SharePlex Guide

Administer SharePlex

Administer, Administration, s, SharePlex

SMARTIN

Browser built in viewer
PDF Viewer
Universal Document Viewer
Google Docs View
Google Drive View
Download Document [pdf]

File Info : application/pdf, 335 Pages, 4.68MB

Document

SharePlex 9.2.8 Admin Guide

SharePlex® 9.2.8
Administrator Guide

© 2020 Quest Software Inc. ALL RIGHTS RESERVED.
This guide contains proprietary information protected by copyright. The software described in this guide is furnished under a software license or nondisclosure agreement. This software may be used or copied only in accordance with the terms of the applicable agreement. No part of this guide may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording for any purpose other than the purchaser's personal use without the written permission of Quest Software Inc.
The information in this document is provided in connection with Quest Software products. No license, express or implied, by estoppel or otherwise, to any intellectual property right is granted by this document or in connection with the sale of Quest Software products. EXCEPT AS SET FORTH IN THE TERMS AND CONDITIONS AS SPECIFIED IN THE LICENSE AGREEMENT FOR THIS PRODUCT, QUEST SOFTWARE ASSUMES NO LIABILITY WHATSOEVER AND DISCLAIMS ANY EXPRESS, IMPLIED OR STATUTORY WARRANTY RELATING TO ITS PRODUCTS INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT. IN NO EVENT SHALL QUEST SOFTWARE BE LIABLE FOR ANY DIRECT, INDIRECT, CONSEQUENTIAL, PUNITIVE, SPECIAL OR INCIDENTAL DAMAGES (INCLUDING, WITHOUT LIMITATION, DAMAGES FOR LOSS OF PROFITS, BUSINESS INTERRUPTION OR LOSS OF INFORMATION) ARISING OUT OF THE USE OR INABILITY TO USE THIS DOCUMENT, EVEN IF QUEST SOFTWARE HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Quest Software makes no representations or warranties with respect to the accuracy or completeness of the contents of this document and reserves the right to make changes to specifications and product descriptions at any time without notice. Quest Software does not make any commitment to update the information contained in this document.
If you have any questions regarding your potential use of this material, contact:
Quest Software Inc.
Attn: LEGAL Dept
4 Polaris Way
Aliso Viejo, CA 92656
Refer to our Web site (https://www.quest.com) for regional and international office information. Patents
Quest Software is proud of our advanced technology. Patents and pending patents may apply to this product. For the most current information about applicable patents for this product, please visit our website at https://www.quest.com/legal. Trademarks
Quest, the Quest logo, SharePlex, and Join the Innovation are trademarks and registered trademarks of Quest Software Inc. For a complete list of Quest marks, visit https://www.quest.com/legal/trademark-information.aspx. All other trademarks and registered trademarks are property of their respective owners.
SharePlex Administrator Guide Updated - 6/24/2020 Version - 9.2.8

SharePlex 9.2.8 Administrator Guide

Contents

About this Guide

directory

save state
temp oos

Information about active and inactive configurations.
Information about the current state of SharePlex when a configuration is active, such as the object and sequence caches.
Used by the copy and append features and other SharePlex sync-related processes.
Stores the transactions that contain out-of-sync operations when the SP_OPO_SAVE_OOS_ TRANSACTION parameter is enabled.

The sp_cop process
The sp_cop program coordinates the SharePlex replication processes: (Capture, Read, Export, Import, Post) and the SharePlex queues, and it initiates all of the other background processes that perform specific tasks. It also maintains communication with other systems in the replication network. In general, most SharePlex users have little interaction with sp_cop other than to start and stop it. Once started, sp_cop runs in the background.
l Only a SharePlex Administrator (member of the SharePlex admin group) can start or stop sp_cop. l sp_cop must be started on all source and target systems involved in replication. l Start sp_cop as soon as (or before) applications access the data on the source system, so that all of the
SharePlex processes are ready to start processing transactions. That way, Capture can keep pace with the changes that are made to the source data.
The sp_ctrl process
Use sp_ctrl to issue the commands that start, stop, configure, direct and monitor SharePlex activities. The sp_ ctrl program interacts internally with the sp_cnc (command and control) process, which is the child process of sp_cop that executes the commands. Users do not interact with sp_cnc itself.

SharePlex replication processes
SharePlex replicates data through a series of replication processes that are started by the main SharePlex process, sp_cop.

SharePlex 9.2.8 Administrator Guide 25

l The Capture process: The Capture process reads the transaction records on the source system for changes to objects that are configured for replication by SharePlex. The Capture process writes the data to the capture queue, where it accumulates until the Read process is ready for it. When data is being replicated from more than one datasource, there is a separate Capture process for each one, each functioning concurrently and independently. The Capture process is named sp_ocap (Oracle Capture).
l The Read process: The Read process operates on the source system to read data from the capture queue and add routing information to it. After processing the data, the Read process sends it to the Export queue. The Read process is named sp_ordr.
l The Export process: The Export process operates on the source system to read data from the export queue and send it across the network to a target system. By default, there is one Export process for each target system. For example, if there are two target systems, there are two Export processes. The Export process is the first part of the Export/Import transport pair, which moves data between systems over a TCP/IP network. The Export process is named sp_xport.
l The Import process: The Import process is the second half of the Export/Import transport pair. The Import process operates on a target system to receive data and build a post queue. There is one Import process on a target system for each Export process that sends data to that target. For example, if there are two source systems (each with an Export process) replicating data to a single target system, there are two Import processes on that target. The Import process is named sp_mport. NOTE: It is possible to replicate data between databases on the same system. In this case the Export and Import processes are not created. The Read process places data directly into a post queue on that system.
l The Post process: The Post process operates on a target system to read the post queue and apply the replicated operations to the target database, file, message queue or topic. There is a Post process for each post queue on a target system. Multiple Post processes can operate simultaneously on a system. The Post process is sp_opst_mt (Oracle Post) or sp_xpst (Open Target Post).
All communication and movement of data by SharePlex is handled by an internal messaging and transport system, using an asynchronous stream protocol with TCP/IP connections that is very efficient for large data transfers. This method ensures optimal performance, reliability and restart capabilities, while conserving communication bandwidth. SharePlex can replicate over any TCP/IP network.
SharePlex 9.2.8 Administrator Guide 26

SharePlex queues
Queues store the replicated data as it is transported from the source system to the target system. Queues are part of a checkpoint recovery system that facilitates safe, asynchronous transport of data. Data travels through the queues in the sequence in which it was generated. Data is not read-released (deleted) from one queue until it is written to the next one. Data accumulates in the queues on the source and target systems if the network, system, or database slows down or fails, or when a replication process stops. When the problem or outage is resolved, SharePlex resumes processing from the point where it stopped. SharePlex replication uses the following queues:
l Capture Queue: The capture queue resides on the source system and stores captured data for further processing by SharePlex. There is one capture queue for each datasource that is being replicated. A capture queue is identified by the datasource, for example o.fin1.
l Export Queue: The export queue resides on the source system. It holds data that has been processed by SharePlex and is ready for transport to the target system. By default, there is one export queue on a source system regardless of the number of active configurations or target systems. A default export queue is identified by the name of the source system on which it resides, for example, SysA. You can instruct SharePlex to create additional namedexport queues for more complex replication strategies.
l Post Queue: The post queue resides on the target system. It holds data that is ready for Post to write to the target database, file, or message queue or topic. On each target system, there is one post queue for the replication stream between a datasource and its target. For example, if DatabaseA and DatabaseB are both replicating to DatabaseC, there are two post queues. A default post queue is identified by the name of the source system plus the datasource and the target, for example SysA (o.DatabaseAo.DatabaseB). You can instruct SharePlex to create additional named post queues for more complex replication strategies.
NOTE: All SharePlex queue files are created and maintained in the rim sub-directory of the SharePlex variabledata directory.

SharePlex Installed Objects
Much of the replication process is controlled and tracked through a series of internal objects that are installed into the source or target database during the installation of SharePlex. They are essential for SharePlex to operate, so do not alter them in any way. NOTE: Not all objects are used for all databases. Most are used for Oracle databases. If you do not see an object in your database, it is not relevant to the database, or the information is stored internally within the SharePlex configuration. If you see an object that is in your database but not in this list, it is not being used in the current release.

Table
DEMO_SRC DEMO_DEST SHAREPLEX_ACTID

Object type
Table Table Table

Description
Used as the source table for the SharePlex demonstrations. Used as the target table for the SharePlex demonstrations. Used by Capture to checkpoint its state.

SharePlex 9.2.8 Administrator Guide 27

Table

Object type

Description

SHAREPLEX_ANALYZE Table

Used by the analyze command.

SHAREPLEX_CHANGE_ Table OBJECT
SHAREPLEX_COMMAND Table

Used by users to stop and resume replication for an object. Used for the flush, abort and purge commands.

SHAREPLEX_CONFIG SHAREPLEX_DATA

Table Table

Used by the activation and Capture processes to mark the start of a new activation.
Used by the SharePlex wallet for Oracle TDE replication.

SHAREPLEX_ DATAEQUATOR
SHAREPLEX_ DATAEQUATOR_ INSERT_TEMP
SHAREPLEX_ DATAEQUATOR_ UPDATE_TEMP
SHAREPLEX_ DATAEQUATOR_ DELETE_TEMP
SHAREPLEX_DDL_ CONTROL

Table Table

Used by the compare and repair commands and the Post process to synchronize their operations.
Used as a temporary table by the compare and repair commands.

Table

Used as a temporary table by the compare and repair commands.

Table

Used as a temporary table by the compare and repair commands.

Table

Used to refine control of DDL that is enabled for replication by the SP_OCT_REPLICATE_ALL_DDL parameter.

SHAREPLEX_JOBID

Sequence Used by the sp_cnc process and the compare, repair, and copy commands to provide a unique job ID.

SHAREPLEX_JOBS

Table

SHAREPLEX_JOB_ STATS
SHAREPLEX_JOBS_ CONFIG
SHAREPLEX_LOB_ CACHE
SHAREPLEX_LOBMAP

Table Table Table Table

SHAREPLEX_LOGLIST Table

Used by the sp_cnc process and the compare, repair, and copy commands to store information about a job. Used by the sp_cnc process and the compare, repair, and copy commands to store information about a job. Used by the disable jobs and enable jobs commands.
Used by the Capture process when processing VARRAYs stored as LOB.
Used by the Capture process to map LOBIDs and rows when a table with LOB columns does not have PK/UK logging enabled.
Used by the Capture process to track inactive RAC instances.

SHAREPLEX_MARKER Table

Used by the Read process when PK/UK logging is not enabled.

SHAREPLEX_OBJMAP
SHAREPLEX_ PARTITION_CACHE

Table Table

Used by the activation and Capture processes to define the objects in replication.
Used by the Capture process to map Oracle partition IDs to tables in replication.

SharePlex 9.2.8 Administrator Guide 28

Table
SHAREPLEX_SYNC_ MARKER SHAREPLEX_TRANS or SHAREPLEX_OPEN_ TRANS

Object type Table
Table

Description
Used by the copy command and the Read and Post processes to sync their operations. Used by the Post process to store checkpoints and to mark transactions that were applied in a master-to-master configuration.

How SharePlex replication works
To replicate data, SharePlex reads the stream of transaction data on the source system and captures changes that are made to objects that are specified in a configuration file. In the configuration file, you specify which data to replicate and the target to which it is applied. You activate a configuration file to start replication. This is done by means of the activate config command in sp_ctrl within a sequence of steps that also includes synchronizing the source and target data for the first time. When a configuration is active, SharePlex replicates only the changes that are made to the objects specified in the configuration file, not entire data records, which provides a fast and reliable replication solution. For more information see:
l Configure data replication on page 57
l Start replication on your production systems on page 186
From the information that it has about a transaction operation, SharePlex creates one or more messages that are sent from the source system to the target system. A message can reflect a SQL operation or an internal SharePlex operation, but most of the time it is an INSERT, UPDATE, DELETE, COMMIT, TRUNCATE or a supported DDL operation.
NOTE: Large operations like those on LONG or LOB columns can require more than one message because a message has a size limitation. Other operations, such as array inserts of small records, have the inverse effect: There could be one record for numerous operations. For example, an array insert of 70,000 rows might be recorded in the transaction stream as only 700 messages, depending on the data. In general, unless you are replicating numerous changes to those kinds of data types, you can assume that the number of messages shown in the status output for a process or queue approximately corresponds to the same number of SQL operations. The Post process reads messages from the post queue and applies the replicated data changes to the target. In the case of a database target, Post constructs SQL statements to apply the data. In the case of nondatabase targets, Post outputs data records in the format required by the target, for example a file or messaging queue or topic. The following explains the default ways that SharePlex builds SQL statements on the target system.
l If the change is an INSERT, SharePlex uses all of the columns in the row to build an INSERT statement.
l If the change is a DELETE, SharePlex uses only the key to build a WHERE clause to locate the correct row. In the case of Oracle, if a table lacks a key, SharePlex simulates one by using the values of all of the columns, except LONG and LOB columns. You can specify columns to use as a key when you create the configuration file. In the case of SQL Server, all configured objects must have a primary key.

SharePlex 9.2.8 Administrator Guide 29

l If the change is an UPDATE, SharePlex uses the key plus the values of the changed columns to build a WHERE clause to locate the correct row. Before applying changes to the database, the Post process compares a pre-image of the values of the source columns to the existing values of the target columns. The pre-image (also known as the before image) is the value of each changed column before the UPDATE. If the pre-image and the existing target values match, confirming a synchronized state, Post applies the changes. If not, then Post logs the operation to an error file and SharePlex returns an "out-ofsync" error.
l If the change is an UPDATE or DELETE statement that affects multiple rows on the source machine, SharePlex issues multiple statements on the target to complete the task. For example, an UPDATE tableA set name = `Lisa' where rownum < 101 statement actually sends 100 UPDATE statements to the target, even though only one statement was issued on the source.
Understand the concept of synchronization
The concept of synchronization applies mainly to table-to-table replication, where Post performs integrity checks to make certain that only one row in the target matches the row change that is being replicated. It does not apply to file, messaging targets, and change-history targets, which contain a record of every operation replicated by Post, some of which may be identical over time. The Post process does not perform integrity checks on those targets.
Characteristics of synchronized tables
The basic characteristics of synchronized source and target tables are as follows (unless the transformation feature is used).
l If a row exists in the source database, it exists in the target. l Corresponding columns in source and target tables have the same structure and data types. l Data values in corresponding rows are identical, including the values of the key. Ensuring data integrity is the responsibility of the Post process. Post applies a WHERE clause to compare the key values and the before values of the SQL operations that it processes. Post uses the following logic to validate synchronization between source and target tables:
l Post applies a replicated INSERT but a row with the same key already exists in the target. Post applies the following logic: o If all of the current values in the target row are the same as the INSERT values, Post considers the rows to be in-sync and discards the operation. o If any of the values are different from those of the INSERT, Post considers this an out-ofsync condition. NOTE: You can configure Post so that it does not consider non-key values when posting an INSERT. See the SP_OPO_SUPPRESSED_OOS parameter in the SharePlex Reference Guide.
SharePlex 9.2.8 Administrator Guide 30

l Post applies a replicated UPDATE but either cannot find a row in the target with the same key value as the one in the UPDATE or Post finds the correct row but the row values do not match the before values in the UPDATE. Post applies the following logic:
o If the current values in the target row match the after values of the UPDATE, Post considers the rows to be in-sync and discards the operation.
o If the values in the target row do not match the before or after values of the UPDATE, Post considers this an out-of-sync condition.
NOTE: You can configure Post so that it returns an out-of-sync message if the current values in the target row match the after values of the UPDATE. See the SP_OPO_SUPPRESSED_OOS parameter in the SharePlex Reference Guide.
l A DELETE is performed on the source data, but Post cannot locate the target row by using the key. When Post constructs its DELETE statement, it includes only the key value in its WHERE clause. If the row does not exist in the target, Post discards the operation.
Hidden out-of-sync conditions
Post only verifies the integrity of the rows that are being changed by its current SQL operation. It does not verify whether other rows in that table, or in other tables, are out of synchronization in the target database. A hidden out-of-sync condition may not show up until much later, when a change to the affected row is eventually replicated by SharePlex or a discrepancy is detected in the course of using that data.
Example of a detectable out-of-sync condition Someone logs into the target and updates the COLOR column in the target table from "blue" to "red" in Row1. Then, an application user on the source system makes the same change to the source table, and SharePlex replicates it to the target. In the WHERE clause used by Post, the pre-image for the target table is "blue," but the current value in the target row is "red." Post generates an out-of-sync error alerting you to the out-of-sync condition. Example of a hidden out-of-sync condition Someone logs into the target and updates the COLOR column in the target table from "blue" to "red" in Row2, but the change is not made to the source table and is not replicated. The two tables are now outof-sync, but Post does not return an error message, because there is no replication performed on that row. No matter how many subsequent updates are made to other columns in the row (SIZE, WEIGHT), the hidden out-of-sync condition for the COLOR column persists (and users on the target have inaccurate information) until someone updates the COLOR column in the source table. When that change is replicated, only then does Post compare the pre-images and return an error message. The majority of time, the cause of out-of-sync data is not anything done wrong by replication, but rather DML applied on the target, an incomplete backup restore, or some other hidden out-of-sync condition, which goes undetected until replication affects the row. Solving out-of-sync conditions can be time-consuming and disruptive to user activity. Once replication is started, it is recommended that you:
l Prevent write access to the target tables, so that DML and DDL cannot be applied to them. l Use the compare command to compare source and target data regularly to verify synchronization and
detect hidden out-of-sync conditions. You can use the repair command to repair any out-of-sync rows. For more information about these commands, see the SharePlex Reference Guide.
SharePlex 9.2.8 Administrator Guide 31

How SharePlex responds to an out-of-sync condition
You can decide how you want SharePlex to respond to transactions that generate an out-of-sync error: l The default Post behavior when a transaction contains an out-of-sync operation is to continue processing other valid operations in the transaction to minimize latency and keep targets as current as possible. Latency is the amount of time between when a source transaction occurs and when it is applied to the target. Different factors affect the amount of latency in replication, such as unusually high transaction volumes or interruptions to network traffic. Post logs the SQL statement and data for the out-of-sync operation to the ID_errlog.sql log file, where ID is the database identifier. This file is in the log sub-directory of the variable-data directory on the target system. l You can configure Post to stop when it encounters an out-of-sync condition by setting the following parameter to 1: l Oracle targets: SP_OPO_OUT_OF_SYNC_SUSPEND l Open Target targets: SP_OPX_OUT_OF_SYNC_SUSPEND l You can configure Post to roll back and discard a transaction if any operation in that transaction generates an out-of-sync error. The entire transaction is logged to a SQL file, but not applied to the target.You can edit the SQL file to fix the invalid DML and then run the SQL file to apply the transaction. This feature is enabled by setting the SP_OPO_SAVE_OOS_TRANSACTION to 1.
Strategies for information availability
With SharePlex, you can put a replica database to work as a reliable, continuously updated alternate database that can be used in many different ways. The following strategies enable you to get the right data to the people who need it, when they need it. NOTE: Support for these topologies may vary depending on the type of database involved.
SharePlex 9.2.8 Administrator Guide 32

Figure 1: SharePlex replication strategies at a glance
Reporting instances
Targets maintained by SharePlex are ideal for offloading report and query processing because they are accessible while being kept up-to-date, and they can be optimized with keys and indexes designed for optimal query performance. You can run reports all day long, without complaints about performance from your OLTP users. Even during busy reporting times such as the end of the month or quarter, application response time will be unaffected by heavy reporting. And, your organization's decision-makers will appreciate the accuracy of the data reflected in the reports. For more information, see Configure replication to share or distribute data on page 114.
SharePlex 9.2.8 Administrator Guide 33

Broadcast and cascade
When many remote users access or use data stored in a primary database, you can move their processing to one or more secondary databases that are kept current through SharePlex replication. That way, you can keep the primary database and system optimized for transactions. SharePlex also can cascade data through an intermediary system to remote systems, providing access for remote users who have no direct network connection to the primary database. For more information, see Configure replication through an intermediary system on page 143
Data warehousing
SharePlex can replicate from numerous source systems to one target system. This configuration is ideal for consolidating data in a data warehouse or a data mart so that information is available enterprise-wide for queries and reports. You have control over the data that is replicated and the option to transform any data to conform to a different target structure. These capabilities enable you to populate your data warehouse with the specific, timely information that users need to make good decisions. For more information, see Configure replication to maintain a central datastore on page 118.
High availability and disaster recovery
SharePlex can be used to maintain duplicate databases over local or wide-area networks. Production can move to the alternate sites in an emergency or in a planned manner when routine maintenance is performed on the primary server. SharePlex replication enables the secondary database to be used for queries and reporting. For more information, see Configure replication to maintain high availability on page 148.
Peer-to peer
SharePlex supports replication among multiple source databases where applications on each system can make changes to the same data while SharePlex maintains synchronization. In this strategy, the databases are usually mirror images of each other, with all objects existing in their entirety on all systems. Although similar in benefit to a high-availability strategy, the difference between the two is that peer-to-peer allows concurrent changes to the same data, while high availability permits changes to the secondary database only in the event that the primary database goes offline. A few ways to use peer-to-peer replication are to maintain the availability and flexibility of a database by enabling access from different locations or to distribute heavy online transaction processing volumes among multiple access points. For more information, see Configure peer-to-peer replication on page 123.
Test before you deploy
Before you implement SharePlex on production systems, make certain to perform tests in a mirror of the production environment to ensure that you configured SharePlex properly to support your requirements. Testing can uncover issues such as configuration errors and unexpected environmental issues, for example network or resource issues that affect replication performance or availability. Additionally, it is assumed that your organization has in place an infrastructure that supports the use of enterprise applications such as SharePlex. These include, but are not limited to, the following:
SharePlex 9.2.8 Administrator Guide 34

l Availability and use of the database and SharePlex documentation l Training programs for users l Rollout and upgrade plans that ensure minimal interruption to business. When SharePlex is
implemented as part of an application's infrastructure, it is strongly recommended to test new application functionality in conjunction with SharePlex in a non-production environment. l Database or system maintenance procedures that consider SharePlex dependencies, such as the proper shutdown of SharePlex processes and the preservation of unprocessed transaction records and replication queues to accommodate system or database maintenance. l Sufficient security that prevents unauthorized persons from accessing SharePlex data records or making configuration changes. The SharePlex Professional Services team can help you prepare for, install, and deploy SharePlex in your environment.
SharePlex 9.2.8 Administrator Guide 35

2
Run SharePlex
This chapter contains instructions for running SharePlex on UNIX, Linux, and Windows
Contents
Run SharePlex on Unix and Linux Startup sequence on Unix and Linux Start SharePlex on Unix and Linux Identify SharePlex processes on Unix and Linux Stop SharePlex on Unix and Linux Shutdown Considerations on Unix and Linux
Run SharePlex on Windows Startup Sequence on Windows Start and stop SharePlex on Windows Set SharePlex startup status on Windows Set SharePlex process priority on Windows Identify SharePlex processes on Windows Shutdown Considerations on Windows
Run SharePlex on Unix and Linux
On Unix and Linux systems, you start SharePlex by running the sp_cop program. After you activate a configuration, sp_cop spawns the necessary child replication processes on the same system. Each instance of sp_cop that you start is a parent to its own set of child replication processes. The sp_cop process must be started on each system that is part of the replication configuration. You can start sp_cop in one of two ways:
l From the operating system command line. l At system startup as part of the startup file. IMPORTANT: Run SharePlex from either the korn (ksh) or C shell (csh) shell.
SharePlex 9.2.8 Administrator Guide 36

l Do not use the Bourne shell (sh), because the way it handles background processes is not compatible with SharePlex. If you must use the Bourne shell, switch shells to ksh or csh to run SharePlex, then exit the shell and return to the Bourne shell.
l If using an Exceed X window emulator, switch from the default shell of POSIX to the ksh shell, then run sp_cop from the ksh shell only.

Startup sequence on Unix and Linux
When you start systems that are involved in replication, start the components in this order:
1. Start the system. 2. Start the source and target databases. 3. Start SharePlex. 4. Start sp_ctrl. 5. Verify that the SharePlex processes are started by issuing the lstatus command in sp_ctrl.
sp_ctrl> lstatus 6. Allow users on the system.

Start SharePlex on Unix and Linux
To start SharePlex, you must log onto the system as a SharePlex Administrator. Your user name must be assigned to the SharePlex admin group in the /etc/group file. For more information, see Assign SharePlex users to security groups on page 183. Table 1: SharePlex startup syntax

Method From root, with full path CD to the product directory
From a startup script

Startup syntax
$ /productdir/bin/sp_cop [-uidentifier] &
$ cd /productdir/bin $./sp_cop [-uidentifier] &
#!/bin/ksh cd productdir\bin nohup sp_cop [-uidentifier] &

Table 2: Description of SharePlex startup syntax

Argument & nohup
-uidentifier

Description
Causes SharePlex to run in the background. Directs the startup of SharePlex to continue in the background after the current user logs out.
Starts sp_cop with a unique identifier. Use this option when there are multiple

SharePlex 9.2.8 Administrator Guide 37

Argument

Description
instances of sp_cop running on a system, which is required for some SharePlex configurations. For more information, see Run multiple instances of SharePlex on page 43. Some suggestions for identifier are:
l the SharePlex port number (such as -u2100) l the identifier of the database for which replication is running (such as -
uora12c) l any descriptive identifier (such as -utest)

Identify SharePlex processes on Unix and Linux
Every session of sp_cop has a process ID number. The ID is returned after startup and then the command prompt reappears. If a configuration was activated during a former session of sp_cop, replication begins immediately. Without an active configuration, sp_cop runs passively in the background. On Unix and Linux systems, you can use the ps -ef | grep sp_ command to view the SharePlex processes that are running.
l The sp_cop process is the root process. l The following child processes are spawned by sp_cop on a source system:
o Command and Control process (sp_cnc) o Capture (sp_ocap) o Read (sp_ordr) o Export (sp_xport) l The following child processes are spawned by sp_cop on a target system: 1. Command and Control process (sp_cnc) 2. Import (sp_mport) 3. Post (sp_opst_mt if the database is Oracle or sp_xpst if the database is Open Target)
Each child process has the same -uidentifier as its parent sp_cop process. This makes it easier to identify related processes when multiple session of sp_cop are running.

Stop SharePlex on Unix and Linux
To stop SharePlex, issue the shutdown command in sp_ctrl. This is a graceful shutdown that saves the state of each process, performs a checkpoint to disk, read/releases buffered data, and removes child processes. Data in the queues remains safely in place, ready for processing when sp_cop starts again. The shutdown process can take some time if SharePlex is processing large operations. You can use the force option with the shutdown command to forcefully shut down replication if necessary. It terminates sp_cop immediately, bypassing normal shutdown procedures. See the SharePlex Reference Guide for more information about this command.

SharePlex 9.2.8 Administrator Guide 38

Shutdown Considerations on Unix and Linux
You can safely shut down SharePlex for a short time while there is still transactional activity. The next time you start SharePlex, replication resumes at the correct place in the redo logs or the archive logs, if needed. However, the best practice is to leave SharePlex running while there is transactional activity. Otherwise, SharePlex may need to process a large volume of redo backlog when you start it again, and there will be latency between the source and target data. If the redo logs wrap and the archive logs cannot be accessed, resynchronization of the source and target data may be the only option. Take this possibility into account whenever you stop SharePlex while redo is still being generated.
NOTE: If you want to shut down both SharePlex and the database, shut down SharePlex first. Otherwise, SharePlex will interpret that the database is failing and generate a warning message. As an alternative to stopping SharePlex, you can use the stop command in sp_ctrl to stop individual SharePlex replication processes as needed. See the SharePlex Reference Guide for more information about this command.
Run SharePlex on Windows
On Windows, SharePlex runs as a service named SharePlex port#, where port# is the port number that is used by a particular instance of SharePlex. The service can be installed during the installation of SharePlex or at a later time by running the SharePlex SpUtils progam. See the SharePlex Reference Guide for more information about SpUtils.
NOTE: When directed in the documentation or other instructions to start or stop sp_cop on the Windows platform, it means to start or stop the SharePlex service.
Startup Sequence on Windows
When you start systems that are involved in replication, start the components in this order:
1. Start the system. 2. Start the source and target databases. 3. Start the NuTCRACKER service in the Windows Administrative Tools control panel. This starts the
MKS Toolkit® operating environment, which must be running in order to run SharePlex. See the SharePlex Installation Guide for more information about the MKS Toolkit. 4. Start the SharePlex port# service. 5. Start sp_ctrl from the bin sub-directory of the SharePlex product directory. 6. Verify that the SharePlex processes are started by issuing the lstatus command in sp_ctrl.
sp_ctrl> lstatus
Start and stop SharePlex on Windows
On Windows, the only way to start or stop SharePlex is by starting or stopping the SharePlex service. To control the service, you can use the SpUtils utility or the Windows Administrative Tools control panel.
SharePlex 9.2.8 Administrator Guide 39

NOTE: Starting the SharePlex service automatically starts the SharePlex replication processes (Capture, Read, Export, Import and Post). If you do not want those processes to start when the service starts, they must be stopped with the stop command in sp_ctrl before you shut down the SharePlex service. They will remain stopped until you issue the start command for each process that you want to start.
To control the SharePlex service with SpUtils You can use SpUtils to install, start, stop, and check the status of the SharePlex service.
1. Log onto Windows as a SharePlex Administrator. Your user name must be assigned to the SharePlex admin group. For more information, see Assign SharePlex users to security groups on page 183.
2. View the Windows services to make certain the NuTCRACKER service is running. Start it, if needed. 3. Run the SpUtils utility from the desktop shortcut or the Programs menu. 4. Click the SharePlex Services tab of the utility. 5. Select the port number of the SharePlex instance that you want to control. 6. Do one of the following:
l Click Install to install the service. If Install is not shown, the service exists. l Click Start to start the service. When Current State shows that the service is running, you can
close the utility. If Start is not shown, the service is either running already or not installed. l Click Stop to stop the service. If Stop is not shown, the service is either stopped already or
not installed. NOTE: You can also use the shutdown command in sp_ctrl to stop the SharePlex service.
To control the SharePlex service from Windows The SharePlex service appears as an entry named SharePlex port# in the Services console of the Administrative Tools control panel. Right-click the service name to start or stop the service.
Set SharePlex startup status on Windows
You can change the timing of SharePlex startup. By default, it is Automatic, meaning the service starts when the system starts. You may need to contact a System Administrator to perform this task.
1. Open the Services console of the Administrative Tools control panel. 2. Right-click the SharePlex service name, then click Properties. 3. From the Startup type list, select the desired startup option. For assistance, click the Help link
in the dialog.
Set SharePlex process priority on Windows
IMPORTANT! Do not use the standard Windows Task Manager to set process priority for SharePlex. On Windows systems, the NuTCRACKER operating environment controls the SharePlex processes. The SpUtils utility has an option that enables you to run the Windows Task Manager with the required elevated privileges to set the process priority. This procedure must be performed by a SharePlex Administrator. Consult with the System Administrator to select an appropriate value based on the requirements of all software running on the system.
SharePlex 9.2.8 Administrator Guide 40

To set SharePlex priority
1. Run SpUtils. 2. Select the TaskMgr tab. 3. Right click the SharePlex instance that you want to prioritize. 4. Select Set Priority, then select the desired priority level.
Identify SharePlex processes on Windows
In the Windows Task Manager, SharePlex appears as Sp_Copsrv.exe, representing the SharePlex sp_cop process. The operating system controls the parent Sp_Copsrv.exe service. The parent Sp_Copsrv.exe process spawns child Sp_Copsrv.exe processes -- one for each replication process (Capture, Read, Export, Import, Post, sp_ctrl, and so forth.). For a standard uni-directional configuration replicating through default queues to one target system, there are following processes on a Windows system: On the source system:
l One parent Sp_Copsrv.exe process. l One Sp_Ocap or Sp_capture (Capture) process plus one child Sp_Copsrv.exe process. l One Sp_Ordr (Read) process plus one child Sp_Copsrv.exe process. l One Sp_Xport (Export) process plus one child Sp_Copsrv.exe process. l If there are any additional SharePlex processes running, such as sp_ctrl, there is an additional Sp_
Copsrv.exe process for each one.
On the target system:
l One parent Sp_Copsrv.exe process. l One Sp_Mport (Import) process plus one child Sp_Copsrv.exe process. l One Sp_Opst_Mt (Oracle Post) or Sp_Xpst (Open Target Post) process plus one child Sp_
Copsrv.exe process. l If there are any additional SharePlex processes running, such as sp_ctrl, there is an additional Sp_
Copsrv.exe process for each one.
If there are no active replication configurations, the SharePlex processes do not start when you start the service, and just the parent Sp_Copsrv.exe will be running. To identify the parent Sp_Copsrv.exe process in the Windows Task Manager, look for the one that is using the largest amount of memory. The child Sp_Copsrv.exe processes consume less memory than the parent process. To identify which replication process is associated with a child Sp_Copsrv.exe process, look in the SharePlex Event Log for the message stating when the replication process started. This entry provides the PID for that process and the PID of the associated Sp_copsvr.exe process.
Shutdown Considerations on Windows
You can safely shut down SharePlex for a short time while there is still transactional activity. The next time you start SharePlex, replication resumes at the correct place in the redo logs or the archive logs, if needed.
SharePlex 9.2.8 Administrator Guide 41

However, the best practice is to leave SharePlex running while there is transactional activity. Otherwise, SharePlex may need to process a large volume of redo backlog when you start it again, and there will be latency between the source and target data. If the redo logs wrap and the archive logs cannot be accessed, resynchronization of the source and target data may be the only option. Take this possibility into account whenever you stop SharePlex while redo is still being generated.
NOTE: If you want to shut down both SharePlex and the database, shut down SharePlex first. Otherwise, SharePlex will interpret that the database is failing and generate a warning message. As an alternative to stopping SharePlex, you can use the stop command in sp_ctrl to stop individual SharePlex replication processes as needed. See the SharePlex Reference Guide for more information about this command.
SharePlex 9.2.8 Administrator Guide 42

3
Run multiple instances of SharePlex
This chapter shows you how to configure and run multiple instances of SharePlex on one machine. For example, when replication is configured from multiple source systems to a central target system, you can run multiple instances of sp_cop to isolate the replication streams.
Contents
Run multiple instances of SharePlex from separate installations Run multiple instances of SharePlex from one installation
Run multiple instances of SharePlex from separate installations
IMPORTANT! This topic assumes that there is no active configuration. It assumes you are configuring multiple instances of sp_cop as part of an initial setup of the replication environment. This method provides a one-to-one relationship between the binaries and the variable-data directory. This procedure creates autonomous SharePlex instances, with nothing in common between them. You start, control and maintain each SharePlex instance separately, and there are no special setup requirements. This method has the following benefits:
l Processes are easily isolated. You do not have to set environment variables to point to the correct port and variable-data directory.
l You can upgrade or perform other maintenance one product directory at a time, or choose not to perform those tasks.
l You can run the same or different versions of SharePlex on the same system. The disadvantages are:
l You must install and upgrade each installation separately. l More disk space is required to store the product files.
SharePlex 9.2.8 Administrator Guide 43

l In startup and shutdown scripts, and other places where you must map environment variables to SharePlex components, you must map them for each installation.
To set up multiple instances of SharePlex in this configuration: l Install each one separately. There should be one product directory and one variable-data directory per installation. l Install each one on a different TCP/IP port number. l IMPORTANT! Make certain to create a different database account for each installation.
To install SharePlex, see the SharePlex Installation Guide.
Run multiple instances of SharePlex from one installation
IMPORTANT! This topic assumes that there is no active configuration. It assumes you are configuring multiple instances of sp_cop as part of an initial setup of the replication environment. This method provides a one-to-many relationship between a set of binaries and two or more variable-data directories. In this configuration, you create multiple variable-data directories and link each one to a unique port number, each running a separate instance of sp_cop. This method has the following benefits:
l You install and upgrade only one installation of SharePlex. Maintenance procedures are performed for only one installation.
l You conserve disk space, because you only store one set of SharePlex binaries and installed files. l The customization of the SharePlex monitoring scripts only need to be done once, in one place. For
more information, see Run monitor scripts on Unix on page 214. l Startup and shutdown scripts only need to be created and run for one set of binaries. The disadvantages are: l Processes must be directed to each instance. You must set environment variables for each instance,
start sp_cop with the correct identifier for each instance, and set a port connection in sp_ctrl to ensure that commands are directed to the correct instance. l Upgrades apply to all instances of SharePlex. l All sp_cop instances are the same version of SharePlex.
How to run multiple sp_cop instances on Unix and Linux
To run multiple instances SharePlex on the same Unix or Linux machine, you run multiple instances of the sp_ cop program, each running on a different port number. You link each sp_cop to a different variable-data directory. Each variable-data directory is identified by the port number of its sp_cop. Connection information to the source or target datastore is linked to each sp_cop instance.
1. Assign port numbers
Assign each instance of sp_cop a unique port number.
SharePlex 9.2.8 Administrator Guide 44

l For each sp_cop instance, obtain one port number that will be used by that instance for both TCP and UDP communication.
l For each sp_cop instance, use the same port number for the remote instances of sp_cop that will process the same replication data stream as the instance you are configuring.
The sp_cop process listens on each system on TCP and UDP ports for communications from SharePlex processes on other systems in the network, such as a command or exchange between the Export and Import processes. If the ports are different, sp_cop on one system cannot connect to the sp_cop on another system to send or receive messages.
2. Create variable-data directories
For each instance of sp_cop, create a variable-data directory, and assign each to one of the port numbers you obtained. The variable-data directory contains the environment that is unique to a SharePlex instance.
1. Install SharePlex according to the instructions in the SharePlex Installation Guide. At the end of the installation, you should have one product directory, one variable-data directory associated with a port number, and one database account. This is your base instance of SharePlex.
2. Log in as a root user. 3. Shut down sp_cop if it is running. 4. Copy the original variable-data directory (with its sub-directories) to a new variable-data directory for
each instance of sp_cop that you want to run. Include the port number in each name, as shown in the following examples. cp -p -r /splex/vardir/splex2100 /splex/vardir/splex2101 cp -p -r /splex/vardir/splex2100 /splex/vardir/splex2102
3. Define the port numbers in the SharePlex environment
For each variable-data directory that you created, perform this procedure to set the port number that you reserved for this instance of SharePlex.
1. Export the SP_SYS_VARDIR variable to point to one of the new variable-data directories, for example splex2101 in the preceding example. ksh shell: export SP_SYS_VARDIR=/full_path_of_variable-data_directory csh shell: setenv SP_SYS_VARDIR=/full_path_of_variable-data_directory
2. Export the SP_COP_TPORT and SP_COP_UPORT variables to point to the port number of the variabledata directory that you exported. ksh shell: export SP_COP_TPORT=port export SP_COP_UPORT=port csh shell: setenv SP_COP_TPORT port setenv SP_COP_UPORT port
3. Log in as SharePlex Administrator.
SharePlex 9.2.8 Administrator Guide 45

4. Run the clean_vardir.sh script. The script removes duplicate replication queues and restores each one to a fresh state. See clean_vardir.sh in the SharePlex Reference Guide for more information.
5. In the rim sub-directory of the exported variable-data directory, delete the shstinfo.ipc and shmaddr.loc files. (These files may not exist if sp_cop has never been started for this variable-data directory.)
6. Repeat these steps for each additional variable-data directory.
4. Establish connections to the source or target datastore
For each sp_cop instance, establish the connections that SharePlex will use to access the source or target data of this SharePlex instance.
1. Export the SP_SYS_VARDIR variable to point to one of the new variable-data directories, for example splex2101 in the example. ksh shell: export SP_SYS_VARDIR=/full_path_of_variable-data_directory csh shell: setenv SP_SYS_VARDIR=/full_path_of_variable-data_directory
2. Run the appropriate Database Setup utility for the database. For more information, see Database Setup Utilities in the SharePlex Reference Guide.
3. Repeat these steps for each additional variable-data directory.
5. Start sp_cop instances
You can now run separate instances of sp_cop and sp_ctrl as needed.
1. Export the SP_SYS_VARDIR environment variable to point to the variable-data directory of the first sp_ cop instance. ksh shell: export SP_SYS_VARDIR=/full_path_of_variable-data_directory csh shell: setenv SP_SYS_VARDIR=/full_path_of_variable-data_directory
2. Start sp_cop with the -u option, where port is the port assigned to the sp_cop instance. /splex/proddir/bin/sp_cop -u port &
3. In sp_ctrl, use the port command to set the session to the port number of the sp_cop instance you want the commands to affect. ./sp_ctrl port number
4. Repeat these steps for each instance of sp_cop that you want to run.
NOTE: If you receive an error message similar to the following, find out if someone else started a session of sp_ cop using the same port number and variable-data directory. If permissible, kill the processes associated with that session, then start sp_cop again.
Error cleaning up previous shared memory segment ###.
Cannot delete because there are users attached.
Check if SharePlex processes are running and kill them if necessary.
SharePlex 9.2.8 Administrator Guide 46

How to run multiple sp_cop instances on Windows
To run multiple instances of SharePlex on the same Windows machine, you install and run a SharePlex service for each instance, each with a different port number and variable-data directory.
1. Install the instances
1. Install SharePlex on the source and target systems according to the directions in the SharePlex Installation Guide. This creates your initial instance of SharePlex.
2. Repeat the installation procedure for each SharePlex instance that you want to create, with the following differences: l Use the same product directory for each instance. l Select or create a different variable-data directory for each instance. l Select a different port number for each instance. The installation program locates available ports, which you can override if necessary. l When prompted for a license key, use the same license key for all instances. Make certain to select the correct port each time. l When you run the Database Setup utility, use the correct port number for the instance, and use a different name for each database account.
2. Start the SharePlex instances
All instances are controlled from the SpUtils utility. 1. Log onto Windows as a SharePlex Administrator (member of the spadmin group). 2. Run the SpUtils utility. 3. Select the SharePlex Services tab. 4. Select the port number for the instance of SharePlex that you want to start. 5. Under SharePlex Service Status, click Start. This starts the service for that instance. 6. When the Current State text box shows that the service has started, you can start another instance of SharePlex. 7. Run Sp_Ctrl for the instance. 8. In Sp_Ctrl, use the port command to set the session to the port number of the sp_cop instance you want the commands to affect. sp_ctrl > port number
SharePlex 9.2.8 Administrator Guide 47

4
Execute commands in sp_ctrl
This chapter contains instructions for using the sp_ctrl command interface to execute commands that configure, control, and monitor SharePlex. The sp_ctrl program resides in the bin sub-directory of the SharePlex product directory.
NOTE: See the SharePlex Reference Guide for more information about the commands shown in this topic.
Contents
How to run sp_ctrl Define a default port for sp_ctrl Define a default host for sp_ctrl Set a default editor for sp_ctrl Command guidelines Issue Commands on a remote system Issue commands for clustered systems Add a scroll bar to output on Windows
How to run sp_ctrl
You can run sp_ctrl on any machine where SharePlex is installed. The sp_cop program must be running in order to run sp_ctrl; otherwise it displays an error message similar to this:
Your tcp port is not set properly or "sp_cop" is not running.
Start sp_ctrl
There are two ways to run sp_ctrl: l from the command shell of the operating system to issue one command, for example: $ /productdir/bin/sp_ctrl command [on host]
SharePlex 9.2.8 Administrator Guide 48

l by running the sp_ctrl command interface to issue one or more commands, for example: $ /productdir/bin/sp_ctrl sp_ctrl>command [on host]
where:
l productdir is the SharePlex product (installation) directory. l command is the SharePlex command. l on host represents one of the command options that allow you to issue a command on the local
machine to control SharePlex on a remote machine (if supported by the command), as shown in the following example.
$ /productdir/bin/sp_ctrl status on host:port On Windows systems, you can run sp_ctrl from the Sp_Ctrl desktop shortcut or Windows Programs menu. The sp_ctrl command line allows a total of 256 characters, including spaces.

sp_ctrl prompt
The sp_ctrl prompt appears in one of two ways, depending on whether or not you set a default host and port number.

Prompt sp_ctrl> sp_ctrl(this_host:3304) >

Description
Basic sp_ctrl prompt
Prompt when a default system and port are set by issuing the host and port commands

Exit sp_ctrl
To exit the sp_ctrl command-line interface, issue the exit or quit command. On Windows systems, you can simply close the sp_ctrl command prompt window. The exit or quit command only closes the sp_ctrl session. It does not stop the SharePlex replication processes.

Define a default port for sp_ctrl
If there is only one instance of SharePlex on a system, sp_ctrl detects the port number. However, if you configured more than one SharePlex instance on a system (where each one runs on a different port number) you must use the port command to set the session of sp_ctrl to the instance for which you want to issue commands.
sp_ctrl > port number For more information, see Run multiple instances of SharePlex on page 43. For more information about the port command, see the SharePlexReference Guide.

