handling
If a permanent I/O error occurs during a real I/O operation, you restart the
spooled files as follows:
•
Printer files - from the beginning of the current page (the last skip to channel
•
Punch files - from the beginning of the spool file.
•
Reader files - from the beginning of the real deck (the operator must normally
place the card deck back in the reader).
•
Console files - do not restart. CP ends console spooling and purges the
console file. The virtual machine continues without further console spooling.
o.
Note: If an I/O error occurs while punching accounting cards, CP will repunch
only the error card and not the whole file as in the case of normal punch spool files.
5-4
VM/SP Operator's Guide
/
§~(]OO~ nU1~
~c
r-aJ U1C·~U(1)U1S
__________________________________________________________________________________
~
Spool Buffers
The buffers used in virtual machines while collecting and writing spool data are
each one page (4096 bytes) long, and contain both the data to be transcribed and
all CCWs necessary for operating the unit record devices that perform the transcription. System failures that end system operation can cause virtual machine
operators to lose up to one page of spooling data.
Buffers used for the temporary storage of spool data on its way between auxiliary
storage and the user's virtual machine are allocated from a pool of virtual pages
that belong to CPo Direct access devices provide the auxiliary storage necessary for
CP spool buffering.
CP uses unit record devices to spool input and output operations. However, certain
features or functions that pertain to this group of machines are not supported.
Consult the VM/SP restrictions listed in the VM / SP Planning Guide and Reference
Spool Files
Each spool file in the system has a number of attributes that are assigned to it,
either explicitly or by default, at the time that it is created. These attributes and
their values are as follows:
Filename and Filetype: Each of these consist of alphameric fields of up to eight
characters. The CHANGE or CLOSE command can change these fields.
Spoolid: This is a system-assigned number between 1 and 9,900. It is automatically assigned when the file is closed, and is unique to that spooled file. To identify a given file, it is necessary to specify the userid of the file's owner, the device
type, and the spoolid number. In most instances, the userid defaults to the ID of
the user issuing the given command. Since the ID number, rather than the filename
and filetype, is used as an identifier, duplicate user-assigned names do not present
an identification problem.
Owner's User ID: This is the ID of the current owner of the file .
. Originating User ID: This is the ID of the file's creator. This is the same as the
current owner, unless the originator has used the SPOOL or TRANSFER commands to send the file to another user.
Number of Copies: This is the number requested for an output spool file; it is a
number between 1 and 255. Unless specified by the user or operator, it defaults to
1.
Date and Time: This is the date and time that the file was created. Users create
most files from virtual spool devices and the date and time indicate when the originator of the file closed the device.
Number of Records: This is an eight-digit number indicating the number of
logical record (printer or card images) in the file and is an indication of the size of
the file.
Section 5. Operator Spooling Functions
5-5
L __ .....__._______ .. _. ____._____._________________ ..
Distribution Code: This is an eight-character value that is normally assigned to
each user by the system administrator, although users may alter their assigned
value. It is printed in large block letters on the separator pages, and punched in the
separator cards. It may assist you in distributing output to users.
Hold Status: A spool file may be in "user hold," "system hold," or both. User
hold can be removed by the user, but only the operator can remove system hold.
Either hold status prevents a spool file from being printed or punched.
3800 Status: There are several attributes that apply only to printer files on the
3800 Printing Subsystem. See the discussion of the SPOOL command in the
VM/SP CP Command Reference for General Users, and the START command in
"START" on page 3-179 for additional information.
Output Classes
In addition to the attributes described above, a file queued for output on a real unit
record device always has an output class associated with it. A single alphameric
character (A through Z, 0 through 9, and the special character *) controls what real
device the file is to be printed or punched on, and the relative priority and sequence
of output on the device. Although each file is assigned a single class, you can designate each real spooling output device to handle from one to four classes. The
device processes only files that have a class code that corresponds to one of its
own, and processes these files in the order you specify. For example, if you assign
a printer the classes A, D, and 2, it always processes any printer file with class A
before it searches the printer output queue for a file with class D, and all class D
files are printed before any file with class 2. You can assign a class of * to a real
output unit record device to allow that device to process any class spool file.
The output class for a file is assigned at the time of the file's creation and is the
class that is associated with the virtual device that created it. When a user logs on
to the system, the class associated with the device is the one defined in his user
directory entry for that device; however, he may alter this class at any time with the
SPOOL command. After files are closed and are awaiting output, their class can be
altered with the CHANGE command either by the file's owner or by the spooling
operator. The spooling operator can alter the standard assignments of output
classes of a real output device with the START command.
Output priorities can also be rearranged by altering the hold status of a file. The
system operator can hold or release the files of specific users, all files of a given
type, or all files in the system. The hold function can ensure the immediate processing of high priority output files, while allowing high volume or low priority jobs
to be deferred for output until periods of low system usage.
5-6
VM/SP Operator's Guide
Output Forms
An output file has two form numbers associated with it - the user form and the
operator form. Each form is one to eight characters long. Any combination of
letters, numbers, or special characters may be used.
The user form is assigned by the user when he creates the spool file. He can later
change it.
The operator form is assigned by the system when the spool file is created. The
system looks in a table for the file's user form. The table provides the corresponding operator form. If the user form is not found in the table, the operator
form is set equal to the user form. The table is created by your installation's system
programmer using the SYSFORM macro. You can change the operator form of
any spool file with the CHANGE command.
When you start a real spooling printer or punch, you can specify the form number
that device is to process. Spool files with the specified operator form number will
be processed if the class matches and they are not held. This is called MANUAL
mode.
You can also specify AUTO move when you start the real spooling printer or
punch. In this mode, the system will process files with ally form number. The
system will send you the following message and sound the audible alarm each time
a new form is selected:
IPRTI
PUN
raddr MOUNT REQ CLASS a ... FORM form AUTO
I I
SEP
NOSEP
raddr
is the real address of the printer or punch.
a ...
are the classes that the device is processing.
form
is the form number for which a mount is requested.
The system then waits for you to make any necessary manual device adjustments
(such as mounting new forms). You then press START on the device, or enter the
command "ST ART raddr" at the console. The printer or punch device will then
print or punch the file.
All other files with the same form number will be printed or punched without
further intervention.
Some special printer forms require careful alignment on the printer (pre-printed
forms, for example). SETUP mode is provided to assist you in performing this
alignment. If you start the printer in SETUP mode, all of the steps described above
for AUTO mode are followed. However, when you press START on the printer,
only the first page of the file prints, with letters replaced by X's and numbers
replaced by 9's. You can then manually adjust and align the printer. Each time
Section 5. Operator Spooling Functions
5 -7
you press the START button, another page of the file is printed, and you can
further adjust the forms. When the forms are properly aligned, enter the command
"START raddr" at the console. The file will then print in its entirety.
At any time during the SETUP process, you may use the FLUSH command to terminate the process, or you can use a BACKSPAC command to return to the previous page of output.
Spooling Commands
The commands shown in Figure 5-1 control VM/SP spooling operations.
User Classification
Command
BACKSPAC
CHANGE
CLOSE
DRAIN
FLUSH
FREE
HOLD
LOADBUF
ORDER
PURGE
REPEAT
SPACE
SPOOL
SPTAPE
START
TAG
TRANSFER
Figure
5-1.
General User
X
X
X
X
Spooling Operator
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
CP Spooling Commands
This list represents all the CP commands that pertain to spooling operations. Users
having privilege classes D (the spooling operator) and G (the general user) can
issue these commands. Users with only the G privilege class cannot invoke class D
commands.
The spooling operator has responsibility for the following unit record spooling
operations. He must:
5-8
1.
Verify that the special VM/SP user identification (ID) card precedes a user's
physical deck of cards to associate the cards with the correct user's virtual
machine (See "Spooled Card Input" on page 5-9.).
2.
Make sure that sufficient blank cards are available for the punch unit, and that
the paper in the pri.nter is the proper size, width, and number of copies, and
that the proper print train is mounted and the print buffer is properly loaded.
3.
Separate punched and printed output by user identification (userid) code.
VM/SP Operator's Guide
§LQ)OO~ n1119
r-QJJ U1c-u:noU1s
=-:J
4.
Start, stop, restart, or rearrange the sequence schedules of spool files to be
printed or punched.
The spooling commands are discussed (along with other operator commands) in
Section 3, "CP Commands."
NOLOG Password
The NOLOG password is reserved for users who are not assigned any privilege
class. These users cannot log on to a virtual machine. The user with the NOLOG
password can submit jobs to the CMS Batch facility via the real card reader. The
user with the NOLOG password can also be the recipient of spool files sent by any
other logged-on user' only if the other user issued a SPOOL FOR command to the
userid having the NOLOG password.