SharePlex 9.2.8 Administrator Guide 49

Define a default host for sp_ctrl
To define a default machine for all interactive sp_ctrl sessions, use the host command. This command enables you to enter a series of commands without using the onhost option for each one.
sp_ctr >host hostname The command changes the sp_ctrl prompt to include the host name:
sp_ctrl (sysA) > The host setting applies only to the sp_ctrl session in which it is set.
Set a default editor for sp_ctrl
You can set the default editor that sp_ctrl runs when you issue a command that requires input to an ASCII text file, such as a configuration file. By default, sp_ctrl runs vi on Unix and Linux and WordPad on Windows. The default text editors are tested and proven to work with SharePlex. If you change the default editor, the new editor must be a native ASCII text editor. Do not use a word processing program or other non-ASCII program, even if you can save a file to ASCII from that program.
Change the editor on Unix or Linux
Set the EDITOR variable in either of these ways: l Before you start sp_ctrl. This sets the editor only for that session of sp_ctrl. l In the shell startup script on the local machine. This sets the editor permanently, until changed in the startup script. You can override this setting on a per-session basis.
Syntax - ksh shell export EDITOR=name_of_editor
Syntax - csh shell setenv EDITOR name_of_editor
Change the editor on Windows
To change the text editor for sp_ctrl on a Windows system, you must add the EDITOR variable in the Windows Registry. IMPORTANT! Incorrect Registry settings can adversely affect the operation of SharePlex. Consult a System Administrator if you need help with this procedure.
1. Shut down the SharePlex service. 2. Open the Run dialog. The location varies with the Windows version. 3. In the Run dialog, type regedit to run the Registry Editor.
SharePlex 9.2.8 Administrator Guide 50

4. Expand the SharePlex node: \HKEY_LOCAL_MACHINE\Software\Wow6432node\Quest Software\SharePlex
5. Right click the port number of the SharePlex instance to which you want to add a variable, then select New, then String Value.
6. Under the Name column, right click the new variable, then select Rename. 7. Type the correct name. 8. Double click the new variable. 9. Under Value Data, enter the string for the new variable and then click OK. 10. Exit the Registry.
Command guidelines
Observe the following when issuing commands: l To issue commands for a machine, sp_cop must be running on that machine. l Enter the syntax exactly as shown in the command description in the SharePlex Reference Guide. l The maximum string length of a SharePlex command is 255 characters, including spaces. To work around this operating-system limitation, use the edit command. l Use the redo command to execute the previous command again without having to retype it. This command is useful when you are making frequent status checks with the information commands, for example using the qstatus command to monitor changes in queue volume. l To view descriptions and syntax for SharePlex commands from within the sp_ctrl interface, issue the help command. To view just the syntax for a command, issue the usage command. l Use the usage command to view the syntax for a SharePlex command. You can enter the entire command or just the first few keywords. For example, type usage compare to view syntax for both the compare using and compare commands. l Use the edit command to edit a previously issued command. l Use the authlevel command to determine your authorization level for issuing SharePlex commands on a system.
For more information, see About the SharePlex security groups on page 183.
Issue Commands on a remote system
To issue a command that will affect a remote machine and to script commands that include a login name, password, port number, or combination of those items, use one of the [onhost] command options. These options are available for most commands. The following table describes the command options for remote connection using the [onhost] options. These options enable you to issue the command on a remote machine and to script commands that include a login name, password, port number, or combination of those items.
SharePlex 9.2.8 Administrator Guide 51

Option on host
on host:portnumber on login/password@host on login/password@host:portnumber

Description
Execute the command on a remote system (one other than the one where the current sp_ctrl session is running). You are prompted for login credentials for the remote system. If used, must be the last component of the command syntax. Example: sp_ctrl(sysB)>status on SysA
Execute the command on a remote system when a remote login and port number must be provided. If used, must be the last component of the command syntax. Example: sp_ctrl(sysB)>status on SysA:8304
Execute the command on a remote system when a remote login, password, and host name must be provided. If used, must be the last component of the command syntax. Example: sp_ctrl(sysB)>status on john/spot5489@SysA
Execute the command on a remote system when a remote login, password, host name, and port number must be provided. If used, must be the last component of the command syntax. Example: sp_ctrl(sysB)>status on john/spot5489@SysA:8304

Issue commands for clustered systems
To issue sp_ctrl commands on clustered systems, use the name set with the SP_SYS_HOST_NAME parameter as the host in the [on host] options when connecting from a remote system, or set it as the default for sp_ctrl by using the host command. For more information about configuring SharePlex within a cluster, see the SharePlex Installation and Setup Guide.
Add a scroll bar to output on Windows
Scrolling might be necessary to view the entire output of certain sp_ctrl commands. If your Command Prompt window does not have a scroll bar, you can change the window properties to produce one.
1. Click the Command Prompt icon at the top left corner of the console, then select Properties from the menu.
2. In the Command Prompt Properties dialog box, click the Layout tab. 3. In the Screen Buffer Size option box, increase the Height value until it produces a scroll bar in the
Command Prompt window. 4. Click OK to apply the settings.

SharePlex 9.2.8 Administrator Guide 52

5. In the Apply Properties To Shortcut dialog box, select one of the options, either to apply the new properties to the current command session or to apply them as the new default for all sessions.
6. Click OK to close the dialog box.
SharePlex 9.2.8 Administrator Guide 53

5
Set SharePlex parameters
SharePlex parameters control and tune various aspects of replication.
Contents
View and set parameters Where parameter information is stored
View and set parameters
A SharePlex Administrator (defined as a member of the SharePlex Admin group) can change parameters that are designated as user-configurable and is the only user authorized to do so.
View parameters
Use the list param command in sp_ctrl to view user-configurable SharePlex parameters. It displays parameter names, current settings, default values (if the parameter has been changed), and set-at points. The set-at point indicates when changes to a parameter will take effect. Possible set-at points are:
l Live means a change takes effect immediately. l Restart Process means a change takes effect after the affected SharePlex process is restarted. l Restart Cop means a change takes effect after sp_cop is restarted. Additional options are available for viewing: l All SharePlex parameters. l Only parameters whose values have changed. l Parameters relating to a specific SharePlex module. To view descriptions of the SharePlex parameters, see the SharePlex Reference Guide.
SharePlex 9.2.8 Administrator Guide 54

Set parameters
Parameters can be set in the following ways:
l With the set param command through the sp_ctrl interface. This is the preferred method because the new value remains intact no matter how many times replication stops and starts. The syntax is: set param parameter_name value Example:
l set param SP_OCT_REPLICATE_ALL_DDL 1 l As environment variables on Unix and Linux systems prior to starting sp_cop. The new value remains in
effect only for that session of sp_cop. Parameters for the Capture, Read, Export, Import, and Post processes can be set on a per-process basis when there are multiple instances of a process for an instance of SharePlex.
Set SharePlex parameters through sp_ctrl
The recommended way to change a SharePlex parameter is to use the set param command in sp_ctrl. To restore a parameter's setting to its default value, use the reset param command. To view descriptions of the SharePlex commands, see the SharePlex Reference Guide.
Set SharePlex parameters as environment variables
On Unix and Linux, a SharePlex parameter can be set as an environment variable. The environment variable overrides the setting in the param-defaults files, but only for the session of sp_cop for which it was set. If you shut down sp_cop and restart it without resetting the environment variable, SharePlex uses the default setting in the param-defaults file. NOTE: On Windows systems, SharePlex environment variables must be set through the Windows Registry. To set a SharePlex parameter as an environment variable on Unix and Linux systems, use one of the following commands. Set the environment variable before starting sp_cop or, if sp_cop is running, restart sp_cop for the new setting to take effect.
ksh shell: $ export parameter_name=value
csh shell: $ setenv parameter_name value
Because of the temporary nature of environment variables, avoid using them if possible; instead make your changes with the set param command. When you rely on environment variables, especially when there are multiple users of SharePlex, you incur the risk of someone forgetting to set them (or using an incorrect value) when they restart sp_cop. That can have a significant, negative impact on replication and can result in the need to resynchronize the data.
Where parameter information is stored
The following files store SharePlex parameter settings:
SharePlex 9.2.8 Administrator Guide 55

l The param-defaults file stores default settings that were set by developers for optimal replication performance under most conditions. The param-defaults file resides in the data sub-directory of the SharePlex product directory. The data in this file does not change unless a new version of SharePlex is installed. IMPORTANT:Never edit this file.
l The paramdb file stores user-defined parameter settings -- values that were changed from their defaults by a SharePlex Administrator using the set param command. Also stored in this file are the SharePlex license key for the local system, the Share- Plex Oracle user, and the SharePlex user's password. The paramdb resides in the data sub-directory of the SharePlex variable-data directory. It starts out empty, and as SharePlex Administrators change parameter values, those values are added to it. Userdefined parameter values override SharePlex default values when SharePlex is running. All of the settings in the paramdb file remain intact when a new version of SharePlex is installed.
SharePlex 9.2.8 Administrator Guide 56

6
Configure data replication
SharePlex uses a configuration file to determine which tables to replicate and where to send the replicated data. This file also provides any special processing instructions, such as column mapping and data filtering. This chapter contains the information that you need to know in order to create a configuration file. Once you understand the basics of how to create a configuration file and route data to your targets, you can move on to any of the more advanced configurations as needed. Documentation for those configurations is included in this guide.
Contents
Ensure compatible source-target mapping Create a configuration file How to qualify object names How to specify case-sensitive names Database specifications in a configuration file Target specifications in a configuration file Routing specifications in a configuration file Configuration examples by datasource and target Capture from multiple local datasources Use Wildcards to specify multiple objects Define a unique key Filter DML operations Map source and target columns Build a configuration file using a script
Ensure compatible source-target mapping
The following guidelines help you ensure that the source and target objects that you want to map in a replication configuration are compatible.
SharePlex 9.2.8 Administrator Guide 57

Object names
For most replication strategies, the name and/or owner of a source object can be different from that of its target object. SharePlex replicates to the correct object because you specify it by owner and name within the configuration file. For high-availability configurations, the owner and name of a source table should be the same as the owner and name of the target table.
Source and target rows
Corresponding source and target rows must contain the same values to accurately reflect the source, unless transformation is being used. One database in the configuration can have more or fewer rows than another database in the configuration. You can control which rows are replicated through horizontally partitioned replication. For more information, see Configure horizontally partitioned replication on page 93.
Source and target columns
Corresponding source and target columns must: l Contain compatible data types (same type, size, precision). l Have identical names, unless you use column mapping in the configuration file. For more information, see Map source and target columns on page 78. l Be the same letter case if one of the databases is case-sensitive, but the source or target of that database is not case-sensitive. You can use column mapping to work around this issue. For more information, see Map source and target columns on page 78.
A target table can have more columns than the source table. l If the source and related target column names are identical, SharePlex will ignore the extra columns in the target table. l If the source and target column names are not identical, SharePlex maps a one-to-one relationship in the order that the columns are defined in each table (for example, map the first column in the source to the first column in the target, map the second column to the second column, and so forth). l To avoid Oracle errors if the extra (non-mapped) columns are NOT NULL, define default values for those columns. For more information, see Map source and target columns on page 78.
A target table can have fewer columns than the number of columns in the source table, but you must use vertically partitioned replication to replicate only that subset of the source columns that matches the rows of the target. For more information, see Configure vertically partitioned replication on page 104.
Create a configuration file
To configure SharePlex to replicate data, you create a configuration file. This is an ASCII text file in which you specify:
SharePlex 9.2.8 Administrator Guide 58

l The data that you want SharePlex to replicate, including any filtering or partitioning of rows or columns l The target name and type l The system (and database if applicable) to which the data must be delivered.
Only a SharePlex Administrator or operator has the authority to create a configuration file. When your configuration file is completed, you activate the configuration with the activate config command to begin replication. For more information, see Start replication on your production systems on page 186.

Create a configuration file
You can create a configuration file by hand in sp_ctrl or, if your data structure supports it, you can automate the creation of a configuration file with a script. Create the configuration file on the system from which data is to be replicated, typically the source system, but strategies such as active-active replication require configuration files on more than one system.

To create a configuration file in sp_ctrl
1. Run sp_ctrl from the bin sub-directory of the SharePlex product directory.
2. In sp_ctrl, issue the create config command. sp_ctrl> create config config_name This command opens a file in the default text editor that is set for the operating system. NOTE: You can change the default editor that sp_ctrl uses. For more information, see Set a default editor for sp_ ctrl on page 50.
3. Complete the configuration file. For more information, see Structure of a configuration file on page 60. IMPORTANT! All configurations must reside in the config sub-directory of the SharePlex variable-data directory. Configuration files outside this directory cannot be activated. SharePlex places configurations in this directory by default when you create them through the sp_ctrl interface with the create config command. If you create the configuration directly through a text editor, make certain to save it to the config sub-directory.

To build a configuration file with a script NOTE: Valid for Oracle only SharePlex provides the following scripts that can be used to automate the building of your configuration file if the source and target object names are identical. These scripts support Oracle Database source and targets only.

Option config.sql
build_config.sql

Description

To get more information

Builds a configuration file that includes all tables and sequences that are in the database. Source and target object names must be identical.

See Configuration Scripts in the SharePlex Reference Guide.

Builds a configuration file that includes all tables in a schema. Source and target object names must be identical.

See Configuration

SharePlex 9.2.8 Administrator Guide 59

Option

Description

To get more information
Scripts in the SharePlex Reference Guide.

Structure of a configuration file

A basic configuration file looks like the following: # comment: basic SharePlex configuration file

datasource_specification

#source specification

target specification

routing map

source_owner.object1

target_owner.object1

routing_map

source_owner.object2

target_owner.object2

routing_map

source_owner.object3

target_owner.object3

The basic components of a configuration file are as follows.

routing_map

Component # Comments
Datasource specification
Source specification

Description

Syntax examples

Lines that describe the file or provide other information about the contents to viewers, but are not used by SharePlex. Precede each comment line with a pound (#) sign. Comments may be entered anywhere in the configuration file.

# This is a comment.

Syntax that specifies the source database. This component must always be the first noncommented line of a configuration file. It has the following syntax elements, all on the same line with no spaces:

Datasource:o.SID

l The keyword Datasource followed by a colon (:). The word "Datasource" is not case-sensitive.
l The database specification. For more information, see Database specifications in a configuration file on page 64.

The fully qualified name of a supported source object, in the form of owner.object. The owner can be a schema or a database, depending on how the database stores objects logically. See:
How to qualify object names on page 62

src_owner.table src_owner.sequence

SharePlex 9.2.8 Administrator Guide 60

Component

Description

Syntax examples

How to specify case-sensitive names on page 63
You can use wildcards to specify multiple objects. Owner names cannot be wildcarded. For more information, see Use Wildcards to specify multiple objects on page 72.

Target specification

The target to which the replicated data is applied. Targets supported by SharePlex are:
l The fully qualified name of a table (or tables) in a relational database. You can use wildcards to specify multiple objects. Owner names cannot be wildcarded. For more information, see:

tgt_owner.table tgt_owner.sequence !file[:tgt_owner.table] !jms[:tgt_owner.table] !kafka[:tgt_owner.table] !cdc:tgt_owner.table

l How to qualify object names on page 62
l How to specify case-sensitive names
l Use Wildcards to specify multiple objects on page 72
l Target specifications in a configuration file on page 65 on page 1
l An Oracle sequence (or wildcard specification).
l A file that contains XML or SQL records.
l A JMS queue or topic.
l A Kafka topic.
l A change-history table that maintains a record of all changes made to a source table (also known as change data capture)

Routing map

One or more routes that send the data to the system that contains the target object specified with the target specification. A route consists of the following:
1. The name of the machine that hosts the target.

host@o.SID host@o.PDBalias host@o.tns_alias host@r.database_name host host@c.SID Compound routing map: host@o.SID+host@r.database[...]

SharePlex 9.2.8 Administrator Guide 61

Component

Description

Syntax examples

2. (Database targets only) The @ symbol followed by the target database specification. For more information, see Database specifications in a configuration file on page 64.

The database specification is absent if the target is JMS, Kafka, or a file.
There can be no spaces between any characters in the routing map.
For more information, see Routing specifications in a configuration file on page 66.

How to qualify object names
So that SharePlex can determine the correct objects to capture from, and post to, you must qualify the object names in the configuration file in the same way that the database stores them logically. The general way this is indicated in SharePlex syntax is: owner.object Where:
l owner is the schema or database that contains the object (or objects, if wildcarded), depending on how that container is defined by the database.
l object is the name of the object or a wildcard specification to specify multiple objects.
When defining source or target objects in the configuration file, follow these guidelines for specifying the owner component:

Database Aurora MySQL Oracle PostgreSQL SQL Server SAP ASE SAP HANA Teradata

Fully qualified object name database_name.object_name database_name.object_name schema_name.object_name schema_name.object_name schema_name.object_name schema_name.object_name schema_name.object_name database_name.object_name

SharePlex 9.2.8 Administrator Guide 62

How to specify case-sensitive names
This topic shows you how to specify case-sensitive names in the configuration file, for example when specifying table names or if you need to specify column names explicitly in a column mapping.

Case-sensitive object names
If the owner or name of an object is case-sensitive in the database, you must enclose that name within quotes in the SharePlex configuration file. IMPORTANT: This applies whether the database itself requires a case-sensitive name to be within quotes, such as Oracle, or whether the database accepts names that are spelled out in their case-sensitive form without quotes, like SQL Server.

To enforce case-sensitive object names Specify the name in its correct case and enclose it within double quotes.

Correct way

l This is how to specify an object where both the owner and object names are both case-sensitive: "Owner"."Object"

l This is how to specify an object where only one of the components is case-sensitive: owner."Object" or "Owner".object The name that is not case-sensitive can be specified in any case.

Examples of both ways:

Datasource o.oraA sales."Emp"

"Sales"."Emp"

sysB@o.oraB

Incorrect way
This is not correct, because both components are within one set of quotes: "Sales.Employees"

Case-sensitive column names
Ordinarily, column names are not specified in the configuration file, unless source column names need to be mapped to different target column names by means of a column mapping (see Map source and target columns on page 78). However, if the names of any pair of source and target columns have difference cases, you may need to include them in a column mapping to enforce their case sensitivity. Whether or not a column mapping is required depends on the target type: Oracle or Open Target.
To enforce case-sensitive column names to Oracle targets The Oracle Post process does not perform case conversion of column names automatically. If the case is different between source and target columns, you must use a column map to map the case of the source names to the case of the target names. To get Post to enforce the case, specify the name in its correct case and enclose it within double quotes.

SharePlex 9.2.8 Administrator Guide 63

This is an example of case-sensitive column name mapping in a column map:
Datasource o.oraA sales.emp(ID,"first","last") sales.emp(ID,"First","Last") sysB@o.oraB
To enforce case-sensitive column names to Open Target The Open Target Post process performs case conversion of column names automatically. If replicating to target columns that have a different case from their source columns, no column mapping is needed.

Database specifications in a configuration file

A database specification is required in the following components of the configuration file: l the datasource (source datastore) specification l routing map (target datastore and location) specification

Database Oracle source
Open Targe targets Oracle change-history target

Database type notation* Database Identifiers

Depending on the Oracle database configuration, use

one of the following. This is the string that SharePlex

will use to connect to the database.

l The Oracle SID of a regular (non-CDB) Oracle database, as in o.ora12.

l The TNS alias of a pluggable database (PDB) in an Oracle container database (CDB), as in o.pdb1.

l The global TNS alias of an Oracle RAC cluster, as in o.rac1. SharePlex connects to an Oracle RAC instance through this TNS alias, which is mapped locally on each node to the Oracle SID of the local Oracle instance. For more information about creating this alias, see Preinstallation instructions for Oracle cluster in the SharePlex Installation Guide.

Use to specify the name of an Open Target (non-

Oracle) target database, as in r.mydb. IMPORTANT!

Use the actual database name. Do not use the ODBC

datasource name (DNS) or database instance name.

If the database name is case-sensitive, specify it that

way.

Use in a routing map to specify the Oracle SID, TNS

alias, or global RAC TNS alias of an Oracle change

history database, as in c.ORA12CH. In this

configuration, SharePlex applies all source

SharePlex 9.2.8 Administrator Guide 64

Database

Database type notation* Database Identifiers
transactions as INSERTs to the target tables, to maintain a history of every operation performed. For more information, see Configure replication to a change history target on page 109.

* NOTE: The dot is required.

Target specifications in a configuration file

The following table shows how to specify a target table or non-table target in a configuration file.

Target Database table Database sequence File
JMS
Kafka

Target Specification tgt_owner.table

Description
The fully qualified name of a database table. For more information, see How to qualify object names on page 62.

tgt_owner.sequence !file[:tgt_owner.table]
!jms[:tgt_owner.table]

The fully qualified name of a sequence. For more information, see How to qualify object names on page 62
The !file designator directs Post to write change operations to a file in SQL or SML format. The file name is applied internally by SharePlex. Optionally, you can specify the fully qualified name of a target table if the data will be consumed by a process that ultimately applies it to a database table.
The !jms designator directs Post to write change operations to a JMS queue or topic in XML format. The queue or topic name can be defined by using the target command. Optionally, you can specify the fully qualified name of a target table if the data will be consumed by a process that ultimately applies it to a database table.

!kafka[:tgt_owner.table]

The !Kafka designator directs Post to write change operations to a Kafka topic in XML or JSON format. The topic name can be defined by using the target command.
Optionally, you can specify the fully qualified name of a target table if the data will be consumed by a process that ultimately applies it to a database table.

SharePlex 9.2.8 Administrator Guide 65

Target Change history table

Target Specification !cdc:tgt_owner.table

Description
The !cdc designator directs Post to insert every data change to the table as a new row, rather than overlay the old data with new data. Specify the fully qualified name of the change history table. For more information, see Configure replication to a change history target on page 109.

Routing specifications in a configuration file
The following instructions show you how to build a routing map based on where you want to send the source data. A routing map sends replicated data to the correct target on the correct target system, or systems. For details about the components of these configurations, see: Database specifications in a configuration file Target specifications in a configuration file

Routing to one target

A simple routing map sends replicated data from one source object to one target object. datasource_specification

src_owner.table

tgt_owner.table2

host2[@database_specification]

src_owner.table

tgt_owner.table3

host3[@database_specification]

Routing to a cloud service
There are special routing requirements for database targets that are hosted by a cloud service such as EC2 and RDS on Amazon AWS, Azure SQL in Microsoft Azure, and Compute Virtual Machines in Oracle Cloud Infrastructure. Whether the service is Infrastructure as a Service (IaaS) or Platform as a Service (PaaS) makes a difference in how you install and configure SharePlex. The following explains these requirements.

IaaS targets

If replicating to a database target hosted in an IaaS cloud service, specify the full endpoint URL as the target host in the routing map.
datasource_specification

src_owner.table

tgt_owner.table2

endpointURL@database_specification

src_owner.table

tgt_owner.table3

endpointURL@database_specification

For example, the following routing map routes to a cloud database on Amazon EC2:

SharePlex 9.2.8 Administrator Guide 66

ec2-12-345-678-910.compute-1.amazonaws.com@o.myora Alternately, you can map the private IP address of the cloud service to a short name in the local hosts file, and then specify that name as the host in the routing map, for example:
shortname@o.myora
PaaS targets
If replicating to a database target hosted in a PaaS cloud service, there are special installation, setup, and routing requirements. Because SharePlex cannot be installed directly on a PaaS cloud server, you must install SharePlex on either the source server or an intermediary server, from which Post connects to the target cloud database. For more information, see Installation and setup for cloud-hosted databases in the SharePlex Installation and Setup Guide.

Routing to multiple targets

A compound routing map sends replicated data from one source object to multiple target objects. It enables you to specify the source and target objects once for all routes, rather than type a separate configuration entry for each route. Only one target specification can be used in a compound routing map, so all of the target objects must be identical as follows:
l All are of one type: All the same database object type or all a JMS queue or all a JMS topic or all a Kafka topic, or all a file (but no combination of these).
l All have the same fully qualified name, including any table specifications in a JMS, Kafka, or file target specification.
l All have identical column mappings or key mappings, if used. For more information about these mappings, see:
l Define a unique key on page 75
l Map source and target columns on page 78
NOTES:
l Certain routing limitations apply when using vertically partitioned replication. For more information, see Configure vertically partitioned replication on page 104.
l If any target has a different qualified name from the other targets of the same source object, you must use a simple routing map for that target.

datasource_specification

src_ owner.table

tgt_owner.table

host1[@database_specification]+host2[@database_specification][...]

Routing between objects on the same system
You can replicate between the following:

SharePlex 9.2.8 Administrator Guide 67

l For Oracle, you can replicate between objects that are in the same database or in different databases on the same system. You can replicate between objects that have the same name, so long as their owners are different.
When SharePlex replicates between objects on the same system, it does not create Import and Export processes. You can force SharePlex to create Import and Export processes by using the following routing map. If you do not need the Import or Export processes, omit the host* portion of the routing map.

Configuration with replication to objects in the same or different database on the same system

datasource_specification src_owner.table tgt_owner.table

host*host[@database_specification]

Routing Limitations
l By default, SharePlex supports replication to a maximum of 19 direct target systems. That is the maximum number of processes that can read the export queue. To replicate to more than 19 targets, use named export queues. With each additional queue that you add, you can replicate to 19 additional targets. For more information, see Configure named export queues on page 85.
l Each instance of sp_cop on a system permits a maximum of 1024 different routes. This limitation includes each route that uses a different named post queue (see Configure named post queues on page 89.) If your replication strategy requires more than 1024 routes, consider using one or more intermediary systems to divide the routes among multiple sp_cop instances. For more information, see Configure replication to share or distribute data on page 114.
l By default, each sp_cop instance allows a total of 25 queues on a system. There will always be one capture queue on a source system and one post queue on a target. Therefore, you can have as many as 24 named export queues on a source system and 24 named post queues on a target system. If a system serves as both a source and target, you will have both a capture queue and a post queue. That allows you to create up to 23 named queues of either type (or a mix of both). If system memory permits, you can change the number of allowed queues by setting the SP_QUE_MAX_QUEUES parameter. See the SharePlex Reference Guide for more information about this parameter.
Configuration examples by datasource and target
These are examples of basic configuration files according to each possible datasource type and target type.
Replicate from a regular Oracle instance to a regular Oracle instance
This configuration applies to on-premises and IaaS cloud deployments. See the System Requirements in the SharePlex Installation and Setup Guide for supported database versions and platforms.

SharePlex 9.2.8 Administrator Guide 68

Datasource:o.SID src_owner.table

tgt_owner.table

host@o.SID

Example

The following example replicates table SCOTT.EMP from Oracle instance oraA to target table SCOTT.EMP2 in Oracle instance oraB on target system sysprod.

Datasource:o.oraA SCOTT.EMP SCOTT.EMP2

sysprod@o.oraB

Replicate from Oracle to target Oracle in PaaS cloud

To replicate from an on-premises or IaaS-based Oracle source to a target Oracle database hosted in a PaaS cloud, the SharePlex target components (Import and Post) must run on the source server or on an intermediary server. Post connects through a remote connection using a TNS alias. To set up this topology, see Installation and setup for cloud-hosted databases in the SharePlex Installation and Setup Guide.
Datasource:o.SID

src_owner.table

tgt_owner.table

source_or_intermediary_host@o.SID

Example

The following example replicates table SCOTT.EMP from Oracle instance oraA to target table SCOTT.EMP2 in the PaaS cloud Oracle instance oraB. Post runs on intermediary target system sysprod2.

datasource:o.oraA

SCOTT.EMP

SCOTT.EMP2

sysprod2@o.oraB

Replicate from a regular Oracle instance to an Open Target database

This configuration applies to on-premises and IaaS cloud deployments. See the System Requirements in the SharePlex Installation and Setup Guide for supported database versions and platforms.
Datasource:o.SID

src_owner.table

tgt_owner.table

host@r.database_name

Example

The following example replicates table SCOTT.EMP from Oracle instance oraA to target table Scott2.Emp2 in Open Target database mydb on target system sys2. The target table is case-sensitive.

Datasource:o.oraA SCOTT.EMP "Scott2"."Emp2"

sys2@r.mydb

SharePlex 9.2.8 Administrator Guide 69

Replicate from a regular Oracle instance to a file in XML or SQL format

Datasource:o.SID

src_owner.table

!file

host

Example
The following example replicates table SCOTT.EMP from Oracle instance oraA to a file on target system sysprod.
Datasource:o.oraA SCOTT.EMP !file sysprod

Replicate from a regular Oracle instance to a JMS queue or topic

Datasource:o.SID

src_owner.table

!jms

host

Example
The following example replicates table SCOTT.EMP from Oracle instance oraA to a JMS queue on target system sysprod.
Datasource:o.oraA SCOTT.EMP !jms sysprod

Replicate from a regular Oracle instance to a Kafka topic

Datasource:o.SID

src_owner.table

!kafka

host

Example
The following example replicates table SCOTT.EMP from Oracle instance oraA to a Kafka topic on target system sysprod.
Datasource:o.oraA SCOTT.EMP !kafka sysprod

SharePlex 9.2.8 Administrator Guide 70

Replicate from and to an Oracle pluggable database (PDB) in a container database (CDB)*

src_owner.table

tgt_owner.table

host@o.PDBalias

Example

This example replicates table SCOTT.EMP from an Oracle PDB that uses the TNS alias of aliasA to target table SCOTT.EMP in an Oracle PDB that uses the TNS alias of aliasB on target system sysprod.

Datasource:o.aliasA SSCOTT.EMP SCOTT.EMP

sysprod@o.aliasB

* You can also replicate data from an Oracle PDB to any other supported target. For more information, see Configure capture and delivery on page 83.

Replicate to maintain a change history target

src_owner.table

!cdc:tgt_owner.table

host@c.SID

Example

The following example replicates table SCOTT.EMP from Oracle instance oraA to change-history target table SCOTT.EMP2 in Oracle instance oraB on target system sysprod.

Datasource:o.oraA

SCOTT.EMP

!cdc:SCOTT.EMP2

sysprod@c.oraB

For more information, see Configure replication to a change history target on page 109.

Capture from multiple local datasources
You can use one instance of SharePlex to capture from multiple datasources on a system. All of the configurations can be active at the same time. NOTE: SharePlex does not support multiple active configuration files for the same datasource, but it does support multiple active configuration files if each replicates a different datasource.

SharePlex 9.2.8 Administrator Guide 71

To capture from multiple datasources 1. Create a configuration file for the first datasource. In each routing map, include a named export queue. For more information, see Configure named export queues on page 85. 2. Create a configuration file for the second datasource. In each routing map, specify a named export queue, but make certain it is different from any of the queues named in the first configuration file. It is important that data from one datasource does not process through the export queues of the other datasource. 3. Create additional configurations with dedicated named export queues, if needed. 4. When you activate the configuration files, use a separate sp_ctrl session for each one. For more information, see How to activate multiple configuration files on page 191.
Use Wildcards to specify multiple objects
You can use wildcard characters to specify multiple objects of a schema in one entry of the configuration file. SharePlex replicates any objects that satisfy the wildcard, except those that you explicitly exclude. NOTE: Only object names can be wildcarded. Owner names cannot be wildcarded.
Requirements and limitations of wildcard support
l The schemas that contain wildcarded object names must exist on the source and target before the configuration is activated.
l (Oracle) The objects themselves do not have to exist before the configuration is activated, provided the correct parameters are set to required values. For more information, see Control Oracle DDL replication on page 152.
l Wildcards are not allowed anywhere in a configuration entry that includes the following: o Vertically partitioned replication. For more information, see Configure vertically partitioned replication on page 104. o Horizontally partitioned replication. For more information, see Configure horizontally partitioned replication on page 93. o (Oracle) Key definition. For more information, see Define a unique key on page 75. o (Oracle) Column mapping. For more information, see Map source and target columns on page 78.
The tables that use these features must be specified in the configuration file separately.
Supported wildcard syntax
SharePlex supports the following SQL wildcards
SharePlex 9.2.8 Administrator Guide 72

l Percent (%) wildcard to specify a string. (See the Examples on page 74.)
l Underscore (_) wildcard to specify a single-character.
l For table names that contain a percent sign or an underscore character (for example emp_salary), SharePlex recognizes the backslash (\) as an escape character to denote the character as a literal, not a wildcard.

Specify wildcarded names in the configuration file

Use this template for help when specifying a wildcarded name in the configuration file.

Configuration with wildcarded object names

datasource_specification expand src_owner.wildcard_name [not (list)]

tgt_owner.wildcard_name routing_map

Description of syntax elements Component expand
src_owner.wildcard_name
not (list)

Description
Indicates that the specification contains wildcard characters that must be expanded. When SharePlex detects the expand keyword, it queries the database for all objects that match the criteria in the wildcard specification. Without this required keyword, the wildcard characters are assumed to be part of an explicit object name, and no wildcard expansion is performed. NOTE: Leave a space between expand and the start of the source object specification.
l src_owner is the owner of the source objects. Owner names cannot be wildcarded. If wildcards are used in the owner name, SharePlex assumes that they are part of the owner (schema) name.
l wildcard_name is the wildcarded name of the source objects.
Rules: Oracle: The names of the target objects must be identical to those of the source objects, but the objects may belong to different owners.
An exclusion list that defines objects to omit from the wildcard expansion. Use this option to exclude objects that you do not want to be replicated. NOTE: This not keyword does not have the same meaning as the SQL wildcard NOT operator.
l The not keyword and parentheses are required elements.

SharePlex 9.2.8 Administrator Guide 73

Component tgt_owner.wildcard_name routing_map

Description
l list is a comma-separated list of tables owned by the same owner, either wildcarded or explicit. Example: not (spo%, gen%, product).
Leave a space before and after the not keyword. A space is allowed after each comma in the list.
NOTE: If an object that satisfies a wildcard is listed elsewhere in the configuration file, that entry overrides the processing or routing specified in the wildcarded entry. In this case, a not clause is not needed. See the Examples.
l tgt_owner is the owner of the target objects.
l wildcard_name is the wildcarded name of the target objects.
The target specification must be in the form of owner.%. Partially expanded target wildcarded names are not supported, such as owner.tab%.
Any valid routing map. For more information, see Routing specifications in a configuration file on page 66

Validate a wildcard specification
To confirm that a wildcard specification will produce the specific list of tables that you want to replicate, issue the verify config command in sp_ctrl before you activate the configuration. This command produces a list of the objects that SharePlex will capture and replicate, as well as any problems that occurred. For more information about this command, see the SharePlex Reference Guide.

Examples

Examples of valid wildcard specifications

Example 1: The following wildcard specification directs SharePlex to activate all tables owned by scott, where the table name is like prod% except if the table name is like %temp%. All tables that match this description are replicated to tables of the same names on the target in the hal schema. Note that SharePlex automatically upshifts the names, so that it actually activates all tables where the table name is like 'PROD%' but not like '%TEMP%'.

Datasource:o.sidA expand scott.prod% not (%temp%)

hal.%

sysa@o.sidB

Example 2: The following example shows how you can specify special handling for one of the tables in a wildcarded specification, in this case the photo table. All tables but photo are routed through the default post queue. The separate entry for the photo table overrides the wildcarded entry and processes the photo table through a named post queue. For more information, see Configure named post queues on page 89.

SharePlex 9.2.8 Administrator Guide 74

Datasource:o.sidA

cust.%

cust.photo cust.photo

hostB@o.oraB hostB:lobQ@o.oraB

The following are additional examples of valid wildcard specifications

Datasource:o.sidA expand scott.%test%

scott.%

sysa@o.sidB

Datasource:o.sidA expand scott.%t__t%

fred.%

sysa@o.sidB

Datasource:o.sidA expand scott.% not (spo%, gen%, prodct)

scott.%

sysa@o.sidB

Datasource:o.sidA expand scott.prod% not (%temp%)

hal.%

sysa@o.sidB

Examples of invalid wildcard specifications

The following example contains a wildcarded schema, which is not permitted.

Datasource:o.sidA expand rob%.%test%

scott.%

sysa@o.sidB

The following example contains a partially wildcarded target object name, which is not permitted.

Datasource:o.sidA expand scott.%test%

scott.%obj%

sysa@o.sidB

Define a unique key
NOTE: Valid for Oracle tables only. If a table was not created with a primary or unique key, you can specify columns to use as a key when you specify the object in the configuration file. SharePlex uses the specified columns as a unique key in its WHERE clause to locate target rows for posting.
NOTES:
l Without a primary or unique key, SharePlex uses all of the columns of a table (or all of the columns in a column partition) as a key, which slows replication performance.
l When a key definition is specified for a table that has a PRIMARY or UNIQUE key, the key definition overrides the defined key. This can be useful if you do not want any of the existing keys to be used by SharePlex.
The columns that you specify as a key must meet the following criteria:
l They cannot be LONG or LOB columns.
l They must be able to uniquely identify a row. Otherwise, replication could return out-of-sync errors or post to incorrect target rows.
l They must be part of the column partition if the table is configured for vertically partitioned replication. When using the exclude column notation in vertical partitioning, the excluded columns cannot be used in the key definition. For more information, see Configure partitioned replication on page 93.

SharePlex 9.2.8 Administrator Guide 75

l Include the columns in a supplemental log group. Otherwise, SharePlex must query the database for their values.
l Create an index on the target table and add the index to the SharePlex hints file, located in the variabledata directory, which directs the Post process to use the index.

Syntax for key definition

To create a key definition, type a space after the source object and use the following syntax, including the parentheses. src_owner.table !key (column_list) where:
l !key is a required keyword.
l column_list is a list of columns to include in the key. Separate column names with commas. A space after the comma is optional.

datasource_specification src_owner.table !key (col_name, col2_name, ...) tgt_owner.table

host@o.SID

Example
Datasource:o.ora1 scott.tab !key(name,ID)

scott.tab2

sysB@oraB

Filter DML operations
You can configure SharePlex to filter out the following DML from replication when wildcarding is being used. l Oracle DML type (INSERT, UPDATE, DELETE) l DML related to Oracle sequences and Oracle SQL*Loader direct-path loads.
Filter out a DML type
You can configure SharePlex to filter any type of DML operation so that the operation is not replicated to the target table. DML filtering is compatible with most other SharePlex configuration syntax.
Configure a DML filter
To configure a DML filter, add the following syntax to the source table specification. Leave a space between the table specification and the filter specification. You can specify multiple operation types to filter. Any additional syntax for other features, such as a column list or key definition, must follow the DML filter specification.
!dml(DML_type[,DML_type][,...]) Where DML_type is one of the following:

SharePlex 9.2.8 Administrator Guide 76

DML_type input i d u

Operation type INSERT DELETE UPDATE

Examples

Example 1 The following example filters DELETE operations from being replicated to the target table.

Datasource:o.ora

scott.emp !dml(d)

scott.emp

prodsys@o.sysdb

Example 2 The following example filters DELETEs and INSERTs so that only UPDATEs are replicated to the target table. This example also shows how a DML filter is compatible with a column mapping specification.

Datasource:o.ora scott.stock !dml(d,i) (ID, name)

scott.stock (SKU, prod)

sys2@o.sysdb

View current DML filters
Use the verify config command to view the DML that is being filtered for each table in the configuration file. This command can be used for an active or inactive configuration file.
sp_ctrl> verify config myconfig 7: "SCOTT"."EMP" "SCOTT"."EMP" prodsys@o.proddb Filter out >>>>> DELETES Unique Key : (EMPLOYEE_ID)
Restrictions
l The copy and compare commands do not support tables that include DML filtering in their specifications.
l If there are multiple specifications of a source table in the configuration file, the DML filter specification must be identical for all of them. Multiple specifications of the same source table occur in the following instances:
l Routing a source table to multiple targets without using a compound routing map. For more information, see Routing specifications in a configuration file on page 66.
l Combining full-table replication with horizontally partitioned replication. For more information, see Configure horizontally partitioned replication on page 93.

SharePlex 9.2.8 Administrator Guide 77

Filter DML related to specific Oracle objects from replication
You can prevent SharePlex from replicating sequences and SQL*Loader direct-path loads. By default the replication of these objects is enabled.

Filter out this object Sequences SQL*Loader direct-path loads

Set this parameter SP_OCT_REPLICATE_SEQUENCES SP_OCT_REPLICATE_DLOAD

Value 0 0

Map source and target columns
When source and target column names are different, you can specify an explicit column mapping in the configuration file, to ensure that Post applies row data to the correct target columns.

Guidelines for using column mapping
l To use column mapping, you must map every column in a source table to a column in the target table, even if only some source and target names are different. When some columns are mapped but not others, the entry is considered to be a column partition for vertically partitioned replication, and data changes for non-listed columns are not replicated.
l If the spelling case is different between the source and target column names, enclose them within quotes.
l You can use horizontally partitioned replication and column mapping for the same source table, but you cannot combine column mapping with vertically partitioned replication.
l A target table can have more columns than the source table, but there must at least be a target column for every source column.
l ALTER TABLE to add a column to a table that is configured with column mapping is not supported.

Configure column mapping

The following syntax creates a column map. For more information, see Configure data replication on page 57. datasource_specification

src_owner.table (src_col,src_col,...) tgt_owner.table (tgt_col,tgt_col,...)

routing_map

Configuration component datasource_specification
src_owner.table and tgt_

Description
The datasource specification. For more information, see Database specifications in a configuration file on page 64. The specifications for the source and target tables, respectively. For

SharePlex 9.2.8 Administrator Guide 78

Configuration component owner.table (src_col,src_col,...)
(tgt_col,tgt_col,...)
routing_map

Description
more information, see Create a configuration file on page 58.
A list of the source columns. Follow these rules to specify a column list:
l A column list must be enclosed within parentheses. l Separate each column name with a comma. A space after the
comma is optional. l The maximum length of a column list is 174820 bytes (the
maximum line length allowed in a configuration file).
A list of the target columns.
l List the target columns in the same logical order as their corresponding source columns. This is required regardless of the actual order of the target columns in the table, so that SharePlex builds the correct map in the object cache. For example, a change to the second column in the source list is replicated to the second column in the target list.
l The syntax rules for the source list also apply to the target list.
The routing map. For more information, see Routing specifications in a configuration file on page 66.

Configuration example
This example contains no case-sensitive columns. Datasource o.oraA sales.prod (ID,name,vendor) mfg.prod (UPC,product,supplier)

sysB@o.oraB

This example contains case-sensitive columns. Datasource o.oraA sales.prod (ID,"name",vendor) mfg.prod (UPC,"product",supplier)

sysB@o.oraB

Build a configuration file using a script
SharePlex provides the following scripts to automate the building of a configuration file to specify Oracle source objects.
l config.sql: configure all tables and optionally all sequences in the database. l build_config.sql: configure multiple or all tables in a schema

SharePlex 9.2.8 Administrator Guide 79

Supported databases
Oracle

Use config.sql
The config.sql script enables you to build a configuration that lists all of the tables, and optionally all of the sequences, in all of the schemas of a database. This script saves time when establishing a high-availability replication strategy or other scenario where you want the entire database to be replicated to an identical secondary database.
Conditions for using config.sql
l Source and target table names must be the same.
l The script does not configure objects in the SYS, SYSTEM, and SharePlex schemas. These schemas cannot be replicated since they are system and/or instance-specific.
l The script does not support partitioned replication. You can use the copy config command to copy the configuration file that the script builds, then use the edit config command to add entries for tables that use partitioned replication. Activate the new configuration file, not the original one.
l You can use the edit config command to make any other changes as needed after the configuration is built.
To run config.sql
1. Change directories to the config sub-directory of the SharePlex variable-data directory. The config.sql script puts configurations in the current working directory, and SharePlex configurations must reside in the config sub-directory. cd /vardir/config
2. Log onto SQL*Plus as SYSTEM. 3. Run config.sql using the full path from the util sub-directory of the SharePlex product directory.
@ /proddir/util/config.sql
Refer to the following table when following the prompts.

Prompt

What to enter

Target machine

The name of the target machine, for example SystemB.

Source database SID

The ORACLE_SID of the source (primary) Oracle instance, for example oraA. Do not include the o. keyword. The ORACLE_SID is case-sensitive.

Target database SID

The ORACLE_SID of the target (destination) Oracle instance, for example oraB. Do not include the o. keyword. The ORACLE_SID is case-sensitive.

Replicate Enter y to replicate sequences or n not to replicate sequences. sequences

SharePlex 9.2.8 Administrator Guide 80

Prompt
SharePlex oracle username

What to enter
The name of the SharePlex user in the source database. This entry prevents the SharePlex schema from being replicated, which would cause replication problems. If a valid name is not provided, the script fails.

NOTE: The name assigned by SharePlex to the configuration is config.file. If you run the script again to create another configuration file, it overwrites the first file. To preserve the original file, rename it before you create the second one.

Next steps:
l If any tables or owners are case-sensitive, open the configuration file with the edit config command in sp_ctrl, then use the text editor to enclose case-sensitive table and owner names within double-quote marks, for example "scott"."emp". The script does not add the quote marks required by Oracle to enforce case-sensitivity. sp_ctrl> edit config filename
l To ensure that the configuration is in the correct location, issue the list config command. If the name of the configuration is not shown, it was created in the wrong directory. Find the file and move it to the config sub-directory of the variable-data directory. sp_ctrl> list config

Use build_config.sql
The build_config.sql script enables you to build a configuration that contains multiple (or all) tables in a schema. It is an interactive script that prompts for each component of the configuration step by step. Instead of entering the information for each object and the routing individually, you can use a wildcard to select certain tables at once, or you can select all of the tables in the schema.
Conditions for using build_config.sql
l Source and target table names must be the same.
l The script does not support sequences. Before you activate the configuration that the script builds, you can use the edit config command in sp_ctrl to add entries for sequences.
l The script does not support partitioned replication. You can use the copy config command to copy the configuration that the script builds, then use the edit config command to add entries for the tables that use partitioned replication. Activate the new configuration, not the original.
l The script does not configure objects in the SYS, SYSTEM, and SharePlex schemas. These schemas cannot be replicated since they are system and/or instance-specific.
l You can run build_config.sql for different schemas, then combine those configurations into one configuration by using a text editor. Make certain to eliminate all but one Datasource:o.SID line, which is the first non-commented line of the file. Do not move the file out of the config sub-directory.
l You can use the edit config command to make any other changes as needed after the configuration is built.

SharePlex 9.2.8 Administrator Guide 81

To run build_config.sql
1. Change directories to the config sub-directory of the SharePlex variable-data directory. The build_ config.sql script puts configurations in the current working directory, and SharePlex configurations must reside in the config sub-directory. cd /vardir/config
2. Log onto SQL*Plus as SYSTEM. 3. Run build_config.sql using the full path from the util sub-directory of the SharePlex product directory.
@ /proddir/util/build_config.sql
Refer to the following table when following the prompts.

Prompt

What to enter

Target machine

The name of the target machine, for example SystemB.

Source database SID

The ORACLE_SID of the source (primary) Oracle instance, for example oraA. Do not include the o. keyword. The ORACLE_SID is case-sensitive.

Target database SID The ORACLE_SID of the target (destination) Oracle instance, for example oraB. Do not include the o. keyword. The ORACLE_SID is case-sensitive.

Owner of the source The owner of the source tables. database tables

Owner of the target The owner of the target tables. database tables

Table name to include (blank for all)

Do one of the following:
l Press Enter to accept the default, which selects all tables that belong to the source owner.

l Enter a wildcard (%) character and a string to select certain tables, for example %e_salary%.

l Enter an individual table name.

Name of the output A name for the configuration. The script gives the file a .lst suffix, for example Scott_

file to create

config.lst.