Spooled Card Input
Spooled input from a real card reader requires a CP ID (identification) card before
the VM/SP system can accept it for processing.
Note: Load cards for only one user at a time. VM/SP does not look for ID cards
in the middle of the card deck.
The CP ID card must be inserted in front of the card deck. Then place the deck in
the empty feed hopper. Then ready the reader device and push the EOF button.
The reader immediately processes the cards (if no errors are encountered) and
places the file on a direct access device. The file now resides in the virtual machine
card reader that was identified by the CP ID card. Note that a file is transferred
from the real reader to the virtual reader regardless of whether that user is logged
on or off. The user may log on at any time to read the file.
CP 10 Card Format
10
USERIO
userid
[CLASS n]
[NAME
1.
Begin punching in column 1.
2.
The first field must be one of the following:
filename
{dsname
filetype}]
10
USERIO
3.
Separate all fields by one or more blanks.
4.
The second field must be the user's identification (userid), limited to eight
characters.
Section S. Operator Spooling Functions
5-9
5.
The third field (optional) is CLASS n (if class is assigned); one or more spaces
must separate the word CLASS and the class designation (n). (A through Z
and 0 through 9 are valid class designations.) If the class option is not coded,
the file defaults to class A. If the class option is coded, it must immediately
follow userid.
6. The fourth field (optional) is the keyword NAME, followed by filename and
filetype (or dsname alone). If only a dsname is specified, it can be 24 characters long. If both filename and file type are coded, they are each restricted to a
maximum of eight characters and must be separated by a blank.
An invalid or missing user CP ID card on the front of a card deck, when read,
causes an error message on the system console.
Spool Printer and Punch Output
Printed and punched output from virtual machine users is directed to the appropriate real unit record device. The operator (class D) can control this spooling
output by combinations of output class, form number, hold status, and other spool
command operands as discussed earlier.
Unless you used the NOSEP option to "START" a device, VM/SP prints a twopage header and a one-page trailer between all output on printers and punches a
four-card separator between all output on punch devices.
The printer header gives the pertinent spool file characteristics as well as real
device data. The header prints the owner userid and distribution code in block
letters on two pages. The header does not print between multiple copies of the file.
The card separator gives the owner's userid and the distribution code. The distribution code is expanded to punch each character in four columns with two separating blartks so that visual interpretation of the punched data is easier.
The printer trailer prints the sequence number in large block letters. The sequence
number is maintained for each printer in the system. It starts at 1 and increases by
1 for each file prin,ted on that real printer. When it reaches 999, it recycles to 1
again. The sequence number is also printed on the header pages. It will assist you
in separating and distributing output.
The SYSPCLAS macro instruction is available to classify printed output with a
classification title. This classification title is printed on the output separator page
and optionally at the top or bottom of each page of output.
Because output for a user file starts on a real device, the following message occurs:
{ PRT}
PUN
raddr
PR INT ING}
PUNCHING
{ REPEATED
USERID FILE
userid file
CLASS RECORDS RECLEFT CPY DIST SEQ
a typ norecs
recleft nnn dist
555
raddr
is the real device address.
5-10
VM/SP Operator's Guide
/
userid
is the identification of the owner of the spool file.
file
is the spool file spoolid number.
a
is the spool file class.
typ
is the originating device typ (PRT, PUN or CON).
norecs
is the logical record count (lines or cards) for the file.
recleft
is the number of records left to be printed or punched. This number is initially the same as "norecs" but decreases as the file is printed or punched.
nnn
is the current copy number being produced where 001 is the last copy.
dist
is the spool file distribution code.
555
is the sequence number of this file on this device. It is printed in large block
letters on the header and trailer pages.
The PRINTING/PUNCHING message occurs when the file output first starts and
the REPEATED message occurs for subsequent copies of the same file as they
start.
When the card deck is read in, the operator receives the following message:
RDR raddr
READING
}
{ HAS READ ...
USERID FILE CLASS RECORDS
userid file c RDR norecs
raddr
is the real address of the card reader.
userid
is the user identification of the owner of the file from the ID card at the front
of the deck.
file
is the spool file spoolid number assigned.
c
is the spool file class from the ID card at the front of the deck.
Section 5. Operator Spooling Functions
5-11
Spooling /Func1l:ions
norecs
is the number of cards read. Initially zero, this number increases as the file is
read.