Next steps:
l If any tables or owners are case-sensitive, open the configuration with the edit config command in sp_ ctrl, then use the text editor to enclose case-sensitive table and owner names within double-quote marks, for example "scott"."emp". The script does not add the quote marks required by Oracle to enforce case-sensitivity. sp_ctrl> edit config filename
l To ensure that the configuration is in the correct location, issue the list config command. If the name of the configuration is not shown, it was created in the wrong directory. Find the file and move it to the config sub-directory of the variable-data directory. sp_ctrl> list config

SharePlex 9.2.8 Administrator Guide 82

7
Configure replication to and from a container database
SharePlex supports replication to and from Pluggable Databases (PDB) in Oracle multitenant container databases (CDB). This support is available on Unix and Linux platforms only.
Contents
Configure capture and delivery PDB configuration examples
Configure capture and delivery
SharePlex can replicate data from one PDB to: l another PDB in the same CDB l a PDB in a different CDB l a regular (non-PDB) target
SharePlex can replicate data from a regular source database to a PDB in a target Oracle CDB. In one configuration file, you can replicate to any number of target PDBs in the same CDB or a different CDB. To capture from a PDB
l In the configuration file, specify the TNS alias of a PDB as the datasource. For example, if the TNS alias is pdb1, the datasource specification is: Datasource: o.pdb1
l You can replicate from as many pluggable databases (PDBs) in the same CDB as desired: Create a separate configuration file for each PDB. Because each PDB is a different datasource, all configurations can be active at the same time.
SharePlex 9.2.8 Administrator Guide 83

l If replicating from more than one PDB on a system, use named export queues to separate the data streams from each one. This allows you to issue SharePlex commands that affect configurations, such as purge config or abort config, for one configuration without affecting the other configurations. For more information, see Configure named export queues on page 85
To replicate to a PDB Specify the TNS alias of the target PDB in the routing map, as shown in the following example where pdb2 is the target:
sys02@o.pdb2

PDB configuration examples

Example 1: This example shows two configuration files, one replicating from pdb1 and the other replicating from pdb2, both replicating data to pdb3.

Datasource: o.pdb1 hr.emp hr2.emp2

sys02@o.pdb3

Datasource: o.pdb2 sales.cust sales2.cust2

sys02@o.pdb3

Example 2: This example shows one configuration file replicating from pdb1 to pdb2 and pdb3, both targets being on different systems.

Datasource: o.pdb1 hr.sal hr2.sal2 hr.sal hr3.sal3

sys02@o.pdb2 sys03@o.pdb3

SharePlex 9.2.8 Administrator Guide 84

8
Configure named queues
This chapter contains instructions for using the advanced SharePlex configuration options of named queues. These options provide an additional level of flexibility to divide and parallelize data to meet specific processing and routing requirements. Before proceeding, make certain you understand the concepts and processes in Configure data replication.
Contents
Configure named export queues Configure named post queues
Configure named export queues
A named export queue is an optional, user-defined export queue that is attached to its own Export process. SharePlex creates each named Export queue and associated Export process in addition to the default export queue-process pair. When SharePlex creates a named export queue-process pair, it also creates a dedicated Import process, post queue, and Post process on the target to contain that data stream. You direct SharePlex to create one or more named export queues when you create your configuration file. Any data that is not configured for processing through a named export queue is processed through the default export queue.
SharePlex 9.2.8 Administrator Guide 85

Supported sources
Oracle
Supported targets
All
Benefits of named export queues
Use named export queues to isolate the replication of: l Individual configurations: By default, SharePlex sends data from all active configurations through one export queue-process pair per target system, but the use of named Export queues enables you to separate each of those replication streams into its own export queue and Export process. In this way, you ensure that purge config or abort config commands that are issued for one configuration do not affect any of the others.
SharePlex 9.2.8 Administrator Guide 86

l Selected database objects: You can use a named export queue to isolate certain objects such as tables that contain LOBs. Because each named export queue has its own Import process, post queue, and Post process on the target, you are able to isolate the data the entire way from source to target. For more information about the benefits of named post queues, see Configure named post queues on page 89.
Additional benefits:
l You can stop the Export or Import process for one data stream, while allowing the others to continue processing.
l You can set SharePlex parameters to different settings for each export queue-process pair. This enables you to tune the performance of the Export processes based on the objects replicating through each one.

Considerations when using named export queues
l Make certain that each queue name is unique.
l You can combine named export queues with default export queues. Tables in the configuration with a standard routing map (targetsys@database_spec without a named queue specification) are replicated through a default export queue.
l All tables with referential integrity to one another must be in the same export queue.
l SharePlex has a maximum number of allowed queues. For more information, see Routing specifications in a configuration file on page 66.
NOTE: If Post returns the error message "shs_SEMERR: an error occurred with the semaphore" on a Windows system, the number of semaphores may need to be increased to accommodate the queues that you created. For more information, see Post stopped on page 239.

Configure a named export queue

Use the following syntax to define a routing map that includes a named export queue. source_host:export_queuename*target_host[@database]

Configuration with named export queue in routing map

Datasource: o.SID

src_owner.table

tgt_owner.table

source_host:export_queue*target_host[@database_specification]

Routing component source_host export_queue

Description
The name of the source system.
The name of the export queue. Queue names are case-sensitive on all platforms. Use one word only. Underscores are permissible, for example:

SharePlex 9.2.8 Administrator Guide 87

Routing component
target_host database specification

Description
sys1:export_q1*sys2@o.myora
The name of the target system. One of the following for the datasource: o.oracle_SID r.database_name

One of the following if the target is a database: o.oracle_SID o.tns_alias o.PDBname r.database_name c.oracle_SID

NOTES:
l Allow no spaces between any components in the syntax of the routing map. l For more information about the components of a configuration file, see Configure data
replication on page 57.

Examples

The following configuration files show two different datasources that are being replicated to two different

databases on the same target system. Each datasource is routed through a named export queue.

Datasource:o.oraA

scott.emp scott.sales

sysA:QueueA*sysB@o.oraC sysA:QueueA*sysB@o.oraC

Datasource:o.oraB scott.prod scott.cust

scott.prod scott.cust

sysA:QueueB*sysB@o.oraD sysA:QueueB*sysB@o.oraD

The following shows how to separate a table that contains LOBs from the rest of the tables by using named

export queues.

Datasource:o.oraA

scott.cust scott.sales scott.prod scott.emp_LOB

sysA:QueueA*sysB@o.oraC sysA:QueueA*sysB@o.oraC sysA:QueueA*sysB@o.oraC sysA:QueueB*sysB@o.oraC

Alternatively, you could simply define a named export queue for the LOB table and allow the remaining tables to

be processed through the default export queue.

Datasource:o.oraA

scott.cust scott.sales scott.prod scott.emp_LOB

sysB@o.oraC sysB@o.oraC sysB@o.oraC sysA:lobQ*sysB@o.oraC

SharePlex 9.2.8 Administrator Guide 88

How to identify named export queues
You can view named export queues through sp_ctrl: l Use the qstatus command to view all queues on a system. l Use the show command to view all Export processes with their queues.
See the SharePlex Reference Guide for more infomation about theses commands.
Configure named post queues
A named post queue is an optional component of the routing map in the configuration file. A named post queue is a user-defined post queue with its own Post process, which together operate in parallel to the default post queue and Post process. You can define one or more named post queue-process pairs to establish a set of parallel Post replication streams.
Supported sources
Oracle
Supported targets
All
Benefits of named post queues
You can use named post queues to isolate data from different tables into two or more separate Post streams. By using named post queues, you can improve posting performance by isolating objects such as the following that cause processing bottlenecks:
l large tables l objects that have LOB columns. Named post queues are recommended for objects that contain LOBs. l objects that involve large transactions. l any objects whose operations you want to isolate. Process the remaining objects through additional named post queues, or use the default post queue. Objects in the configuration file with a standard routing map (host@target) are replicated through a default post queue. You can use horizontal partitioning to divide the rows of very large tables into separate named post queues as an added measure of parallelism. You can set SharePlex parameters to different settings for each queue-process pair. This enables you to tune the performance of the Post processes based on the objects replicating through each one.
SharePlex 9.2.8 Administrator Guide 89

Considerations when using named post queues
l The analyze config command can help you determine how to organize your tables into named queues, based on any dependencies they have and their individual transactional activity. Run the command over a period of time that captures typical database activity, and then view the command output.
l Assign each post queue a unique name. l If objects are linked by relational dependencies, process all of those objects through the same named
post queue. If interdependent objects are not replicated through the same post queue, parent and child operations may be applied out of order and will cause database errors. As an alternative to processing interdependent objects through the same queue, you can disable their referential constraints on the target. This may be acceptable, because the constraints are satisfied on the source system and then replicated to the target. l When using multiple Posts, the target objects might not be changed in the same order as the corresponding source objects, possibly causing the target database to be inconsistent with the source database at any given point in time. l If you implement named post queues for objects in an active configuration (thus changing the routing) SharePlex locks those objects to update its internal directions.
SharePlex 9.2.8 Administrator Guide 90

l SharePlex has a maximum number of allowed queues. For more information, see Routing specifications in a configuration file on page 66.
NOTE: If Post returns the error message "shs_SEMERR: an error occurred with the semaphore" on a Windows system, the number of semaphores may need to be increased to accommodate the queues that you created. For more information, see Post stopped on page 239.

Configure a named post queue

If you are using named export queues, SharePlex creates a named post queue-process pair for each one by default. If you are not using named export queues, use the following syntax to define a named post queue in the configuration file by adding the :queue component to the routing map:
host:queue@target

Configuration with named post queue in routing map

Datasource: o.SID src_owner.table tgt_owner.table

host:queue[@database_specification]

Routing component host queue
database_specification

Description
The name of the target system.
The unique name of the post queue. Queue names are case-sensitive on all platforms. One word only. Underscores are permissible, for example:
sys2:post_q1@o.myora
One of the following for the datasource: o.oracle_SID

One of the following if the target is a database:
l o.oracle_SID o.tns_alias o.PDBname r.database_name c.oracle_SID

NOTES: l Allow no spaces between any components in the syntax of the routing map. l For more information, see Configure data replication on page 57.
Examples
The following configuration creates one post queue named Queue1 that routes data from table scott.emp and another post queue named Queue2 that routes data from table scott.cust.

SharePlex 9.2.8 Administrator Guide 91

Datasource:o.oraA scott.emp scott.cust

scott.emp scott.cust

sysB:Queue1@o.oraC sysB:Queue2@o.oraC

The following shows how a named post queue is specified when you are routing data in a pass-through

configuration using an intermediary system. For more information, see Configure replication to share or

distribute data on page 114.

Datasource:o.oraA

scott.emp

sysB*sysC:Queue1@o.oraC

How to identify a named post queue
A named post queue is identified by the datasource (source of the data) and one of the following: l the name of an associated named export queue (if the Import is linked to a named export queue) l the user-assigned post-queue name (if the Import is linked to a default export queue).
You can view named post queues through sp_ctrl: l Use the qstatus command to view all queues on a system. l Use the show command to view all Post processes with their queues.
See the SharePlex Reference Guide for more infomation about theses commands.

SharePlex 9.2.8 Administrator Guide 92

9
Configure partitioned replication
This chapter contains instructions for using the advanced SharePlex configuration options of horizontally partitioned and vertically partitioned replication. These options provide an additional level of flexibility to divide, parallelize, and filter data to meet specific requirements. Before proceeding, make certain you understand the concepts and processes in Configure data replication.
Contents
Configure horizontally partitioned replication Configure vertically partitioned replication
Configure horizontally partitioned replication
Use horizontally partitioned replication to divide the rows of a table into separate processing streams. You can use horizontally partitioned replication to:
l Replicate a subset of rows to a target, while retaining the rest of the rows in the source. l Replicate different subsets of rows to different targets. l Divide the replication of a source table into parallel Post queues for faster posting to the target table.
Supported sources
Oracle
Supported targets
All
SharePlex 9.2.8 Administrator Guide 93

Overview of horizontally partitioned replication
To configure horizontally partitioned replication for a table, the steps are: 1. Define row partitions and link them to a partition scheme. l A row partition is a subset of rows in a source table that you want to replicate as a group. l A partition scheme is a logical container for row partitions. 2. Specify the name of the partition scheme in the SharePlex configuration file to include the partitions in replication.
Partition types
The row partitions in a partition scheme can be based on one of the following: l Columns: A column-based partition scheme contains row partitions defined by a column condition. A column condition is a WHERE clause that defines a subset of the rows in the table. l Hash: A hash-based partition scheme contains row partitions defined by a hash value that directs SharePlex to distribute rows evenly across multiple queues.
About column-based partition schemes
You can use row partitions based on column conditions for the following purposes: l Use a single row partition to replicate only a subset of the rows of a table. For example, you can replicate only those rows where the value of the YEAR column is greater than 2014. The partition scheme in this case could be named "Since2014" or "Recent." l Use multiple row partitions to divide the rows of a table so that each set of rows replicates to a different target. For example, a table named CORPORATE.SALES could have two row partitions named "East" and "West." The column conditions are defined accordingly, where the rows that satisfy REGION = EAST replicate to one location and the rows that satisfy REGION = WEST replicate to a different location. The partition scheme could be named "Sales_by_region."
SharePlex 9.2.8 Administrator Guide 94

l Use multiple row partitions to divide the rows of a table into parallel processing streams (parallel Export-Import-Post streams) for faster posting to a target table. For example, you can improve the flow of replication to a heavily updated target table. The use of column conditions for this purpose is appropriate only if the table contains a column that enables you to split the processing evenly among parallel Post processes.
About hash-based partition schemes
You can use row partitions based on a hash value to divide the rows of a table into parallel processing streams (parallel Export-Import-Post) for faster posting to a target table. The advantage of using a hash value over column conditions to create partitions is that the rows are divided evenly and automatically by SharePlex, without the need to reference table columns in WHERE clauses. However, unlike column-based partition schemes, you cannot use the SharePlex compare or repair commands for hash-based partition schemes.
Combine partitioned replication with full-table replication
You can combine horizontally partitioned and vertically partitioned replication for maximum control over how information is distributed. For example:
l A corporate headquarters maintains a master corporate database. l Each of the corporation's four regional offices maintains its own database. l Corporate headquarters uses vertically partitioned replication to share some column data of a table with
the regional offices, but does not share any columns that contain sensitive data. l The rows of the table are horizontally partitioned into four groups (East, West, North, South) for
replication, so that each region receives only the record changes that apply to it. Horizontally partitioned replication can be used in conjunction with full-table replication for the same table, for example to route groups of rows to different reporting systems and all rows to a backup system.
Limitations of use
Hash-based partitioning does not support the following:
SharePlex 9.2.8 Administrator Guide 95

l The compare and repair commands. l Index-organized tables (IOT) and tables that contain LOBs or LONGs. l Operations that delete or update a key value and then reinsert the same key value. This can cause
unique constraint violations because of different rowids. l Column-based partitioning on the same table. Hash-based partitioning also does not support operations that cause rows to migrate into a different partition. Examples of such operations are: o Update a value so that it moves to a new row partition o Table reorganization o Split a table partition or combine two partitions o Export or import the table o ALTER TABLE with the MOVE option o ALTER TABLE with the SHRINK SPACE option o FLASHBACK TABLE o Redefine a table by using dbms_redefinition o UPDATE to a non-partitioned table that changes row size so that the data does not fit into the
current block o DELETE of a row in a non-partitioned table and then re-insert. NOTE: If Post returns the error message "shs_SEMERR: an error occurred with the semaphore" on a Windows system, the number of semaphores may need to be increased to accommodate the queues that you created. For more information, see Post stopped on page 239.
Define partition schemes and row partitions
Use the add partition command to create row partitions and assign them to a partition scheme.
To partition rows based on column conditions Issue add partition for each row partition that you want to create in a given partition scheme. When you create the first row partition, SharePlex creates the partition scheme as well. sp_ctrl> add partition to scheme_name set condition = column_condition and route = routing_map [and name = name] [and tablename = owner.table] [and description = description]
To partition rows based on a hash value Issue add partition once to specify the number of hash partitions to create. sp_ctrl> add partition to scheme_name set hash = value and route = value
add partition command syntax
NOTE: After you specify add partition with to scheme_name and the set keyword, all other components can be in any order.
SharePlex 9.2.8 Administrator Guide 96

Component to scheme_name
set condition = column_condition
hash = value route = routing_map

Description
to is a required keyword indicating the row partition is being added to scheme_name. scheme_name is the name of the partition scheme. The partition scheme is created by the first add partition command that you issue, which will also specify the first set of rows to partition. If you are making heavy use of horizontal partitioning, it may help to establish naming conventions for your partition schemes.
Required keyword that starts the definition of the row partition.
Creates a row partition based on a column condition. The condition must be in quotes. Use standard WHERE conditional syntax such as ((region_ id = West) and region_id is not null). See How to create a valid column condition for additional information. The condition and hash components are mutually exclusive.
Creates a row partition based on a hash value. The specified value determines the number of row partitions in the partition scheme. The condition and hash components are mutually exclusive.
The route for this partition. This can be one of the following: Partition based on a column condition: Specify any standard SharePlex routing map, for example: sysB@o.myora or sysB:q1@o.myora or sysB@o.myora+sysC@o.myora (compound routing map). If the target is JMS, Kafka, or a file, then the target should be specified as x.jms, x.kafka, or x.file, for example: sysA:hpq1@x.kafka. To route a partition to multiple target tables that have different names, do the following:
l Issue a separate add partition command for each different target name. Use the tablename option to specify the name.
l In the configuration file, specify any of these target tables as the target table in the entry that uses this partition scheme. SharePlex will detect the other names when the configuration is activated.
l Set the SP_ORD_FIRST_FIND parameter to 0 so that SharePlex checks all of the column conditions in the partition scheme. By default SharePlex assumes that any given row change will satisfy only one column condition in the partition scheme. For more information, see the SharePlex Reference Guide.
Partition based on a hash: Use the following format to direct SharePlex to create a named post queue for each partition: host:basename|#{o.SID | r.database} where:

SharePlex 9.2.8 Administrator Guide 97

Component name = name tablename = owner.table description = description

Description
l host is the name of the target system.
l basename is the base name that is assigned to all queues.
l |# directs SharePlex to number the queues sequentially by appending the base name with an integer, starting with 1 to the value set with hash.
l o.SID for an Oracle target or r.database for an Open Target target.
(Recommended) A short name for this partition. This option is only useful for partitions based on column conditions. A name eliminates the need to type out long column conditions in the event that you need to modify or drop the partition in the future.
(Optional) Use this option when there are multiple target tables and one or more have different names. Issue a separate add partition command for each name. The table name must be fully qualified. If case-sensitive, the name must be specified in quotes. Example: add partition to scheme1 set name = p1 and condition = "C1 > 200" and route = sysb:p1@o.orasid and tablename = myschema.mytable
(Optional) Description of this partition.

Examples
Partitions based on a column condition
Route different sets of rows through different post queues: sp_ctrl> add partition to scheme1 set name = q1 and condition = "C1 >= 200" and route = sysb:q1@o.orasid sp_ctrl> add partition to scheme1 set name = q2 and condition = "C1 < 200" and route = sysb:q2@o.orasid
Route different sets of rows to different target systems and different table names from the source: sp_ctrl> add partition to scheme1 set name = east and condition = "area = east" and route = sys1e@o.orasid and tablename = ora1.targ sp_ctrl> add partition to scheme1 set name = west and condition = "area = west" and route = sys2w@o.orasid and tablename = ora2.targ

Partitions based on a hash
Divide rows into four partitions, each processing through a different post queue: sp_ctrl> add partition to scheme1 set hash = 4 and route = sysb:hash|#@o.ora112

How to create a valid column condition
The following are guidelines for creating column conditions. These guidelines do not apply to row partitions that are created with a hash value.

SharePlex 9.2.8 Administrator Guide 98

Choose appropriate columns
The types of columns on which you base your column conditions vary per datasource: Base column conditions on columns whose values will not change, such as PRIMARY or UNIQUE key columns. The objective is to avoid a partition shift, where changes made to the conditional columns of a partition can cause the underlying data to satisfy the conditions of a different (or no) partition. Partition shift case 1: The column value is updated so that the new value no longer satisfies any column condition:
l SharePlex performs the operation, but future operations on that row are not replicated. The reason: the row no longer satisfies a column condition.
l The source and target tables of the original partition are now out of synchronization, but Post returns no errors.
Partition shift case 2: A row that satisfies one column condition gets updated to meet a different condition:
l Post cannot find a matching target row. The reason: the original change was not replicated because it did not meet the column condition.
l Post returns an out-of-sync error.
You can use the following method to repair the out-of-sync rows that are caused by changes to the values of column conditions:
l Use the compare command to repair the out-of-sync rows. For more information about this command, see the SharePlex Reference Guide.
Additionally, you can ensure that data is replicated properly by setting the following parameter on the source prior to activating the configuration file.
l Set the SP_ORD_HP_IN_SYNC parameter to a value of 1. When this parameter is enabled, if an UPDATE changes a column (conditional column) value resulting in a row that no longer satisfies the correct condition, SharePlex corrects the row. Enabling this parameter causes some performance degradation, depending on how many tables are configured for horizontally partitioned replication. See the SharePlex Reference Guide for more information and a list of conditions corrected by this parameter.
NOTE: If you are using a column other than a key to base the column condition on, and you notice reduced performance with horizontally partitioned replication enabled, add a log group for that column.
Use supported data types
SharePlex supports the following data types in column conditions:
l NUMBER l DATE l CHAR l VARCHAR l VARCHAR2 l LONG VARCHAR
NOTES:
SharePlex 9.2.8 Administrator Guide 99

l For the dates, SharePlex uses MMDDSYYYYHH24MISS. For example: hiredate<`1111 2011000000'
l Horizontally partitioned replication does not support the following: l data types other than the ones listed in this section. This also excludes large types like LOBs and object types such as VARRAYs and abstract data types. l Oracle TO_DATE function l UPDATEs or INSERTs on LONG columns larger than 100k l Sequences l TRUNCATEs of an Oracle partition
Use standard conditional syntax
The following list shows the conditional syntax that SharePlex supports in a column condition, where:
l value can be a string or a number. Enclose strings and dates within single quote marks (`west'). Do not use quote marks for numbers .
l column is the name of a column in the table that you are configuring to use horizontally partitioned replication.
column = value not (column = value) column > value value > column column < value column <= value column >= value column <> value column != value column like value column between value1 and value2 not (column between value1 and value2 ) column is null column is not null
Conditions can be combined into nested expressions with parentheses and the AND, OR, and NOT logical connectives.
Example column conditions
not (col1 = 5) (col2 = 5) and not (col3 = 6) ((col1 is not null) and (col2 = 5))
Additional guidelines
l NULLs are replicated by SharePlex in cases such as this one: not (department_id = 90). If department_ id is NULL, it is replicated. To avoid replicating records with NULLs, include the column is not null syntax as part of the condition, for example: not (department_id = 90) and department_id is not null.
SharePlex 9.2.8 Administrator Guide 100

l If parentheses are not used to indicate operator precedence, SharePlex supports operator precedence in the same order that SQL does. For example, a condition not x and y behaves the same way as (not x) and y. A condition x and y or z behaves the same as (x and y) or z. When a condition includes parentheses, the explicit precedence is respected.
l If the condition column is a VARCHAR column and the values used to define the partitions are string literals, the entire condition must be enclosed in double quotes, as in the following example: add partition to scheme set route=route and condition="C2 = 'Fred'"
l If the column name must be enclosed in quotes, then the entire condition must be enclosed in quotes, as in the following example: add partition to scheme set route=route and condition="\"c2\" > 0"
l Do not:
l include references to other tables in the column condition.
l exceed the 1024 bytes maximum defined storage.
l During the activation of a configuration that refers to partition schemes, SharePlex verifies the syntax in the column conditions of those schemes. If any syntax is incorrect, the activation fails. SharePlex prints an error to the Event Log that indicates where the error occurred.

Specify partition schemes in the configuration file

Use one configuration file for all of the data that you want to replicate from a given datasource, including tables that will have full-table replication and those that will use partitioned replication. For more information about how to create a configuration file, see Configure data replication on page 57. To configure entries for horizontally partitioned replication, use the following syntax.
Datasource: o.SID

src_owner.table

tgt_owner.table

!partition_scheme

routing_map

Component o.database src_owner.table and tgt_ owner.table !partition_scheme
! routing_map

Description
The datasource designation. Use the o. notation for an Oracle source. For database, specify the ORACLE_SID.
The specifications for the source and target tables, respectively.
The name of the partition scheme to use for the specified source and target tables. The ! is required. The name is case-sensitive. Compound routing of multiple partition schemes is not supported, for example !schemeA+schemeB. Create a separate entry for each partition scheme that you want to use for the same source table. See Examples.
A placeholder routing map. It is required only if a route that you used in a partition scheme is not listed somewhere in the configuration file. SharePlex requires every route to be in the configuration file even if it is

SharePlex 9.2.8 Administrator Guide 101

Component

Description
listed in a partition scheme.
l NOTES:
l This option is valid only for partitions based on a column condition.
l If using named queues, list each queue route with this option
l If routing the partition scheme to different targets, list each one with this option. You can use a compound routing map if the names of all target tables are identical.
See Examples.

Examples To specify a partition scheme:

Datasource: o.mydb scott.emp

scott.emp_2

!partition_emp

To specify multiple partition schemes for the same source table:

Datasource: o.mydb scott.emp scott.emp

scott.emp_2 scott.emp_3

!partition_schemeA !partition_schemeB

To specify a placeholder routing map:

! targsys1 ! targsys2@o.ora2+targsys3@o.ora3 This placeholder is only required for partitions based on column conditions.

View the partitions and schemes
Use the view partitions command to view the row partitions in one partition scheme or all partition schemes in a horizontally partitioned replication configuration.
To view row partitions
1. Run sp_ctrl on the source system. 2. Issue the following command with either option, depending on whether you want to view all partitions or
just those for a particular partition scheme. sp_ctrl> view partitions for {scheme_name | all}

SharePlex 9.2.8 Administrator Guide 102

The following example shows both a hash-based partition scheme and a column-based partition scheme.

sp_ctrl> view partitions all

Scheme ----------HASH4 TEST_CT TEST_CT

Name ------------hash highvalues lowvalues

Route -----------------------------sys02:hash|#@o.ora112 sys02:highvalues@o.ora112 sys02:lowvalues@o.ora112

Hash -----4

Condition --------------ROWID sales>=10000 sales<10000

Hash4 hash-based partition scheme
l The Scheme column shows a partition scheme named HASH4. l The Name column shows that the name for the partition definition is hash. l The Route column shows that the partitions are created automatically and that the target is o.ora112. l The Hash column has a value of 4, which indicates that this is a hash-based partition scheme with four
partitions. l The Condition column shows that the type of hash algorithm that is being used is the default of rowid,
rather than block.
TEST_CT column-based partition scheme
l The Scheme column shows a partition scheme named TEST_CT. There are two entries for this name, indicating that it contains two partitions.
l The Name column shows the name of each partition, which by default is the name of the post queue or the value set with the Name option of the add partition command.
l The Route column shows that the names of the post queues are based on the parition name and that the target is o.ora112.
l The Hash column is empty for a column-based partition scheme. l The Condition column shows the column condition that creates the row partition.

To view partition post queues The qstatus command on the target shows the post queues that are associated with horizontally partitioned replication.

Queues for TEST_CT column-based partition scheme sp_ctrl sys02> qstatus

Queues Statistics for sys02

Name: highvalues (o.ora11-o.ora112) (Post queue)

Number of messages:

0 (Age

0 min; Size

Backlog (messages):

0 (Age

0 min)

1 mb)

Name: lowvalues (o.ora11-o.ora112) (Post queue)

Number of messages:

0 (Age

0 min; Size

Backlog (messages):

0 (Age

0 min)

1 mb)

SharePlex 9.2.8 Administrator Guide 103

Queues for HASH4 hash-based partition scheme: Queues Statistics for sys02

Name: hash1 (o.ora11-o.ora112) (Post queue)

Number of messages:

0 (Age

0 min; Size

Backlog (messages):

0 (Age

0 min)

Name: hash2 (o.ora11-o.ora112) (Post queue)

Number of messages:

0 (Age

0 min; Size

Backlog (messages):

0 (Age

0 min)

Name: hash3 (o.ora11-o.ora112) (Post queue)

Number of messages:

0 (Age

Backlog (messages):

0 (Age

0 min; Size 0 min)

Name: hash4 (o.ora11-o.ora112) (Post queue)

Number of messages:

0 (Age

Backlog (messages):

0 (Age

0 min; Size 0 min)

1 mb) 1 mb) 1 mb) 1 mb)

Make changes to partition schemes
The following commands or parameters are available to manage partition schemes. For more information, see the SharePlex Reference Guide.

Task

Command/Parameter Description

Modify a partition

modify partition command

Modifies any of the attributes of a row partition definition.

Remove a partition scheme

drop partition scheme command

Removes the partition scheme and all row partitions within it.

Change hash SP_OCF_HASH_BY_ Change the hash algorithm from the default of rowid-based to

algorithm

BLOCK

block-based. Set to 1 to enable block-based algorithm.

Configure vertically partitioned replication
Use vertically partitioned replication to replicate a subset of the columns of a table. For example, you can replicate data changes made to C1, C2, C3, and C4, but not changes made to C5 and C6, as shown in the diagram.

SharePlex 9.2.8 Administrator Guide 104

Supported sources
Oracle
Supported targets
All
Guidelines for using vertically partitioned replication
Follow these guidelines when creating a configuration file that includes vertically partitioned replication. l Vertically partitioned replication is appropriate for reporting and other data sharing strategies, but it is not appropriate for high availability environments. Once you configure a table for vertically partitioned replication, SharePlex does not recognize the other columns, so data in those columns is not replicated.
SharePlex 9.2.8 Administrator Guide 105

l You can combine horizontally partitioned and vertically partitioned replication for maximum control over which information is distributed, and to where. For example: A company has a headquarters and regional divisions. The headquarters maintains the corporate database, and each region maintains a regional database. The headquarters uses vertically partitioned replication to share some of the column data of a table to those locations, while retaining other sensitive data at headquarters. Row changes made to the shared columns are further partitioned horizontally, for replication to the appropriate regional database.
l A table cannot be configured to replicate some columns to one target system and all columns to another (combination of vertically partitioned replication and full-table replication). You can, however, configure full-table replication to an identical table on one target, and then configure vertically partitioned replication from that target to a second target that contains the table that requires only the partition columns.
l A target table can, but does not have to, contain all of the same columns as its source table. The target can contain just the columns being replicated from the source table. The names of corresponding source and target columns do not need to be the same. Corresponding columns must contain the same data types (same type, size, precision).
Configure vertically partitioned replication
To configure vertically partitioned replication, you specify either a column partition or an exclusion column partition in the configuration file:
l A column partition specifies the columns that you want to include in replication. Only data changes that are made to the specified columns get sent to the target.
l An exclusion column partition specifies columns to be excluded from replication. No data from those columns is replicated to the target.
Follow these rules to specify either type of column partition:
l There can be one partition per source table. A column partition and an exclusion partition are mutually exclusive.
l A column list must be enclosed within parentheses. l Separate each column name with a comma. A space after the comma is optional. l The maximum length of a partition is 174820 bytes (the maximum line length allowed in a configuration
file). Therefore, the actual number of columns that you can list depends on the length of each name. l The columns can be contiguous or non-contiguous in the source table. For example, you can replicate
the first, third and seventh columns of a table. l Key columns are not required to be included in the partition. l If using horizontally partitioned and vertically partitioned replication together for this table, all of the
columns in the partition scheme must be part of the column condition. l Use one configuration file for all of the data that you want to replicate from a given datasource, including
tables that will have full-table replication and those that will use partitioned replication.
To configure entries for vertically partitioned replication, use the following syntax. For more information about how to create a configuration file, see Configure data replication on page 57.
SharePlex 9.2.8 Administrator Guide 106

datasource_specification # table specification with column partition

src_owner.table (src_col,src_col,...)

tgt_owner.table [(tgt_col,tgt_col,...)]

# table specification with exclusion column partition

src_owner.table !(src_col,src_col,...)

tgt_owner.table

routing_map routing_map

Configuration component src_owner.table and tgt_ owner.table (src_col, src_col,...) !(src_col,src_col,...)
(tgt_col,tgt_col,...)
routing_map

Description
The specifications for the source and target tables, respectively.
Specifies a column partition that lists the columns to include in replication. No other column data is replicated, including data in columns that are added after the start of replication (assuming DDL replication is enabled).
Specifies an exclusion column partition that lists the columns to exclude from replication. All other column data is replicated, including data in columns that are added after the start of replication (assuming DDL replication is enabled). NOTE: When using an exclusion column partition, the corresponding source and target column names must be identical, and the excluded columns cannot be used in a key definition. For more information, see Define a unique key on page 75.
The target columns. Use this option to map source columns to target columns that have different owners or names. If the source and target columns have identical owners or names, the target columns can be omitted. To map source columns to target columns, follow these rules:
l The syntax rules for the source column partition also apply to the target column list.
l The target columns must have identical definitions as their source columns, except for their names.
l List the target columns in the same logical order as their corresponding source columns. This is required regardless of the actual order of the target columns in the table, so that SharePlex builds the correct correlations in the object cache. For example, a change to the second column in the source list is replicated to the second column in the target list.
The routing map for the column partition. The routing map can be one of the following:
l If using horizontally partitioned replication for the source table, specify a partition scheme, as in: !partition_scheme.

SharePlex 9.2.8 Administrator Guide 107

Configuration component

Description
l If not using horizontally partitioned replication for the source table, specify a routing map as follows:
l Use a simple routing map like sysB@o.myora if replicating the column partition to one target. A route with a named export or post queue is supported. For more information, see: Configure named export queues on page 85 Configure named post queues on page 89
l Use a compound routing map like sysB@o.myora+sysC@o.myora2 if replicating the column partition to multiple target systems. IMPORTANT! A compound routing map must be used, rather than listing multiple targets in separate entries, because only one column condition per source table can be listed in the configuration file. To use a compound routing map, the owners and names of all of the target tables must be identical. For more information, see Routing specifications in a configuration file on page 66.

Configuration examples
The following is a vertically partitioned replication configuration replicating to multiple targets by using a compound routing map. To use a compound routing map for this source table, all targets must be named scott.sal.

Datasourceo.oraA scott.emp (c1,c2)

scott.sal

sysB@o.oraB+sysC@o.oraC

The following is a vertically partitioned replication configuration replicating to a single target where the target columns have different names from those of the source.

Datasourceo.oraA scott.emp (c1,c2)

scott.sal (c5,c6)

sysB@o.oraB

The following configuration file is not valid because it repeats the same column partition of scott.emp (c1, c2) twice in the configuration file.

Datasourceo.oraA scott.emp (c1,c2) scott.emp (c1,c2)

scott.cust (c1,c2) scott.sales (c1,c2)

sysB@o.oraB sysC@o.oraC

SharePlex 9.2.8 Administrator Guide 108

10
Configure replication to a change history target
This chapter contains instructions for how to configure SharePlex to maintain a change-history target. SharePlex enables you to maintain this history, while also replicating the same data set to maintain up-to-date targets.
Contents
Overview of the change-history target Configure change history
Overview of the change-history target
A change history target differs from a replication target in that a change history target maintains a record of every change that occurs to a source object or objects, rather than simply maintaining a mirror of the current state of the source data. While regular replication overlays current target data with change data, change history inserts the change data to the target as a new record. The old data is preserved as a step-by-step record of change. The historical data can be queried and analyzed for such purposes as data mining or resolving customer disputes. By using SharePlex to maintain change history on a secondary server, you can offload the overhead from the production database. Such overhead includes the SQL work of adding the history rows, the extra storage of those rows, and the query activity against the historical data.
NOTE: File, JMS, and Kafka targets support change history by default, because every source change is written as a separate XML record. There is no overlaying of old data with new. Metadata that is supported for these targets is included automatically when Post writes the XML. For a list of supported metadata, see the target command in the SharePlex Reference Guide.
Capabilities
This replication strategy supports the following:
SharePlex 9.2.8 Administrator Guide 109

l Identical or different source and target names l Use of vertically partitioned replication l Use of horizontally partitioned replication l Use of named export and post queues l Combination of regular replication and change-history replication of the same source object(s)
Supported sources
Oracle
Supported targets
Oracle target
Operations supported
SharePlex supports adding a change history row for these operations: l INSERT l UPDATE l DELETE l TRUNCATE l ALTER TABLE to DROP COLUMN (NOTE: for ADD COLUMN, Post adds a column to the table, but does not create a change history row.) l ALTER TABLE to MODIFY the data type of a column
Operations not supported
l Changes made to UDT or VARRAY columns l DBMS_LOB operation that is used to change a part of a LOB column (The value stored for that column
on the target will not be the complete LOB column.)
How SharePlex maintains change history
In a change history configuration, each target table serve as a history table that records every change made to the source data as a continuous series of rows. Each new change row that SharePlex inserts includes the following:
SharePlex 9.2.8 Administrator Guide 110

l the values of the key columns
l the after image of the changed columns. For inserts and updates, the after image consists of the new values of the columns that were changed (or added in the case of an insert). For deletes, the after image consists of the key values plus the other columns set to null.
l (optionally) a set of metadata values that provide context for the change. For example, there is metadata that captures the userid of the user who made the change and the source system where the change was made (useful when change data is tracked from multiple databases).
SharePlex can be configured to include the before image of update operations in the history or to control which operation types are included in the history. For example, you could include only updates and deletes.

Configure change history
To configure change history, you use special syntax in the SharePlex configuration file and, optionally, configure filter rules and other attributes to customize the history to your needs.

Create a change-history configuration file

1. Make certain that SharePlex is installed and the system is prepared according to the instructions in the SharePlex Installation Guide.

2. Create the Oracle target history tables with the same name and structure as the source tables whose history they will track, but omit all constraints on all columns.
IMPORTANT: The Oracle target tables must not have PRIMARY KEY, FOREIGN KEY, UNIQUE, NOT NULL, or CHECK constraints, nor can columns be defined with a DEFAULT value. Because this is a history of changes, a row may have the same image as another row that has the same key. Post does not perform integrity checks on a change-history target.

3. Disable triggers on the target tables.

4. Allow no DML or DDL to be performed on the target tables except by SharePlex.

5. On the source system, create a configuration file using the following syntax. For more information about how to create a configuration file, see Configure data replication on page 57.
datasource_specification

src_owner.table

!cdc:tgt_owner.table

host@c.SID

where:

l Datasource:o.SID is the ORACLE_SID of the source Oracle instance. l src_owner.table is the fully qualified name of a source object (owner.object) or a wildcarded
specification. l !cdc: identifies the target as a change-history table. l tgt_owner.table is the fully qualified name of the target history table or a wildcarded specification. l host is the target system. l c.SID specifies the target Oracle instance.

SharePlex 9.2.8 Administrator Guide 111

6. (Optional) Run the following script on the target tables to add default metadata columns with their default names. Post automatically populates the default metadata columns without any additional configuration. You can customize the script to meet your requirements. product_dir/util/add_change_tracking_columns.sql NOTES: l The script only adds the default columns. To add optional columns, or to change a column name, use the target command to add them to the Post configuration. For a list of default and optional metadata columns, see the target command in the SharePlex Reference Guide. l The default columns are automatically added to new tables that are added to the SharePlex change history configuration.
Additional change history configuration options
This section describes how you can customize the SharePlex change history configuration.
Customize column names
You can use the target command with the colname option to customize the name of any target metadata column. For instructions, see the SharePlex Reference Guide.
Add the before image to each change row
You can include the before image of updates in the target table by setting the SP_OPO_TRACK_PREIMAGE parameter to U. This parameter causes Post to insert two rows to the target table for every change made to the tracked source table: one for the after image and one for the before image. The before image is composed of the key values plus the before values of the columns that were changed, unless the SP_OCT_USE_SUPP_KEYS parameter is used. When before images are enabled, the SHAREPLEX_SOURCE_OPERATION column values for the two records will be:
UPDATE BEFORE UPDATE AFTER NOTE: The before row will not include the before image of any LOB columns, because the redo log does not contain the before image of LOBs. You can override the global setting of SP_OPO_TRACK_PREIMAGE at the table level by using the set cdc preimage option of the target command. For more information about SP_OPO_TRACK_PREIMAGE and the target command, see the SharePlex Reference Guide.
Include all columns of an operation in the history
This option is valid for Oracle data only. To include the values of all table columns in each target history record, rather than only the changed columns, configure the following:
SharePlex 9.2.8 Administrator Guide 112

1. Turn on supplemental logging for all columns of the source tables that are being tracked. For example: Alter table emp ADD SUPPLEMENTAL LOG DATA (ALL) COLUMNS;
2. Set the SP_OCT_USE_SUPP_KEYS parameter to 1. 3. Set the SP_OCT_INCLUDE_UNCHANGED_COL to 1. NOTE: When both SP_OCT_USE_SUPP_KEYS and SP_OPO_TRACK_PREIMAGE are enabled, the before image includes all column values as they were before the change.
Disable change history of an operation type
To disable the change history of a DML operation type, set the SP_OPO_TRACK_OPERATIONS parameter to the appropriate value or values. Separate values with a slash (/). For example, to maintain change history only for inserts and updates, change the parameter to I/U.The default is I/U/D which sends all DML operation types to the history records.
Set rules and filters
You can use the set rule option of the target command to apply conditions on columns to control whether a change is applied to the target history table. For example, you can specify a rule that if column 1 and column 3 are changed, then apply the operation and discard any changes that apply to other columns. For instructions, see the SharePlex Reference Guide.
Include COMMITs
By default, the COMMIT record is not included in the history tables. To configure Post to insert a row for every COMMIT, set the SP_OPO_TRACK_COMMITS parameter to 1.
SharePlex 9.2.8 Administrator Guide 113

11
Configure a replication strategy
This chapter contains instructions for configuring SharePlex to support different replication objectives. Production implementations can vary widely from basic configurations with one source and target, to multiple instances of SharePlex with named queues,multiple targets, partitioned data, and more. It is difficult to foresee and document every possible way that an organization may want to deploy SharePlex. The goal of this documentation is to present instructions for setting up the basic deployment types in a way that is clear enough for you to be able to combine them and expand upon them to suit your needs. Additional deployment assistance is available through our Professional Services organization.
Contents
Configure replication to share or distribute data Configure replication to maintain a central datastore Configure peer-to-peer replication Configure replication through an intermediary system Configure replication to maintain high availability
Configure replication to share or distribute data
These instructions show you how to set up SharePlex for the purpose of sharing or distributing data from one source system to one or more target systems. This strategy supports business requirements such as the following:
l reporting to support real-time decision making l data sharing to support research and transparency requirements l data integration throughout an enterprise l customer service inquiries and other query-intensive applications l data auditing and archiving
SharePlex 9.2.8 Administrator Guide 114

Supported sources
Oracle
Supported targets
All
Capabilities
This replication strategy supports the following: l Replication to one or more target systems l Replication between databases on the same system l Replication between schemas in the same database (Oracle) l Identical or different source and target names l Use of vertically partitioned replication l Use of horizontally partitioned replication l Use of named export and post queues l Use of transformation
Requirements
l Prepare the system, install SharePlex, and configure database accounts according to the instructions in the SharePlex Installation Guide.
l No DML or DDL should be performed on the target tables except by SharePlex. Tables on the target system that are outside the replication configuration can have DML and DDL operations without affecting replication.
l If sequences are unnecessary on the target system, do not replicate them. It can slow down replication. Even if a sequence is used to generate keys in a source table, the sequence values are part of the key columns when the replicated rows are inserted on the target system. The sequence itself does not have to be replicated.
IMPORTANT! These instructions assume you have a full understanding of SharePlex configuration files. They use abbreviated representations of important syntax elements. For more information, see Configure data replication on page 57.
Conventions used in the syntax
In the configuration syntax in this topic, the placeholders represent the following:
SharePlex 9.2.8 Administrator Guide 115

l source_specification[n] is the fully qualified name of a source object (owner.object) or a wildcarded specification.
l target_specification[n] is the fully qualified name of a target object or a wildcarded specification. l host is the name of a system where SharePlex runs. Different systems are identified by appending a
letter to the names, like hostB. l db is a database specification. The database specification consists of either o. or r. prepended to the
Oracle SID, TNS alias, or database name, as appropriate for the connection type. A database identifier is not required if the target is JMS, Kafka, or a file.
IMPORTANT!
l Configure data replication on page 57.

Replicate within the local system
Replication on the same system supports the following configurations:
l Within one Oracle instance, replicate to different tables within the same schema or to the same table in different schemas.
l Replicate to from an Oracle instance to any SharePlex-supported target on the same system.
On the Windows platform, SharePlex does not support replication between Oracle databases that reside on the same system, but you can replicate to Open Target targets on the same system.

Configuration options

datasource_specification source_specification1 source_specification2

target_specification1 target_specification2

hostA[@db] hostA[@db]

Example

This example shows how you can replicate data to the same Oracle instance, to a different Oracle instance

(Unix and Linux only), and to different target types, all on the same local system.

Datasource:o.oraA

hr.emp

hr.emp2

hostA@o.oraA

hr.sal

hr.sal2

hostA@o.oraB

fin.*

hostA@r.mss

act.*

!file

hostA

Configuration when using SharePlex Manager
Replication from and to the same machine omits an Export process. However, SharePlex Manager expects an export queue to exist. If using this configuration with SharePlex Manager, you must explicitly configure an export queue as follows. The hostA* component in the routing map creates the export queue and an Export process, which sends the data to an Import process, then the post queue.

SharePlex 9.2.8 Administrator Guide 116

datasource_specification source_specification1 source_specification2

target_specification1 target_specification2

hostA*hostA[@db] hostA*hostA[@db]

Replicate to a remote target system

Configuration options

datasource_specification source_specification1 source_specification2

target_specification1 target_specification2

hostB[@db] hostB[@db]

Example

The last line in this example shows how you can replicate data to different target types on the same remote

target system.

Datasource:o.oraA

hr.emp

hr.emp2

hostB@o.oraB

hr.sal

hr.sal2

hostB@o.oraB

fin.*

!file

hostB

Replicate to multiple target systems
This topology is known as broadcast replication. It provides the flexibility to distribute different data to different target systems, or all of the data to all of the target systems, or any combination as needed. It assumes the source system can make a direct connection to all of the target systems. All routing is handled through one configuration file. For more information, see Configure replication through an intermediary system on page 143.

Configuration options

If the target specification is identical on all targets

If the target specification of a source object is identical on all target systems, you can use a compound routing map, rather than type a separate entry for each route. For more information, see Configure data replication on page 57.

datasource_specification

source_specification1

target_specification1

hostB[@db]+hostC[@db][+...]

source_specification2

target_specification2

hostC[@db]+hostD[@db][+...]

SharePlex 9.2.8 Administrator Guide 117

If the target specification is not identical on all targets
l When the target specification of a source object is different on some or all target systems, you must use a separate configuration entry to specify each one that is different.
l You can use a compound routing map for routes where the target specifications are identical.

datasource_specification source_specification1 source_specification1

target_specification1 target_specification2

hostB[@db] hostC[@db]

Example

NOTE: This example does not cover all possible source-target combinations. The last entry in this example

shows the use of horizontally partitioned replication to distribute different data from the sales.accounts table to

different regional databases.

Datasource:o.oraA

hr.emp

hr.emp2

hostB@o.oraB

hr.emp

hr."Emp_3"

hostC@r.mssB

hr.emp

!jms

hostX

cust.%

hostD@o.oraD+hostE@o.oraE

sales.accounts

!regions

Configure replication to maintain a central datastore
These instructions show you how to set up SharePlex for the purpose of consolidated replication: replicating from multiple source systems to one central target system. This strategy supports business requirements such as the following:
l Real-time reporting and data analysis l The accumulation of big data into a central datastore/mart or warehouse
Supported sources
Oracle
Supported targets
Oracle and Open Target
Capabilities
This replication strategy supports the following:

SharePlex 9.2.8 Administrator Guide 118

l Identical or different source and target names l Use of vertically partitioned replication l Use of horizontally partitioned replication l Use of named export and post queues
Requirements
l Prepare the system, install SharePlex, and configure database accounts according to the instructions in the SharePlex Installation Guide.
l No DML or DDL should be performed on the target tables except by SharePlex. Tables on the target system that are outside the replication configuration can have DML and DDL operations without affecting replication.
l Each source system must replicate a different set of data to the central target. If any source systems replicate the same data to the central target system, it is considered to be active-active replication. For more information, see Configure peer-to-peer replication on page 123.
l If sequences are unnecessary on the target system, do not replicate them. It can slow down replication. Even if a sequence is used to generate keys in a source table, the sequence values are part of the key columns when the replicated rows are inserted on the target system. The sequence itself does not have to be replicated.
Deployment options
You have two options for deploying SharePlex to replicate from many source systems to one target system.
l One instance of SharePlex processes all of the incoming data from all sources. For more information, see Deploy with one instance of SharePlex on the target system on page 119.
l Multiple instances of SharePlex each process the incoming data from a different source. For more information, see Deploy with multiple instances of SharePlex on the target system on page 120.
In either deployment, if any source system cannot make a direct connection to the target system, you can use cascading replication for that route to enable SharePlex to cascade the data an intermediary system that allows connection to the target. For more information, see Configure replication through an intermediary system on page 143. NOTE: The SharePlex compare and repair commands cannot be used in a cascading configuration.
Deploy with one instance of SharePlex on the target system
You can use one instance of SharePlex to process all incoming data on the target. For each source system, SharePlex creates an Import process on the central target system when replication starts. That, in turn, creates post queues and Post processes for each source-target replication stream, all controlled by one sp_cop process. You can control each source-target stream separately, but the post queues all share the same SharePlex variable-data directory on the target system. A deployment with a single variable-data directory has the following potential risks:
SharePlex 9.2.8 Administrator Guide 119

l An event that interrupts processing to and from the disk that contains the variable-data directory will affect all replication streams.
l Any cleanup utilities that you use will affect all of the replication streams, because the cleanup is performed across the entire variable-data directory.
l A purge config command that is issued on one source system also deletes the data that is replicated from the other source systems, because the purge affects the entire variable-data directory. The use of named post queues eliminates this risk, but adds complexity to the naming, monitoring and management of the SharePlex objects in the deployment.
To use this deployment l Install SharePlex in the normal manner, with one port number and one variable-data directory on each system (sources and target). l Make certain that when you install SharePlex, you create a database account for SharePlex for each installation. l IMPORTANT! Use the same port number for SharePlex on all systems.
Deploy with multiple instances of SharePlex on the target system
You can deploy multiple instances of SharePlex on the target, one for each source system. A SharePlex instance is composed of the following elements:
l A unique sp_cop process l A unique variable-data directory l A unique port number on which sp_cop runs l A unique database account that the processes of that instance use to interact with the database. By running multiple, distinct instances of SharePlex, you can isolate each source-target replication stream from the others. It enables you to: l Avoid contention problems that can occur if multiple processes must compete for access to the same
queues in a single variable-data directory.
SharePlex 9.2.8 Administrator Guide 120

l Purge one configuration, or clean up and resynchronize one replication stream, while allowing the others to continue processing data.
l Place the variable-data directories on separate disks so that problems with one disk do not affect the variable-data directories on the other disks.

To use this deployment Install on the target system first, if possible. This enables you to establish a port number for each variable-data directory, which you can then refer to when you set up SharePlex on the corresponding source system.
Steps on the target system
Select either of the setup options presented in Run multiple instances of SharePlex on page 43. These procedures will guide you through the steps to establish independent instances of SharePlex on the target. If you already installed SharePlex on the target, a variable-data directory, database account, and port number already exist. You can dedicate that SharePlex instance to one of the source systems, and then create additional instances on the target per those instructions.
Steps on the source systems
Install one instance of SharePlex on each source system, as directed in the SharePlex Installation Guide. Match the port numbers of those instances to the port numbers of their associated target variable-data directories. If you already installed SharePlex on the source systems, you can change the port numbers as needed. For more information, see Set the SharePlex port number on page 316.

Configuration

Create a configuration file on each source system that replicates the objects from that system to the central target. For more information about how to create a configuration file, see Configure data replication on page 57.
datasource_specification

source_specification where:

target_specification

central_host[@db]

SharePlex 9.2.8 Administrator Guide 121

l source_specification is the fully qualified name of a source object (owner.object) or a wildcarded specification.
l target_specification is the fully qualified name of a target object (owner.object) or a wildcarded specification.
l central_host is the target system.
l db is a database specification. The database specification consists of either o. or r. prepended to the Oracle SID, TNS alias, or database name, as appropriate for the connection type. A database identifier is not required if the target is JMS, Kafka or a file.

Example
This example shows data from datasource oraA on hostA and datasource oraB on hostB replicating to oraC on system hostC.

Data from hostA

Datasource:o.oraA

hr.*

fin.*

fin*

hostC@o.oraC hostC@o.oraC

Data from hostB
Datasource:o.oraA cust.* mfg.*

hr.* mfg.*

hostC@o.oraC hostC@o.oraC

Recommended target configuration
Each source system in a consolidated configuration sends a discrete data stream that flows to its own Post process on the target. You can assign a unique identifier of your choosing to each source system, and then configure the Post process to include that identifier in each insert or update that it posts on the target. By identifying rows in this manner, your environment is prepared to support the SharePlex compare and repair commands (which require a source ID) as well as any other work that may require the selection or identification of rows by their source. The compare and repair processes will use the source ID value to select only the rows that are valid for that source.
To configure each Post to write a source ID
1. Create or alter the target table to include a column named SHAREPLEX_SOURCE_ID. This is the column that will contain the source ID value. NOTE: You can change this name by running the target command with the set metadata option, before continuing further. See the SharePlex Reference Guide for more information.
2. Choose a unique ID for each of the source systems. Any single alphanumeric string is permitted.
3. On the target, run sp_ctrl for each Post process.

SharePlex 9.2.8 Administrator Guide 122

4. For each Post process, issue the target command with the set source option. This command sets the source ID that will be posted by that Post process. The following example shows the command for three Post processes: sp_ctrl> target sys4 queue Q1 set source east sp_ctrl> target sys4 queue Q2 set source central sp_ctrl> target sys4 queue Q3 set source west
Configure peer-to-peer replication
These instructions show you how to set up SharePlex for the purpose of maintaining multiple databases, where applications on each system can make changes to the same data, while SharePlex keeps all of the data synchronized through replication. This is known as peer-to-peer, or active-active, replication. In this strategy, the databases are usually mirror images of each other, with all objects existing in their entirety on all systems. Although similar in benefit to a high-availability strategy, the difference between the two is that peer-to-peer allows concurrent changes to the same data, while high availability permits changes to the secondary database only in the event that the primary database goes offline. This strategy supports the following business requirements:
l Maintain the availability of mission-critical data by operating multiple instances in different locations. l Distribute heavy online transaction processing application (OLTP) loads among multiple points
of access. l Limit direct access to an important database, while still enabling users outside a firewall to make
updates to their own copies of the data.
An example of peer-to-peer replication is an e-commerce company with three identical databases. When users access the application from a web browser, the web server connects to any of those databases sequentially in a round-robin configuration. If one of the databases is unavailable, the server connects to a different available database server. Thus the configuration serves not only as a failover resource, but also as a means of distributing the load evenly among all the peers. Should the company need to produce business reports, user access to one of the databases can be stopped temporarily, and that database can be used to run the reports.
NOTE: Data changes made in peer-to-peer replication are prevented from looping back from one machine to another because Capture ignores transactions performed on the local system by the Post process. Peer-to-peer replication is not appropriate for all replication environments. It requires a major commitment to database design that might not be practical when packaged applications are in use. It also requires the development of conflict resolution routines to prioritize which transaction SharePlex posts to any given database if there are multiple changes to the same data at or near the same time.
Supported source-target combinations
Oracle to Oracle
Capabilities
This replication strategy supports the following:
SharePlex 9.2.8 Administrator Guide 123

l Use of named export and post queues This replication strategy does not support the following:
l Replication of LOBs. If tables with LOBs are included in replication the LOBs will be bypassed by conflict resolution, causing the potential for data to be out of synchronization.
l Column mapping and partitioned replication is not appropriate in a peer-to-peer configuration.
Requirements
l Every table involved in peer-to-peer replication must have a primary key or a unique key with no nullable columns. Each key must uniquely identify the same owner.table.row among all of the databases that will be involved in replication, and the logging of the key columns must be enabled in the database. See additional requirements in this topic.
l Prepare the system, install SharePlex, and configure database accounts according to the instructions in the SharePlex Installation Guide.
l Enable supplemental logging for primary keys, unique keys, and foreign keys on all databases in the peer-to-peer configuration.
l Enable archive logging on all systems. l You must understand the concepts of synchronization. For more information, see Understand the
concept of synchronization on page 30.
Overview
In peer-to-peer replication, DML changes are allowed on copies of the same tables in different databases, usually on different systems, while SharePlex keeps them all current through replication. If a record is changed in more than one database at (or near) the same time, conflicts can occur, and conflict-resolution logic must be applied to resolve the discrepancy.
What is a conflict?
A conflict is defined as an out-of-sync condition -- source and target tables are not identical. You can predict that out-of-sync (conflict) situations will occur when a DML statement constructed by SharePlex fails to execute on a row in the target table because of the following reasons:
l Post applies a replicated INSERT but a row with the same key already exists in the target. Post applies the following logic: o If all of the current values in the target row are the same as the INSERT values, Post considers the rows to be in-sync and discards the operation. o If any of the values are different from those of the INSERT, Post considers this an out-ofsync condition. NOTE: You can configure Post so that it does not consider non-key values when posting an INSERT. See the SP_OPO_SUPPRESSED_OOS parameter in the SharePlex Reference Guide.
SharePlex 9.2.8 Administrator Guide 124

What causes a conflict in peer-to-peer replication?

To understand how SharePlex determines a conflict, refer to the following examples of normal and conflict situations. In the examples, three systems (SysA, SysB and SysC) are used.

The following tables are used in the example:

Scott.employee_source

jane.employee_backup

The column names and definitions are identical:

EmpNo number(4) not null, SocSec number(11) not null, EmpName char(30),

Job Salary Dept

char(10), number(7,2), number(2)

The values for both tables in a synchronized state are:

EmpNo (key) 1 2 3 4

SocSec 111-22-3333 111-33-4444 000-11-2222 000-44-7777

EmpName Mary Smith John Doe Mike Jones Dave Brown

Job Manager Data Entry Assistant Manager

Salary 50000 20000 30000 45000

Dept 1 2 3 3

Example of peer-to-peer replication without a conflict
1. At 9:00 in the morning, UserA on SysA changes the value of the Dept column to 2, where EmpNo is 1. SharePlex replicates that change to SysB and SysC, and both databases remain synchronized.
2. At 9:30 that same morning, UserB on SysB changes the value of Dept to 3, where EmpNo is 1. SharePlex replicates that change to SysA and SysC, and the databases are still synchronized.
Now the row looks like this:

SharePlex 9.2.8 Administrator Guide 125

EmpNo (key) 1

SocSec 111-22-3333

EmpName Mary Smith

Job Manager

Salary 50000

Dept 3

Example of peer-to-peer replication with an UPDATE conflict
1. At 11:00 in the morning, UserA on SysA updates the value of Dept to 1, where EmpNo is 1. At 11:02 that morning, the network fails. Captured changes rest in the export queues on all systems.
2. At 11:05 that morning, before the network is restored, UserB on SysB updates the value of Dept to 2, where EmpNo is 1. The network is restored at 11:10 that morning. Replication data transmission resumes.
3. When SharePlex attempts to post the change from UserA to the database on SysB, it expects the value in the Dept column to be 3 (the pre-image), but the value is 2 because of the change made by UserB. Because the pre-images do not match, SharePlex generates an out-of-sync error.
4. When SharePlex attempts to post the change from UserB to SysA, it expects the value of the column to be 3, but the value is 1 because of the change made by UserA. SharePlex generates an out-of-sync error.
5. When SharePlex attempts to post the changes made by UserA and User B to the database on SysC, both of those statements fail because the pre-images do not match. SharePlex generates an out-of-sync error.
NOTE: For more information, see Appendix A: Peer-To-Peer Diagram on page 331.

Deployment
To deploy peer-to-peer replication, perform the following tasks:
1. Evaluate the data for suitability to a peer-to-peer environment. Make any recommended alterations. For more information, see Evaluate the data on page 126.
2. Configure SharePlex so that data from each system replicates to all other systems in the peer-to-peer environment. For more information, see Configure replication on page 130.
3. Develop conflict resolution routines that provide rules for how Post handles conflicts. For more information, see Develop conflict resolution routines on page 131.
4. Create a conflict resolution file. SharePlex refers to this file to determine the correct procedure to use when a conflict occurs. For more information, see List the routines in conflict_resolution.SID on page 140.

Evaluate the data
To successfully deploy SharePlex in a peer-to-peer configuration, you must be able to: l isolate keys l prevent changes to keys l control sequence generation l control trigger usage l eliminate cascading deletes

SharePlex 9.2.8 Administrator Guide 126

l designate a trusted host l define priorities
These requirements must be considered during the architectural phase of the project, because they demand cooperation with the application. Consequently, many packaged applications are not suitable for a peer-to-peer deployment because they were not created within those guidelines. Following are more detailed explanations of each of the requirements.
Keys
The only acceptable key in peer-to-peer replication is a primary key. If a table has no primary key but has a unique, not-NULL key, you can convert that key to a primary key. LONG columns cannot be part of the key. If you cannot assign a primary key, and you know all rows are unique, you can create a unique index on all tables. The primary key must be unique among all of the databases in the peer-to-peer replication network, meaning:
l it must use the same column(s) in each corresponding table in all databases. l key columns for corresponding rows must have the same values.
The primary key must be created to contain enough information about a row so there can be no question about the uniqueness of that row, and so that there will be a conflict if a replicated operation would violate uniqueness. The primary key value cannot be changed. Supplemental logging of primary and unique keys must be enabled in the database. Using only a sequence as the primary key probably will not suffice for peer-to-peer replication. For example, suppose the sample table uses sequences to generate values for key column EmpNo. Suppose UserA gets the next sequence value on SysA and inserts a row for "Jane Wilson." UserB gets the next sequence value on SysB and also inserts a row for "Jane Wilson." Even if the sequence numbers are different on each system, so there are no unique key violations on the replicated INSERTs, data integrity is compromised because there are now two entries for "Jane Wilson" in the databases, each with a different key. Subsequent UPDATEs will fail. The solution is to include other unique columns in the key, so that there is enough information to ensure uniqueness and ensure a conflict that can then be resolved through resolution logic.
Sequences
SharePlex does not support peer-to-peer replication of sequences. If the application uses sequences to generate all or part of a key, there must be no chance for the same range of values to be generated on any other system in the peer-to-peer configuration. You can use a sequence server or you can maintain sequences separately on each server and make sure you partition a unique range to each one. Quest recommends using n+1 sequence generation (where n = the number of systems in replication). Depending on the type of application, you can add a location identifier such as the system name to the sequence value in the primary key to enforce uniqueness.
Triggers
DML changes resulting from triggers firing on a source system enter the redo log and are replicated to the target system by SharePlex. If the same triggers fire on the target system, they return out-of-sync errors. To handle triggers in a peer-to-peer configuration, you can do either of the following:
SharePlex 9.2.8 Administrator Guide 127

l Disable the triggers. l Keep them enabled, but alter them to ignore the SharePlex user on all instances in the peer-to-peer
configuration. SharePlex provides the sp_add_trigger.sql script for this purpose. This script puts a WHEN clause into the procedural statement of the trigger that tells it to ignore the Post process. For more information about this script, see the SharePlex Reference Guide.
ON DELETE CASCADE constraints
ON DELETE CASCADE constraints can remain enabled on all instances in the peer-to-peer replication configuration, but you must set the following parameters to direct Post to ignore those constraints:
l SP_OPO_DEPENDENCY_CHECK parameter to 2 l SP_OCT_REDUCED_KEY parameter to 0 l SP_OPO_REDUCED_KEY parameter to 0 (although in other replication scenarios this parameter can
be set to different levels, it must be set to 0 in a peer-to-peer configuration)
Balance values maintained by using UPDATEs
Applications that use UPDATE statements to record changes in quantity, such as inventory or account balances, pose a challenge for peer-to-peer replication. The following example of an online bookseller explains the reason why. The bookseller's Inventory table contains the following columns.
Book_ID (primary key) Quantity Suppose the following sequence of events takes place:
1. A customer buys a book through the database on one server. The quantity on hand reduces from 100 books to 99. SharePlex replicates that UPDATE statement to the other server. (UPDATE inventory SET quantity = 99 WHERE book_ID = 51295).
2. Before the original UPDATE arrives, another customer buys two copies of the same book on another server (UPDATE inventory SET quantity = 98 WHERE book_ID = 51295), and the quantity on that server reduces from 100 books to 98.
3. When the Post process attempts to post the first transaction, it determines that the pre-image (100 books) on the first system does not match the expected value on the second system (it is now 98 as a result of the second transaction). Post returns an out-of-sync error.
A conflict resolution procedure could be written, but how would the correct value be determined? The correct value in both databases after the two transactions should be 97 books, but no matter which of the two UPDATE statements is accepted, the result is incorrect. For this reason, peer-to-peer replication is not recommended for applications maintaining account or inventory balances using UPDATEs. If you can use a debit/credit method of maintaining balances, you can use INSERT statements (INSERT into inventory values "n",...) instead of UPDATE statements. INSERT statements do not require a before-and-after comparison with a WHERE clause, as do UPDATE statements. If your application must use UPDATE statements, you can write a conflict resolution procedure to determine the absolute (or net) change resulting from different UPDATE statements on different systems. For example, in the case of the preceding online bookseller example, when the first customer's purchase is replicated to the second system, the following conflict resolution procedure fires:
if existing_row.quantity <> old.quantity then old.quantity - new.quantity = quantity_change; update existing_row set quantity = existing_row.quantity - quantity_change;
SharePlex 9.2.8 Administrator Guide 128

The conflict resolution logic tells SharePlex that, if the quantity value of the existing row in the target database (98) does not equal the old value (pre-image of 100), then subtract the new value (the replicated value of 99) from the pre-image to get the net change (1). Then, issue an UPDATE statement that sets the Quantity column to 98-1, which equals 97. When the second user's change is replicated to the first system, the same conflict resolution procedure fires. In this case, the net change (pre-image of 100 minus the new value of 98) is 2. The UPDATE statement on this system also results in a value of 97, which is 99 (the existing row value after the first customer's purchase) minus the net change of 2. The result of this procedure's logic is that the Quantity columns on each system are updated to 97 books, the net effect of selling three books. The following example illustrates this concept using an account balance within a financial record:
account_number (primary key) balance
1. Suppose a row (an account) in the example table has a balance of $1500 on SysA. CustomerA makes a deposit of $500 on that system. The application uses an UPDATE statement to change the balance to $2000. The change is replicated to SysB as an UPDATE statement (such as UPDATE...SET balance=$2000 WHERE account_number=51295).
2. Before the change arrives, CustomerA's spouse makes a withdrawal of $250 on SysB, and the application updates the database on that system to $1250. When CustomerA's transaction arrives from SysA and Post attempts to post it to SysB, there is a conflict, since the pre-image from the source system is $1500, but the pre-image on the target is $1250 because of the spouse's transaction -- not a match.
You can write a conflict resolution routine to accommodate this kind of transaction by calculating the absolute (or net) change in the account, then using that value to resolve the conflict. For example:
if existing_row.balance <> old.balance then old.balance - new.balance = balance_change; update existing_row set balance = existing_row.balance - balance_change; The result of this procedure would be to update the account balance to $1750, the net effect of depositing $500 and withdrawing $250. On SysB, the routine directs SharePlex to subtract the new (replicated) balance of 2000 from the old balance of 1500 for a net change of -500. The UPDATE statement sets the balance value to 1250 (-500) = 1750, the correct value. On SysA, the replicated value of 1250 is subtracted from the old balance of 1500 to get the net change of 250. The UPDATE statement subtracts that value from the existing balance of 2000 to get the correct value of 1750.
Priority
When the environment is established to avoid or resolve conflict when SharePlex searches for the correct row to change, the only remaining conflict potential is on fact data -- which change to accept when the values for the same column in the same row differ on two or more systems. For this, your application must be able to accept the addition of timestamp and source columns, with source being the name of the local system for the table. The following explains how those columns play a vital role when using a conflict resolution routine to establish priority.
Trusted source
You must assign a particular database or server to be the prevailing, or trusted, source for two reasons:
l The conflict resolution routine has the potential to get quite large and complex the more systems you have. There are bound to be failures that require resynchronization at some point. One of the systems in the configuration must be considered the true source from which all other systems will be resynchronized if necessary.
SharePlex 9.2.8 Administrator Guide 129

l You can write your conflict resolution routines so that operations from the trusted source system take priority over conflicting operations from other systems. For example, changes on the server at corporate headquarters could take priority over the same changes made by a branch office.
Timestamp It is recommended that you include a timestamp column in the tables and assign priority in the conflict resolution routine to the earliest or latest timestamp. However, the timestamp must not be part of a key, or it will cause conflicts. SharePlex cannot locate rows if a key value changes -- and the key value will change if one of the columns is a timestamp. For timestamp priority to work, you must make sure all of the servers involved agree on the date and time. Tables on servers in different time zones can use Greenwich Mean Time (GMT). The default date format for SharePlex conflict resolution is MMDDYYYY HH24MISS. Tables with default dates must use that format, or conflict resolution will return errors. Before creating a table with a default date, use the following command to change the date format in SQL*Plus.
ALTER SESSION SET nls_date_format = 'MMDDYYYYHH24MISS'
Configure replication
The configuration files on the systems in a peer-to-peer configuration are identical with the exception of the datasource specification and the routing.
Conventions used in the syntax
In the configuration syntax in this topic, the placeholders represent the following items in the environment. This documentation assumes three systems, but there can be more.
l hostA is the first system. l hostB is the second system. l hostC is the third system. l ownerA.object is the fully qualified name of an object on hostA or a wildcarded specification. l ownerB.object is the fully qualified name of an object on hostB or a wildcarded specification. l ownerC.object is the fully qualified name of an object on hostC or a wildcarded specification. l oraA is the Oracle instance on hostA. l oraB is the Oracle instance on hostB. l oraC is the Oracle instance on hostC. IMPORTANT! l See Configure data replication on page 57 for more information about the components of a
configuration file.
SharePlex 9.2.8 Administrator Guide 130

Configuration on hostA

Datasource:o.oraA

ownerA.object

ownerB.object

hostB@o.oraB

ownerA.object ownerA.object

ownerB.object ownerC.object

hostB@o.oraB hostC@o.oraC

ownerA.object

ownerC.object

hostC@o.oraC

Note: If all owner names and table names are the same on all systems, you can use a compound routing map for each of these configuration files. For example, the compound routing for replication from hostA is as follows:

Datasource:o.oraA

owner.object

hostB@o.oraB+hostC@o.oraC

Configuration on hostB

Datasource:o.oraB ownerB.object ownerB.object ownerB.object ownerB.object

ownerA.object ownerA.object ownerC.object ownerC.object

hostA@o.oraA hostA@o.oraA hostC@o.oraC hostC@o.oraC

Configuration on hostC

Datasource:o.oraC ownerC.object ownerC.object ownerC.object ownerC.object

ownerA.object ownerA.object ownerB.object ownerB.object

hostA@o.oraA hostA@o.oraA hostB@o.oraB hostB@o.oraB

Example
Datasource:o.oraA hr.emp hr.sal cust.%

hr.emp hr.sal cust.%

hostB@o.oraB hostB@o.oraB hostB@o.oraB

Develop conflict resolution routines
To create conflict resolution routines, you write PL/SQL procedures that direct the action of SharePlex when a conflict occurs. Business rules vary widely from company to company, so it is impossible to create a standard set of conflict resolution rules and syntax that apply in every situation. You will probably need to write your own routines. It is good practice to write more than one procedure, such as making site or system priority the primary

SharePlex 9.2.8 Administrator Guide 131

routine and timestamp a secondary routine. SharePlex invokes one routine after another until one succeeds or there are no more procedures available. SharePlex provides the following tools that can be used as a basis for your routines:
l A generic PL/SQL interface that you can use to write basic routines based on DML operation types. For more information, see How to write a routine using the SharePlex generic interface on page 132.
l Packaged routines that perform basic conflict resolution based on a key or column value, which can be used as a backup measure in case the custom routines fail. For more information, see How to use the SharePlex prepared routines on page 137.
IMPORTANT!
l This documentation provides guidelines, examples and templates to assist you, but do not use them as your own routines.
l Test your conflict resolution routines before you put them into production to make sure they work as intended, and to make sure that one routine does not counteract another one.
l By default, SharePlex does not stop for out-of-sync conditions. If failed attempts at conflict resolution are not resolved, the databases can become more and more out of synchronization. Check the Event Log frequently to monitor for out-of-sync warnings by using the show log command in sp_ctrl. See the SharePlex Reference Guide for more information about show log and other SharePlex commands.
l Updates are occasionally made to the conflict resolution logic, so refer to the Release Notes and documentation for your version of SharePlex for any additional information that augments or supersedes these instructions.
How to write a routine using the SharePlex generic interface
SharePlex provides a generic conflict resolution PL/SQL package that can be used to pass information to and from the procedural routines that you write. NOTE: Custom routines are supported for Oracle to Oracle source-target combinations only. Before you get started, understand the following guidelines:
l The same PL/SQL package is used for both generic conflict resolution and transformation (its name is sp_cr). Use either generic conflict resolution or transformation for a table, but not both. Transformed tables cannot be compared by SharePlex and conflict resolution cannot succeed. If both are used, SharePlex only calls the transformation routine. If appropriate, you can use generic conflict resolution and transformation for different tables in the same configuration. For more information, see Configure data transformation on page 166.
l Conflict resolution cannot be used for DDL changes. l Any table to be accessed through PL/SQL for conflict resolution requires implicitly granted privileges
from the owner of the object to SharePlex. l Conflict resolution does not support changes to LONG or LOB columns.
NOTE: If you ran the SharePlex conflict resolution demonstration in the SharePlex Installation and Setup Guide, you can view a sample generic conflict resolution routine by viewing the od_employee_gen routine that was installed in the database used for the demonstration.
SharePlex 9.2.8 Administrator Guide 132

Procedure interface
Follow this template to create your procedure. (table_info in outsplex.sp_cr.row_typ, col_values insplex.sp_cr.col_def_tabtyp)
where:
l splex is the SharePlex schema. l sp_cr is the name of the package that contains the PL/SQL record and table structures. l row_typ is the name of the PL/SQL record that passes in/out variables (see Package definition). l col_def_type is the name of the PL/SQL table that stores column information (see col_def_type table).
Package definition
SharePlex defines PL/SQL record and table structures in a public package named sp_cr in the SharePlex database schema. The package uses the following parameters.
"CREATE or REPLACE PACKAGE %s.sp_cr AS "TYPE row_typ IS RECORD "(src_host VARCHAR2(32), "src_ora_sid VARCHAR2(32), "src_ora_time VARCHAR2(20), "source_rowid VARCHAR2(20), "target_rowid VARCHAR2(20), "statement_type VARCHAR2(6), "source_table VARCHAR2(78), "target_table VARCHAR2(78), \n\t" "oracle_err NUMBER, "status NUMBER, "action NUMBER, "reporting NUMBER ");
IN variables
For each row operation that causes a conflict, SharePlex passes this metadata information to your procedure.

Variable src_host
src_ora_sid src_ora_time source_rowid

Description
The name of the source system (where the operation occurred). It is case-sensitive and is passed using the same case as on the source system, for example SysA. If there are named post queues in use on the target system, this variable consists of the name of the post queue, for example postq1.
The ORACLE_SID of the source database. It is case-sensitive and is passed in the same case as in the oratab file, Windows Registry or V$PARAMETER table.
The timestamp of the change record in the source redo log.
The row ID of the source row. It is passed as a literal within single quotes, for example `123456'.

SharePlex 9.2.8 Administrator Guide 133

Variable target_rowid
statement_type source_table
target_table oracle_err

Description
The row ID of the corresponding row in the target database. SharePlex obtains the row ID by querying the target database. It is passed as a literal within single quotes, for example `123456'. If the row cannot be found using the PRIMARY key, the value is NULL.
A letter, either I, U or D, indicating whether the operation is an INSERT, UPDATE or DELETE statement.
The owner and name of the source table, expressed as owner.table. This value is case-sensitive and matches the way the table is named in the database. It is passed within double quotes, for example "scott"."emp."
The owner and name of the target table, expressed as owner.table. This value is case-sensitive and matches the way the table is named in the database. It is passed within double quotes, for example "scott"."emp."
This is different, depending on whether the procedure is being used for conflict resolution or transformation. Transformation: SharePlex passes a value of 0 for this variable. This variable is only used for conflict resolution. Conflict resolution: The Oracle error number that caused the conflict.

OUT variables These variables direct the action of SharePlex based on whether the procedure succeeded or failed).

Variable status
action

Description
Defines whether or not the procedure succeeded. You must specify a value for this parameter.
l A value of 0 implies successful execution. It acts differently, depending on whether the procedure is used for conflict resolution or transformation. Transformation: Post does not write any SQL. SharePlex does not write any error messages to the Event Log when transformation succeeds. It continues its processing by reading the next replicated operation in the post queue.
l Conflict resolution: A value of 0 directs SharePlex to proceed with the SQL statement. SharePlex does not write any log entries to the Event Log when conflict resolution succeeds.
l A value of 1 implies that the procedure was unsuccessful. In this case, the action SharePlex takes depends on what you specified as the action variable.
l (Transformation only) A value of 7 implies unsuccessful execution and instructs the Post process to stop.
Defines the action that you want SharePlex to take. This is different, depending on whether the procedure is used for transformation or

SharePlex 9.2.8 Administrator Guide 134

Variable reporting

Description
conflict resolution. Transformation: You must specify a value of 0 for this parameter, which directs SharePlex NOT to post the SQL statement. Your transformation routine is responsible for posting the results of the transformation either to the target table or another table. The outcome of this action depends on what you specify for the reporting variable Conflict resolution: Specifies the action to take as a result of an unsuccessful conflict resolution procedure. You must specify a value for this parameter.
l A value of 0 directs SharePlex NOT to post the SQL statement. The outcome of this action depends on what you specify for the reporting variable. In addition, it directs SharePlex to try the next conflict resolution procedure that you listed in the conflict resolution file, if one exists.
l The value of 1 is reserved for internal SharePlex use. Do not use it.
l A value of 2 directs SharePlex to try the next conflict resolution procedure that you listed in the conflict resolution file, if one exists.
Determines how SharePlex reports unsuccessful procedural results. You must specify a value for this parameter.
l A value of 0 directs SharePlex NOT to report an error or write the failed SQL statement to the SID_errlog.sql log.
l Values 1 and 2 are reserved for internal SharePlex use. Do not use them.
l A value of 3 directs SharePlex to write the failed SQL statement to the SID_errlog.sql log and report an error to the Event Log.

col_def_type table

SharePlex creates a col_def_tabtyp PL/SQL table for each replicated operation. This table stores column information. It is different depending on whether the procedure is used for transformation or conflict resolution.

l Transformation: For each row operation, SharePlex writes column information to col_def_type.
l Conflict resolution: For each row operation that causes a conflict, SharePlex writes column information to col_def_tabtyp.

All fields are passed by SharePlex to your routine, although not all will have values if SharePlex cannot locate the row. Following is the data type that is used to populate the col_def_tabtyp table.

type col_def_typ is record

(column_name user_tab_columns.column_name%type

,data type

user_tab_columns.data type%type

SharePlex 9.2.8 Administrator Guide 135

,is_key boolean

,is_changed

boolean

,old_value

varchar2(32764)

,new_value

varchar2(32764)

,current_value varchar2(32764)

);

type col_def_tabtyp is table of col_def_typ

Description of col_def_tabtyp

Column column_name data type is_key is_changed
old_value
new_value current_value

Description
Tells your procedure the name of the column that was replicated from the source table, for example emp_last_name. This value is not casesensitive.
Tells your procedure the data type of the data in the replicated column, for example VARCHAR2. This value is always in capital letters.
Tells your procedure whether or not the column is a key column. If it is a key column, SharePlex passes a value of TRUE. If the column is not part of a key, SharePlex passes a value of FALSE.
Tells your procedure whether or not the column value has changed. If it is changed, SharePlex passes a value of TRUE. If the column is not changed, SharePlex passes a value of FALSE.
l For INSERTs, is_changed is TRUE for non-NULL values, because none of the columns existed in the database. If a NULL value is inserted, is_changed is FALSE.
l For UPDATEs, is_changed is TRUE for non-key columns. For key columns, is_changed normally is FALSE, but SharePlex will pass a value for a changed key column. Conflict resolution only: If a key value also was changed on the target system, SharePlex will not be able to locate the correct row, and conflict resolution could fail.
l For DELETEs, is_changed is always FALSE, because SharePlex replicates only the key values for a DELETE statement.
Tells your procedure the old value of the replicated column, before it was changed on the source system. This column is NULL for INSERTs, because the row did not exist in the target database before the INSERT. Conflict resolution only: This is the pre-image against which SharePlex compared the source and target columns as part of its synchronization check for UPDATEs and DELETEs. If the old value passed by SharePlex does not match the current_value value obtained from the target row, then there is a conflict.
Tells your procedure the new value of the replicated column, as changed on the source system.
Tells your procedure the current value of the column in the target table. If SharePlex cannot locate the target row, the value is NULL.

SharePlex 9.2.8 Administrator Guide 136

Example entries in col_def_tabtyp table per operation type The following tables illustrate the possible outcomes of each type of operation.

INSERT operation

column_name C1 C2 C3

is_changed TRUE TRUE FALSE

old_value NULL NULL NULL

new_value bind bind NULL

current_value1 is_key

NULL NULL NULL

FALSE TRUE TRUE | FALSE

1 When an INSERT fails, it is because a row with the same PRIMARY key already exists in the target database. SharePlex does not return the current value for INSERTs.

UPDATE operation

column_name C1 C2 C3

is_changed TRUE FALSE TRUE

old_value bind bind bind

new_value bind NULL bind

current_value1, 2 NULL | target_value NULL | target_value NULL | target_value

is_key FALSE TRUE TRUE

1 (Conflict resolution) When an UPDATE fails, it is because SharePlex cannot find the row by using the PRIMARY key and the pre-image. If the row cannot be found, SharePlex searches for the row by using only the PRIMARY key. If SharePlex finds the row, it returns the current value for the key column as well as the changed columns. If SharePlex cannot find the row by using just the PRIMARY key, then SharePlex returns a NULL. 2 (Transformation) For an UPDATE, SharePlex cannot locate a row using the PRIMARY key and the preimages, because the pre-images are different due to transformation. As an alternative, it searches for the row using just the PRIMARY key. If it finds it, SharePlex returns the current value for the key column as well as the changed columns. If it cannot locate the row using just the PRIMARY key, then current_value is NULL

DELETE operation

column_name C1

is_changed FALSE

old_value bind

new_value NULL

current_value1 NULL

is_key TRUE

1 When a DELETE fails, it is because SharePlex could not find the row by using the PRIMARY key. Therefore, SharePlex returns a NULL.

How to use the SharePlex prepared routines
SharePlex provides optional prepared routines for use in conjunction with custom routines. These options can be used with basic and generic conflict resolution formats. There are no limitations on column types. Supplemental logging of primary and unique keys must be enabled in the database.

SharePlex 9.2.8 Administrator Guide 137

Considerations
Review the following considerations before implementing SharePlex prepared routines.

!UpdateUsingKeyOnly

This routine works for UPDATE operations. It provides conflict resolution that relies solely on the key value of the changed row.

NOTE: This routine can be used only with an Oracle to Oracle source-target combination.

Supported source-target: This routine can be used only with Oracle to Oracle source-target combinations.

Normally, when SharePlex builds a SQL statement to post data, the WHERE clause uses both the key and the pre-image of the columns that changed to ensure synchronization. The !UpdateUsingKeyOnly routine directs SharePlex to post the data even though the pre-image values do not match, assuming the keys match.

If appropriate, this routine can be used as the sole routine for UPDATEs, but with the understanding that it does not include logic that assigns priority, such as system or time priority, in case of multiple concurrent UPDATEs. To avoid out-of-sync errors, Quest recommends using !UpdateUsingKeyOnly in conjunction with other, more specific routines, relying on !UpdateUsingKeyOnly as a final option if the custom routines fail.

IMPORTANT: !UpdateUsingKeyOnly must be the last entry in the list of routines, thus assigning it last priority.

In the following example, when there is a conflict for owner.table1 during an UPDATE, SharePlex calls the two custom routines first (in order of priority) and then calls the !UpdateUsingKeyOnly routine.

owner.table1 owner.table1

u owner.procedure_up_A u owner.procedure_up_B

owner.table1

u !UpdateUsingKeyOnly

The !UpdateUsingKeyOnly name is case sensitive. It must be typed exactly as shown in these instructions, with no spaces between words. Do not list an owner name with this routine in the configuration file. For more information, see List the routines in conflict_resolution.SID on page 140.

For INSERT and DELETE operations, custom logic must be used.

!HostPriority
This prepared conflict resolution routine works for INSERT, UPDATE, and DELETE operations. It provides hostbased conflict resolution by assigning priority to the row change that originated on the trusted source system. To define the trusted source, set the SP_OPO_TRUSTED_SOURCE or SP_OPX_TRUSTED_SOURCE parameter to the name of the source system.
Resolution logic

Operation Resolution Action

INSERT

If the source is the one specified with SP_OPO_TRUSTED_SOURCE or SP_OPX_TRUSTED_ SOURCE, convert the INSERT to an UPDATE and overwrite the existing row. Otherwise, discard the change record and do nothing to the target row.

UPDATE

(Oracle only) If the source is the one specified with SP_OPO_TRUSTED_SOURCE, overwrite the existing row using an UPDATE and use only the key columns in the WHERE clause. Otherwise, discard the change record and do nothing to the target row.

DELETE Ignore the out-of-sync error and do nothing to the target row.

SharePlex 9.2.8 Administrator Guide 138

Syntax in conflict resolution file owner.table {I | U | D} !HostPriority For more information, see List the routines in conflict_resolution.SID on page 140.

!MostRecentRecord
This prepared routine works for INSERT, UPDATE, and DELETE operations. It provides time-based conflict resolution by assigning priority to the most recent row change, as determined by a timestamp. NOTE: This routine can be used only with an Oracle to Oracle source-target combination. To capture the timestamp, tables using this routine must have a non-NULL timestamp column that is updated with every INSERT and UPDATE on the table. If the timestamp column in the DML, or in the existing row, is NULL, this routine cannot resolve the conflict. This routine requires the SP_OCT_REDUCED_KEY parameter to be set to 0 on the source system, so that all of the pre-image values of UPDATES are available to the Post process.

Resolution logic

Operation Resolution Action

INSERT and UPDATE

l If the timestamp of the replicated operation is greater than the timestamp of the row in the target, overwrite the existing row using an UPDATE and use only the key columns in the WHERE clause.
l If the timestamp of the replicated operation is less than or equal to the timestamp of the row in the target, discard the change record and do nothing to the target row.

DELETE Ignore the conflict (out-of-sync message).

Syntax in conflict resolution file owner.table {I | U | D} !MostRecentRecord(col_name)
Where col_name is the timestamp column to be used by the routine. See List the routines in conflict_resolution.SID on page 140.

!LeastRecentRecord
This prepared routine works for INSERT, UPDATE, and DELETE operations. It provides time-based conflict resolution by assigning priority to the least recent row change, as determined by a timestamp. NOTE: This routine can be used only with an Oracle to Oracle source-target combination. To capture the timestamp, tables using this routine must have a non-NULL timestamp column that is updated with every INSERT and UPDATE on the table. If the timestamp column in the DML, or in the existing row, is NULL, this routine cannot resolve the conflict.

SharePlex 9.2.8 Administrator Guide 139

Resolution logic

Operation Resolution Action

INSERT and UPDATE

l If the value of the timestamp column of the replicated operation is greater than or equal tothe timestamp column of the row in the target, discard the replicated operation and do nothing to the target row.
l If the timestamp column of the replicated operation is less than the timestamp column of the row in the target, overwrite the existing row using an UPDATE and use only the key columns in the WHERE clause.

DELETE Ignore the conflict (out-of-sync message).

Syntax in conflict resolution file owner.table {I | U | D} !LeastRecentRecord(col_name)
Where col_name is the timestamp column to be used by the routine. See List the routines in conflict_resolution.SID on page 140.

List the routines in conflict_resolution.SID
After you create the conflict resolution procedure(s), construct the conflict resolution file. This file tells SharePlex which procedures to use for which objects and operation types, and in which order.

Where to find the conflict resolution file
A blank conflict_resolution.SID file, where SID is the ORACLE_SID of the target instance, was included in the data sub-directory of the SharePlex variable-data directory when SharePlex was installed. Use the file on the target system. If this file does not exist, you can create one in ASCII format in an ASCII text editor. It must be named conflict_ resolution.SID, where SID is the ORACLE_SID of the target instance. Note: the SID is case-sensitive. IMPORTANT! There can be only one conflict_resolution.SID file per active configuration.

How to make entries in the conflict resolution file

Use the following template to link a procedure to one or more objects and operation types.

owner.object

{i | u | d | iud}

owner.procedure

where:

l owner.object is the owner and name of a target object, or a wildcarded entry. (See Syntax rules on page 141)
l i| u | d is the type of operation that creates the conflict that is resolved with the specified procedure. You can specify any or all operation types, for example id or iud. Upper or lower case are both valid.
l owner.procedure is the owner and name of the conflict resolution procedure that will handle the specified object and operation type.

SharePlex 9.2.8 Administrator Guide 140

Syntax rules
l There must be at least one space between the object specification, the operation type specification, and the procedure specification.
l You can use the LIKE operator and a SQL wildcard (%) to specify multiple objects by using a search string. (See the Example.)
l You can use an underscore (_) to denote a single-character wildcard. For table names that contain an underscore character (for example emp_sal), SharePlex recognizes the backslash (\) as an escape character to denote the underscore as a literal and not a wildcard, for example: like:scott.%\_corp\_ emp. If you are not using the LIKE operator, the backslash escape character is not required if an object name contains an underscore.
l The order in which you list the procedures in the conflict resolution file determines their priority of use (in descending order). If you list a table-specific procedure, SharePlex uses it before procedures that are specified with a wildcarded object name.
l You a comment line anywhere in the file. Start a comment line with the pound symbol (#).

Example conflict resolution file

scott.sal

IUD

like:scott.%\_corp\_emp

IUD

like:scott.%\_corp\_emp

IUD

like:scott%

IUD

scott.cust

How it works:

scott.sal_cr scott.emp_cr1 scott.emp_cr2 scott.emp_cr3 scott.sal_cr

l The scott.sal_cr routine is used for the scott.sal table before the scott.emp_cr1 procedure is used for that table.
l The scott.emp_cr1 procedure is used before the scott.emp_cr2 procedure for all tables meeting the search criteria, and so forth.
l For scott.cust, a procedure is called for UPDATEs before the other routines are used for all operations.

How to specify SharePlex prepared routines in the conflict resolution file

To use the SharePlex prepared routines for all tables in the replication configuration, use the !DEFAULT parameter instead of specifying an owner and object name.

A custom routine takes priority over a SharePlex prepared routine. A prepared routine is used only if the custom routine fails. This is true regardless of the order in which the !DEFAULT-associated routine appears in the file.

The !DEFAULT parameter is case-sensitive. It must be typed in all capital letters.

In the following example, the !UpdateUsingKeyOnly procedure is used for UPDATEs and DELETEs for all tables, including james.table1, although the user-defined procedures listed for james.table1 take precedence.

!DEFAULT !DEFAULT

!UpdateUsingKeyOnly

james.table1 james.table1 james.table1

james.procedure_upd

james.procedure_ins

james.procedure_del

SharePlex 9.2.8 Administrator Guide 141

How to change the conflict resolution file while replication is active
You can change the conflict resolution file any time during replication to add and remove tables and procedures. After you change the conflict resolution file, stop and re-start the Post process.

Log information about resolved conflicts
You can configure the Post process to log information about successful conflict resolution operations if you are using the SharePlex prepared routines. This feature is disabled by default. NOTE: This feature can be used only with Oracle to Oracle source-target combinations.

To enable the logging of conflict resolution
1. Run sp_ctrl on the target system. 2. Issue the following command:
sp_ctrl> set param SP_OPO_LOG_CONFLICT {1 | 2} l A setting of 1 enables the logging of conflict resolution to the SHAREPLEX_CONF_LOG table. l A setting of 2 enables the logging to the SHAREPLEX_CONF_LOG table and also directs Post to log the following to the Post process log: o conflict resolution routine called o tablename o source rowid o whether or not the conflict was resolved NOTE: A setting of 2 may affect the performance of Post as a result of making the query.
3. Restart Post.
Post logs the information to a table named SHAREPLEX_CONF_LOG. The following describes this table.