The READING message occurs when the ID card is successfully read. The HAS
READ message occurs when the entire file is read and the reader stops.
An invalid or missing user identification card on the front of a card deck, when
read, causes an error message on the system console.
When the last file for an output device is processed and the device becomes idle,
the following message is produced:
{ PRT}
PUN
raddr WAITING CLASS c ... FORM form
MANUAL}
AUTO
{SETUP
}
{SEP
NOSEP
raddr
is the address of the device.
c ...
is the class(es) the device will process.
form
is the form number the device will process.
Spool Files for Virtual Machines Running Batch Jobs
Usually, when an installation has only one real reader, printer, or punch, the unit
record devices must be shared by all virtual machines. Before VM/SP can process
any virtual machine CP spool files and direct them to a real device, those files must
be closed. For virtual machines executing batch jobs and using operating systems
other than CMS, usually the operator who initially program loaded the virtual
machine operating system must intervene to issue the CP CLOSE command.
A feature that automatically closes CP spool files is available for OS/VS 1 and
DOS/VSE with VSE/ AF. Once output files are closed, they can be processed by
VM/SP without operator intervention.
Without this feature, CP spool files are not sent to the real printer or punch until
the virtual machine operator intervenes.
Using the eMS Batch Facility
The batch facility is a VM/SP programming facility that runs under CMS. It
allows a VM/SP user to run jobs in batch mode by sending jobs from either his
own virtual machine or the real card reader to a virtual machine dedicated to
running batch jobs underbthe batch facility. This dedicated machine is generally set
up at a terminal in the installation's computer room and is controlled by the system
operator.
5-12
VM/SP Operator's Guide
,/
c----------------------
The batch facility virtual machine runs continuously, executing all jobs spooled to
its virtual card reader from other virtual machines or from the real card reader.
The batch operator need pay no attention to the batch machine once he has
started, and disconnected it.
Starting the Batch Virtual Machine
The system operator starts the batch virtual machine by logging on with a batch
userid and loading CMS using the CP IPL command.
The Batch Use rid
Every installation in which the CMS batch facility is available should establish one
or more common userids for the Batch Monitor virtual machine. Users can then
spool their files for execution to the card reader for that batch userid.
It is the operator's responsibility to log on the VM/SP system using the batch
use rid that has been established for his installation.
The userid established for the batch facility virtual machine must have a read/write
disk in its directory at virtual address 195. The 195 disk is erased when it is
accessed as the A-disk at the beginning of each job.
Invoking the Batch Facility
The batch facility virtual machine is invoked by the batch operator when he issues
the CP IPL command with P ARM BATCH or CP IPL followed by the
CMSBATCH command. The latter method follows:
ipl cms
eMS mm/dd/yy WED 17.58.48
cmsbatch
Y/S (19E) RIO.
THE FOLLOWING NAMES ARE UNDEFINED:
BATEXIT1 BATEXIT2
R; T=0.14/0.39 08:47:40
WAITING FOR THE READER
The operator may now disconnect the batch machine terminal, if he wishes, using
the CP DISCONN command. The batch facility will IPL itself after each job is
executed.
All virtual machine (CMS) console output is automatically spooled to a file to be
printed after the program output at the real system printer. All commands entered
through the virtual reader are displayed on the console to allow them to appear in
the console output file. If the batch terminal is disconnected, only CP and batch
initialization messages are displayed at the terminal.
If an installation wishes to use a saved system in running batch jobs, the operator
must enter the name of the saved system in the CMSBATCH command line.
Section 5. Operator Spooling Functions
5 -13
§ ~] (Q)(!J ~ DrD 91
L___ _
r- nJl ril C1c
D
A
Create accounting records for logged on
users, and reset accounting data. ACNT
also closes the spool file that is accumulating accounting records.
ADSTOP
G
Halt execution at a specific virtual
machine instruction address.
ATTACH
B
Logically connect a real device to a
virtual machine for that machine's exclusive use or logically connect a DASD
device for CP access and control. With
CHANNEL operand, dedicate all devices
on a particular channel to a specific user.
Appendix A. Summary of CP Commands
A-I
COMMAND
FUNCTION
TYPE*
IBMDEFINED
PRIVILEGE
CLASS
ATTN
G
Make an attention interruption pending
for the virtual machine console.
AUTOLOG
<0>
A,B
Log on any virtual machine defined in the
directory.