Column
CONFLICT_ NO CONFLICT_ TIME
CONFLICT_ TABLE CONFLICT_ TYPE CONFLICT_ RESOLVED

Column Definition

Information that is logged

NUMBER NOT The unique identifier of the resolved conflict. This value is generated from

NULL

the shareplex_conf_log_seq sequence.

TIMESTAMP

The timestamp of the conflict resolution

DEFAULT

SYSTIMESTAMP

VARCHAR2 (100)

The name of the target table that was involved in the conflict

VARCHAR2(1) The type of conflict, either I for insert, U for update, or D for delete

VARCHAR2(1) NOT NULL

Indicator of whether the conflict was resolved or not. Y = yes, the conflict was resolved N = no, the conflict was not resolved. Unresolved conflicts are logged to

SharePlex 9.2.8 Administrator Guide 142

Column

Column Definition

Information that is logged

the ID_errlog.sql file, where ID is the source database identifier.

TIMESTAMP_ VARCHAR2(50) The name of the column that contains the timestamp that Post compared

COLUMN

to determine which record was most recent.

INCOMING_ DATE TIMESTAMP

The timestamp that the row was inserted, updated, or deleted on the source system

EXISTING_ DATE TIMESTAMP

The current timestamp of the row in the target database. This applies only if the SP_OPO_LOG_CONFLICT parameter is set to 2, which directs Post to query the target database to get this value.

PRIMARY_ KEYS

VARCHAR2 (4000)

The names of the primary key columns

MESSAGE

VARCHAR2 (400)

A message that states which row won in the conflict. The row that wins depends on which conflict resolution routine was used. For example, the following message is returned when the !MostRecentRecord routine is used and the most recent record is the source record:

Incoming timestamp > existing timestamp. Incoming wins, overwrite existing.
If the target record was the most recent one or has the same timestamp as the source record, then the message would be:

Incoming timestamp <= existing timestamp. Existing wins, discard incoming.

SQL_

LONG

STATEMENT

CONFLICT_ VARCHAR2(1) CHECKED

The final SQL statement that got executed as a result of the conflict resolution
Indicates whether or not someone reviewed the conflict. The default is N for No. The person who reviews the conflict can change this value to Y.

Configure replication through an intermediary system
These instructions show you how to set up cascading replication, also known as multi-tiered replication. This strategy replicates data from a source system to an intermediary system, and then from the intermediary system to one or more remote target systems. Cascading replication can be used to support various replication objectives as a workaround in such conditions as the following:
l Your replication strategy exceeds the 1024 routes that are allowed directly from a given source system: You can send data to the intermediary system and then broadcast to the additional targets from there.
l The source has no direct connection to the ultimate target, because of firewall restrictions or other factors. You can cascade to a system that does allow remote connection from the source system.
To use a cascading strategy, the source machine must be able to resolve the final target machine name(s), although the ability to make a direct connection is not required.

SharePlex 9.2.8 Administrator Guide 143

Supported sources
Oracle
Supported targets
Oracle Oracle and Open Target (final target)
Capabilities
This replication strategy supports the following: l Replication to one or more target systems l Identical or different source and target names l Use of vertically partitioned replication l Use of horizontally partitioned replication l Use of named export and post queues
Requirements
l Prepare the system, install SharePlex, and configure database accounts according to the instructions in the SharePlex Installation Guide. IMPORTANT! Create the same SharePlex user on all systems if you will be using SharePlex to post to a database on the intermediary system.
l Disable triggers that perform DML on the target objects. l No DML or DDL should be performed on the target tables except by SharePlex. Tables on the target
system that are outside the replication configuration can have DML and DDL operations without affecting replication. l If sequences are unnecessary on the target system, do not replicate them. It can slow down replication. Even if a sequence is used to generate keys in a source table, the sequence values are part of the key columns when the replicated rows are inserted on the target system. The sequence itself does not have to be replicated.
DDL Replication Support
DDL replication from source to target through an intermediary system is supported in accordance with the information found in the DDL that SharePlex supports chapter of the Administration Guide, with the following exceptions:
SharePlex 9.2.8 Administrator Guide 144

l DDL initiated on the intermediary system, as opposed to the source, will cause inconsistencies leading to Post errors and should be avoided unless the DDL is synchronized across all systems.
l All systems must be monitored to ensure that latency or errors on the intermediary system do not cause inconsistencies.
IMPORTANT! These instructions assume you have a full understanding of SharePlex configuration files. They use abbreviated representations of important syntax elements. For more information, see Configure data replication on page 57.
Conventions used in the syntax
In the configuration syntax in this topic, the placeholders represent the following: l source_specification[n] is the fully qualified name of a source object (owner.object) or a wildcarded specification. l target_specification[n] is the fully qualified name of a target object or a wildcarded specification. l host is the name of a system where SharePlex runs. Different systems are identified by appending a letter to the names, like hostB. l db is a database specification. The database specification consists of either o. or r. prepended to the Oracle SID, TNS alias, or database name, as appropriate for the connection type. A database identifier is not required if the target is JMS, Kafka, or a file.
IMPORTANT! l Configure data replication on page 57.
Deployment options
To cascade data, you have the following options: l If there is a database on the intermediary system, you can configure SharePlex to post to that database and then capture the data again to replicate it to one or more remote targets. l If there is not a database on the intermediary system, you can configure SharePlex to import, queue, and then export the data to one or more remote targets. There is no Capture process on the system. This is known as a pass-through configuration. It passes the data directly from the source system to the target(s).
SharePlex 9.2.8 Administrator Guide 145

Cascade with posting on intermediate system
To use this configuration: l SharePlex database accounts must exist on all systems and must be the same name on all systems. This account is usually created when SharePlex is installed. See the SharePlexInstallation Guide for more information. l Triggers must be disabled in the intermediary database, as well as on the target system. l Oracle DDL replication is not supported from an Oracle database on the intermediary system to the target systems. It is supported only from the source system to the intermediary system. l You create two configuration files: one on the source system, and one on the intermediary system. l Enable archive logging on the source and intermediary systems in case the redo logs wrap before Capture is finished with them.
Configuration options on source system
This configuration replicates from the source system to the database on the intermediary system. NOTE: In this template, hostB is the intermediary system.
SharePlex 9.2.8 Administrator Guide 146

datasource_specification source_specification1 source_specification2

target_specification1 target_specification2

hostB@o.SID hostB@o.SID

Example on source system

Datasource:o.oraA hr.emp hr.sal cust.%

hr.emp2 hr.sal2 cust.%

hostB@o.oraB hostB@o.oraB hostB@o.oraB

NOTE: In this same configuration, you could route data from other source objects directly to other

targets, without cascading through the intermediary system. Just specify the appropriate routing on a

separate line.

Configuration options on intermediary system

This configuration captures the data from the database on the intermediary system, then replicates it to the target system(s). The tables that were the target tables in the source configuration are the source tables in this configuration. The target can be any supported SharePlex target.

datasource_specification

source_specification1 source_specification2

target_specification1 target_specification2

hostC[@db][+...] hostD[@db][+...]

Example on intermediary system

Datasource:o.oraB hr.emp hr.sal cust.%

hr.emp2 hr.sal2 cust.%

hostC@o.oraC hostD@o.oraD+hostE@r.mssE !cust_partitions

NOTE: The last entry in this example shows the use of horizontally partitioned replication to distribute different

data from the sales.accounts table to different regional databases. For more information, see Configure

horizontally partitioned replication on page 93.

Required parameter setting on intermediary system
(Oracle intermediary database) Set the SP_OCT_REPLICATE_POSTER parameter to 1 if the intermediary database is Oracle. This instructs the Capture process on the intermediary system to capture the changes posted by SharePlex and replicate them to the target system. (The default is 0, meaning that Capture ignores Post activity on the same system.) In sp_ctrl, issue the following command. The change takes effect the next time Capture starts.
set param SP_OCT_REPLICATE_POSTER 1

Cascade with pass-through on intermediary system
To use this configuration:

SharePlex 9.2.8 Administrator Guide 147

l Create an Oracle instance and an ORACLE_SID specification in the oratab file (Unix and Linux systems) or the Windows Registry. The database can be empty.
l DDL replication is not supported. l You create one configuration file, which is on the source system.

Configuration options on source system

NOTE: In this template, hostB is the intermediary system.

datasource_specification

source_specification1 source_specification2 source_specification3

target_specification1 target_specification2 target_specification3

hostB*hostC[@db] hostB*hostD[@db][+hostB*hostE[@db][+...] hostB*hostX[@db]+hostY[@db]

l The hostB*host syntax configures the pass-through behavior.
l If using a compound routing map where all data passes through the intermediary system first, make certain to use the hostB* component in each target route.
l You can also use a compound routing map where data from a source object is replicated directly to one target, and also through the intermediary system to another target, as in the third line of this configuration file.

Example
Datasource:o.oraA hr.emp hr.emp cust.%

hr.emp2 hr."Emp_3" cust.%

hostB*hostC@o.oraC hostB*hostC@r.mssB hostB*hostD@o.oraD+hostE@o.oraE

Configure replication to maintain high availability
These instructions show you how to set up SharePlex for the purpose of high availability: replicating to a secondary Oracle database that is a mirror of the source database. This strategy uses bi-directional replication with two SharePlex configurations that are the reverse of each other. The configuration on the secondary (standby) machine remains in an activated state with the Export process on that system stopped in readiness for failover if the primary machine fails. This strategy supports business requirements such as the following:
l Disaster recovery
l Continuous operation of business applications throughout maintenance cycles or mechanical failures
In this strategy, SharePlex operates as follows:
l Under normal conditions, SharePlex replicates changes from the primary database to the secondary database.

SharePlex 9.2.8 Administrator Guide 148

l When the primary system or database is offline and users are transferred to the secondary system, SharePlex captures their changes and queues the data on that system until the primary system is restored.
l When the primary system is restored, SharePlex updates it with those changes and then resumes capture and replication from the primary database.
Supported sources
Oracle
Supported targets
Oracle
Capabilities
This replication strategy supports the use of named export and post queues. NOTE: Column mapping and partitioned replication is not appropriate in a high availability configuration. Source and target objects can have different names but this makes the management of a high-availability structure more complicated.
Requirements
l Prepare the system, install SharePlex, and configure database accounts according to the instructions in the SharePlex Installation Guide.
l All objects must exist in their entirety on both systems. l The target objects must have the same structure and qualified names as their source objects. l Enable archive logging on all systems. l Create a script that denies INSERT, UPDATE and DELETE operations to all users except SharePlex. For failover purposes, the following are required: l Make the applications used on the primary system available on the secondary system. l Copy non-replicated database objects and critical files outside the instance to the secondary system. l Create a script that grants INSERT, UPDATE and DELETE privileges to all users, which can be run
during a failover procedure. l Create a script that enables constraints on the secondary system to be used during a failover procedure. l Develop a failover procedure for relocating users to the secondary system. NOTE: If you use an Oracle hot backup to create the secondary instance, keep the script. It can be modified to re-create the primary instance.
SharePlex 9.2.8 Administrator Guide 149

Conventions used in the syntax
In the configuration syntax in this topic, the placeholders represent the following:
l hostA is the primary system. l hostB is the secondary system. l ownerA.object is the fully qualified name of an object on hostA or a wildcarded specification. l ownerB.object is the fully qualified name of an object on hostB or a wildcarded specification. l oraA is the Oracle instance on hostA. l oraB is the Oracle instance on hostB. IMPORTANT! l See Configure data replication on page 57 for more information about the components of a
configuration file.

Configuration
A high availability configuration uses two configurations that are the reverse of each other. To replicate all objects in the database, you can use the config.sql script to simplify the configuration process. For more information, see Configuration Scripts on page 1.

Configuration on the source system (primary system)

Datasource:o.oraA ownerA.object

ownerB.object

hostB@o.oraB

Configuration on the target system (secondary system)

Datasource:o.oraB ownerB.object

ownerA.object

hostA@o.oraA

Make the system ready for failover
1. On the secondary system (the one that will initially be the passive system) run sp_ctrl and then issue the following command to stop the Export process on the secondary system so that nothing accidentally happening on the secondary system (such as a scheduled job changing data) gets replicated back to the primary system. This is the required state of SharePlex on that system until there is a need for a role switch between systems.
2. Perform initial synchronization and startup. You will activate the source configuration during this procedure. For more information, see Start replication on your production systems on page 186.

SharePlex 9.2.8 Administrator Guide 150

3. Making sure the Export process is stopped on the secondary system, activate the configuration on that system. The configuration on the secondary machine remains in an activated state, but the stopped Export process and lack of user activity ensure that the system remains static in readiness for failover.
4. Monitor the SharePlex instance that is linked to the secondary Oracle instance to make sure no nonSharePlex DDL or DML changes were performed. You can do this as follows: View the status of the export queue on the secondary system using the qstatus command in sp_ctrl. The queue should be empty, because the Capture process on a system ignores the Post process on that system. If there are any messages in the export queue, it means those transactions originated on the secondary system or the SP_OCT_REPLICATE_POSTER parameter was mistakenly enabled. See the SharePlex Reference Guide for more information about SharePlex commands and parameters.
5. Maintain backups of replication files.
Perform recovery procedures
If a system fails in your high-availability environment, you can move replication to a secondary system and then move it back to the primary system when it is restored. For more information, see Recover replication after Oracle failover on page 288.
SharePlex 9.2.8 Administrator Guide 151

12
Configure DDL replication
This chapter contains the information that you need to know in order to replicate Oracle DDL operations that are supported by SharePlex.
Contents
DDL that SharePlex supports Control Oracle DDL replication Filter DDL Replication Best practices for ALTER TABLE DDL DDL logging and error handling
DDL that SharePlex supports
SharePlex supports DDL replication for Oracle databases only. SharePlex replicates certain Oracle DDL changes that are written to the redo logs. Changes that bypass the redo logs are not replicated. For details on the DDL that SharePlex supports, see the SharePlexSharePlex Release Notes. In a cascading configuration, DDL replication is supported from source to target through an intermediary system. However, DDL initiated on the intermediary system could cause inconsistencies leading to Post errors and must be avoided. For more information, see Configure replication through an intermediary system on page 143. DDL against objects owned by the SharePlex user or against system-owned objects is not replicated.
NOTE: Because the source and target databases can be of different versions in SharePlex replication, the source and target objects can be different. When DDL is applied to the target, it may fail if the operation is forbidden on the target but allowed on the source. For a list of objects for which DDL is supported, see the SharePlex Release Notes.
Control Oracle DDL replication
By default, SharePlex replicates some Oracle DDL for objects that are listed in the active configuration (explicitly or by wildcard), but you can expand this support with parameter settings. IMPORTANT!
SharePlex 9.2.8 Administrator Guide 152

l For the most current information about supported DDL and requirements or limitations, see the SharePlex Release Notes provided with this release. That information may supercede what is documented here.
l DDL replication is supported only to Oracle targets, except for ALTER TABLE to ADD COLUMN or DROP COLUMN, which are supported for all SharePlex targets.
SharePlex provides default DDL support for objects in the configuration file. You can expand this support through parameter settings. See the SharePlex Release Notes for detailed information about the DDL that is supported by SharePlex.
Default support for Oracle DDL
SharePlex provides some basic DDL support by default.
DDL for existing objects
By default SharePlex replicates: l TRUNCATE TABLE l ALTER TABLE to: ADD, MODIFY, DROP, SPLIT, COALESCE, MOVE, TRUNCATE, EXCHANGE PARTITION/SUBPARTITION ADD, MODIFY, or DROP columns when: l the affected object exists in the source and target at the time of activation and l its name is listed in the configuration file (explicitly or through wildcard).
This functionality is controlled by the SP_OCT_REPLICATE_DDL parameter. The valid values are as follows: 0 (disable replication of both ALTER TABLE and TRUNCATE) 1 (enable ALTER replication only) 2 (enable TRUNCATE replication only) 3 (enable replication of ALTER and TRUNCATE)
DDL for objects added after activation
By default, the SharePlex Auto-Add feature is also enabled to provide DDL support for tables and indexes that are created after activation. When SharePlex detects a CREATE statement for one of these objects and its name satisfies a wildcard in the configuration file, SharePlex does the following:
l replicates the CREATE to add the object to the target l adds the object to replication l maintains that object through future DDL and DML changes The Auto-Add feature is controlled by the SP_OCT_AUTOADD_ENABLE parameter, which is set to 1 (enabled) by default. See the SharePlex Reference Guide for details about this parameter.
SharePlex 9.2.8 Administrator Guide 153

Optional DDL on objects in replication
You can enable the replication of the following DDL when it is issued on objects that are in the configuration file. To enable the replication of a DDL command, set the associated parameter to 1.

DDL command CREATE / DROP TRIGGER CREATE / DROP SYNONYM GRANT

Set this parameter to 1 SP_OCT_REPLICATE_TRIGGER SP_OCT_REPLICATE_SYNONYM SP_OCT_REPLICATE_GRANT

See the SharePlex Reference Guide for details about these parameters.

Optional Auto-Add support for Oracle DDL
You can expand Auto-Add support to include any of the object types listed in the following table. The object is added to replication if its name satisfies a wildcard specification in the active configuration file.
To enable auto-add of individual object types 1. Make certain the SP_OCT_AUTOADD_ENABLE parameter is set to 1. 2. Set the appropriate parameter to 1, using the following table as your guide.

DDL to auto-add

Parameter to set to 1

CREATE / DROP SEQUENCE SP_OCT_AUTOADD_SEQ

CREATE /DROP MATERIALIZED VIEW*

SP_OCT_AUTOADD_MV

Additional requirements
Set the SP_SYS_TARGET_ COMPATIBILITY parameter to at least 8.6.3.
Set the SP_SYS_TARGET_ COMPATIBILITY parameter to at least 8.6.2.

* SharePlex does not replicate materialized views to materialized views. SharePlex converts a CREATE MATERIALIZED VIEW to a CREATE TABLE, applies the CREATE TABLE to the target, and then replicates the DML that populates the view. SharePlex replicates DROP MATERIALIZED VIEW, but not ALTER MATERIALIZED VIEW. See the SharePlex Reference Guide for details about these parameters.

Expanded DDL support for objects outside replication
You can configure SharePlex to replicate DDL for certain objects that are not listed in the configuration file. SharePlex replicates the DDL statements, but does not replicate any data change operations made to the objects because they are not part of active replication. Therefore, SharePlex does not maintain synchronization of these objects on the target. The objects must exist in the source and target prior to configuration activation.

SharePlex 9.2.8 Administrator Guide 154

NOTES:
Expanded DDL replication supports not only tables and sequences but also a wide range of other objects such as procedures, functions, users, and views, which are not part of replication. Some of these objects may have underlying objects that are in replication. In those cases, Expanded DDL replication applies not only to the object that is outside the replication configuration, but also to the underlying objects that are in replication.
SharePlex does not support the Oracle Flashback Table feature. If the SP_REPLICATE_ALL_DDL parameter is enabled (value of 1), SharePlex may try to replicate the flashback DDL, which will return an error. To perform Flashback Table on a table that is in replication, use the following procedures to work around this issue:
1. Remove source objects from replication 2. Perform the flashback 3. Add or change objects in an active configuration
To replicate DDL for objects outside the replication configuration Set the SP_OCT_REPLICATE_ALL_DDL parameter to 1. See the SharePlex Reference Guide for details about this parameter. See the SharePlex Release Notes for more information about supported DDL for objects that are not in the configuration file.

Filter DDL Replication
You can filter the objects for which DDL is replicated when the SP_OCT_REPLICATE_ALL_DDL parameter is enabled. This is the DDL that is not in the replication configuration.
NOTE: DDL filtering is only allowed for objects outside the replication configuration. All DDL performed on objects inside the replication configuration must be replicated to keep the source and target metadata consistent so that DML succeeds. DDL filtering is controlled in the SHAREPLEX_DDL_CONTROL table that is installed in the SharePlex schema.

Name ------------DDL_PARAMETER DDL_CODE SCHEMA_FILTER OBJECT_FILTER

Type -----------NUMBER NUMBER VARCHAR2(32) VARCHAR2(32)

Each row in the SHAREPLEX_DDL_CONTROL table defines a filter based on what you specify in each of the following columns:

SharePlex 9.2.8 Administrator Guide 155

l SCHEMA_FILTER filters DDL by a schema name.
l OBJECT_FILTER filters DDL by an object name. l DDL_CODE filters by the code number of a DDL type. See DDL codes.
A null value in the DDL_CODE column means that the filter applies to all of the DDL types A null in the SCHEMA_FILTER or OBJECT_FILTER column means that the filter applies to any schema or object name.
NOTE: The DDL_PARAMETER column is not an active column as of this release of SharePlex.

To filter DDL Insert a row into the table with the desired values in the active columns.
Examples
The following filters out of replication the DDL for ALTER TABLE: INSERT INTO SPLEX.SHAREPLEX_DDL_CONTROL (DDL_CODE, SCHEMA_FILTER, OBJECT_FILTER) values (15,null,null);
The following filters out of replication all DDL for all objects with names that begin with TEST_ in any schema: INSERT INTO SPLEX.SHAREPLEX_DDL_CONTROL (DDL_CODE, SCHEMA_FILTER, OBJECT_FILTER) values (null,null,'TEST_%');
The following filters out of replication the DDL for CREATE TABLE for the "Sales" schema and objects with names that begin with "TEST_":
INSERT INTO SPLEX.SHAREPLEX_DDL_CONTROL (DDL_CODE, SCHEMA_FILTER, OBJECT_FILTER) values (1,'Sales','TEST_%');

DDL codes

DDL Type CREATE TABLE ALTER TABLE DROP TABLE ASSOCIATE STATISTICS DISASSOCIATE STATISTICS COMMENT TABLE, COMMENT ON COLUMNS TRUNCATE

DDL_CODE 1 15 12 168 169 29 85

CREATE INDEX

ALTER INDEX

DROP INDEX

CREATE SEQUENCE

ALTER SEQUENCE

SharePlex 9.2.8 Administrator Guide 156

DDL Type DROP SEQUENCE
CREATE CLUSTER DROP CLUSTER
CREATE USER ALTER USER DROP USER
CREATE_ROLE ALTER_ROLE DROP_ROLE GRANT REVOKE
CREATE SYNONYM DROP SYNONYM
CREATE VIEW ALTER VIEW DROP VIEW
CREATE TYPE ALTER TYPE DROP TYPE CREATE TYPE BODY DROP TYPE BODY
CREATE FUNCTION ALTER FUNCTION DROP FUNCTION

DDL_CODE 16
4 8
51 43 53
52 79 54 17 18
19 20
21 88 22
77 80 78 81 83
91 92 93
SharePlex 9.2.8 Administrator Guide 157

DDL Type
CREATE PROCEDURE ALTER PROCEDURE DROP PROCEDURE
CREATE PACKAGE ALTER PACKAGE DROP PACKAGE CREATE PACKAGE BODY ALTER PACKAGE BODY DROP PACKAGE BODY
CREATE DIRECTORY DROP DIRECTORY

DDL_CODE
24 25 68
94 95 96 97 98 99
157 158

Best practices for ALTER TABLE DDL
The following are best practices for the replication of Oracle ALTER TABLE operations.
Tables with VARRAY or ABSTRACT types
Do not add a VARRAY column or abstract data type column if you plan to issue an ALTER TABLE to drop a column or set a column unused soon thereafter. SharePlex must query the database to obtain information about this data type. If the second DDL was performed before SharePlex was able to process the first DDL, the query will fail because the metadata is already changed.
Tables with system-specific metadata
If some metadata is system-specific, such as the storage parameters of database objects, there may be unexpected results when DDL on that metadata is replicated. For example, SharePlex replicates all of the storage parameters for a source Oracle object, even though only some of them were changed with the ALTER TABLE command. If the source and target objects were not created with the same storage parameters, one of two things can happen: either the target table will assume the storage of the source table or, if the DDL is not supported by the target, an error will be generated.
Example: consider a source table with MAXEXTENTS 525 and MINEXTENTS 20, and a target table with MAXEXTENTS 505 and MINEXTENTS 4. If the MAXEXTENTS of the source object is changed to unlimited, SharePlex will replicate both the MAXEXTENTS change and the non-changed MINEXTENTS of 20. This causes Oracle error 01570, because MINEXTENTS cannot be larger than the extents currently allocated. Alternatively, if the MINEXTENTS is changed to 1 on the source, but MAXEXTENTS is

SharePlex 9.2.8 Administrator Guide 158

not changed, SharePlex replicates both, which results in target parameters of MAXEXTENTS 525 and MINEXTENTS 1.
Tables that are renamed
When ALTER TABLE RENAME is issued on a source table that is currently in replication, SharePlex changes the name of the table in the active configuration file by commenting out the old configuration line and adding a new line at the end of the configuration file. If the source and target table names are the same, both are changed to the new name. Otherwise, just the source name is changed. The following is an example:When ALTER TABLE RENAME is issued on a source table that is currently in replication, SharePlex changes the name of the table in the active configuration file by commenting out the old configuration line and adding a new line at the end of the configuration file. If the source and target table names are the same, both are changed to the new name. Otherwise, just the source name is changed. The following is an example:
# Table scott.table1 renamed to scott.table2 August 5, 2003 10:14 scott.table2 scott.table2 sysA@o.ora555 Whether the Post process stops on RENAME operations or not depends on the setting of the SP_OPO_STOP_ ON_DDL_ERR parameter.
Tables with system-generated interval partitions/subpartitions
Because the database generates the names of system-generated interval partitions/subpartitions, the names of those partitions on the source will not match the names of their corresponding partitions on the target. Set the SP_OCT_TRUNC_PARTITION_BY_ID parameter to 1 to ensure that SharePlex truncates the correct partition when it replicates an ALTER TABLE to TRUNCATE a system-generated interval partition. This setting directs SharePlex to identify the partition by using the partition ID, rather than by using the partition name that is specified in the original ALTER TABLE command. Post maps the partition ID to the correct partition name on the target table. For more information, see the SharePlex Reference Guide. To support the replication of system-named interval partitions/subpartitions, both the source and target must be SharePlex version 8.6.4 or later. SharePlex does not support TRUNCATE of a system-generated sub-partition if the sub-partition is empty.
ALTER TABLE...MOVE
ALTER TABLE DDL commands that change the rowid of a table can affect subsequent DML operations if the primary or unique keys of the tables in replication are not being logged. When the keys are not logged, SharePlex fetches their values based on the rowid. Any operation that changes the rowid, such as ALTER TABLE...MOVE, can cause the wrong key values to be used for subsequent DML operations.
DDL logging and error handling
Both Capture and Post log the DDL that they process. SharePlex also prints replicated DDL to the SharePlex Event Log, but it may be truncated. Only the Post DDL log contains complete DDL statements. SharePlex stores the DDL logs in the log subdirectory of the variable-data directory on the source and target systems.
SharePlex 9.2.8 Administrator Guide 159

By default, Post stops on DDL errors. An error usually indicates that the database component for which the DDL was executed on the source system does not exist in the target database. The default setting of the SP_OPO_ STOP_ON_DDL_ERR parameter stops Post on DDL errors, so that subsequent DML on that object does not fail. This enables you to correct the problem and keep the databases synchronized. For more information about this parameter, see the SharePlex Reference Guide.
Table 3: SharePlex DDL log naming conventions

DDL log type Capture Oracle target
Open Target target

Naming convention
o.ORACLE_SID_ocap_ddl_log_number.log
o.ORACLE_SID_machine_name_opo_ddl_log_ number.log
r.database_name_machine_name_xpst_ddl_log_ number.log

Example
o.ora12_ocap_ddl_01.log
o.ora12_server2_opo_ddl_ 01.log
r.mssdb1_server3_xpst_ddl_ 01.log

SharePlex 9.2.8 Administrator Guide 160

13
Configure error handling
This chapter contains an overview of the tools that SharePlex provides to handle errors that are returned by the Post process.
Contents
Continue to post when there is a DML error Continue to Post when there is a DDL error Increase the number of retries on error Handle transactions that contain out-of-sync operations
Continue to post when there is a DML error
SharePlex provides a way for Post to continue processing after it encounters a DML error, rather than stop.
Continue posting on Oracle and SharePlex errors
Valid for Oracle targets When SharePlex posts to an Oracle target, you can configure Post to ignore specific Oracle DML errors and specific SharePlex error messages and continue processing. Post determines which messages to ignore based on the list in the oramsglist file. The file is installed with a small list of errors by default, but you can remove any of them as desired. When Posts ignores an error, it writes the error to the SharePlex Event Log. Post also logs the error and the SQL statements that caused the error to the Error log. This log is named SID_errlog.sql log file and is stored in the data directory in the SharePlex variable-data directory. For more information, see View events and errors on page 211. NOTE: There are certain errors that Post will not ignore, and it will stop for those messages even if they are listed in the oramsglist file.
SharePlex 9.2.8 Administrator Guide 161

IMPORTANT: Use caution when using this feature. It could result in hidden out-of-sync conditions. Enable this parameter only if your target users cannot tolerate replication lag and it is acceptable to have some out-of-sync data. Check the SID_errlog.sql log frequently to see if there were errors that could cause replication problems.
To configure Post to continue on errors
1. On the target system, change directories to the data sub-directory of the SharePlex variable-data directory.
2. Find the oramsglist file. 3. If replication is not active, open the file in a text editor. If replication is active, make a copy of the file and
then open the copy in the editor.
4. Increase the number on the first line by the number of errors that you are adding. This number must be equal to the total number of errors that are in the file. For example, in the following file there are 10 errors listed.
ora@sys1dad > vi oramsglist 10 604 900 902 908 909 910 911 932 960 1026
5. Starting at the end of the file, add the number of each Oracle or SharePlex error, one per line as shown in the preceding example. The messages need not be in numerical order.
6. Save and close the file. 7. Stop Post (if running).
sp_ctrl> stop post
8. If you edited a copy of the oramsglist file, save the copy to the original name of oramsglist. 9. Change the value of the SP_OPO_CONT_ON_ERR parameter to 1. Or change the value to 2 to also
continue posting on table errors listed in the oramsglist file. See the SharePlex Reference Guide for a description of the SP_OPO_CONT_ON_ERR parameter.
sp_ctrl> set param SP_OPO_CONT_ON_ERR 1
10. Start Post. sp_ctrl> start post
Continue posting on ODBC errors
Valid for Open Target When SharePlex posts to an Open Target target, you can configure Post to ignore ODBC errors and continue processing. Post writes the error to the SharePlex Event Log. Post also logs the error and the SQL statements
SharePlex 9.2.8 Administrator Guide 162

that caused the error to the Error log. This log is named ID_errlog.sql log file, where ID is the database identifier. The file is stored in the data directory in the SharePlex variable-data directory. For more information, see View events and errors on page 211.
IMPORTANT: Use caution when using this feature. It could result in hidden out-of-sync conditions. Enable this parameter only if your target users cannot tolerate replication lag and it is acceptable to have some out-of-sync data. Check the SID_errlog.sql log frequently to see if there were errors that could cause replication problems.

To configure Post to continue on errors
1. On the target system, change directories to the data sub-directory of the SharePlex variable-data directory.
2. Look for one of the following files, depending on the database. These files are installed empty.

File name postgresmsglist sqlservermsglist sybasemsglist tdmsglist mysqlmsglist

Supported database Postgres
Microsoft SQL Server SAP Adaptive Server Enterprise (ASE) Teradata Oracle MySQL

NOTE: There are certain errors for which Post will stop, even if you list those errors in the message file. 3. If replication is not active, open the file in a text editor. If replication is active, make a copy of the file and
then open the copy in the editor. 4. Starting at the end of the file, add the number of each error, one per line as shown in the example. The
messages need not be in numerical order. Example:
sqlservermsglist:
8102
8180
544
2627
3621 5. Save and close the file. 6. Stop Post (if running).
sp_ctrl> stop post 7. If you copied the original file, save it back to its original name. 8. Change the value of the SP_OPX_CONT_ON_ERR parameter to 1.
sp_ctrl> set param SP_OPX_CONT_ON_ERR 1

SharePlex 9.2.8 Administrator Guide 163

9. Start Post. sp_ctrl> start post
Continue to Post when there is a DDL error
By default, Post stops on DDL errors. An error usually indicates that the database component for which the DDL was executed on the source system does not exist in the target database. The default setting of the SP_OPO_ STOP_ON_DDL_ERR parameter stops Post on DDL errors, so that subsequent DML on that object does not fail. This enables you to correct the problem and keep the databases synchronized. For more information about this parameter, see the SharePlex Reference Guide.
Increase the number of retries on error
Post will retry certain failed operations when there is the possibility that they will succeed with another attempt. The main operations that Post will retry are TNS write failures, connection failures, or locks on tables when Post needs to apply a TRUNCATE. To increase the likelihood that the failed operations are successful, you can increase the SP_OPO_RETRIES_ MAX parameter so that Post tries the operation more times. At the same time, increase the SP_OPO_RETRY_ DELAY_TIME parameter to increase the time interval between the attempts. That gives the lock or other blocking operation enough time to be resolved between attempts. If the Post process is set to continue on error (SP_SYS_SUSPEND_ON_ERROR=0) or if the error message is listed in the oramsglist file, Post moves on to the next transaction in the queue. In all other cases, Post stops after it reaches the maximum allowed attempts. IMPORTANT: Reducing this parameter can cause the data to accumulate in the queues, possibly causing them to exceed the available disk space. For more information, see SP_OPO_RETRIES_MAX in the SharePlex Reference Guide.
Handle transactions that contain outof-sync operations
You can configure Post to handle out-of-sync conditions. The following explains the default and alternate behaviors.
Default Post handling of out-of-sync errors
The default Post behavior when a transaction contains an out-of-sync operation is to continue processing other valid operations in the transaction to minimize latency and keep targets as current as possible. Latency is the amount of time between when a source transaction occurs and when it is applied to the target. Different factors affect the amount of latency in replication, such as unusually high transaction volumes or interruptions to network traffic.
SharePlex 9.2.8 Administrator Guide 164

Post logs the SQL statement and data for the out-of-sync operation to the ID_errlog.sql log file, where ID is the database identifier. This file is in the log sub-directory of the variable-data directory on the target system.
Stop on out-of-sync errors
You can configure Post to stop when it encounters an out-of-sync condition by setting the following parameter to 1:
l Oracle targets: SP_OPO_OUT_OF_SYNC_SUSPEND l Open Target targets: SP_OPX_OUT_OF_SYNC_SUSPEND If you use this feature, make certain to monitor replication frequently. If Post stops, latency increases and data accumulates in the queues. For more information, see the parameter documentation in the SharePlex Reference Guide.
Roll back the transaction if it generates out-ofsync errors
You can configure Post to roll back and discard a transaction if any operation in that transaction generates an out-of-sync error. The entire transaction is logged to a SQL file, but not applied to the target.You can edit the SQL file to fix the invalid DML and then run the SQL file to apply the transaction. This feature is enabled by setting the SP_OPO_SAVE_OOS_TRANSACTION to 1. For more information, see the parameter documentation in the SharePlex Reference Guide.
SharePlex 9.2.8 Administrator Guide 165

14
Configure data transformation
This topic contains instructions for using the transformation feature of SharePlex. Transformation enables SharePlex to manipulate data before, or instead of, posting it to a target.
Contents
Overview of transformation Considerations when using transformation Deploy transformation
Overview of transformation
Transformation directs the Post process to call a PL/SQL procedure (de fined as a transformation routine) instead of applying a SQL operation to the target database. Transformation enables replicated data to be manipulated before, or instead of, posting to a target. For example, if a source table and its target table are dissimilar in construction -- like when a person's first and last name are in one column in the source table but in separate columns in the target table -- you can write a transformation routine to convert the data for those columns so that replication succeeds. You can use transformation routines to convert data types, units of measurement, or character sets. You can use them instead of database triggers to reduce I/O overhead, and for many other business requirements. When you specify transformation for a table, Post takes no action on the replicated data. Instead, it passes data values to your transformation routine, enabling you to control both the form and destination of the data with the procedure. You can post to the target table, post to an alternate location, or both. Therefore, when writing your routine, is your responsibility to include in your procedure the necessary SQL operations for posting.
Supported sources
Oracle
SharePlex 9.2.8 Administrator Guide 166

Supported targets
Oracle
Supported replication strategies
Transformation is a convenient way to use SharePlex to transfer data from one table to another without concern for maintaining identical structure or data. This makes it practical for reporting, broadcast, and data mart and warehousing applications. Transformation is not suitable for peer-to-peer or high-availability replication environments. High availability requires identical databases that are kept synchronized by replication. For peer-to-peer replication SharePlex must be able to detect and resolve conflicts when there are concurrent changes to the same record. When data is transformed, SharePlex cannot compare before and after values to verify synchronization and detect conflicts.
Supported operations
Transformation supports only INSERT, UPDATE and DELETE operations. You can do the following when developing procedures:
l You can create one procedure for all three operation types, or you can create a procedure for each operation type.
l You can use one procedure for all tables, or use different procedures for different tables. SharePlex allows this through the use of wildcards to specify the tables.
If a transformation routine is specified for an individual table, and the table also is part of a group of tables for which another routine is specified, only the table-specific routine is used for that table when the associated DML operation occurs.
Considerations when using transformation
Because transformation changes data and because SharePlex does not post the data, transformation changes the behavior of replication. It is a customization of SharePlex processing. Review the following considerations before implementing transformation to ensure that your transformation procedures succeed.
Privileges
Any table that will be accessed through PL/SQL for transformation requires implicitly granted privileges from the owner of the object to SharePlex.
Keys
A PRIMARY or UNIQUE key is required for all tables using transformation. SharePlex locates the target row for UPDATEs and DELETEs by using the key, which enables it to return values to your transformation routine from
SharePlex 9.2.8 Administrator Guide 167

the target table for UPDATE operations. Do not allow keys to be changed on the target system, or SharePlex will not be able to locate the row to pass values to your routine.
Test your routines
Test your transformation routines before you put them into production to make sure they work as intended, and to make sure that one routine does not counteract another one. When data is transformed, SharePlex cannot compare before and after values to verify synchronization, which is a measure of whether the routines are performing correctly. The only way to confirm synchronization is to use the compare command with the key option. This option restricts the comparison to just the key values and is not a complete confirmation of synchronization. It only detects missing or extra rows. It does not (and cannot) indicate whether values in nonkey columns are properly correlated to those in the source columns, because the target data was transformed. For those reasons, the repair command cannot be used to resynchronize data. You must devise your own resynchronization procedures based on your company's business rules and the database environment.
Dates
The default date format for SharePlex transformation is MMDDYYYY HH24MISS. Tables with default dates must use that format, or transformation will return errors. Before creating a table with a default date, use the following command to change the date format in SQL*Plus.
ALTER SESSION SET nls_date_format = 'MMDDYYYYHH24MISS'
Other considerations
l Transformation does not support changes to LOB and LONG columns. l The processing overhead for passing data to your procedure, combined with that of executing the
procedure itself, degrades overall performance on the target system compared to normal replication and posting. l The same PL/SQL package is used for both generic conflict resolution and transformation (its name is sp_cr). Use either generic conflict resolution or transformation for a table, but not both. Transformed tables cannot be compared by SharePlex and conflict resolution cannot succeed. If both are used, SharePlex only calls the transformation routine. If appropriate, you can use generic conflict resolution and transformation for different tables in the same configuration. For more information, see Configure peer-to-peer replication on page 123.
Deploy transformation
Deployment of transformation involves the following steps. l Create configuration entries for the source and target tables to be transformed. There are no special configuration procedures for tables that use transformation. Configure them as you would any other table. For more information, see Create a configuration file on page 58. l Create transformation routines. For more information, see Create transformation routines on page 169. l Create a transformation file, which directs SharePlex to call the transformation routines. For more information, see Create the transformation file on page 174.
SharePlex 9.2.8 Administrator Guide 168

Create transformation routines
Write your transformation routines with dynamic PL/SQL procedural language. Use parameters and record and table structures defined in the public package named sp_cr. This package was installed in the SharePlex schema in the database. The package uses the following parameters.

Variable src_host

SharePlex 9.2.8 Administrator Guide 169

Variable src_ora_sid src_ora_time source_rowid target_rowid
statement_type source_table
target_table oracle_err

Description
The ORACLE_SID of the source database. It is case-sensitive and is passed in the same case as in the oratab file, Windows Registry or V$PARAMETER table.
The timestamp of the change record in the source redo log.
The row ID of the source row. It is passed as a literal within single quotes, for example `123456'.
The row ID of the corresponding row in the target database. SharePlex obtains the row ID by querying the target database. It is passed as a literal within single quotes, for example `123456'. If the row cannot be found using the PRIMARY key, the value is NULL.
A letter, either I, U or D, indicating whether the operation is an INSERT, UPDATE or DELETE statement.
The owner and name of the source table, expressed as owner.table. This value is case-sensitive and matches the way the table is named in the database. It is passed within double quotes, for example "scott"."emp."
The owner and name of the target table, expressed as owner.table. This value is case-sensitive and matches the way the table is named in the database. It is passed within double quotes, for example "scott"."emp."
This is different, depending on whether the procedure is being used for conflict resolution or transformation. Transformation: SharePlex passes a value of 0 for this variable. This variable is only used for conflict resolution. Conflict resolution: The Oracle error number that caused the conflict.

OUT variables These variables direct the action of SharePlex based on whether the procedure succeeded or failed).

Variable status

SharePlex 9.2.8 Administrator Guide 170

Variable action
reporting

Description
l A value of 1 implies that the procedure was unsuccessful. In this case, the action SharePlex takes depends on what you specified as the action variable.
l (Transformation only) A value of 7 implies unsuccessful execution and instructs the Post process to stop.
Defines the action that you want SharePlex to take. This is different, depending on whether the procedure is used for transformation or conflict resolution. Transformation: You must specify a value of 0 for this parameter, which directs SharePlex NOT to post the SQL statement. Your transformation routine is responsible for posting the results of the transformation either to the target table or another table. The outcome of this action depends on what you specify for the reporting variable Conflict resolution: Specifies the action to take as a result of an unsuccessful conflict resolution procedure. You must specify a value for this parameter.
l A value of 0 directs SharePlex NOT to post the SQL statement. The outcome of this action depends on what you specify for the reporting variable. In addition, it directs SharePlex to try the next conflict resolution procedure that you listed in the conflict resolution file, if one exists.
l The value of 1 is reserved for internal SharePlex use. Do not use it.
l A value of 2 directs SharePlex to try the next conflict resolution procedure that you listed in the conflict resolution file, if one exists.
Determines how SharePlex reports unsuccessful procedural results. You must specify a value for this parameter.
l A value of 0 directs SharePlex NOT to report an error or write the failed SQL statement to the SID_errlog.sql log.
l Values 1 and 2 are reserved for internal SharePlex use. Do not use them.
l A value of 3 directs SharePlex to write the failed SQL statement to the SID_errlog.sql log and report an error to the Event Log.

col_def_type table
SharePlex creates a col_def_tabtyp PL/SQL table for each replicated operation. This table stores column information. It is different depending on whether the procedure is used for transformation or conflict resolution.

SharePlex 9.2.8 Administrator Guide 171

All fields are passed by SharePlex to your routine, although not all will have values if SharePlex cannot locate the row. Following is the data type that is used to populate the col_def_tabtyp table.

type col_def_typ is record

(column_name user_tab_columns.column_name%type

,data type

user_tab_columns.data type%type

,is_key boolean

,is_changed

boolean

,old_value

varchar2(32764)

,new_value

varchar2(32764)

,current_value varchar2(32764)

);

type col_def_tabtyp is table of col_def_typ

Description of col_def_tabtyp

Column column_name data type is_key is_changed
old_value

SharePlex 9.2.8 Administrator Guide 172

Column
new_value current_value

Description
because the row did not exist in the target database before the INSERT. Conflict resolution only: This is the pre-image against which SharePlex compared the source and target columns as part of its synchronization check for UPDATEs and DELETEs. If the old value passed by SharePlex does not match the current_value value obtained from the target row, then there is a conflict.
Tells your procedure the new value of the replicated column, as changed on the source system.
Tells your procedure the current value of the column in the target table. If SharePlex cannot locate the target row, the value is NULL.

Example entries in col_def_tabtyp table per operation type The following tables illustrate the possible outcomes of each type of operation.

INSERT operation

column_name C1 C2 C3

is_changed TRUE TRUE FALSE

old_value NULL NULL NULL

new_value bind bind NULL

current_value1 is_key

NULL NULL NULL

FALSE TRUE TRUE | FALSE

1 When an INSERT fails, it is because a row with the same PRIMARY key already exists in the target database. SharePlex does not return the current value for INSERTs.

UPDATE operation

column_name C1 C2 C3

is_changed TRUE FALSE TRUE

old_value bind bind bind

new_value bind NULL bind

current_value1, 2 NULL | target_value NULL | target_value NULL | target_value

is_key FALSE TRUE TRUE

SharePlex 9.2.8 Administrator Guide 173

DELETE operation

column_name C1

is_changed FALSE

old_value bind

new_value NULL

current_value1 NULL

is_key TRUE

1 When a DELETE fails, it is because SharePlex could not find the row by using the PRIMARY key. Therefore, SharePlex returns a NULL.

Create the transformation file
To direct SharePlex to call transformation routines instead of posting SQL operations, use the transformation.SID file, where SID is the ORACLE_SID of the target database. Before executing a SQL operation, Post reads this file to determine if there is a transformation routine that it must call.

Where to find this file
A blank transformation.SID file, where SID is the ORACLE_SID of the target instance, was included in the data sub-directory of the SharePlex variable-data directory when SharePlex was installed. Use the file on the target system. If this file does not exist, you can create one in ASCII format in an ASCII text editor. It must be named transformation.SID, where SID is the ORACLE_SID of the target instance. Note: the SID is case-sensitive. IMPORTANT! There can be only one transformation.SID file per active configuration.

How to make entries in the file

Use the following template to link a procedure to one or more objects and operation types.

owner.object

{i | u | d | iud}

owner.procedure

where:

l owner.object is the owner and name of a target object, or a wildcarded entry. (See Syntax rules)
l i| u | d is the type of operation to be transformed by the specified procedure. You can specify any or all operation types, for example id or iud. Upper or lower case are both valid.
l owner.procedure is the owner and name of the procedure that will handle the specified object and operation type.

Syntax rules
l There must be at least one space between the object specification, the operation type specification, and the procedure specification.
l You can use the LIKE operator and a SQL wildcard (%) to specify multiple objects by using a search string. (See the Example.)
l You can use an underscore (_) to denote a single-character wildcard. For table names that contain an underscore character (for example emp_sal), SharePlex recognizes the backslash (\) as an escape character to denote the underscore as a literal and not a wildcard, for example: like:scott.%\_corp\_ emp. If you are not using the LIKE operator, the backslash escape character is not required if an object name contains an underscore.
l You a comment line anywhere in the file. Start a comment line with the pound symbol (#).

SharePlex 9.2.8 Administrator Guide 174

Example transformation file

scott.sal like:scott.%\_corp\_emp like:scott% scott.cust

IUD scott.sal_tr IUD scott.emp_tr1 IUD scott.emp_tr2 U scott.sal_tr

How to change the file during replication
You can change the transformation file any time during replication to add and remove tables and procedures. After you change the file, stop and re-start the Post process.

SharePlex 9.2.8 Administrator Guide 175

15
Configure security features
SharePlex provides a number of security features that help protect replicated data on the local system and during transfer across a network. This chapter provides guidelines for the configuration and use of these features.
Contents
Secure data with SSL/TLS Host Authentication Secure data with SSH Encrypt data between Export and Import Assign SharePlex users to security groups
Secure data with SSL/TLS
SharePlex provides a feature to enable SSL/TLS connections for all network traffic. This will encrypt data on the network between SharePlex instances and between SharePlex and the command line interface. If SSL/TLS is enabled, SharePlex will only accept SSL/TLS connections. For all SharePlex instances that replicate to each other, either all must have SSL/TLS enabled or all must have SSL/TLS disabled. The SSL/TLS configuration includes a network password. This password must be the same across all of your SharePlex instances. NOTE: For TLS connections, SharePlex supports TLS 1.2 (or later). To change the SSL/TLS configuration:
l Shutdown sp_cop on all nodes l Run "sp_security --setup" on all nodes l Start sp_cop on all nodes
SharePlex 9.2.8 Administrator Guide 176

sp_security
Use the sp_security utility to enable, disable or view the SSL/TLS settings for SharePlex network communication.

Enable SSL/TLS
IMPORTANT! SSL/TLS must be either enabled with a common network password or disabled on all SharePlex installations.

To enable SSL/TLS Run "sp_security --setup", select the SSL/TLS option, and then enter a network password.
% sp_security --setup

Security Setup Wizard --------------------This wizard will walk you through setting up the SharePlex network security.

Setup configuration for '/home/shareplex/var110/' and Port 2100 [N]: Y

Choose your network security model. Please note the following: * Cop must be down when the security model is changed, or when the
network password is changed * The same model must be used among all SharePlex nodes replicating to
each other * For security model [1], the same network password must be set on all
SharePlex nodes replicating to each other

[1] Use basic SSL/TLS connections [2] Use non-SSL/TLS connections (default prior to SharePlex 9.1.3)

Security model: 1

Please enter a network password that will be used for authentication among the SharePlex nodes. All SharePlex nodes that replicate data to each other must have the same network password.

Network password: Please re-enter the network password

Network password:

Security settings:

Configuration for '/home/shareplex/var110/' and Port 2100:

Security model Network password

: SSL/TLS : stored for unattended startup

SharePlex 9.2.8 Administrator Guide 177

SSL key file password SSL key file SSL cert file
Setup complete!

: stored for unattended startup : key.pem : cert.pem

Disable SSL/TLS
IMPORTANT! SSL/TLS must be either enabled with a common network password or disabled on all SharePlex installations.

To disable SSL/TLS Run "sp_security --setup" and select non-SSL/TLS connections.
% sp_security --setup

Security Setup Wizard --------------------This wizard will walk you through setting up the SharePlex network security.

Setup configuration for '/home/shareplex/var110/' and Port 2100 [N]: Y

[1] Use basic SSL/TLS connections [2] Use non-SSL/TLS connections (default prior to SharePlex 9.1.3)

Security model: 2

Security settings:

Configuration for '/home/shareplex/var110/' and Port 2100:

Security model

: Un-encrypted

Setup complete!

View current SSL/TLS configuration
To view the current SSL/TLS configuration Run "sp_security --show".

SharePlex 9.2.8 Administrator Guide 178

% sp_security --show

Security settings:

Configuration for '/home/shareplex/var110/' and Port 210:

Security model

: Un-encrypted

Host Authentication
SharePlex provides host authorization security that verifies whether or not SharePlex processes on specific remote systems are authorized to connect to the local system for service and command requests. To implement host authorization, you create an ASCII text file named auth_hosts in the data sub-directory of the SharePlex variable-data directory and then populate it with the names of systems being granted connection permission.

Requirements
l If used, the auth_hosts file must contain valid entries. If this file exists but is empty or contains invalid entries, SharePlex sends an error message similar to the following example to the Event Log: unauthorized connection attempt.
l If an auth_hosts file does not exist on a system, SharePlex accepts all requests from all systems that attempt to connect to sp_cop.
l The name of the local system must be the first non-commented line of this file, or host authorization will not function.
l All entries, including comments, must end with a return.
To configure the auth_hosts file NOTE: Begin comment lines with a pound character (#).
1. Run an ASCII text editor such as vi (Unix and Linux), NotePad (Windows), or WordPad (Windows) to open a blank file. If you are using a Unix and Linux text editor, change directories to the data subdirectory of the SharePlex variable-data directory before you run the editor.
2. On the first non-commented line, enter the full machine name of the local system, for example:Localhost.mycorp.com.
3. On the next non-commented line, enter one of the following:

Value all hostname

Description
Grants connection authorization to processes on all remote systems.
Grants connection authorization to the specified host. Enter the fully qualified machine name, for example remotehost.mycorp.com. Specify as many host names as needed, each on its own line.

4. Save the file as auth_hosts in the data sub-directory of the SharePlex variable-data directory. If running multiple instances of sp_cop, make certain to save the file to the correct variable-data directory.

SharePlex 9.2.8 Administrator Guide 179

Example
Note the name of the local host is on the first non-commented line. #Comment: first line is local host name. Localhost.mycorp.com #Comment: remaining lines are remote hosts. remotehost.mycorp.com remotehost2.mycorp.com remotehost3.mycorp.com
Secure data with SSH
SharePlex uses the SSH® Secure ShellTM utility to provide encryption for network services such as secure remote login and other services over an insecure network.
Requirements
l Purchase and install the SSH software. SSH is not included with SharePlex. l Using SSH with SharePlex requires the use of local port forwarding (also known as tunneling) within the
SSH configuration. Port forwarding allows you to establish a secure SSH session and then tunnel TCP connections through it. l SharePlex can be configured to work with SSH software between a source system and one target system. If a source replicates to multiple targets, only one of the routes can be configured with SSH. l This feature is supported on Unix and Linux. To set up SharePlex to use SSH 1. On the source and target systems, choose an available local port to be used as the tunnel port. For peerto-peer and high availability replication, the port must be the same number on both systems. For other replication strategies, choose a different port on each system.
SharePlex 9.2.8 Administrator Guide 180

2. On the source system, issue the following command from the command prompt. This command connects to the target system to set up the tunnel. $ ssh -L source_port:target_host:target_port userid@target_host -N -f where: l -L specifies that the specified port on the local host (acting as the client) is to be forwarded to the remote host and port. l source_port is the port number on the source system. l target_host is the name of the target system. l target_port is the port on the target system. l userid is your Unix and Linux user ID. You will be prompted for the password. l -N specifies not to execute a remote command. This is used just to forward a port (protocol version 2 only). l -f forces the SSH shell to work in the background just before command execution. If this argument is omitted, the terminal window you are using must be kept open. SSH cannot be started with nohup. Refer to your SSH documentation for more information about these commands.
3. (If using multiple SharePlex instances) On the source system, export the correct variable-data directory for the instance of sp_cop for which you are setting up SSH. ksh shell: export SP_SYS_VARDIR=/full_path_of_variable-data_directory csh shell: setenv SP_SYS_VARDIR=/full_path_of_variable-data_directory
4. On the source system, start sp_cop. 5. On the source system, run sp_ctrl from the bin subdirectory of the product directory. 6. In sp_ctrl, set the SP_XPT_USE_LOCALHOST parameter in one of the following ways.
l If there is only one target system, set the parameter with the following syntax: sp_ctrl> set param SP_XPT_USE_LOCALHOST 1
l If there are multiple targets, use the following command to set up a tunnel to the target that will use SSH. Replication to the other target systems will connect directly in the normal fashion. sp_ctrl> set param SP_XPT_USE_LOCALHOST to host 1
where: host is the name of the target system that will use the tunnel. 7. In sp_ctrl, use the list param command with the modified option to verify the parameter setting. If the
setting is correct, you can activate a configuration at this point. sp_ctrl> list param modified 8. If there is an active configuration, stop and then start sp_cop to make the new parameter setting active. To stop sp_cop: sp_ctrl> shutdown /productdir/bin/sp_cop & To start sp_cop: $ /productdir/bin/sp_cop &
SharePlex 9.2.8 Administrator Guide 181

Encrypt data between Export and Import
If you do not want to use SSL/TLS but still want to encrypt data between Export and Import, you might do so using this feature. SharePlex can be configured to encrypt replicated data across the network. SharePlex uses Advanced Encryption Standard (AES) encryption.
Encryption guidelines
Encryption must be enabled on the source and target systems. You enable encryption and set the size of the key through the Export process. You configure the Import process to ensure that encryption is enabled on the source, so that no data is sent across the network unless it is encrypted. When configuring encryption, follow these guidelines:
l Use one encryption key for all Export processes in the SharePlex instance. l To use encryption, SharePlex must be version 9.1 or later.
Encryption procedure
On the source system
1. Set the Export parameter SP_XPT_ENABLE_AES to 1. This enables encryption. sp_ctrl> set param sp_xpt_enable_aes 1
2. Run the create encryption key command to create the key. sp_ctrl> create encryption key
The following is an example key: E5F5D4CBA329D2C86B5D7ABA096C18600595490129F55A1422AAB0248B28D0E4 3. (Optional) Set the SP_XPT_AES_KEY_LENGTH parameter to increase the key size. The create encryption key command returns a randomly generated, 256-bit AES key. By default, SharePlex uses 128 bits of that length to encrypt the data. To increase the key length that SharePlex uses, set the SP_XPT_AES_KEY_LENGTH parameter to 192 or 256 bits. When you increase the length, the key is harder to hack but requires more CPU power.
sp_ctrl> set param sp_xpt_aes_key_length {192 | 256} Example: set param sp_xpt_aes_key_length 256 4. Run the set encryption key command. This adds the key to the Export configuration. sp_ctrl> set encryption key key_value Example: set encryption key E5F5D4CBA329D2C86B5D7ABA096C18600595490129F55A1422AAB0248B28D0E4 5. Restart Export to activate the settings. sp_ctrl> stop export sp_ctrl> start export
SharePlex 9.2.8 Administrator Guide 182

On the target system 1. Set the SP_IMP_ENABLE_AES parameter to 1. This prevents SharePlex on the target from accepting data that is not encrypted. 2. Run the set encryption key command with the same key value that you set for Export. The key values on the source and target must match. sp_ctrl>set encryption key key_value Example: set encryption key E5F5D4CBA329D2C86B5D7ABA096C18600595490129F55A1422AAB0248B28D0E4 3. Restart Import to activate the settings. sp_ctrl> stop import sp_ctrl> start import
View the encryption key
Issue this command on the source and target systems to ensure that both key values match. sp_ctrl> show encryption key
Assign SharePlex users to security groups
Contents
Overview
The SharePlex security groups provide access control to the SharePlex command and control system. Without proper configuration of these groups, anyone with permissions on the system can use the commands that view, configure, and control data replication.
About the SharePlex security groups
To monitor, control, or change SharePlex replication, a person must be assigned to one of the SharePlex security groups on the systems where he or she will be issuing commands. Each group corresponds to an authorization level, which determines which SharePlex commands a person can issue. To execute a command, a user must have that command's authorization level or higher. Use the authlevel command to determine your authorization level for issuing SharePlex commands on a system.
SharePlex 9.2.8 Administrator Guide 183

Description of the SharePlex security groups
Refer to the following table to determine the group and authorization level that you want to grant each SharePlex user.

User Authorization Levels and Roles

Auth User type level

User group

User roles

1 Administration spadmin* You need at least one user with Administrator rights on each source and target system.
Can issue all SharePlex commands. Commands that can only be issued by a SharePlex Administrator are:

l startup, shutdown l all configuration commands relating to an active configuration l all parameter commands except list param l start capture l stop capture l abort capture l truncate log

The SharePlex Administrator user must be in the Oracle dba group. For Oracle RAC and ASM 11gR2 and above, the user must also be in the Oracle Inventory group. For example: $ useradd g spadmin G dba,oinstall. The membership in Oracle Inventory group must be listed explicitly in the etc/group file.
On Unix and Linux, unless you install SharePlex as a root user, the SharePlex Administrator user and the SharePlex admin group must exist prior to installation.

Operator

spopr Can issue all SharePlex commands except those listed above.

Viewer

spview Can view lists, status screens, and logs to monitor replication only.

NOTE: The default name for the SharePlex administrator group is spadmin, but you can designate any group or specify any name for that group during installation.

Create and populate SharePlex groups on Unix and Linux
Where and when to create the SharePlex groups on Unix and Linux depends on whether you install SharePlex as a root or non-root user.
l If you install as non-root, create the groups in the /etc/group file before you run the SharePlex installer. In a cluster, create them on all nodes.*

SharePlex 9.2.8 Administrator Guide 184

l If you install SharePlex as a root user, you can direct the installer to create the groups in the /etc/group file. If you install in a cluster, the installer creates the groups on the primary node, but you must create them yourself on the other nodes. * The groups must exist because the installer adds the SharePlex Administrator user to the spadmin group during the installation process. In a cluster, this user is only added to the primary node. You must add the SharePlex Administrator user to the other nodes.
To create the groups in /etc/group # groupadd spadmin # groupadd spopr # groupadd spview
To assign a user to a group 1. Open the /etc/group file. 2. Add the Unix or Linux user name to the appropriate group. To assign a list of user names to a group, use a comma-separated list (see the following example). spadmin:*:102:spadmin,root,jim,jane,joyce,jerry If the password field is null, no password is associated with the group. In the example, the asterisk (*) represents the password, "102" represents the numerical group ID, and spadmin is the group. The group ID must be unique. 3. Save the file.
Users can verify their authorization levels by issuing the authlevel command in sp_ctrl.
Create and populate SharePlex groups on Windows
On Windows, the SharePlex groups are created in the Windows User Accounts control panel by the SharePlex installer. To assign users to these groups, use that control panel after you install SharePlex. Users can verify their authorization levels by issuing the authlevel command in sp_ctrl.
SharePlex 9.2.8 Administrator Guide 185

16
Start replication on your production systems
This chapter contains instructions for the initial startup of replication from a source database to a target database on production systems.
Contents
What is activation? Activation commands Requirements for activating a configuration Test the configuration before activation Frequently Asked Questions about activation How to activate multiple configuration files Activate replication with an Oracle hot backup on an active database Activate replication with an Oracle hot backup on a quiet database Activate replication with Oracle transportable tablespaces Activate replication with cold copy/transfer methods Activate replication from Oracle to Open Target
What is activation?
When you activate a configuration, through the activate config command in sp_ctrl, SharePlex does the following:
l Activates (read) the configuration file to build a series of internal structures that identify objects and routes. Only one configuration can be active for any given datasource at a time. Configurations for different datasources on a system can be active at the same time. For example, you can activate a configuration for each Oracle instance.
l Starts the processes that maintain the capture and replication of source transactions.
SharePlex 9.2.8 Administrator Guide 186

l (Oracle only) Reconciles replicated data with the copied data. SharePlex applies transactions that occurred after the copy was taken and discards transactions that occurred before the copy (and thus were applied by the copy and would be redundant if applied by replication). The reconcile process is only required for procedures that start replication while the source database is active.
The activation of a configuration generally proceeds as follows.
1. Assign an activation ID SharePlex assigns an activation ID number to each configuration activation and its associated replication processes and queues. A configuration can be activated many times, and this ID keeps track of each one.
2. Create an object cache SharePlex builds an object cache that records the standard metadata needed to support replication: the name, size, and type of columns, NOT-NULL constraints, and whether a column is part of a key. For tables using partitioned replication, additional information is stored.
3. Add a configuration change marker SharePlex places a configuration-change marker in the data stream. This marker directs sp_cop to generate a new set of replication processes and queues. If another configuration is active for the same datasource, the marker deactivates it, causing the removal of the old processes and queues after the data they contain is posted.
4. Lock the tables, add the activation marker, unlock (Oracle only) SharePlex locks the tables that are listed in the configuration file so that it can obtain information about them while they are in a read-consistent state. As many tables can be locked concurrently as there are locking threads available. When SharePlex locks a table, it places an activation marker in the data stream that tells the Capture process to start (or stop) replicating that table. NOTE: If an application uses NOWAIT locking on tables in the replication configuration, the NOWAIT could fail if it attempts to obtain a lock on an object that is already locked because it is being activated. SharePlex locks the following:
l All tables added to replication (new and reactivated configurations) l All tables removed from replication (reactivated configurations) l All tables where routes changed (reactivated configurations) Each table is locked for a very short time, just long enough to activate a table. Replication of each table begins as soon as its activation is complete. Should one or more table fail to activate, SharePlex continues with the activation of the other tables. Users can access the data in a source table when the activation lock is released.
Activation commands
Use sp_ctrl commands to activate, deactivate and view information about a configuration activation, as well as to reconcile ongoing changes with a copy. For more information about these commands, see the SharePlex Reference Guide.
SharePlex 9.2.8 Administrator Guide 187

Purpose

Command

Activate a configuration file

activate config Starts the activation and retains control of the sp_ctrl command line until the activation is finished.

View configuration statistics

show config
Shows statistics for active configurations. It displays the status of a configuration (active or inactive), the datasource, the date and time the configuration was activated, and the number of objects that are configured.

Deactivate a configuration file

deactivate config or abort config The purposes of these commands are different, but both deactivate a configuration file.
l deactivate config gracefully terminates replication for an active configuration. It stops all Capture activity for the configuration, posts all data currently in the queues, and removes the associated SharePlex processes and queues.
l abort config is a forceful deactivation. It stops all replication activity for the configured datasource on the source and target systems, whether or not data in the queues has been posted.

Deactivating or aborting a configuration stops replication. If users continue making changes to the configured objects, the source and target data can go out of synchronization.

Reconcile replicated changes with the copy

reconcile
(Valid for an Oracle source only) Coordinates the results of ongoing replication with a copy of the source data that is applied to the target system, so that changes that occurred before the copy are discarded.

View replication status

status

Shows a summary of the status of replication to help you ensure that processes are

running and to check for errors, warnings or notices.

View queue status

qstatus Shows statistics for the capture, post, and export queues.

Requirements for activating a configuration
Activation of a configuration is an event that requires a series of actions to be taken in a timely manner, uninterrupted. Therefore, it is important to have all of your preparations done ahead of time, leaving nothing in question. You can activate a configuration if you performed the minimal requirements outlined in this topic. NOTE: These instructions apply to Oracle source databases, unless otherwise noted.

SharePlex 9.2.8 Administrator Guide 188

Required authorization level
Only a SharePlex Administrator can activate a configuration. Additional users should be assigned to monitor SharePlex and perform basic operational procedures. For more information, see Assign SharePlex users to security groups on page 183.

Required setup
l Before you activate a configuration, make certain that the objects that you want to replicate exist in the source database.
l If a table will be partitioned, create those partitions before you activate a configuration to begin replication processing. Partitioning a table while it is actively replicating causes SharePlex to lose the identifying information it has compiled, and DML from that table partition will not be replicated. You can add a partition to a table already in replication, but you will need to reactivate the configuration to update that table in the SharePlex object cache.

Prerequisites
Make certain you satisfy the following prerequisites before you activate a configuration.

Requirement Understand how to start and stop the sp_cop program. Understand how to issue SharePlex commands.

Documentation to Read Run SharePlex Execute commands in sp_ctrl

Understand the commands you will use during activation.
Make certain your SharePlex configuration and setup are complete and any optional features are included in the configuration or setup.

Activation commands
Configure data replication Configure partitioned replication Configure named queues Configure security features Configure data transformation Configure error handling

Prepare the database to support replication.

See "Set up an Oracle environment for replication" in the Installation and Setup Guide for an Oracle Source

Plan and configure SharePlex to support your replication Configure a replication strategy strategy

Start the source and target databases.

Database documentation

Unless your applications only generate SharePlexsupported DDL, prevent DDL operations, including TRUNCATE, during activation. Where permitted, DML changes are the only permissible changes during activation.

Database documentation

SharePlex 9.2.8 Administrator Guide 189

Test the configuration before activation
Before you activate a configuration in production, perform a test activation by using the verify config command. If there are syntax errors, misspellings or duplicate entries in the configuration file, the entire activation will fail. This command will test the configuration to find and report the following conditions:
l Verify the syntax of the entries in the configuration file. l Report an error if the source object is not supported for replication by SharePlex. l Report if a host name specified in a route is unreachable. l Report if there are duplicate specifications for a single object. l Report if an object specification will be skipped and the reason why. l List the qualifying objects included under wildcard specifications. The verify config command does not verify how long the activation will take, nor will it verify the target objects or database connection (as represented by the database identifier listed in the routing map.) For more information, see the verify config command in the SharePlex Reference Guide.
Frequently Asked Questions about activation
What happens during activation?
During activation, SharePlex gets the information it needs to identify and understand the objects that are configured for replication and build routing maps. For more information, see What is activation? on page 186.
How long does activation take?
The length of time that activation takes varies, depending on the size, number and structure of the configured objects.
Do users have to stop access to the data?
An Oracle database can remain available for transactions. Each source table is briefly locked so SharePlex can build its internal object information. Then the lock is released. This happens very quickly and may not even be noticeable by users. However, if a business application has NOWAIT locking, the SharePlex lock may cause the application to fail if it attempts to obtain a lock on an object that is being activated.
Can DDL be performed during activation?
No. The definition of objects cannot be changed during activation.
SharePlex 9.2.8 Administrator Guide 190

Can I activate multiple configurations for the same datasource?
Yes. You can create different configurations for the same datasource. You must use multiple instances of SharePlex (different sp_cop processes and variable-data directories, running on different ports). See Run multiple instances of SharePlex on page 43
Can I activate configurations for different datasources at the same time?
Yes. For more information, see How to activate multiple configuration files on page 191.
Can I test a configuration before I activate it?
Yes. This is highly recommended. Use the verify config command. For more information, see Requirements for activating a configuration on page 188.
Can I interrupt an activation to make changes?
No. Activation is meant to be an uninterrupted procedure that initiates replication while maintaining the same series of transactional events as those appearing in the redo log. You can use the abort config command to terminate an activation, but you may then need to clean out the queues and resynchronize the source and target data again. Ideally, you should have a tested configuration file ready to be activated and be prepared to issue commands to the database and to SharePlex.
Can I activate against a quiet database?
Yes. A quiet database is optional to activate Oracle capture. The Oracle database must be mounted and open, however.
Can I increase the number of activation threads?
Yes, but only for an Oracle source. The number of threads is controlled globally by the SP_OCF_ THREAD_COUNT parameter, which must be set before you issue the activate config command. You can override this value for a particular activation by using the threads option when you issue the activate config command.
How to activate multiple configuration files
You can activate one configuration per datasource on a system. For example, if there are ConfigA, ConfigB and ConfigC for the same datasource, you can activate only one of them at a time. Activating another configuration for the same datasource automatically deactivates the first one. However, if ConfigA replicates data from one datasource, while ConfigB replicates data from a different datasource, you can activate both of those configurations at the same time. Replication for those configurations can operate concurrently.
To activate multiple configuration files The activation process retains control of the sp_ctrl interface until the activation is finished. Because an activation can take a significant amount of time, you can activate different configurations in different sessions of
SharePlex 9.2.8 Administrator Guide 191

sp_ctrl. Activate the first configuration, then open another session of sp_ctrl to activate the second one. Open as many sessions of sp_ctrl as you have configurations to activate.
The number of activation threads can be controlled for an Oracle source. To set the number of threads, use the SP_OCF_THREAD_COUNT parameter, which is a global setting for the instance of SharePlex. When activating more than one configuration concurrently, you can override this value for a particular activation by using the threads option when you issue the activate config command.

Activate replication with an Oracle hot backup on an active database
Use this procedure to use an Oracle hot backup to establish a target Oracle instance and activate replication without quieting the source database. This procedure involves using the reconcile command to ensure that transactions which occurred after the point of backup are applied to the target, while eliminating redundant replicated transactions that were already captured by the backup.

Preliminary considerations
Read these points before you proceed.

Supported databases
Oracle source and Oracle target

Supported replication strategies
All replication strategies are supported with the following limitations:

Limitation applies to:

Description

Consolidated replication (many sources to one target)

To establish consolidated replication, the use of a hot backup from all source systems is not possible. A backup from one source will override the data that was applied by a backup from a different source. You can use a hot backup of one of the source instances to establish a target instance, and then use another copy method to apply the objects from the other source instances. Possible methods include:

l Export/import. For more information, see Activate replication with cold copy/transfer methods on page 202.
l Transportable tablespaces. For more information, see Activate replication with Oracle transportable tablespaces on page 200.

Peer-to-peer To establish peer-to-peer replication, you must:

1. Quiet all of the systems except the trusted source system for the duration of this procedure.
2. Move all users to the trusted source system, and then follow this procedure.

SharePlex 9.2.8 Administrator Guide 192

Limitation applies to:
Windows systems

Description
Only after this procedure has been performed on all of the secondary systems may users may resume activity on them.
To use a hot backup between Windows systems, the target system must have an instance already created containing an identical ORACLE_SID and directory structure created with the Oracle creation tools. Oracle runs as a service on Windows, and the Registry entries must exist before starting the database recovery process. The database can start empty, because the hot backup will populate it.

Requirements
l [Unix and Linux systems] Verify that the ORACLE_SID and ORACLE_HOME in the oratab file are correct for the instance you will be establishing with the hot backup. The SID must be the SID used in the routing map in the configuration file that you will be activating.
l Read the requirements before you start this procedure. For more information, see Requirements for activating a configuration on page 188.
l Make certain a SharePlex database account exists in the source database (only). This account usually is created when SharePlex first is installed. See the SharePlex Installation and Setup Guide for more information.
l Before you start, review this procedure and see the SharePlex Reference Guide for more information about the commands that are used.

Troubleshooting
If the configuration fails to activate, you can find information about the failure in these places: l Use the show log command to view the event_log. l View the activation process log, which is a file named SID_oconf##.log in the log sub-directory of the SharePlex variable data directory.
See also Solve Database Setup problems on page 228.

Procedures
There are two procedures for activation with a hot backup, depending on your replication strategy. Activation with hot backup: all strategies except cascading Activation with hot backup: cascading replication
Activation with hot backup: all strategies except cascading
Use this procedure for all replication strategies except cascading replication where SharePlex will be posting to a database on the intermediary system.

SharePlex 9.2.8 Administrator Guide 193

1. On the source and target systems, go to the bin sub-directory of the SharePlex product directory, and start sp_cop and sp_ctrl. In a cluster, the source is the primary node where the cluster VIP is running.
2. On both systems, verify that the SharePlex processes are running. sp_ctrl> status
3. On the target system (primary node of a target cluster), stop the Post process. This allows replicated data to accumulate in the post queue until the database has been recovered and reconciled. sp_ctrl> stop post
4. On the source system, run the Oracle hot backup. 5. When the backup is finished, activate the configuration on the source system.
sp_ctrl> activate config filename 6. On the source system, monitor activation status.
NOTE: The command retains control of sp_ctrl until activation is finished. 7. When activation is complete, switch log files on the source system.
On-premises database: svrmgr1> alter system switch logfile;
Amazon RDS database: Run the Amazon RDS procedure rdsadmin.rdsadmin_util.switch_logfile.
8. Do one of the following:
l To recover the database to a sequence number, make a note of the highest archive-log sequence number.
l To recover the database to a Oracle System Change Number (SCN), pick an SCN to recover to on the target database.
9. On the target system, do one of the following: l If recovering to a sequence number, recover the database from the hot backup using the UNTIL CANCEL option in the RECOVER clause, and cancel the recovery after Oracle has fully applied the log from the previous step. l If recovering to a SCN, recover the database from the hot backup using the UNTIL CHANGE scn option in the RECOVER clause, and cancel the recovery after Oracle has applied the logs matching the SCN from the previous step.
10. On the target system, open the database with the RESETLOGS option. 11. On the target system, run Database Setup for Oracle on the database. When prompted for the SharePlex
database user, enter n to choose the existing user and password (these were copied in the backup). Would you like to create a new SharePlex user [y]. n
NOTES: l SharePlex can remain running during the setup process. l Database Setup Utilities in the SharePlex Reference Guide.
12. [Optional] If you are using named post queues and are unsure of the queue names, issue the qstatus command. sp_ctrl> qstatus
SharePlex 9.2.8 Administrator Guide 194

13. On the target system, issue the reconcile command as follows, depending on the recovery option you chose. If you are using named post queues, issue the command for each one.
l If recovering to a sequence number, substitute the sequence number of the log that you noted previously. sp_ctrl> reconcile queue queuename for datasource-datadest seq sequence_ number Example: reconcile queue SysA for o.oraA-o.oraA seq 1234
l If recovering to a SCN, substitute the SCN that you noted previously. sp_ctrl> reconcile queue queuename for datasource-datadest scn scn_number Example: reconcile queue SysA for o.oraA-o.oraA scn 0123456789
NOTE: The command retains control of sp_ctrl until the reconcile process is finished. 14. On the target system, run the cleanup.sql script to truncate the SharePlex internal tables. Instructions for
running this script are in the SharePlex Reference Guide. 15. On the target system, disable triggers on the tables, or run the sp_add_trigger.sql utility script so that the
triggers ignore the SharePlex user. 16. On the target system, disable check constraints and scheduled jobs that perform DML. 17. [Partitioned replication only] If you are using vertically partitioned or horizontally partitioned replication
for any tables, delete the unneeded columns and rows from those tables. 18. [High availability only] On the target (secondary) system, stop Export.
sp_ctrl> stop export 19. [High availability and peer-to-peer only] On the target (secondary) system, activate the configuration so
that SharePlex is ready in the event of failover. sp_ctrl> activate config filename
20. On the target system, start the Post process. The two instances are now in synchronization, and SharePlex will continue replicating to maintain synchronization. sp_ctrl> start post
21. [Optional] If this was only a partial backup, drop the tablespaces that were not copied over during the hot backup.
Activation with hot backup: cascading replication
Use this procedure for cascading replication where SharePlex will be posting to a database on the intermediary system. You will apply the backup to the intermediary system first (represented as "sysB"), and then to the target system (represented as "sysC").
1. On all systems, go to the bin sub-directory of the SharePlex product directory, and start sp_ cop and sp_ctrl
2. On all systems, verify that the SharePlex processes are running. sp_ctrl> status
3. On the intermediary and target systems, stop the Post process. This allows replicated data to accumulate in the post queue until the databases are recovered.
SharePlex 9.2.8 Administrator Guide 195

sp_ctrl> stop post

5. On the source system, run the Oracle hot backup to the intermediary and target systems.

6. When the backup is finished, activate the configuration on the source system. sp_ctrl> activate config filename

7. On the source system, view activation status. NOTE: The command retains control of sp_ctrl until activation is finished.

8. When activation is complete, switch log files on the source system. On-premises database: svrmgr1> alter system switch logfile; Amazon RDS database: Run the Amazon RDS procedure rdsadmin.rdsadmin_util.switch_logfile.

9. Make a note of the highest archive-log sequence number.

10. On the intermediary system, recover the database from the hot backup using the UNTIL CANCEL option in the RECOVER clause, and cancel the recovery after Oracle has fully applied the log from the previous step.

11. On the intermediary system, open the database with the RESETLOGS option.

12. On the intermediary system, run Database Setup on the database. When prompted for the SharePlex database user, enter n to choose the existing user and password (these were copied in the backup). Would you like to create a new SharePlex user [y]. n NOTES:

l SharePlex can remain running during the setup process.
l For more information, see Database Setup Utilities in the SharePlex Reference Guide.
13. [Optional] If you are using named post queues and are unsure of the queue names, issue the qstatus command. sp_ctrl> qstatus

14. On the intermediary system, issue the reconcile command for each post queue. For seq sequence_ number, substitute the sequence number of the log that you noted previously.

sp_ctrl> reconcile queue queuename for datasource-datadest seq sequence_number Example: reconcile queue SysA for o.oraA-o.oraA seq 1234

15. On the intermediary system, run the cleanup.sql script to truncate all of the SharePlex internal tables. Instructions for running this script are in the SharePlex Reference Guide.

16. On the intermediary system, disable triggers on the tables, or run the sp_add_trigger.sql utility script so that the triggers ignore the SharePlex user.
17. On the intermediary system, disable check constraints and scheduled jobs that perform DML.

18. On the intermediary system, set the SP_OCT_REPLICATE_POSTER parameter to 1. This directs SharePlex to capture posted changes on that system and replicate them to the target system. sp_ctrl> set param SP_OCT_REPLICATE_POSTER 1

SharePlex 9.2.8 Administrator Guide 196

19. [Partitioned replication only] If you are using vertically partitioned or horizontally partitioned replication for any tables, delete the unneeded columns and rows from those tables. IMPORTANT! Do not start any Post processes yet.
20. On the target system, recover the database from the hot backup using the UNTIL CANCEL option in the RECOVER clause, and cancel the recovery after Oracle has fully applied the log that you reconciled to in the previous steps taken on the intermediary system.
21. On the target system, open the database with the RESETLOGS option. 22. On the target system, run Database Setup on the database. When prompted for the SharePlex database
user, enter n to choose the existing user and password (these were copied in the backup). Would you like to create a new SharePlex user [y]. n NOTE: SharePlex can remain running during the setup process. For more information about Database Setup, see Database Setup Utilities in the SharePlex Reference Guide. 23. On the target system, run the cleanup.sql script to truncate the SharePlex internal tables. Instructions for running this script are in the SharePlex Reference Guide. 24. On the target system, disable triggers on the tables, or run the sp_add_trigger.sql utility script so that the triggers ignore the SharePlex user. 25. On the target system, disable check constraints and scheduled jobs that perform DML. 26. [Partitioned replication only] If you are using vertically partitioned or horizontally partitioned replication for any tables, delete the unneeded columns and rows from those tables. 27. On the intermediary system, activate the configuration file.
sp_ctrl> activate config filename 28. On the intermediary system, monitor activation status.
NOTE: The command retains control of sp_ctrl until activation is finished. 29. On the intermediary and target systems, start the Post process. All instances are now in synchronization,
and SharePlex will continue replicating to maintain synchronization. sp_ctrl> start post
30. [Optional] If this was only a partial backup, drop the tablespaces that were not copied over during the hot backup.
Activate replication with an Oracle hot backup on a quiet database
Use this procedure to use an Oracle hot backup to establish a target Oracle instance and activate replication if user activity can be stopped while the procedure is performed. This procedure can be used for all replication configurations.
Preliminary considerations
Read these points before you proceed.
SharePlex 9.2.8 Administrator Guide 197

Supported databases
Oracle source and Oracle target

Supported replication strategies
All but high-availability. This procedure is not appropriate for a high-availability strategy because it requires the source database to be quieted while the backup is taken and the configuration file is being activated. Certain limitations apply:

Limitation applies to:

Description

Consolidated replication (many sources to one target)

Windows systems

To use a hot backup between Windows systems, the target system must have an instance already created containing an identical ORACLE_SID and directory structure created with the Oracle creation tools. Oracle runs as a service on Windows, and the Registry entries must exist before starting the database recovery process. The database can start empty, because the hot backup will populate it.

Requirements
l [Unix and Linux systems] Verify that the ORACLE_SID and ORACLE_HOME in the oratab file are correct for the instance you will be establishing with the hot backup. The SID must be the SID used in the routing map in the configuration file that you will be activating.
l Read the requirements before you start this procedure. For more information, see Requirements for activating a configuration on page 188.
l Users must stop accessing the production database while the hot backup and configuration activation take place.
l Make certain a SharePlex database account exists in the source database (only). This account usually is created during SharePlex installation. See the SharePlex Installation and Setup Guide for more information.
l Before you start, review this procedure and see the SharePlex Reference Guide for more information about the commands that are used.

Procedure
NOTE: If you are not using cascading replication, ignore all references to an intermediary system. For more information, see Configure replication through an intermediary system on page 143.

SharePlex 9.2.8 Administrator Guide 198

1. On the source system, complete the Oracle hot backup. 2. On the source system, stop user access to the source database by shutting it down and opening it in
restricted mode. 3. On the source system, switch the redo logs.
On-premises database: svrmgr1> alter system switch logfile;
Amazon RDS database: Run the Amazon RDS procedure rdsadmin.rdsadmin_util.switch_logfile.
4. Keep a record of the sequence number of the current log. 5. On all systems, start sp_cop and sp_ctrl from the bin sub-directory of the SharePlex product directory. 6. On all systems, verify that sp_cop and sp_ctrl are running.
sp_ctrl> status 7. On the intermediary and target systems, stop Post. Stopping Post allows replicated data to accumulate in
the post queue until the databases have been recovered. sp_ctrl> stop post
8. On the source system, activate the configuration file. sp_ctrl> activate config filename
9. On the source system, view activation status. NOTE: The command retains control of sp_ctrl until activation is finished.
10. When the activation is finished, allow users to resume access to the source database. 11. List the archive logs on the intermediary and target systems. Delete any logs made after the one for
which you made a record. 12. On the intermediary and target systems, recover the database to the log number that you recorded. Make
sure a full recovery is performed. 13. On the intermediary and target systems, open the database. 14. On the intermediary and target systems, run the Database Setup utility for the target instance. When
prompted for the SharePlex database user, enter n to choose the existing user and password (these were copied in the backup).
Would you like to create a new SharePlex user [y].n NOTES:
l SharePlex can remain running during the setup process. l For more information about Database Setup, see Database Setup Utilities in the SharePlex
Reference Guide. 15. On the intermediary and target systems, run the cleanup.sql script to truncate the SharePlex internal
tables. Instructions for running this script are in the SharePlex Reference Guide. 16. On the intermediary and target systems, disable triggers on the tables, or run the sp_add_trigger.sql
utility script so that the triggers ignore the SharePlex user. 17. On the intermediary and target systems, disable check constraints and scheduled jobs that perform DML.
SharePlex 9.2.8 Administrator Guide 199

18. [Partitioned replication only] If you are using vertically partitioned or horizontally partitioned replication for any tables, delete the unneeded columns and rows from those tables on the intermediary and target systems.
19. [Intermediary system only] On the intermediary system, set the SP_OCT_REPLICATE_POSTER parameter to 1. This directs SharePlex to capture posted changes on that system and replicate them to the target system. sp_ctrl> set param SP_OCT_REPLICATE_POSTER 1
20. On the intermediary system, activate the configuration file. sp_ctrl> activate config filename
21. On the intermediary system, monitor activation status. NOTE: The command retains control of sp_ctrl until activation is finished.
22. When activation is finished, start the Post process on the intermediary and target systems. All instances are now in synchronization, and SharePlex will continue replicating to maintain synchronization. sp_ctrl> start post
23. [Optional] If this was only a partial backup, drop the tablespaces that were not copied over during the hot backup.
Activate replication with Oracle transportable tablespaces
Use this procedure to use the Oracle transportable tablespaces feature to establish a target Oracle instance and activate replication. It enables you to synchronize and resynchronize numerous objects quickly and with minimal downtime. It allows you to export just the metadata (data dictionary) and then copy the data files. This method also moves indexes so that there is no need to rebuild them in the target database, and you can move multiple tablespaces at one time. NOTE: This document does not provide instructions for how to use transportable tablespaces. This procedure should be performed by someone who has a solid understanding of database copy methods.
Preliminary considerations
Read these points before you proceed.
Supported databases
Oracle source and Oracle target
Supported replication strategies
All replication strategies. This procedure may not appropriate for a high-availability strategy if the source database cannot be quieted even briefly.
SharePlex 9.2.8 Administrator Guide 200

Requirements
l Read the requirements before you start this procedure. For more information, see Requirements for activating a configuration on page 188.
l Make certain a SharePlex database account exists in the source database (only). This account usually is created when SharePlex is first installed. See the SharePlex Installation and Setup Guide for more information.
l Before you start, review this procedure and see the SharePlex Reference Guide for more information about the commands that are used.
Naming conventions used
In this procedure, the "source" system is one of the following: l The source system of a single-direction replication configuration, including cascading replication. l All source systems of a consolidated replication configuration. l The trusted source system in a peer-to-peer replication configuration. l The primary node of a cluster (where the cluster VIP is running).
In this procedure, the "intermediary" system only needs to be part of this procedure if SharePlex will be posting to, and capturing from, an intermediary system in a cascading configuration. In this procedure, the "target" system is one of the following:
l The target system of a single-direction replication configuration, including cascading and consolidated replication.
l The secondary systems in a peer-to-peer replication configuration. l The primary node (where the cluster VIP is running) of the target cluster. In this procedure, the SharePlex commands in the procedure apply to all sp_cop instances that apply to the replication strategy you are using (for example, all sp_cop processes on a target in consolidated replication).
Procedure
1. On the source system, set the source tablespaces that you want to copy to READ ONLY. svrmgr1> alter tablespace name read only;
2. On the source system, activate the configuration file. sp_ctrl> activate config filename
3. On the source system, start sp_cop and sp_ctrl from the bin sub-directory of the SharePlex product directory.
4. On the source system, verify that sp_cop and sp_ctrl are running. sp_ctrl> status
5. On the intermediary and target systems, stop Post. Stopping Post allows replicated data to accumulate in the post queue until the databases have been recovered. sp_ctrl> stop post
6. On the source system, export the metadata to an export file.
SharePlex 9.2.8 Administrator Guide 201

7. When the export is finished, copy the datafiles to another location on the source system. This minimizes the impact on the source database of copying the files to the target system.
8. Set the source tablespaces back to read/write mode. svrmgr1> alter Tablespace name read write;
9. If any of the copied datafiles and tablespaces exist in the intermediary or target database, drop them so that the copied files can be applied.
10. Copy the files from the new location on the source system to the intermediary and target systems. 11. On the intermediary and target systems, use the Oracle import utility to import the metadata and the
tablespace definitions. 12. On the intermediary and target systems, set the tablespace(s) to read/write mode. 13. On the intermediary and target systems, open the Oracle instances. 14. On the intermediary and target systems, disable triggers on the tables, or run the sp_add_trigger.sql
utility script so that the triggers ignore the SharePlex user. 15. On the intermediary and target systems, disable check constraints and scheduled jobs that perform DML. 16. [Partitioned replication only] If you are using vertically partitioned or horizontally partitioned replication
for any tables, delete the unneeded columns and rows from those tables on the intermediary and target systems. 17. [Intermediary system only] Set the SP_OCT_REPLICATE_POSTER parameter to 1. This directs SharePlex to capture posted changes on that system and replicate them to the target system. sp_ctrl> set param SP_OCT_REPLICATE_POSTER 1 18. [Intermediary system only] Activate the configuration file.
sp_ctrl> activate config filename 19. [High availability] On the target system, stop the Export process.
sp_ctrl> stop export 20. [High availability and peer-to-peer replication] Activate the configuration on the target system(s).
sp_ctrl> activate config filename 21. Start Post on the intermediary and target systems. SharePlex begins executing the SQL statements that
have been collecting in the post queue, keeping the source and target data in sync. sp_ctrl> start post
22. [Peer-to-peer replication] Allow users to access the databases on all systems.
Activate replication with cold copy/transfer methods
Use this procedure to synchronize the source and target data with the following utilities:
l Import/Export/Data Pump l Store/Restore from tape l FTP
SharePlex 9.2.8 Administrator Guide 202

NOTE: This document does not provide instructions for how to perform the chosen copy method. This procedure should be performed by someone who has a solid understanding of database copy methods.
Preliminary considerations
Read these points before you proceed.
Supported databases
Oracle source and Oracle target
Supported replication strategies
All but high-availability. This procedure is not appropriate for a high-availability strategy because it requires the source database to be quieted while the configuration file is being activated.
Requirements
l [Unix and Linux systems] Verify that the ORACLE_SID and ORACLE_HOME in the oratab file are correct for the instance you will be establishing with the hot backup. The SID must be the SID used in the routing map in the configuration file that you will be activating.
l Read the requirements before you start this procedure. For more information, see Requirements for activating a configuration on page 188.
l Users must stop accessing the production database while the copy and configuration activation take place.
l The target instance must exist. l Make certain SharePlex database accounts exist in the source and target databases. This account
usually is created during installation. See the SharePlex Installation and Setup Guide for more information. l Before you start, review this procedure and see the SharePlex Reference Guide for more information about the commands that are used.
Naming conventions used
In this procedure, the "source" system is one of the following: l The source system of a single-direction replication configuration, including cascading replication. l All source systems of a consolidated replication configuration. l The trusted source system in a peer-to-peer replication configuration. l The primary node of a cluster (where the cluster VIP is running).
In this procedure, the "intermediary" system only needs to be part of this procedure if SharePlex will be posting to, and capturing from, an intermediary system in a cascading configuration. In this procedure, the "target" system is one of the following:
SharePlex 9.2.8 Administrator Guide 203

l The target system of a single-direction replication configuration, including cascading and consolidated replication.
l The secondary systems in a peer-to-peer replication configuration. l The primary node (where the cluster VIP is running) of the target cluster.
In this procedure, the SharePlex commands in the procedure apply to all sp_cop instances that apply to the replication strategy you are using (for example, all sp_cop processes on a target in consolidated replication).
Procedure
1. On the source system, stop user access to the objects that are in the replication configuration. l If deploying consolidated replication, you can either stop access to all of the source systems at once and make the copies at the same time, or you can synchronize each source system one at a time using these instructions. l If deploying peer-to-peer replication, stop access to all databases in the peer group, including the trusted source.
2. Copy the files from the source system to the intermediary and target systems. 3. On the source system, start sp_cop and sp_ctrl. 4. On the source system, activate the configuration file (all files if using consolidated replication).
sp_ctrl> activate config filename 5. On the intermediary and target systems, start sp_cop and sp_ctrl. 6. On the intermediary and target systems, stop Post. Stopping Post allows any data that gets replicated
before the target data is established to collect in the post queue. sp_ctrl> stop post
7. On the source system, allow users to resume access to the source database. 8. On the source system, verify that the sp_cop, Capture, and Read processes are running.
sp_ctrl> status 9. Start and mount the intermediary and target databases, but do not allow users access. 10. On the intermediary and target systems, apply the copy to the database. 11. On the intermediary and target systems, disable triggers on the tables, or run the sp_add_trigger.sql
utility script so that the triggers ignore the SharePlex user. 12. On the intermediary and target systems, disable check constraints and scheduled jobs that perform DML. 13. [Partitioned replication only] If you are using vertically partitioned or horizontally partitioned replication
for any tables, delete the unneeded columns and rows from those tables on the intermediary and target systems. 14. [Intermediary system only] Set the SP_OCT_REPLICATE_POSTER parameter to 1. This directs SharePlex to capture posted changes on that system and replicate them to the target system. sp_ctrl> set param SP_OCT_REPLICATE_POSTER 1 15. [Intermediary system only] Activate the configuration file.
sp_ctrl> activate config filename
SharePlex 9.2.8 Administrator Guide 204