BACKSPAC
D
Restart or reposition the current output
on a real punch or printer.
BEGIN
G
Continue or resume execution of the
virtual machine at either a specific storage
location or at the address in the current
PSW.
CHANGE
S
D
Alter one or more external attributes of a
closed spool file or files.
G
G
Alter one or more attributes of a closed
spool file.
CLOSE
G
Terminate spooling operations on a
virtual card reader, punch, printer, or
console.
COMMANDS
N/A
Any
COUPLE
G
CP
N/A
Any
CPTRAP
C
Create a reader file of selected trace
table, CP interface, and virtual machine
interface entries for problem determination.
DCP
C,E
Display the contents of real storage
locations at the terminal.
DEFINE
R
B
Redefine the status of a 3330V volume.
G
G
Reconfigure your virtual machine.
R
B
Remove a real device from the CP
system. With the CHANNEL operand,
remove a dedicated channel from a user.
G
G
Detach a virtual device from a virtual
machine. Detach a channel from your
virtual machine.
N/A
Any
DETACH
DIAL
J
I
A -2
VM/SP Operator's Guide
USERDEFINED
CLASS
None
Usage
Display the commands and diagnose
codes you are authorized to use.
Connect channel-to-channel devices.
None
None
Execute a CP command while remaining
in the CMS virtual machine environment.
Connect a terminal or display device to
the virtual machine's virtual communication line.
COMMAND
FUNCTION
TYPE*
IBMDEFINED
PRIVILEGE
CLASS
USERDEFINED
CLASS
Usage
DISABLE
A,B
DISCONN
N/A
Any
DISPLAY
G
Display virtual storage on your terminal.
DMCP
C,E
Print the contents of real storage
locations on a user's virtual spooled
printer.
DRAIN
D
Stop spooling operations on a specified
real unit record devices after the file currently being processed has been completed.
DUMP
G
Print the following on the virtual printer:
Virtual PSW, general registers, floatingpoint registers, storage keys, and contents
of specified virtual storage locations.
ECHO
G
Test terminal hardware by redisplaying
data entered at the terminal.
ENABLE
A,B
Enable the previously disabled or
nonenabled devices so users may access
the system.
EXTERNAL
G
Simulate an external interruption for a
virtual machine and return control to that
machine.
FLUSH
D
Halt and immediately purge or hold the
current output on a specified real unit
record device.
FORCE
<0>
A
Force a logoff of any user on the system.
FREE
D
Remove a set of spool files belonging to a
specified user from a system hold status.
HALT
<0>
A
Terminate any active channel program on
a specified real device.
HOLD
D
Place user spool files in a system hold
status.
INDICATE
0
A
Provides a list of statistics for all users
who have the favored execution option.
A
E
Display, at the console, the use of and
contention for system processor and
storage.
Prevent low-speed communication lines
from accessing the system.
None
Disconnect your terminal from your
virtual machine.
Appendix A. Summary of CP Commands
A -3
COMMAND
FUNCTION
TYPE*
IBMDEFINED
PRIVILEGE
CLASS
USERDEFINED
CLASS
Usage
G
G
Indicate resource utilization and contention.
IPL
G
Simulate IPL for a virtual machine.
LINK
G
Provide access to a specific DASD by a
virtual machine.
LOADBUF
D
On a 1403 printer load the Universal
Character Set(UCS) with a specified print
chain/train image. On 3203,3211,3262,
4245, or 4248 printers, load UCS or
Forms Control Buffer(FCB) with a specified image. On 3289 Model 4 printer,
load the Font Offset Buffer (FOB) with
the image print belt and FCB.
LOADVFCB
G
Load virtual forms control buffer for a
virtual 3203, 3262, 3289E, 3211,4245,
or 4248 printer.
LOCATE
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-c043 52.372728, 2009/01/18-15:56:37 Producer : Adobe Acrobat 9.31 Paper Capture Plug-in Modify Date : 2010:03:19 17:37:16-08:00 Create Date : 2010:03:19 17:37:16-08:00 Metadata Date : 2010:03:19 17:37:16-08:00 Format : application/pdf Document ID : uuid:c5849e03-1a2b-4a46-a99d-01a0fe17ab55 Instance ID : uuid:1e1fd523-2df7-4bab-b290-9e9b42df9a5c Page Layout : SinglePage Page Mode : UseNone Page Count : 406EXIF Metadata provided by EXIF.tools