16. [Peer-to-peer] Activate the configuration file on the target systems. 17. Start Post on:
l The intermediary system l The trusted source and all other targets in a peer group l All other targets NOTE: SharePlex will start executing SQL statements that accumulated in the post queue. 18. [Peer-to-peer] On the target systems in the peer group, allow users to resume access to the database.
Activate replication from Oracle to Open Target
Use this procedure to synchronize an Oracle source database with an Open Target database. SharePlex replicates the Oracle data changes and maintains them in the Post queue until the target is established with the copy. When the target is ready, you run the SharePlex reconcile feature, which ensures that Post only applies the operations that occurred after the copy and discards operations that were committed to the source before the copy.
Preliminary considerations
Read these points before you proceed.
Supported databases
Oracle source and any supported target
Supported replication strategies
All
Requirements
l You can use your Oracle RMAN backup system to take a hot backup of your primary instance and recover to an SCN or sequence number in a staging instance.
l This document does not provide instructions for how to perform the chosen copy method. Someone with expertise in database copy methods should perform this procedure. You can use Toad Data point or a third-party tool to extract data from the staging instance into the Open Target database.
l Read the requirements for activating a configuration file. For more information, see Requirements for activating a configuration on page 188.
l Before you start, review this procedure and see the SharePlex Reference Guide for more information about the commands that are used.
SharePlex 9.2.8 Administrator Guide 205

Procedure
1. On the source and target systems, start sp_cop and sp_ctrl from the bin sub-directory of the SharePlex product directory.
2. On source and target systems, verify that the SharePlex processes are running. sp_ctrl> status
3. On the target system, stop the Post process. This allows replicated data to accumulate in the post queue until the target database is instantiated and reconciled. sp_ctrl> stop post
4. Activate the configuration on the source system. sp_ctrl> activate config filename
5. On the source system, monitor activation status. NOTE: The command retains control of sp_ctrl until activation is finished.
6. When activation is complete, start the hot backup to the staging instance. 7. When the hot backup is finished, switch log files on the primary source system twice.
On-premises database: svrmgr1> alter system switch logfile; svrmgr1> alter system switch logfile;
Amazon RDS database: Run the Amazon RDS procedure rdsadmin.rdsadmin_util.switch_logfile twice.
8. Copy the archive logs that were generated by the log switch from the primary instance to the staging instance.
9. Do one of the following: a. If the source is RAC, recover the database on the staging server to the latest SCN of the last archive log that was copied to the staging server. b. If the source is not RAC, recover to the sequence number of the last archive log that was copied to the staging server. NOTE: The next steps apply the replicated changes that occurred after the backup point.
10. Do one of the following: l If the source is RAC, make a note of the SCN that you recovered to on the staging server. l If source is non-RAC, make a note of the log sequence number that you recovered to on the staging server.
11. Using the copy method of your choice, make a copy of the Oracle data from the staging server to the Open Target database. Wait until the copy is finished before proceeding to the next step.
12. [Optional] If you are using named post queues and are unsure of the queue names, issue the qstatus command and make a note of them. sp_ctrl> qstatus
13. On the target system, disable triggers on the target tables. 14. On the target system, disable check constraints and scheduled jobs that perform DML.
SharePlex 9.2.8 Administrator Guide 206

15. On the target system, run sp_ctrl, then issue one of the following reconcile commands. If you are using named post queues, issue the command for each one. l If the source is non-RAC, reconcile to the log sequence number of the log that you noted previously. sp_ctrl> reconcile queue queuename for datasource-datadest seq sequence_ number Example: reconcile queue SysA for o.prod1-r.rep1 seq 1234 l If the source is RAC, reconcile to the SCN that you noted previously. sp_ctrl> reconcile queue queuename for datasource-datadest scn scn_number Example: reconcile queue SysA for o.prod1-r.rep1 scn 0123456789 NOTE: The command retains control of sp_ctrl until the reconcile process is finished.
16. On the target system, start the Post process. The two instances are now in synchronization, and SharePlex will continue replicating to maintain synchronization. sp_ctrl> start post
SharePlex 9.2.8 Administrator Guide 207

17
Monitor SharePlex
This chapter contains an overview of the tools that SharePlex provides to detect errors and monitor the replication processes. Like any mission-critical software, SharePlex should be monitored regularly for situations or events that could interfere with processing, especially those that could result in loss of data synchronization.
Contents
View and terminate SharePlex processes View events and errors Monitor with sp_ctrl commands Run monitor scripts on Unix Run a monitor script on Windows Monitor replication with SNMP
View and terminate SharePlex processes
These instructions show you how to forcefully terminate SharePlex processes in cases where replication must be shut down immediately. In normal circumstances, follow the shutdown instructions in the following topics: Run SharePlex on Unix and Linux Run SharePlex on Windows
View and terminate processes on Unix and Linux
On Unix and Linux systems, you can use the ps -ef | grep sp_ command to view the SharePlex processes that are running.
SharePlex 9.2.8 Administrator Guide 208

l The sp_cop process is the root process. l The following child processes are spawned by sp_cop on a source system:
o Command and Control process (sp_cnc) o Capture (sp_ocap) o Read (sp_ordr) o Export (sp_xport) l The following child processes are spawned by sp_cop on a target system: 1. Command and Control process (sp_cnc) 2. Import (sp_mport) 3. Post (sp_opst_mt if the database is Oracle or sp_xpst if the database is Open Target) Each child process has the same -uidentifier as its parent sp_cop process. This makes it easier to identify related processes when multiple session of sp_cop are running.
To terminate a SharePlex process on Unix and Linux $ killPID
Or... $ kill -9PID
View and terminate SharePlex processes on Windows
On Windows systems, SharePlex is managed by the NuTCRACKER operating environment, rather than the Windows operating system. However, you can view and terminate SharePlex processes in the normal manner, as you would other Windows programs.
View SharePlex processes on Windows
You can view the SharePlex processes on Windows systems in the following ways: l From the Command Prompt console using the tlist program provided with the SharePlex software. l From the Windows Task Manager.
In the Windows Task Manager, SharePlex appears as Sp_Copsrv.exe, representing the SharePlex sp_cop process. The operating system controls the parent Sp_Copsrv.exe service. The parent Sp_Copsrv.exe process spawns child Sp_Copsrv.exe processes -- one for each replication process (Capture, Read, Export, Import, Post, sp_ctrl, and so forth.). For a standard uni-directional configuration replicating through default queues to one target system, there are following processes on a Windows system: On the source system:
l One parent Sp_Copsrv.exe process. l One Sp_Ocap or Sp_capture (Capture) process plus one child Sp_Copsrv.exe process.
SharePlex 9.2.8 Administrator Guide 209

l One Sp_Ordr (Read) process plus one child Sp_Copsrv.exe process. l One Sp_Xport (Export) process plus one child Sp_Copsrv.exe process. l If there are any additional SharePlex processes running, such as sp_ctrl, there is an additional Sp_
Copsrv.exe process for each one. On the target system:
l One parent Sp_Copsrv.exe process. l One Sp_Mport (Import) process plus one child Sp_Copsrv.exe process. l One Sp_Opst_Mt (Oracle Post) or Sp_Xpst (Open Target Post) process plus one child Sp_
Copsrv.exe process. l If there are any additional SharePlex processes running, such as sp_ctrl, there is an additional Sp_
Copsrv.exe process for each one. If there are no active replication configurations, the SharePlex processes do not start when you start the service, and just the parent Sp_Copsrv.exe will be running. To identify the parent Sp_Copsrv.exe process in the Windows Task Manager, look for the one that is using the largest amount of memory. The child Sp_Copsrv.exe processes consume less memory than the parent process. To identify which replication process is associated with a child Sp_Copsrv.exe process, look in the SharePlex Event Log for the message stating when the replication process started. This entry provides the PID for that process and the PID of the associated Sp_copsvr.exe process.
Terminate SharePlex processes on Windows
You can terminate SharePlex processes on Windows systems as follows: l Windows Task Manager: Terminate one SharePlex process at a time l KillSharePlex.bat: Terminate all SharePlex processes at once with a batch file.
To terminate a process with Task Manager 1. Press Control | Alt | Delete. 2. Select Start Task Manager or Task Manager, depending on your Windows version. 3. Select the Processes tab. 4. (Optional) Sort the processes by name. 5. Select the process that you want to kill. 6. Click End Process.
To terminate all processes with KillSharePlex.bat Run KillSharePlex.bat from the bin sub-directory of the SharePlex product directory. NOTE: This tool terminates all SharePlex processes without displaying them. There are no user prompts or confirmation messages.
SharePlex 9.2.8 Administrator Guide 210

View events and errors
SharePlex reports errors and other abnormal conditions in the following ways.
Event Log
SharePlex reports operational errors, notices and warning conditions to the Event Log. This log provides a perpetual step-by-step record of replication activities, errors, and events. The Event Log can help you replay the sequence of events that led up to a problem. Examples of replication events include:
l Start or stop of sp_cop or a replication process l Execution of a command in sp_ctrl. User-issued commands are recorded for every SharePlex command
that is issued. NOTE: A user-issued command appears in the Event Log as a notice, as in the following example:
Notice 08-07-02 16:13:24.641582 23696 1 User command: rjones activate config 1route (from mycomp14) l Database errors l Failure of a network connection or SharePlex process l Start or stop of a utility or script l Login or logout of a user
Each entry in the Event Log includes:
l The date and time of the event. l A description of the event and any related messages (error or non-error). l The event's process ID number, if it is associated with a SharePlex process.
To view the Event Log Use the show log command in sp_ctrl or open the file named event_log in the log sub-directory of the SharePlex variable-data directory. SharePlex provides a script for unattended monitoring of this log. For more information, see Monitor events with sp_eventmon on page 217. NOTE: To control the number of out-of-sync messages that Post logs when a target table is very out-of-sync, use the SP_OPO_SYNC_LOG_FREQUENCY parameter. For more information, see the SharePlex Reference Guide.
Status Database
The Status Database contains a summary of the conditions reported in the Event Log, including events that did not generate an error message or warning at the sp_ctrl user interface. This information alerts you to potential problems and helps you resolve existing ones. The Status Database may refer you to the Event Log for a more detailed explanation of a warning, notice or event.
To view the Status Database Use the show statusdb command in sp_ctrl or open the file in the data sub-directory of the SharePlex variabledata directory.
SharePlex 9.2.8 Administrator Guide 211

Error Log When the Post process detects that source and target tables are out of synchronization, it logs the first 100 SQL statements and data for the out-of-sync transactions to an error file on the target system. You can use this log to determine the extent of the out-of-sync condition, and you can use the SQL statements to repair target tables if the condition is not too severe, after first correcting the cause of the problem.
To view the Error Log Open the ID_errlog.sql file in the log sub-directory of the SharePlex variable-data directory (where ID is the identifier of the SharePlex target, for example a target database).
Process logs When a SharePlex process cannot process a record, the process not only logs the record to the Event Log, but also to its process log file. The process logs are primarily for use in debugging. The name of a process log consists of the datasource identifier (such as the ORACLE_SID), the short name of the process (such as ocap, ord, opo, rcl), the file number, and the file extension (.log). Examples: Capture: ora10_ocap02.log Read: ora10_ord01.log Post: ora10_opo03.log Reconcile: ora10_rcl01.log The aging of old log files is performed in a circular pattern. The numbering begins with 01 and ends with 03. Up to three logs can exist at any time, including the current one. When all three logs are full (50 MB), the process starts overwriting them, beginning with the oldest one.
To view a process log Open the file in the log sub-directory of the SharePlex variable-data directory.
Activation log When you activate a configuration, it generates a log.
To view the activation log Open the file named SID_oconf##.log in the log sub-directory of the SharePlex variable data directory.
Compare/repair log The compare and repair commands log errors, messages and warnings to a log. For more information about these logs, see the compare commands in the SharePlex Reference Guide.
Monitor with sp_ctrl commands
The information commands in sp_ctrl help you monitor different aspects of replication. Issue them frequently to: l Monitor for out-of-sync tables. l Verify that replication processes are running.
SharePlex 9.2.8 Administrator Guide 212

l View the number of replicated messages in the queues. l View the Event Log to view warnings, errors and other notifications. l View process statistics that are helpful for tuning and problem solving. l Detect tables or operations that are slowing down the replication process.
List of information commands

Command append status copy status compare status lstatus job status orainfo qstatus repair status report show
show capture show config show export show import show log show post show read show sql

Auth. level 3 3 3 3 3 3 3 2 3 3
3 3 3 3 3 3 3 3

Description
Displays status and results of the append using and append commands.
Displays status and results of the copy using and copy commands.
Displays the status and results of the compare using and compare commands.
Displays detailed information about the state of SharePlex replication.
Displays current status and history for append, compare, copy and repair commands.
Displays the Oracle database information.
Displays the state of the capture, export and post queues.
Disaplys the status and results of the repair and repair using commands.
Displays append, compare, copy and/or repair history for a table.
Displays the source and destination of the data being processed by each replication process on a system, and displays the status of each process.
Displays brief or detailed statistics for the Capture process for use in tuning and problem solving.
Displays properties of the active configuration.
Displays the number of messages sent to the target system(s).
Displays the number of messages received from the source system(s).
Displays the Even Log, Command Log, Verify Log, Trace Log, or a process log.
Displays brief or detailed statistics for the Post process for use in tuning and problem solving.
Displays brief or detailed statistics for the Read process for use in tuning and problem solving.
Displays the current or last SQL statement processed by the Post process.

SharePlex 9.2.8 Administrator Guide 213

Command show statusdb
show sync status

Auth. level 3
3 3

Description Displays the Status Database, which contains records of important replication events. Displays information about out-of-sync conditions.
Displays an overview of the state of SharePlex replication.

See the SharePlex Reference Guide for details about these commads.

Run monitor scripts on Unix
The SharePlex monitoring scripts notify you of events and conditions that can adversely affect replication on Unix or Linux systems. These scripts provide a monitoring mechanism without the need for frequent status checks through sp_ctrl. They can be run independently or through scheduled jobs. SharePlex provides the following scripts:
l sp_eventmon monitors the SharePlex Event Log and reports errors that you specify in a special file. l sp_logmon monitors how well Capture is keeping pace with the changes entering the redo logs. If
Capture loses pace by a specified number of logs, sp_logmon alerts you before the logs wrap so that you can take corrective action. l sp_ps monitors the SharePlex processes and notifies you if one or more are stopped so that you can correct the problem before the logs wrap or the queues exceed available disk space. l sp_qstatmon monitors the status of the SharePlex queues and sends a warning if the backlog exceeds a threshold (limit) that you define. This enables you to take corrective action before the queues exceed available disk space and replication is adversely affected.
IMPORTANT!
l These scripts run on Unix or Linux systems only. l The monitoring scripts are overwritten with new scripts during patches and upgrades of SharePlex.
Before you install the patch or upgrade, rename your existing scripts so that your customizing is retained. After applying the patches, update the new scripts with your customizing. Do not rename the existing scripts to replace the updated scripts, or you could lose important improvements or fixes.
Requirements for using the monitoring scripts
l These scripts must be run in the ksh shell.
l All monitoring scripts must remain in the directory where they were installed. All but sp_ps are in the .app-modules directory of the SharePlex installation directory. The sp_ps script is in the util directory of the installation directory.
l The scripts must be customized to reflect your environment, such as the type of e-mail or the paging available.

SharePlex 9.2.8 Administrator Guide 214

l To use the monitoring scripts, start sp_cop with the -uname name option, where name can be an identifier of your choice. Suggestions are: l the SharePlex port number l the ORACLE_SID of the instance for which replication is being monitored l the SharePlex Administrator's name
l SharePlex must be running prior to executing a monitoring script. l Verify the ORACLE_HOME (the path to the ORACLE_HOME directory) for each Oracle instance
being monitored. l The monitoring scripts make use of sp_ctrl commands. Before you use the scripts, make a link in the util
sub-directory to the sp_ctrl binary in the bin sub-directory of the SharePlex product directory. Do not copy the binary itself, because that makes it difficult to maintain patches to sp_ctrl. l Users of the monitoring utilities must have the following rights:
l Local access to sp_ctrl and permission to run the script on the system on which the sp_cop to be monitored is running.
l Korn (ksh) shell access and ps permission from the Unix or Linux command line. l Read, write and execute permissions to the directory where the scripts reside. l The permission on the iwgrep utility must be 755. l The monitoring utilities use the mailx program to send e-mail notifications. Before using a script, make sure mailx is configured to send e-mail on all systems where the monitoring scripts will be deployed. l Paging requires that your service provider supports receiving e-mails on your paging device. l To kill any of the processes generated by these scripts, use the kill -9 command. The kill command alone does not kill all of the processes.
Monitor Oracle capture with sp_logmon
The sp_logmon monitoring script helps prevent situations where you must resynchronize your data because the Oracle redo logs wrapped before Capture was finished reading them. It monitors the redo log group to which Oracle currently is writing and determines which log SharePlex is reading. If Capture loses pace by a specified number of logs, sp_logmon generates a warning in the logmon.log file and in an e-mail message, if that option is enabled. This gives you time to correct the cause of the delay and restore the archive logs, if necessary.
Prepare to run sp_logmon
Before running the script, perform the following tasks.
Satisfy requirements
See Requirements for using the monitoring scripts on page 214 before you use this script. NOTE: The script must be run in the ksh shell.
SharePlex 9.2.8 Administrator Guide 215

Define email addresses
To use the e-mail notification feature, define the e-mail address(es) in the script before running it.
1. Open the script in the app-modules directory of the SharePlex product directory. 2. Add any number of address strings after the MailUserName= variable. Use the full e-mail and/or pager
address. Separate multiple entries with a comma, as shown in the following example: MailUserName=scott@company.com,12345678910@pageservice.com
IMPORTANT! If the person modifying the script is someone other than a SharePlex user, he or she needs to have these Oracle privileges:
l CONNECT privileges l SELECT privileges for the V$LOG table l SELECT privileges for the SharePlex internal tables

Run sp_logmon
Run the script from the util sub-directory of the SharePlex product directory, not from app-modules. When you run it from the util directory, you actually make a soft link that runs a utility which first sets up the correct environment before running the script itself.

Syntax nohup sp_logmon -p port -t interval -l integer [-m ] [/dev/null] &
Table 4: Required arguments

Argument nohup sp_logmon -p port -t interval -l integer
&

Description
Directs the script to continue running in the background if the user logs out. This ensures continuous monitoring. The sp_logmon component runs the script.
Sets the port number for the instance of sp_cop that you are monitoring. You can monitor different SharePlex instances by running sp_logmon for each one, using different values for this argument.
Sets the time interval between scans in seconds. The value can be any positive integer.
Sets the maximum permissible number of redo logs between where Oracle is writing and where Capture is reading. This value triggers the warning generated by sp_logmon. Valid values are positive integers from 1 to the number of redo logs in the group.
Runs the script in the background.

Table 5: Optional arguments

Argument /dev/null

SharePlex 9.2.8 Administrator Guide 216

Argument -m

Description
Enables the e-mail/paging option. Without this parameter, sp_logmon only logs errors to the log file.

Monitor events with sp_eventmon
The sp_eventmon monitoring script monitors the SharePlex Event Log (event_log) at set intervals for entries relating to key replication events. You can define the scan interval and the error messages you want the script to detect. Each scan starts where the previous one stopped to keep the impact on the system minimal and prevent duplicate warnings. The sp_eventmon script takes the following actions after each scan of the Event Log:
l When sp_eventmon detects an error that you defined, it prints a notification to the error.splex log file and an e-mail message, if that option is enabled.
l It logs each error, the error Event Log line number of the error, the sp_cop instance name (typically the port number), and the time and date of the error.
The script relies on the iwgrep program, the error_list file (described later), and a marker file named username.mrk (where username is derived from the string that you enter with the -s argument when you run sp_eventmon). These three components must be kept in the same directory as the script, or it will not function. NOTE: The username.mrk file prevents duplicate warning messages from being sent to the log and to your email or pager. Without this file, the script starts scanning the Event Log from the beginning every time it starts. Warnings that were previously generated are sent again.
Prepare to run sp_eventmon
Before running the script, perform the following tasks.
Satisfy requirements
See Requirements for using the monitoring scripts on page 214 before using this script. NOTE: The script must be run in the ksh shell.
Define error messages
The sp_eventmon script scans for events listed in the error_list file, located in the util sub-directory of the SharePlex product directory. View that file for more information about the supported errors. You can add custom error strings to the error_list file by editing it in any ASCII text editor. Open the file and place each error string on a separate line.
Set IW_HOME
The IW_HOME variable in the script must be set to the correct value on each machine. This variable must point to the directory in which the monitoring scripts and iwgrep reside. If the path is not correct:

SharePlex 9.2.8 Administrator Guide 217

1. Open the script in the app-modules directory of the SharePlex product directory. 2. Set the path as shown in the following example:
IW_HOME=/export/home/splex/monscripts
Define e-mail addresses
To use the e-mail notification feature, define the e-mail address(es) in the script before running it. 1. Open the script in the app-modules directory of the SharePlex product directory. 2. Add any number of address strings after the MailUserName= variable. Use the full e-mail and/or pager address. Separate multiple entries with a comma, as shown in the following example: MailUserName=scott@company.com,12345678910@pageservice.com

Run sp_eventmon
NOTES:
l If you are running multiple instances of sp_eventmon, each instance must be run under the name of a different operating system user. Each username.mrk file will have a different username.
l Use the truncate log command in sp_ctrl to truncate the Event Log frequently when you are running the sp_eventmon script. If the log grows too large, the iwgrep program cannot grep from it properly. When you issue the truncate log command, remove the username.mrk file. The next time you run sp_ eventmon it will create a new file. See the SharePlex Reference Guide for more information about the truncate log command.
l When there is an existing Event Log with errors in it and the script is running, issue the truncate log command and then delete the sp_cop_name.mrk file, where sp_cop_name is the value used in the -s argument when the script was run. This file is in the util sub-directory of the SharePlex product directory.

To run sp_eventmon
Run the script from the util sub-directory of the SharePlex product directory, not from app-modules. When you run it from the util directory, you actually make a soft link that runs a utility which first sets up the correct environment before running the script itself.

Syntax nohup sp_eventmon -s 'sp_copname' -t interval -p path [-n name ] [-m] /dev/null &
Table 6: Required arguments

Component nohup sp_eventmon
-s 'sp_copname'

Description
Directs the script to continue running in the background if the user logs out. This ensures continuous monitoring. The sp_eventmon component runs the script.
Sets the name of sp_cop that was used when sp_cop was started with the -u option. The name of sp_cop must be enclosed within single quote marks. You can use this parameter more than once to monitor multiple sp_cop instances on a system. Without this parameter, sp_eventmon will not start.

SharePlex 9.2.8 Administrator Guide 218

Component & -t interval

Description
Runs the script in the background.
Sets the time interval between scans in seconds. The value can be any positive integer.

Table 7: Optional Components

Component -p path /dev/null
-n name -m

Description
Sets the path to the SharePlex variable-data directory. Without this variable, sp_eventmon assumes the default path.
Redirects the notification output to the /dev/null device on the local system so that the monitoring process continues to run in the background and generate output. To have the output appear on screen, omit this argument.
Sets the name of the Event Log if it is something other than the default name "event_log."
Enables the e-mail/paging option. Without this option, sp_eventmon only logs errors to the log file.

Monitor processes with sp_ps
The sp_ps monitoring utility monitors all SharePlex processes, including child processes, associated with a specified sp_cop instance. It scans the processes at regular intervals and reports abnormal conditions to one or more log files. It can monitor multiple installations of SharePlex on one or more systems, and it supports unidirectional and bi-directional (peer-to-peer) configurations.
Prepare to run sp_ps
Before running the script, perform the following tasks.
Satisfy requirements
See Requirements for using the monitoring scripts on page 214 before using this script. NOTE: The script must be run in the ksh shell.

SharePlex 9.2.8 Administrator Guide 219

Set the scan interval
The scan interval specifies how long the sp_ps program waits between checks. The default is 2,000 seconds. To specify a different scan interval, follow these steps.
1. Open the sp_ps file in the app-modules directory of the SharePlex product directory. 2. Set the interval= parameter to the required scan interval. Use any positive integer, for example:
interval=1500
Define email addresses
To use the e-mail notification feature, define the e-mail address(es) in the script before running it.
1. Open the script in the util directory of the SharePlex product directory. 2. Add any number of address strings after the MailUserName= variable. Use the full e-mail and/or pager
address. Separate multiple entries with a comma, as shown in the following example: MailUserName=scott@company.com,12345678910@pageservice.com
NOTE: The e-mail/paging option is enabled by default for sp_ps, but confirm that it was not changed. In the script, MAILOPTION=TRUE enables e-mail notifications and MAILOPTION=FALSE disables them.

Run sp_ps
Run the script from the util sub-directory of the SharePlex product directory.
Syntax nohup sp_ps ['sp_cop -u name'] CONFIGURATION [> /dev/null] [ &]

Table 8: Required arguments

Argument nohup sp_ps 'sp_cop -u name'
CONFIGURATION

Description
Directs the script to continue running in the background if the user logs out. This ensures continuous monitoring. The sp_ps component runs the script.
Use this parameter if you are running more than one sp_cop process. Use it to specify each one of those processes that you want to monitor. This argument must reflect exactly the same name that was used when sp_cop was started with the -u option. It must be enclosed within single quote marks.Without the -uname option, sp_ps assumes you want to monitor the sp_cop that uses the default SharePlex port of 2100.
Specifies the type of configuration of the SharePlex instance being monitored. This value must be entered in CAPITAL letters. Valid values:
SOURCE -- Use for uni-directional replication to monitor the Capture, Read and Export processes on the source system.
TARGET -- Use for uni-directional replication to monitor the Import and Post processes on the target system.
MULTI-SOURCE -- Use for peer-to-peer replication. It directs the script to monitor the Capture, Read, Export, Import and Post processes on each

SharePlex 9.2.8 Administrator Guide 220

Argument
> /dev/null &

Description
system. NOTE: If replicating between source and target tables on the same system, there are no Export or Import processes.
Redirects the notification output to the /dev/null device on the local system so that the monitoring process can continue to run in the background and generate output. To have the output appear on screen, omit this argument
(Ampersand) Runs the script in the background.

Monitor queues with sp_qstatmon
The sp_qstatmon script monitors the status of the capture and post queues for message backlogs. You can configure the script to alert you if the number of messages in a queue exceeds a defined threshold (limit), indicating that there is a potential data, system or network problem. This gives you time to correct the problem before the queues exceed their allocated space on the filesystem. After each analysis of the queues, the sp_qstatmon script prints a notice in the capstat.log file for the capture queue or the poststat.log file for the post queue, as well as an e-mail message if that option is enabled.
Prepare to run sp_qstatmon
Before running the script, perform the following tasks.
Satisfy requirements
See Requirements for using the monitoring scripts on page 214 before using this script. NOTE: The script must be run in the ksh shell.
Assign permission to create temporary files
The script creates some temporary files in the util sub-directory of the SharePlex product directory. Assign write permission to that directory to the sp_qstatmon module.
Define email addresses
To execute sp_qstatmon with e-mail notification, you must first must define the e-mail address(es) in the script. Notification messages are sent to all addresses coded in the script. Unless email notification is enabled, sp_ qstatmon only logs errors to the log file. You can specify as many addresses as you want.
1. Open the sp_qstatmon script in any ASCII text editor. The script is in the .app-modules directory in the SharePlex installation directory.
2. Add the address strings after the MailUserName= variable. Use the full e-mail and/or pager address. Separate multiple entries with a comma, as shown in the following example.
scott@company.com, 12345678910@pageservice.com 3. Save and close the file.

SharePlex 9.2.8 Administrator Guide 221

Run sp_qstatmon
Run the script from the util sub-directory of the SharePlex product directory, not from app-modules. When you run it from the util directory, you actually make a soft link that runs a utility which first sets up the correct environment before running the script itself.

Syntax nohup sp_qstatmon -v path -t n -p port_number [-c integer ] [-d integer ] [-m] > /dev/null &
Table 9: Required arguments

Argument nohup sp_qstatmon -v path
-t n -p port
&

Description
Directs the script to continue running in the background if the user logs out. This ensures continuous monitoring. The sp_qstatmon component runs the script.
Sets the path to the SharePlex variable data directory for the instance of sp_cop that you want to monitor. Without this variable, sp_qstatmon fails and prints an error message requesting a valid path.
Sets the time interval between scans in seconds. This value can be any positive integer.
Sets the port number for the instance of sp_cop that you are monitoring. You can monitor different SharePlex instances by running sp_qstatmon for each one, using different values for this argument.
Runs the script in the background.

Table 10: Optional arguments

Argument /dev/null -c integer -d integer -m

Description
Redirects the notification output to the /dev/null device on the local system so that the monitoring process continues to run in the background and generate output. To have the output appear on screen, omit this argument.
Sets the number of messages in the capture queue at which the script issues a warning message. This value can be any positive integer. Without this parameter, sp_qstatmon defaults to 100 messages.
Sets the number of messages in the post queue at which the script issues a warning message. This value can be any positive integer. Without this parameter, sp_qstatmon defaults to 100 messages.
Enables the e-mail/paging option. Without this parameter, sp_qstatmon only logs errors to the log file.

Run a monitor script on Windows
SharePlex provides the Sp_Nt_Mon script for monitoring processing Windows systems. Sp_Nt_Mon runs in the background to monitor SharePlex and notify you of events and conditions that can adversely affect replication. This script provides a monitoring mechanism without the need for frequent status checks through sp_ctrl. IMPORTANT! Do not rename this script. The Sp_Nt_Mon script detects and reports the following conditions by email:

SharePlex 9.2.8 Administrator Guide 222

l The status of the Capture, Read, Export, Import, Post, and Sp_Copsrv.exe processes. l The volume of the capture and post queues. l Latency between transactions being processed by Oracle and by SharePlex. l Specific messages in the Event Log (described later in this topic) To configure and run Sp_Nt_Mon 1. Run the SpClient utility from the SharePlex programs group. 2. On the SpClient toolbar, click the SpMonitor button. 3. Select the port number of the instance of SharePlex that you want to monitor, then click
Configure Monitor.
SharePlex 9.2.8 Administrator Guide 223

4. Refer to the following table to modify the properties in the sp_nt_mon.txt file. Sp_Nt_Mon reads input from this file. To make comments in this file, start each comment line with the # symbol.

Property

Default Value Description

Time Interval
Send Email alert

60 seconds 1 (enabled)

Sets a time interval, in seconds, between monitoring polls.
Enables or disables email alerts. Email is the only reporting option. 0 = disabled 1 = enabled

SMTP Server Name
Email Address

none none

Specifies the name of the SMTP server for the e-mail account.
Specifies one or more email addresses that will receive monitor notifications. Separate each email address with a space, for example: Email Address: Your.Name@YourCompany.com Other.Name@YourCompany.com

Cop Capture Read Export Import Post

0 (do not monitor)

Specifies processes to monitor. A value of 1 enables monitoring and must be set for each process that you want to monitor. For example, to monitor Capture and Export, set the properties as follows: Cop: 0 Capture: 1 Read: 0 Export: 1 Import: 0 Post: 0

Capture Threshold Post Threshold

0 (do not monitor)

Sets the maximum number of messages that a queue can contain before an alert message is triggered.
Set a positive numeric value for each queue that you want to monitor. Set a value that is well below the amount of disk space that is allocated to this queue. This gives you enough time to add disk space or resolve any problems that are causing a backlog.

Monitor the Event Log

0 (do not monitor)

Enables or disables the monitoring of supported Event Log messages. 0 = disabled 1 = enabled The following messages are monitored:

SharePlex was shutdown

System call error

Queue partition is full

SharePlex 9.2.8 Administrator Guide 224

Property
Monitor Redo Logs Redo Threshold

Default Value Description

Host is down

Internal error

Out of sync

Poster failure

Parameter was changed Activate config

Deactivate config

Bad config file

sp_cop was shutdown

Child process failed

Error

ORA-error number errors

0 (do not monitor)
2 logs

(Oracle only) Enables or disables the monitoring of Capture latency. 0 = disabled 1 = enabled
(Oracle only) Sets a value for the maximum permissible number of logs for Oracle to be ahead of SharePlex in the redo stream. When the value is reached, an alert is triggered. Use when Monitor Redo Logs is enabled.

5. Click Start to run the script. The script starts and the word Start becomes Stop. The script runs continuously to monitor SharePlex until you click Stop.

Monitor replication with SNMP
SharePlex provides agent support for Simple Network Management Protocol (SNMP) on all Unix, Linux, and Windows platforms supported by SharePlex replication.
NOTE: SharePlex provides only agent support for SNMP. It only sends SNMP traps. SharePlex does not provide an SNMP signal daemon (SNMP manager) to intercept the traps. Use the SharePlex SNMP feature only if you have a Network Management Station (NMS) to manage SNMP signals. The SharePlex SNMP agent is named snmptrap and is installed with SharePlex in the bin sub-directory of the SharePlex product directory. Do not run this program.
Enable SNMP
To enable SNMP monitoring of SharePlex replication, set the SP_SLG_SNMP_ACTIVE parameter to 1. By default, the parameter is set to 0 (disabled).

SharePlex 9.2.8 Administrator Guide 225

Configure the SNMP agent
The following parameters configure the SNMP agent to communicate with the NMS. Each parameter must have a value if the SP_SLG_SNMP_ACTIVE parameter is enabled.

Parameter SP_SLG_SNMP_HOST SP_SLG_SNMP_COMMUNITY SP_SLG_SNMP_MJR_ERRNUM SP_SLG_SNMP_MNR_ERRNUM

Value The name of the system (host) to which the traps will be sent The community security string The major error number to be used by the traps The minor error number to be used by the traps

Custom MIB parameters
The following parameters specify required information for a custom MIB.

Parameter
SP_SLG_SNMP_ ENTERPRISE_OID SP_SLG_SNMP_TRAP_OID
SP_SLG_SNMP_TRAP_ PROGRAM

Value
The enterprise object identifier to send with the trap. The default is 1.3.6.1.4.1.3.1.1 .
A custom object identifier to bind to the trap. The default is 1.3.6.1.2.1.1.1.0.
The name of the trap program. The default is iwsnmptrap.

Configure the SNMP traps
The following parameters configure the SNMP agent to send traps for specific replication events. The message or error text for the event is included in the trap and is the same error that appears in the Event Log. To enable an SNMP trap for an event, set the corresponding parameter to a value of 1. By default all traps are disabled (parameter value of 0).

Parameter SP_SLG_SNMP_INT_ERROR SP_SLG_SNMP_SYS_ERROR SP_SLG_SNMP_ERROR SP_SLG_SNMP_OUT_OF_SYNC SP_SLG_SNMP_STARTUP SP_SLG_SNMP_SHUTDOWN SP_SLG_SNMP_LAUNCH SP_SLG_SNMP_EXIT

SharePlex Event SharePlex logic errors and errors that cause processes to exit System-related errors encountered by SharePlex Other SharePlex errors Replication is out of synchronization SharePlex starts up SharePlex shuts down A SharePlex process starts A SharePlex process stops

SharePlex 9.2.8 Administrator Guide 226

18
Prevent and solve replication problems
This section contains solutions to many of the common questions and problems that can arise during replication, and it also suggests preventive measures for avoiding problems.
Contents
Find the solution in the SharePlex Knowledge Base Solve Database Setup problems Solve configuration file problems Solve activation problems Solve replication problems Solve Oracle DDL replication problems Solve queue problems Solve synchronization problems Solve compare command errors Solve other replication problems How to resynchronize source and target tables How to restore Oracle archive logs How to release semaphores after process failure How to resolve disk space shortage How to find the ORACLE_SID and ORACLE_HOME
Find the solution in the SharePlex Knowledge Base
If the issue you are experiencing is not listed in this documentation, search the SharePlex Knowledge Base at: https://support.quest.com. The Knowledge Base provides filtering options and links to other resources that can help you use and troubleshoot SharePlex.
SharePlex 9.2.8 Administrator Guide 227

Solve Database Setup problems
This section helps you diagnose problems that are associated with the SharePlex database account and connection information that was created with the Database Setup utility when SharePlex was installed on the system. NOTE: For more information about Database Setup, see Database Setup Utilities in the SharePlex Reference Guide. If the issue you are experiencing is not listed in this documentation, search the SharePlex Knowledge Base at: https://support.quest.com. The Knowledge Base provides filtering options and links to other resources that can help you use and troubleshoot SharePlex.

Oracle setup issues

Problem

Description

Solution

Incorrect ORACLE_SID and/or ORACLE_ HOME

If SharePlex cannot interact with Oracle, it might be using the wrong ORACLE_SID and/or ORACLE_ HOME.

1. See How to find the ORACLE_SID and ORACLE_HOME on page 270 to determine the Oracle values.
2. Rerun the Database Setup utility. For more information, see Database Setup Utilities in the SharePlex Reference Guide.

Insufficient database privileges

If the Database Setup utility fails, the person who runs it may not have the correct privileges

For more information, see Database Setup Utilities on page 1.

Asterisk as the ORACLE_SID entry

Sometimes, the oratab file has an * (asterisk) symbol instead of a value for the ORACLE_SID.

More than one oratab file (Sun Solaris)

On Solaris systems, the oratab file is typically located in the /var/opt/oracle directory, but because other platforms store the oratab file in the /etc directory, there could be a second oratab in the /etc directory.

Ensure that a valid ORACLE_SID is in the oratab file, and then try running the database setup again.
Either move, rename or delete the secondary oratab file, and then try running the database setup again.

Oracle not running

Oracle must be running and the instance must be open while you run the Database Setup utility. The utility accesses Oracle to establish SharePlex as a user and install its internal tables.

Start Oracle and open the instance.

sp_cop is running

The SharePlex sp_cop process cannot be running while you are running the Database Setup utility.

If it is running, shut it down using the shutdown command in sp_

SharePlex 9.2.8 Administrator Guide 228

Problem

Description

Solution

Oracle library location not correct

ctrl. Run sp_ctrl from the bin subdirectory in the SharePlex product directory.

On Unix and Linux systems, SharePlex expects the Oracle library to be in the $ORACLE_HOME/lib or $ORACLE_HOME/lib32 directory. In some environments, the Oracle library has a different name than what SharePlex expects it to be, or it is installed in a different location than expected (or both). In that case, you will see an error message when you attempt to run the Database Setup utility.

Install the appropriate library from Oracle and then re-start SharePlex (if it is stopped). SharePlex will link to the correct library from that point forward.

ld.so.1: sqlplus: fatal: libsunmath.so.1: can't open file: errno=2" error

On Unix and Linux systems, this error indicates that SharePlex cannot find the libsunmath and libshareplex libraries, even though the link exists in the proper place.

You can use either of these solutions:
l Create a softlink for $ORACLE_ HOME/lib/libsunmath.so.1 in the /usr/lib directory. or...
l In the ECXpert/config/bdg.ini file in the [DB_ENV] section add the following line: LD_LIBRARYPATH=full oracle home path/lib

Wrong user-id

To run Database Setup on Unix and Linux systems, the set-user-id for the Oracle software need to be rwsr-s--x. Those permissions allow non-Oracle users to log into SQL*Plus.

Set the correct values for setuser-id.

Solve configuration file problems
This section reviews problems and solutions associated with the management of configuration files. See also Solve activation problems on page 230. If the issue you are experiencing is not listed in this documentation, search the SharePlex Knowledge Base at: https://support.quest.com. The Knowledge Base provides filtering options and links to other resources that can help you use and troubleshoot SharePlex.
A configuration file was accidentally deleted
You might be able to recover an accidentally deleted configuration file if that configuration was previously active and you did not run db_cleansp since it was activated.

SharePlex 9.2.8 Administrator Guide 229

Solution: View the Event Log to determine the activation ID for that configuration file, then look in the save subdirectory of the SharePlex variable-data directory for a .conf.actid file, where actid is the activation ID you got from the Event Log.

Solve configuration file errors

Error message

Description

Solution

The parameter for `create config' must be a new file name. or... Destination file exists - file must not exist prior to operation.

The name you gave this configuration already exists for another configuration file.

Use a different name. To see a list of configurations on a system, use the list config command.

Couldn't fork editor. or... Editor execution failed.
Destination file exists - file must not exist prior to operation.
Problems in reading or writing file used in
edit -- command aborted.

SharePlex failed to open the default text editor.

Make sure the editor still exists on the system. The default editors are vi on Unix and Linux and WordPad on Windows. To change the default text editor, see Set a default editor for sp_ ctrl on page 50.

The name you specified when

Use a different name. To see a list of

copying the configuration file already configurations on a system, use the

exists in this SharePlex instance.

list config command.

The edit config command could not open the specified file.

Confirm the name of the file you are trying to edit, including the case, and check to see if the file is corrupted.

Destination file exists - file must not exist prior to operation.

The new name you are giving this configuration already exists for another file.

Use a different name. To see a list of configurations on a system, use the list config command.

Invalid file name passed to command. or...
File does not exist.

You could have misspelled the configuration name or used the wrong case.

File access denied check file permissions.

You are not authorized to issue the command.

Verify the name and spelling, including the case, then enter it again. To see a list of configurations on a system, use the list config command.
View your authorization level with the authlevel command in sp_ctrl, then determine its minimum authorization level. The SharePlex Administrator assigns authorization levels. For more information, see About the SharePlex security groups on page 183.

Solve activation problems
This section reviews problems and solutions that may be associated with the activation of a configuration file. If the issue you are experiencing is not listed in this documentation, search the SharePlex Knowledge Base at:

SharePlex 9.2.8 Administrator Guide 230

https://support.quest.com.
The Knowledge Base provides filtering options and links to other resources that can help you use and troubleshoot SharePlex.
NOTE: Many configuration problems can be prevented by using the verify config command in sp_ctrl before you activate a configuration or reactivate one containing new or changed objects. The verify config command reviews the configuration to help ensure that basic requirements for activation and replication have been satisfied. See the SharePlex Reference Guide for more information.

SharePlex cannot locate the configuration file
If SharePlex cannot locate the configuration file that you want to activate, verify its location. Configuration files must reside in the config sub-directory of the SharePlex variable-data directory for activation to succeed. The create config command in sp_ctrl automatically puts configuration files in that directory. It is possible someone moved it. Problems also occur if the configuration file was created without using the create config command. This can happen if the file was created directly through the operating system or if the config.sql or build_config.sql script was used to create it. If the working directory at the time was not the config sub-directory, or if the file was saved to a different directory, the activation will not find it. Solution: Move the configuration file to the config sub-directory of the SharePlex variable-data directory.

Some objects failed to activate
If SharePlex cannot activate one or more objects listed in the configuration file, it will continue to activate the other ones and display the names of the ones that failed in the ID_oconf##.log file. The following are reasons why individual objects fail to activate.

Problem Description

Solution

Invalid objects

You may be trying to replicate objects that are not supported by SharePlex.

To understand the objects and operations that SharePlex supports, see the SharePlex Release Notes that accompany this release.

Invalid target systems

SharePlex could not get routing information.

Verify the names of the targets, and fix any syntax errors in the routing map. For some databases you specify an instance name, while for others you specify the actual database name.
For routing instructions, see:

l Database specifications in a configuration file on page 64
l Routing specifications in a configuration file on page 66

After your fix the routing syntax, activate the affected configuration again.

Syntax

Misspelled names, object names specified

errors and

Run the verify config command to view the errors. For help with configuration syntax, see

SharePlex 9.2.8 Administrator Guide 231

Problem Description

Solution

misspelled without an owner name, and other improper

words

syntax in the configuration specification can

cause activation of an object to fail.

Configure data replication on page 57.

(Oracle) SharePlex cannot lock tables

If activation cannot lock the tables in the configuration file, activation of that table will fail.

Assign the SharePlex database user the privilege to lock tables.

Activation failed completely
There are several things can cause the entire activation of a configuration to fail. Common error messages for configuration activation problems are:
Bad configuration file The Oracle sid SID specified in the config file is invalid. The following are causes and solutions for activation failures.

Problem
Password problems

Description

Solution

If SharePlex is having trouble connecting to the source database and you know the SharePlex account exists, find out if someone changed the password.

Update the connection information. For more information, see Change the SharePlex database account on page 312.

Deactivation followed too closely by activation

In rare cases, if you activate a configuration too soon after a deactivation, the activation fails.

Before you activate a new configuration after deactivating one, wait until you see the following message in the Event Log: Notice: sp_ordr (for o.ora10
queue o.ora10) Deactivated.

No datasource specification

The datasource specification is wrong or incomplete. An error similar to the following is returned:
The datasource specified in the config file is invalid.

Specify as: Datasource:o.SID

(Oracle) Incorrect ORACLE_SID

The wrong ORACLE_SID is specified on the Datasource:o.SID line.

Edit the configuration file to specify the correct ORACLE_SID.
How to find the ORACLE_SID and ORACLE_HOME on page 270.

(Oracle)Insufficient PROCESSES setting

If Oracle Error 20 (ORA-00020 maximum number of processes (string) exceeded) is the cause of an activation failure, it is because Oracle ran out of resources on the source system to allow one or more threads to log on.

Increase the PROCESSES parameter in Oracle or decrease the number of activation threads that you are using. The number of threads is controlled by the threads option of the activate config command.

(Oracle) ORA-

SharePlex cannot access the Data

Make certain that the O7_DICTIONARY_

SharePlex 9.2.8 Administrator Guide 232

Problem
00942: table or view does not exist.

Description Dictionary.

Solution
ACCESSIBILITY parameter in the init.ora file is set to TRUE. (This is the default.) The database must be restarted if the parameter is changed.

The reconcile command is slow to complete
This applies to activation and reconcile of an Oracle source. If you issue the reconcile command when source database activity is low, the command process can, in some circumstances, seem to stall. This happens because the reconcile command is dependent on data continuing to arrive from the source system. If there is no replicated activity on the source system after the point of the hot backup or copy, the reconcile process waits until source activity resumes. This is normal.

Common activation errors
The following are common error messages that you might encounter when activating a configuration.

ACTIVATE CONFIG error messages

Cause

Solution

line n, source object name (T_HFL_1) not of form OWNER.TABLE syntax error in line n.
line n, bad routing spec (o.ora10)
File does not exist.
Attempt to run sp_conf when sp_conf is already active Login parameters not set...
WARNING, not all objects activated successfully.

The owner name may be missing from one or more objects listed in the configuration file.

Specify object names as ownername.objectname.

There is a syntax error on the specified line of the configuration file.

Fix the incorrect syntax. For more information, see Configure data replication on page 57.

There is a syntax error in the routing map.

Make sure the routing map is written correctly. For more information, see Routing specifications in a configuration file on page 66.

SharePlex cannot find the configuration file.

Issue the list config command. If the file you want to activate is not listed, it may not be in the config sub-directory of the variable-data directory. Find the file and move it to that directory, then issue the activate config command again.

The configuration is already in None required. the process of being activated.

A SharePlex account and internal tables do not exist in the source database.

Run the Database Setup utility. For more information, see Database Setup Utilities in the SharePlex Reference Guide.

One or more tables failed to activate.

For more information, see Some objects failed to activate on page 231.

SharePlex 9.2.8 Administrator Guide 233

ACTIVATE CONFIG error messages

Cause

Solution

Check activation log.

Deactivate/flush a nonactive datasource
(Oracle) Currently involved in transaction.

You are attempting to deactivate a configuration that is not active.

No action is necessary if this is the configuration that you wanted to deactivate. To see a list of configurations, use the list config command.

Objects in the configuration file are locked.

SharePlex cannot lock Oracle tables to analyze them if another process has them locked. If the locks are from source transaction activity, try activating at a time when the database is less busy.

Solve replication problems
This section reviews common problems that you could encounter while data is being replicated. If the issue you are experiencing is not listed in this documentation, search the SharePlex Knowledge Base at: https://support.quest.com. The Knowledge Base provides filtering options and links to other resources that can help you use and troubleshoot SharePlex.

General issues

Problem Description

Solution

Network problems

Network factors can cause data to transfer to a target system too slowly, or not at all.

Make certain that all Export and Import processes are running, and ask the network administrator to check the network for causes of slowdowns.

Parameter settings

Some SharePlex parameters, when changed from their default settings, can inhibit replication performance.

To determine if any parameter settings are responsible for reduced performance, issue the list param modified command to view parameters that have non-default settings. Review the documentation for those parameters in the SharePlex Reference Guide to determine whether a particular setting is affecting replication speed.

Oracle Capture-related issues
If Capture is stopped, issue the status command in sp_ctrl to verify whether or not it was stopped intentionally by an authorized SharePlex user. If it was, find out why and make certain that it is not stopped for too long. The following are other reasons why Capture could slow down or stop unexpectedly.

SharePlex 9.2.8 Administrator Guide 234

Problem

Description

Solution

Excessive chaining

Excessive chaining reduces the performance of the source database because Oracle must read multiple blocks for a single row. Excessive chaining reduces Post performance on the target system because block fragmentation in the database decreases the speed at which Post can apply SQL statements.

Reorganize chained tables.

Ineffective redo log configuration

Capture reads archive logs if the redo logs wrap before Capture is finished with them, but this can slow down replication and consume disk space. In most cases, the redo logs should be configured so that SharePlex can avoid using the archive logs.

For more information about how to set up the logs, see the SharePlex Installation and Setup Guide.

SharePlex has low process priority

If Capture, Read, or both slow down during hot backups or other Oracle-intensive operations, view the process-priority settings on the system to determine if SharePlex can be assigned more resources.

Ideally, the SharePlex priority should match Oracle's priority.

Keys are not being logged

If PK/UK logging is not enabled for the tables in replication, SharePlex must query the database to get the key values for update and delete operations.

Enable PK/UK logging.

Temporary tables and FND tables

Temp tables receive numerous changes that Remove temporary tables from replication. cause replication overhead and performance degradation.

archive logs not available

When Capture detects a log wrap and the archive logs are not available, Capture stops and returns a "Log wrap detected" error. It waits a certain amount of time and then starts again, continuing this process until the logs are restored.

Restore the archive logs from the one that Capture last processed or point SharePlex to their location with the SP_OCT_ARCH_LOC parameter. To determine the log that Capture needs, use the show capture command with the detail option. For more information, see How to restore Oracle archive logs on page 266.

Compressed Capture cannot read compressed archive archive logs logs.

Uncompress the current log that Capture needs and all those after it so that Capture can process them. To determine the log that Capture needs, use the show capture command with the detail option.

archive logs in unexpected location

Capture stops if it cannot locate the archive logs it needs. When the redo logs wrap, Capture looks for the archives in Oracle's archive-log list, and if the logs are not there,

If you are storing the archive logs in a location other than the normal Oracle location, make sure this parameter is set to the full path name of the directory containing the archive logs.

SharePlex 9.2.8 Administrator Guide 235

Problem

Description

Solution

Capture looks in the location specified by the SP_OCT_ARCH_LOC parameter.

No access to Capture stops when it cannot read or find the Find out if someone changed permissions

the redo logs logs, then it tries to read the log again.

such that SharePlex does not have permission

to read the redo logs.

No access to the variabledata directory

Capture stops if it is unable to write to the SharePlex logs because of restrictive permissions on the state sub-directory in the variable-data directory, or if there are space restrictions in that directory.

Verify the permissions and space for the variable-data directory.

Oracle Post-related issues
There are a number of things that cause the Post process to slow down. SharePlex generates the operation taking too long message when Post exceeds the internally controlled allowable wait time to apply a SQL statement to the target instance. Often it only takes one table to cause a bottleneck. Use the show sql or show post command to find out which table Post is processing, and then check for the following.

Problem Full table scans
Bitmap indexes on target tables Disk I/O bottleneck

Description

Solution

If a target table does not have a key, Post must perform a full table scan to find the correct row. This slows the Post process.

Do one of the following:
l Add an index or key if possible, and use a hints file. See For more information, see Use Oracle INDEX hints on page 280.

l If you cannot add a key, create a key definition.For more information, see Define a unique key on page 75.

l Set the Oracle DB_FILE_ MULTIBLOCK_READ_COUNT parameter to the maximum number of blocks that can be read in one I/O request. This is defined by the system setting MAX_IO_SIZE/DB_ BLOCK_SIZE. You can increase the DB_BLOCK_BUFFERS parameter as well.

Bitmap indexes are beneficial for

Avoid using bitmap indexes on target tables

queries, but they slow down the DML in replication, or only use them on tables

operations applied by Post.

that do not have frequent DML activity.

Disk I/O bottlenecks on the target system are a common cause of slow

Disk I/O is the nature of the database environment, but you can reduce the bottleneck by placing the redo logs on

SharePlex 9.2.8 Administrator Guide 236

Problem

Description

Solution

Post performance. Post can spend a high percentage of its time waiting for Oracle to commit data. The effect is worse when there is a Capture process reading from the same log device in peer-to-peer replication.

faster hard drives or on a solid-state drive.

High number of buffer gets

Review any table with SQL statements that cause a high number of buffer gets. There should only be two to four buffer gets on an index, depending on its size and whether or not it is a unique index.

If there are more buffer gets than four, the index probably needs to be rebuilt. You can rebuild an index without the need to reactivate the configuration file.

Oracle write-rate bottlenecks

By default, one buffer writer writes all dirty Oracle blocks to disk. Whenever that buffer writer process wakes up to write, it locks portions of shared memory and, in effect, blocks the processes that are either modifying or reading data blocks -- including the Post process.

If this is a constant problem, you can consider increasing the number of writers. See the database documentation.

Small transaction size

Normally, Post performs an internal read/release after it receives each COMMIT, which means it purges that data from the queue as part of the checkpoint recovery process. For smaller transactions, this can cause excessive I/O on the target system and hinder the Post process.

If most of your transactions are small, try changing the value of the Post SP_OPO_ READRELEASE_INTERVAL parameter. See the SharePlex Reference Guide for more information about this parameter.

Sequences not cached

If the sequences in replication are not cached, they add unnecessarily to the replication volume.

Cache sequences. If replicated sequences are part of a key, replicate the tables that contain those keys, and remove the sequences from the replication configuration. You should see significant performance improvement.

Low ulimit

An error similar to the following means that the system file descriptors setting on the target system needs to be increased:
Error 07-24-08 12:11:40.360226 8693 12345 Poster error: /var/quest/ vardir/log/event_log: Too many open files (posting

Solution: Set the ulimit as close to the optimal value of 1024 as possible.
The ulimit can be set either as a system hard limit or a session-based soft limit, as follows:

SharePlex 9.2.8 Administrator Guide 237

Problem

Description

Solution

from ora102, queue prodsys, to ora10b)

l Set a hard limit: (Recommended) A root user and system restart are required to change the hard limit, but the value remains fixed at the correct level to support SharePlex. Consult your System Administrator for assistance.
l Set a soft limit: A soft limit setting stays in effect only for the duration of the sp_cop session for which it was set, and then it reverts back to a default value that may be lower than the hard limit and too low for SharePlex.

Compare process locks

The repair command locks all rows that need repair, or the entire table if there are more than 1000 out-ofsync rows, throughout the repair process. This could block Post if Post tries to apply data to the table being repaired.

If you do not want Post to wait for a repair process to finish, you can kill the compare process. To avoid issues with locks caused by repair processes, consider running the repairs during non-peak hours.

Difference in Capture and Post speed
Full archive log directory

SharePlex reads and processes records from the redo logs faster than it can post those operations to the target database with standard SQL statements.

For more information, see Tune the Post process on page 280.

If Post appears stalled and will not shut down normally, but the only error in the Event Log is sp_opst_ mt - operation taking too long, it could be that the archived log directory on the target system is full. If so, then Oracle cannot create new logs, and it suspends processing. Post stalls because it is waiting for Oracle.

Move some of the old archive logs to another device, or delete them to make room for new ones.

Commit reduction issues
If commit reduction is not working, it may be because records are not being dispatched quickly enough for there to be a valid message available after the commit. If a valid message is found after the commit message, then Post can skip the commit. Otherwise it issues the commit. One way to find out if this is the issue is to stop the Post process completely, run a large amount of small transactions through replication, then start the Post process. If you do not see any commit reduction, then this would rule out this issue as the cause.

SharePlex 9.2.8 Administrator Guide 238

Post stopped
If Post stops, issue the status command in sp_ctrl to find out why. · An idle status means there is no data in the post queue to post. · A stopped by user status means that a SharePlex user stopped the Post process. To find out which user is responsible, view the user issued commands in the Event Log. · A stopped due to error status means a replication or database error caused Post to stop. The following are some potential causes for Post to stop unexpectedly.

Problem

Description

Solution

Correctable database errors

Post stops for database errors that can be corrected, so that you can fix the problem without risk of the data going out of synchronization. The errors and the faulty SQL statement are logged to the databaseID_errlog.sql file in the log subdirectory of the variable-data directory on the target system.

Use the information in the log file to correct the problem, then start Post again. Posting resumes from the point where it stopped.

Noncorrectable database errors

Some database errors, such as out-of-sync conditions, cannot be corrected. In that case, SharePlex reports the error to the Event Log, writes the error and SQL statement to the databaseID_errlog.sql file, and continues processing. Sometimes, errors that cannot be corrected cause Post to stop.

If the errors cannot be corrected but you want Post to continue processing, list the errors in the in the databasemsglist file and then set the SP_OPO_CONT_ON_ERR or SP_OPX_ CONT_ON_ERR parameter to 1, which directs SharePlex to ignore those errors and continue posting. For more information, see Continue to post when there is a DML error on page 161.

Locks on target tables

If a target table is locked, the Post process cannot apply a SQL statement and generates an error message. This message could mean that a user, application, or job is accessing the table and might have caused an out-of-sync condition. Or, for an Oracle target, it could mean that a repair command has locked the table.

Find out why the table was locked and resolve the access issue (unless due to the repair command). You might have to resynchronize the data if DML was performed on the table. For more information, see How to resynchronize source and target tables on page 261.

Configuration deactivation when there are named export queues

When you are using named export queues, and you deactivate the configuration, Post could stop with the following error instead of posting the remaining data from the queues:
sp_opst_mt (for o.qa920-o.qa920 queue q5) 15007 - Can't open poster queue que_NOEXIST: Queue does not exist.

To start Post and finish replication, shut down SharePlex, then start it again.

Queue name Post stops if a queue name is too long. is too long

Make certain the name assigned to named queues in the configuration file are no longer

SharePlex 9.2.8 Administrator Guide 239

Problem

Description

Solution

than 15 characters.

Post generates a semaphore error

If Post returns an error message of "shs_ SEMERR: an error occurred with the
semaphore" on a Windows system, SharePlex requires more semaphores than are available. This usually occurs when named queues or horizontally partitioned replication are being used. You can increase the number of semaphores through the MKS Toolkit.

1. In the Windows Control Panel, open Configure PTC MKS Toolkit.
2. Select Runtime Settings.
3. From Category, select Semaphores.
4. Set the following semaphores: Max Number of Semaphore IDs: 10 Max Number of Semaphores: 512 Max Number of Semaphore Undo entries: 90

(Oracle) New For an Oracle source, tables created after

tables are not activation are automatically added to

added

replication.

The auto-add feature is controlled by parameters. Make certain the correct ones are set to achieve your goals. For more information, see Control Oracle DDL replication on page 152.

(Oracle) No more open cursors

If you see the following error, Post has exceeded the available open cursors.
Warning: sp_opst_mt (for o.oracle-o.oracle queue oracle) Post has opened number cursors. No more available cursors! Exiting
Post requires a certain number of open cursors to the target database.

View the current database OPEN_CURSORS value using the following SQL statement:
select value from V$PARAMETER where name = 'open_cursors' ;
To determine an adequate OPEN_CURSORS value for SharePlex, see Adjust open cursors on page 283.

Other problems and solutions

Problem Description

Solution

SharePlex cannot resolve a machine name

Sometimes machine names cannot be resolved between Unix or Linux and Windows systems.

Add the IP addresses and names of all servers (Unix, Linux, and Windows) in the replication network to the /etc/hosts file on all Unix and Linux machines, and add the same information to the hosts file on all Windows machines.

sp_cop is using too much CPU time

SharePlex may be performing its overhead Increase the idle time of sp_cop with the SP_COP_

activities too frequently.

IDLETIME parameter. See the parameter

documentation in the SharePlex Reference Guide.

Corrupted source table

If a source table is corrupted or there is another reason that you do not want the replicated data to be posted to the target database, you can prevent posting for that

To disable posting for one or more tables, use the SP_OPO_DISABLE_OBJECT_NUM or SP_OPX_ DISABLE_OBJECT_NUM parameter, which can be set to disable posting of both DML and DDL

SharePlex 9.2.8 Administrator Guide 240

Problem Description

Solution

table without removing it from the active configuration or affecting posting for other objects.

operations for specified object IDs. This parameter is disabled by default. When you are ready to begin posting to the target table again, disable the parameter again by resetting it to 0. For more information, see the parameter documentation in the SharePlex Reference Guide.

Only some columns were replicated

If you have a table that is configured for vertically partitioned replication but that table name also satisfies a wildcard, the specific listing (with the vertical partitioning) takes precedence over the wildcarded listing. Vertical partitioning cannot be used with full-table replication for the same table.

Conflict resolution generates compile errors

If you encounter compile problems with your conflict resolution routines, check whether any tables have the same names as their owners. A known issue in PL/SQL prevents the SharePlex conflict resolution logic from compiling the PL/SQL for tables whose names are the same as their owners. Oracle has stated that the issue will not be fixed. See Oracle TAR 2577886.996 for more information.

No action is needed if replication of the specified columns is desired. For more information, see Configure vertically partitioned replication on page 104.
This issue does not affect replication; SharePlex replicates data for tables with identical owner and table names.

Common replication errors
The following table explains many of the common error messages that you might experience during replication.

sp_cop error messages
sp_cop cannot setup; memory segment n in use.
Error: sp_cop can't setup shared memory statistics capability - exiting Error: sp_cop(shs) Cannot delete previous memory segment 303. Please check to see if any SharePlex processes are running (ps -ef | grep sp_). If there are some processes running then kill them and restart sp_cop.
Capture error messages
Capture time limit (300 sec) exceeded.

Cause
Processes that access the queues were still running when you last shut down sp_cop.
There are already one or more SharePlexsp_cop processes pointing to the same variable-data directory.

Solution
Kill those processes. SharePlex processes start with sp_. When all have been killed, sp_cop should start.
To run multiple sessions of SharePlex, you need to use separate variable-data directories for each one. For more information, see Run multiple instances of SharePlex on page 43..

Capture is not processing records, which could

See Oracle Capture-related issues on page 234 for possible causes. If

SharePlex 9.2.8 Administrator Guide 241

sp_cop error messages

Cause

Solution

(Oracle) Log wrap detected
Post error messages operation taking too long. Rowid not found

indicate a problem with the redo log. If, after a specific number of seconds, Capture cannot process a record, it stops, logs the record, and returns this message.

you cannot determine the cause of the problem, call Quest Support before a log wrap occurs.

The redo logs wrapped and Capture cannot locate the archive logs.

If archive logs are available, uncompress and restore them to the archive log directory. SharePlex looks for the archive logs first in the Oracle archive log list, then in the location specified by the SP_OCT_ ARCH_LOC parameter. This parameter should always be set to the correct archived log directory. If you use compression for the archive logs, do not compress them until SharePlex is finished processing them. To determine the current log for SharePlex, issue the show capture command with the [detail] option in sp_ctrl on the source system. You can compress any logs that were generated before the current one. Note that this error also occurs when the archived log is corrupted.

It is taking more than the internally allowed time to apply a SQL statement to the target instance.

For more information, see Oracle Post-related issues on page 236.

SharePlex cannot locate the correct row to update in the target database.

Check for triggers, processes, or users that may have deleted the row on the target. For more information, see Solve synchronization problems on page 249.

Database not available.

Post cannot log into the target database.

Verify that the database is running, and determine whether someone changed the password of the SharePlex database account.

Oracle-related error messages

Can't access OBJ$Table

SharePlex cannot access the Data Dictionary, which it must be able to do in order to replicate.

Check the O7_DICTIONARY_ ACCESSIBILITY Oracle tuning parameter and make sure it is set to TRUE (the default).

Forward and backward operation Messages may be out of

counts do not match...

sequence.

Find out if Oracle was shut down before SharePlex. This can cause SharePlex to return errors, and in

SharePlex 9.2.8 Administrator Guide 242

sp_cop error messages

Cause

Solution

Error: sp_opst_mt (for o.blues920-o.ora9 queue bluesky) 15033 - Failed to execute SQL on table: QA.T_ DEST_1: ORA-00001: unique constraint (.) violated.
SQL Cache error messages

rare cases, corrupt the queues. The proper procedure is to shut down SharePlex and then shut down Oracle. For help resolving this problem, contact Quest Support.

A unique constraint was violated on the source system. The change entered the redo log, but Oracle rolled it back. The rollback also entered the redo log. SharePlex detects the constraint violation on the target.

Ignore this message. The tables are still in synchronization because Oracle rolled back the offending operation. This is an unavoidable error because of the way Oracle handles the violation.

Warning: Too many concurrent transactions. Will disable the SQL Cache capability.

The SQL Cache size is set to 1, and more cursors are still needed. SharePlex disables the SQL Cache feature in this case.

No action is required if this is your intended configuration. For more information, see Tune SQL Caching on page 281.

Warning: Running out of cursors. Number of cursors opened so far is number. Will attempt to decrease SQL Cache size.
or...
Notice: Shrinking SQL Cache size to number per session.

Post detected that it will exceed its maximum number of cursors and is trying to decrease the SQL Cache size.

No action is required unless the value gets to 1 and there are still not enough cursors. For more information, see Tune SQL Caching on page 281.

SQL Cache disabled.
Hints file error messages (Oracle only) 15050 hint file not found

The SQL Cache feature is disabled.

No action is required if this is your intended configuration. For more information, see Tune SQL Caching on page 281.

SharePlex looks for the hints.SID file whether you use it or not. The location of this file is the data subdirectory of the SharePlex variable-data directory. If this file gets moved or deleted, SharePlex returns this error message.

To prevent this message, create a blank hints file in the data subdirectory of the variable-data directory. Name it hints.SID.

15051 missing column in the hint file (either table of

The hints file is not configured correctly.

For more information, see Use Oracle INDEX hints on page 280.

SharePlex 9.2.8 Administrator Guide 243

sp_cop error messages

Cause

Solution

index name)

15052 syntax error for tablename
15053 syntax error for indexname

15054 source table's object_ id not found in object cache

The hints file includes a source table that is not in the active configuration. All tables in the hints file must be listed in the active configuration.

If this table is in the configuration, make certain that the owner name and table name are spelled in the hints file the same way as they are in the configuration.

15055 more than n valid entries were entered into the hints file

The hints file permits only as many table-index combinations as the value set by the SP_OPO_HINTS_ LIMIT parameter.

Either remove some of the tableindex combinations or increase the value of SP_OPO_HINTS_LIMIT. For more information about this parameter, see the Post parameters documentation in the SharePlex Reference Guide.

15056 error allocation memory for hints

This indicates a system-level memory problem; the hints file itself does not demand a significant amount of memory.

If you believe the system memory is sufficient, stop the Post process and start it again. If you are not using the hints file, you can ignore this error.

17000 error opening hint file

SharePlex cannot open the hints file.

Check the file for corruption. If the file is valid, make certain there is sufficient read permission for the Post process. For more information, see Use Oracle INDEX hints on page 280.

Environment-related error messages

sp_opst_mt: pid=num date/time src host/ sid=db01:N2PB /var/quest/vardir/log/ event_ log: Too many open files

The system file descriptors setting is not 1024.

Set the ulimit to 1024.
The ulimit can be set either as a system hard limit or a session-based soft limit, as follows:

SharePlex 9.2.8 Administrator Guide 244

sp_cop error messages

Cause

Solution

06/29/00 08:05 System call error: sp_ocap(que) (for o.QA11 queue o.QA11) No space left on device devname 06/29/00 08:05 Internal error:
sp_ocap (for o.QA11 queue o.QA11) 10705 - writecommit failed que_BUFWRTERR: Error writing buffer to file 06/29/00 08:05 Process exited sp_ocap (for o.QA11 queue o.QA11) [pid = 8692] -exit(1)

SharePlex ran out of space for the queues on the disk.

l Set a soft limit: A soft limit setting stays in effect only for the duration of the sp_cop session for which it was set, and then it reverts back to a default value that may be lower than the hard limit and too low for SharePlex.
For more information, see How to resolve disk space shortage on page 268.

gethostbyname name failed exiting

The local hosts file is not configured properly.

The host name is not specified correctly in the hosts file (/etc/hosts on Unix and Linux or hosts on Windows). If this system is not part of a cluster, make corrections to the name in the file. If this system is part of a cluster, the virtual IP address must be mapped to a host alias in the hosts file. For instructions on configuring SharePlex in a cluster, see the SharePlex Installation and Setup Guide.

Other error messages

Snapshot too old

The read-consistent view

Increase the size of the rollback

required by SharePlex is no segment.

longer available.

Parameter paramname does not exist in the paramdefaults file. Using hard coded default
value. Please make sure that your param-defaults file is the correct version.

SharePlex cannot find a parameter that it needs to reference. You have an older version of the paramdefaults file than the version of SharePlex that you are running. Someone might have updated the SharePlex binaries using a downloaded patch, and may not have installed the latest

Always check for an updated param-defaults file when you manually update SharePlex.

SharePlex 9.2.8 Administrator Guide 245

sp_cop error messages

Cause

Solution

param-default file.

Invalid DATE format detected in record with rowid=rowid, on obj object_id. See capture log for detail.
shs_SHMERR: an error occurred with shared memory.

A user or application entered an invalid date value into the Oracle database that bypassed the database's validity check.
You ran the qview utility without shutting down SharePlex (sp_cop).

Set the SP_OCT_DEF_ parameters to enable SharePlex to fix the format of dates and times if they are not caught by the database checks. See the SP_OCT_DEF parameters documentation in the SharePlex Reference Guide.
Shut down SharePlex and re-run qview.

Solve Oracle DDL replication problems
This section reviews many common problems and solutions encountered when replication of Oracle DDL is active. For more information about SharePlex DDL support, see Configure DDL replication on page 152. If the issue you are experiencing is not listed in this documentation, search the SharePlex Knowledge Base at: https://support.quest.com. The Knowledge Base provides filtering options and links to other resources that can help you use and troubleshoot SharePlex.
DDL is not replicating
Only some Oracle DDL is enabled by default. Other DDL support must be explicitly enabled with parameter settings. The SP_OPO_STOP_ON_DDL_ERR parameter is defaulted to direct the Post process to stop if there is an error applying the DDL, to enable you to correct the problem to keep the databases synchronized. This parameter should be set to 1 (on). When enabled, messages such as the following notify you that DDL was skipped.
Skipping generic 9i DDL operation, schema (bob) could not be set. FAILED DDL Replication for "create user bob." Solution: Make certain the following parameters are set to appropriately. See Control Oracle DDL replication on page 152 for more information about how to use these parameters to configure DDL replication.
l SP_OCT_REPLICATE_DDL l SP_OCT_AUTOADD_ENABLE l SP_OCT_AUTOADD_MV l SP_OCT_AUTOADD_SEQ l SP_OCT_REPLICATE_ALL_DDL

SharePlex 9.2.8 Administrator Guide 246

Replicated DDL is not completely displayed in the Event Log
SharePlex prints replicated DDL to the Event Log, but it truncates statements longer than 2,000 characters. Only the first 2,000 characters are recorded in the log. Solution: None required.

Solve queue problems
This topic helps you solve replication problems that may be related to the SharePlex queues. If the issue you are experiencing is not listed in this documentation, search the SharePlex Knowledge Base at: https://support.quest.com. The Knowledge Base provides filtering options and links to other resources that can help you use and troubleshoot SharePlex.

SharePlex is running out of disk space
Data accumulating in the queues can cause SharePlex to run out of disk space. The following are some possible causes and solutions.

Problem Description

Solution

Stopped replication processes

When replication processes stop, data accumulates in the queues.

If a process was stopped by a user, find out why and then start it as soon as possible so that SharePlex can process the accumulated data. If a process stopped because of an error, view the Event log with the show log command to find out what happened, then resolve the problem so that processing can continue and the queue backlog can be reduced.

Large operations

The Post queue may become large when storing large transactions for which there has not yet been a COMMIT. To allow for rollbacks and data recovery, SharePlex retains data in the post queue until it receives the COMMIT.

View the post queue with the qstatus command. If the value in the Backlog (messages) field remains constant or is shrinking, while the value in the Number of messages field is increasing, Post is waiting for a COMMIT before releasing the data.
Use the show post detail command to verify that Post is processing transactions normally. If possible, set the COMMIT point for your application to 500 to generate smaller SQL statements for SharePlex to process. Also, consider creating a configuration that uses named post queues which isolate tables that are known for long transactions. For more information, see Configure named post queues on page 89.
If the accumulation of messages in the Post queue is threatening to cause disk space issues, you can stop Import until Post can clear out some of the operations from other transactions. For more information, see How to resolve disk

SharePlex 9.2.8 Administrator Guide 247

Problem Description

Solution

space shortage on page 268.

(Oracle) Capture is processing archive logs

If Capture is processing archive logs, the capture queue consumes disk space while the archived records are being processed.

If the location of Capture in the logs is far behind that of the database, the latency between the source and target data may be too large to be acceptable to the target users. Rather than add more disk space to the variable-data directory, it might be more practical to run ora_cleansp to clean up the replication environment and queues, resynchronize the data, and reactivate the configuration.
If SharePlex continues to run out of disk space because Capture is processing the archive logs, the SharePlex Support team can help you tune the performance of Capture and adjust the redo configuration.

Unplanned increase in source transaction activity

An unplanned increase in activity on the source system can cause data to accumulate in the Post queue and approach the maximum amount of allocated disk space.

For more information, see How to resolve disk space shortage on page 268.

There is a "failure to write and open queue" error
If there is a sequence of messages in the Event Log similar to the following, stop and start sp_cop. 1 08-06-12 13:20:17.089485 2384 1 sp_ordr(rim) (for o.user queue o.user) Error opening output queue rv=9 que_open(-,writeuser+ X,0x0a02d364+PR+o.user+sp_ ordr+o.user) Notice 08-06-12 13:20:17.089485 2384 1 sp_ordr (for o.user queue o.user) data route to a02d364.48.7e failed err=100 Error 08-06-12 13:20:17.089485 2384 1 sp_ordr (for o.user queue o.user) 11004 sp_ordr failure writing to queue(s) Notice 08-06-12 13:20:17.089485 2384 1 sp_ordr (for o.user queue o.user) Exit sp_ ordr to retry rim-write. Info 08-06-12 13:20:17.089485 2384 1 Process exited sp_ordr (for o.user queue o.user) [pid = 8183] - exit(1)
The queues are corrupted
If the system that hosts SharePlex fails, one or more SharePlex queues can become corrupted, and there will be errors reading from and writing to them. In that case, the purge config and abort config commands cannot be used, because they rely on the queues to operate. Solution: Contact SharePlex Support to resolve queue corruption issues.

SharePlex 9.2.8 Administrator Guide 248

The post queue seems too large
If the size of the post queue seems too large relative to the number of messages it contains, that is not unusual. The SharePlex post queue actually consists of a number of sub-queues, each approximately corresponding to a user session on the source system. A sub-queue can have one or more files associated with it, each with a default size of 8 MB. If the entire 8 MB of size is not used, the file remains on the system even after the data has been posted and read-released. Solution: The size for the post queue in a status display is the actual disk space used by all of the queue files. You can eliminate obsolete files that have been read-released by using the trim command in the SharePlex qview utility. Thecommand preserves files containing data not yet posted and committed to the target database. See the SharePlex Reference Guide for more information about the qview utility.
Solve synchronization problems
This section reviews the causes and solutions for common synchronization problems. If you try these solutions and are still having problems, contact Quest Support. For more information, see Understand the concept of synchronization on page 30. If the issue you are experiencing is not listed in this documentation, search the SharePlex Knowledge Base at: https://support.quest.com. The Knowledge Base provides filtering options and links to other resources that can help you use and troubleshoot SharePlex.
How SharePlex reports out-of-sync conditions
For all objects except those involved in transformation, SharePlex verifies that the source and target data in a given operation are synchronized before posting the replicated data to the target. SharePlex does not verify synchronization if transformation is being used because:
l Transformation changes the target data, so before and after images cannot be compared. l The transformation routine posts the data, not SharePlex.
When SharePlex determines that source and target data are different, it generates error conditions but continues to post other data from the post queue. To direct Post to stop processing altogether when it detects an out-of-sync condition, change the SP_OPO_OUT_OF_SYNC_SUSPEND (Oracle) or SP_OPX_OUT_OF_ SYNC_SUSPEND (Open Target) parameter. See the Post parameter documentation in the SharePlex Reference Guide. When an out-of-sync condition occurs, the Post process logs a message in the Status Database and also to the Event Log. To view these files in sp_ctrl:
l Status Database: Use the show statusdb or show sync command. l Event Log: Use the show log command.
Use these commands frequently to monitor for out-of-sync errors. The following is an example of how SharePlex reports an out-of-sync condition.
sp_ctrl (irvspxu14:8567)> show sync Out Of Sync Status Database irvspxu14 Count Details
SharePlex 9.2.8 Administrator Guide 249

----- -------------------3 Table "SCOTT"."TG_TEST1" out of sync for queue irvspxu14 since 16-Jun08 17:06:33 3 Table "SCOTT"."TG_TEST2" out of sync for queue irvspxu14 since 17-Jun08 15:47:58 1 Table "SCOTT"."TG_TEST3" out of sync for queue irvspxu14 since 17-Jun08 15:52:03 When data goes out of synchronization, SharePlex logs the failed SQL statements to the database_ID_ errlog.sql file in the data sub-directory of the SharePlex variable-data directory. IMPORTANT: If you see an out-of-sync message in the Status Database and in the Event Log, but there is no record in the database_ID_errlog.sql file for the transaction, do not ignore those messages. They could be associated with a ROLLBACK. Regardless of whether or not a transaction is rolled back, SharePlex still compares the pre-images of the source and target rows. If they are different, that indicates that the data is out of synchronization. Only when a transaction is committed on the source but fails on the target does SharePlex log it to the database_ID_errlog.sql file, to give you a record of the statement that should have been applied as a tool for problem solving and for manually applying the statement if appropriate. Rolled back statements are canceled operations, and therefore not logged on the target.
Detect false out-of-sync conditions
Sometimes an out-of-sync message can be false, and the data is not out-of-sync. You can compare the source and target data by using the compare command, or you can perform the following comparison manually.
To compare with the compare command See the compare commands in the SharePlex Reference Guide.
To verify that data is out-of-sync 1. Get the rowid of the affected row from the database_ID_errlog.sql file in the variable-data directory on the target system. 2. Using the rowid in the WHERE clause, run a SELECT query on the source table to get the row data for this rowid. 3. Run a SELECT query on the target table, using the row data that you got from the source query in the WHERE clause. 4. Compare the query results. If they are different, the rows are out of synchronization. If they are the same, the rows are synchronized.
Common out-of-sync conditions and solutions
The following are common ways in which data goes out of synchronization. In most cases, SharePlex detects out-of-sync conditions and returns an error message, but there are some situations where an out-of-sync condition is hidden, and SharePlex will not return an error.
SharePlex 9.2.8 Administrator Guide 250

Out of sync Description condition

Solution

Incorrect cleanup procedure

If one of the database_cleansp cleanup utilities was run on some, but not all, systems associated with an active configuration, SharePlex perceives an out-ofsync condition.

Determine whether or not the cleanup utility was run on all systems by viewing the Event Log. The log tells you if and when it was run on each system. It also tells you if and when the configuration was activated. You can compare the times for those events to determine what happened.
If the cleanup was not completed on all replication systems for this configuration, run the cleanup utility on all systems. Because the cleanup removes replication queues and processes and deactivates the configuration, you must perform initial synchronization again.

DDL changes

Some DDL-related causes of out-of-sync conditions are:
l Non-replicated DDL changes are made to source objects, but the configuration was not reactivated so that the objects can be reanalyzed.
l DDL that SharePlex replicates also gets performed manually on the target.

For a list of supported DDL, refer to the SharePlex Release Notes. To undo duplicate DDL changes made manually and also by SharePlex:
1. Stop the Post process (it might already be stopped). sp_ctrl(sysB)> stop post
2. Alter the target table to undo the DDL changes.
3. Start Post and let SharePlex post the replicated DDL (which is still in the post queue). sp_ctrl(sysB)> start post

DML changes made directly to the target

If applications or users make DML changes to the tables on the target, the results of those changes will cause hidden out-of-sync conditions until Post attempts to apply a replicated change to the affected rows. When the change is applied, SharePlex returns an out-of-sync error.

Prevent all DML access to the target tables that are in replication.
You can use the compare and repair commands to compare tables for outof-sync rows and then repair those rows. For more information, see the command documentation in the SharePlex Reference Guide.

SharePlex 9.2.8 Administrator Guide 251

Out of sync Description condition

Triggers on target objects

Triggers must be disabled on target objects. The triggers fire on the source system and SharePlex replicates their effects to the target.

Solution
If triggers have fired on the target system, the objects changed by the triggers are out of synchronization and must be resynchronized. For more information, see How to resynchronize source and target tables on page 261.

To disable the effects of triggers on an Oracle target
To disable the effects of triggers on target Oracle objects after the data is resynchronized, you can either of the following:

l Run the sp_add_trigger.sql script, which directs triggers to ignore the SharePlex Oracle user. See the utilities documentation in the SharePlex Reference Guide for more information about the trigger scripts.

l Disable the triggers if they are not needed.

Unnecessary constraints

The only constraints that are necessary on target tables in a one-way replication configuration are primary and unique key constraints. CHECK constraints are not necessary on the target because they are satisfied on the source. FOREIGN KEY and ON DELETE CASCADE constraints are also satisfied on the source, and SharePlex replicates the child operations to the correct tables on the target.

For an Oracle target, you can leave ON DELETE CASCADE constraints enabled if you configure SharePlex to ignore them.

See Set up Oracle database objects for replication on page 1for how to handle constraints on replicated objects.

Duplicate entries in the configuration file.

Duplicate entries, where the source, target, and routing map are identical, cause double posting on the target.

Copy the configuration file to a new file.
Find and remove any duplicates in the new file.
Perform a resynchronization and reactivation. For more information, see How to resynchronize source and target tables on page 261.

SharePlex 9.2.8 Administrator Guide 252

Out of sync Description condition

Solution

Lack of disk space

Data goes out of synchronization if user transactions continue when SharePlex does not have enough room to accommodate them in the queues. This can happen if:

For more information, see How to resolve disk space shortage on page 268.

l The network or target system was unavailable and too much data accumulated in the export queue.
l The target was unavailable and too much data accumulated in the Post queue.
l Capture lost pace with the logging of source transactions. In this case, data accumulates in the capture queue.
l A SharePlex process was stopped, but not restarted.
l The flush command was issued, but Post was not started again.

Changes to column conditions in horizontally partitioned replication

Out-of-sync conditions can result from the use of horizontally partitioned replication in the following cases:
l A column condition value is updated and the new value no longer satisfies the row selection criteria.

l A row that does not satisfy the column condition is updated to satisfy the condition.

Create column conditions so that partition shift does not occur. For more information, see Configure horizontally partitioned replication on page 93.

Not reactivating after a configuration change.

If a table was added to the configuration, but the configuration was not reactivated, operations on that table are not being replicated.
NOTE: For Oracle source tables, the auto-add feature is enabled by default, and any new table whose name satisfies a wildcard in the configuration file gets automatically added to replication. For more information, see Control Oracle DDL replication on page 152.

Resynchronize the affected tables, then reactivate the configuration so that SharePlex can update its object cache. For more information, see How to resynchronize source and target tables on page 261.

Queue corruption

If the SharePlex queues are corrupted, such as if there is a system failure, the data in them can be lost. This requires a resynchronization.

For more information, see How to resynchronize source and target tables on page 261.
(Oracle) To avoid queue corruption during system failure, you can use the parameter SP_QUE_SYNC. See the Queue parameters documentation in the SharePlex Reference Guide.

SharePlex 9.2.8 Administrator Guide 253

Oracle-related out-of-sync conditions and solutions
The following are common synchronization issues that relate specifically to replication between Oracle databases. In most cases, SharePlex detects out-of-sync conditions and returns an error message, but there are some situations where an out-of-sync condition is hidden, and SharePlex will not return an error.

Out of sync Description condition

Solution

DDL changes Some DDL-related causes of out-ofsync conditions are:
l Non-replicated DDL changes are made to source objects, but the configuration was not reactivated so that the objects can be re-analyzed.
l DDL that SharePlex replicates also gets performed manually on the target.

Inadequate or non-existing conflict resolution routines

Conflict resolution procedures are required in a peer to peer (activeactive) configuration. SharePlex uses the conflict resolution procedures to determine which operation to post when the same data change is received from different systems.

Revise and test the conflict resolution procedures, then resynchronize the data. For more information, see How to resynchronize source and target tables on page 261.

Log wrap

If the redo logs wrap before Capture can process the data it needs, the data can go out of synchronization if archived logging is not enabled or if the archive logs needed by Capture were removed. (Normally, Capture would access the archive logs and continue replicating.)

l See Correct the problem first on page 256
l If archiving is not enabled, there are no archive logs for SharePlex to read. Data lost after the log wrap cannot be recovered. Enable archived logging, and resynchronize the data. For more information, see How to resynchronize source and target tables on page 261.

SharePlex 9.2.8 Administrator Guide 254

Out of sync condition
LONG columns
Changes to keys

Description

Solution

l If SharePlex is many logs behind Oracle, consider resynchronizing the data instead of restoring the logs. It might take less time than Capture would take to process the missing records from the archive logs. In addition, it eliminates the possibility that the capture queue could exceed free disk space while processing the archive logs. You can base your decision on the size of the redo logs and the number of tables being replicated, both of which determine how much information Capture must process. Also take into consideration how much latency the users of the target data can tolerate.
l If archive logs are available, copy the appropriate logs back to the archived-log directory on the source system, or use the SP_OCT_ARCH_LOC parameter to point SharePlex to their location.

If a table has no primary or unique key, SharePlex builds a simulated key based on all of the columns except the LONG and LOB columns. If the LONG columns in the target rows are the only columns that contain unique values, multiple rows could meet the criteria for the simulated key. SharePlex could apply the UPDATE or DELETE to the wrong row without the error being detected, causing the table to go out of synchronization without an error message.

If you can create a key from columns that ensure uniqueness on the target table(s), you can avoid this kind of out-of-sync condition. After you create the key, resynchronize and re-activate those objects so that SharePlex can update its object cache.
If adding a primary or unique key is not possible (such as when packaged applications are in use), uniqueness of rows on the target system cannot be ensured.

If tables use automatically generated numerical sequences as keys, and a key value changes, this may cause duplicity on the target system. If the new value already exists as a key in another row on the target system, SharePlex returns a unique-key constraint violation and out-of-sync error. This can happen when you update values using an x +n formula, where n is an incremental increase. It is possible that one of the x +n values will equal an

Create the sequences so that the updates cannot result in a duplication of keys on the target system.

SharePlex 9.2.8 Administrator Guide 255

Out of sync Description condition

Solution

existing value.

DBMS_ SCHEDULER procedures running on source and/or target system

The effects of these procedures, such as objects being created, manipulated and dropped outside replication, may not be visible to replication or may not be supported, resulting in data changes that cause out-of-sync conditions.

Exclude source and target objects from the jobs.

Virtual private database

If replicating data that is configured as a virtual private database, the SharePlex database user may not have the access rights to capture the data. Any changes to that data will not be reflected on the target.

If you do not need that data replicated to the target, you can filter it out of the SharePlex configuration by means of partitioned replication.
If you want the data to be replicated, assign the SharePlex user the correct access rights.

PK/UK logging not enabled

Certain replication problems can be prevented by logging key values. SharePlex fetches key values based on the rowid. Any operation that changes the rowid, such as ALTER TABLE...MOVE, can cause the wrong key values to be used for subsequent DML operations.

SharePlex recommends that both primary key and unique key supplemental logging be set, or that a supplemental log group on unique columns be defined for every table in replication.

Correct the problem first
If data is out-of-synchronization, do the following before you resynchronize the data:
1. Determine why it happened before you resynchronize the data. Otherwise, the problem can repeat itself and result in more data going out of synchronization.
2. Stop the Post process to prevent further errors. If the accumulation of messages in the Post queue is threatening to cause disk space issues, and if there is enough disk space available on the source system, you can stop Import until Post can clear out some of the operations from other transactions. For more information, see How to resolve disk space shortage on page 268.
3. View the Status Database and the Event Log to determine the cause of the problem. 4. Resolve the problem.
Resynchronize data
For more information, see How to resynchronize source and target tables on page 261.

Solve compare command errors
If you are having trouble with any of the compare or repair commands, refer to the following for assistance. If the issue you are experiencing is not listed in this documentation, search the SharePlex Knowledge Base at:

SharePlex 9.2.8 Administrator Guide 256

https://support.quest.com. The Knowledge Base provides filtering options and links to other resources that can help you use and troubleshoot SharePlex.

Problem Description

Solution

Oracle error 904

Error 904 calling oexec in de_select_ prepare_to_fetch.
A comparison failed because the source and target tables being compared are structurally different. The compare and repair commands do not detect and repair out-of-sync conditions caused by unsynchronized DDL operations or tables that are not structurally identical.

Do a DESCRIBE on both tables. It probably will show that the tables do not have an equal number of columns or the data types are different. After you correct the DDL problem, you can run a repair to resynchronize the values in the rows.

"Too many users" error

Can not add DataEquator queue reader
que_TOOMANYUSERS: User table is full.
The maximum number of processes reading from and writing to the SharePlex queues was exceeded. No more than 20 processes can read from and write to the post queue at the same time, including the replication processes and the compare and repair processes. An error is most likely to occur when a repair option is used, because a repair accesses the queue much longer than a comparison without a repair.

There is no workaround or way to adjust the limit, nor is there a way to determine how many compare processes can run concurrently without exceeding the limit.
TIP: To compare multiple tables at the same time, without being restricted by process limitations, use the compare using command. To limit the tables being compared, create a new configuration containing only the ones that you want to compare, and use it for the comparison. (Do not activate that configuration.) All of the tables are compared with one compare using process.

A client process fails to die

When a sp_desvr server process dies, the associated sp_declt client processes usually die. If a process does not die, you can kill it.

A server process fails to die

When you kill a sp_declt client process or it dies on its own -- or if the sp_desvr server process has not communicated with the client -- the sp_ desvr server process usually exits after a certain amount of time, which is controlled by the SP_ DEQ_TIMEOUT parameter.

For more information, see Kill compare processes on page 257.
For more information, see Kill compare processes on page 257.

Kill compare processes
To kill a client process If you need to kill a sp_declt client process, and there are multiple compares running, you can determine which one to kill in one of the following ways:
l By viewing the sp_declt log file -- In the file, look at the Session IDs of the sp_declt processes and find the one that matches the PID of the sp_desvr process that died. That is the sp_declt process to kill. The

SharePlex 9.2.8 Administrator Guide 257

sp_declt Session ID is the same as the PID of the associated sp_desvr process. l By viewing the Event Log -- The Event Log records the startup of each sp_declt client process and its
PID. A subsequent entry in the log records the compare log file to which the process is writing. Within the compare log file's name in that entry is the PID of the server process. For example, in the following sample entry, the sp_declt process PID is 2450. The process writes to compare log ../o734v32a_declt1228-01.log. The 1228 is the PID of the server process. 05/04/01 17:01 Process launched: sp_declt (for o.o734v32a-o.o734v32a- 87056 queue all) [pid = 2450] 05/04/01 17:01 Notice: sp_declt(deq) (for o.o734v32a-o.o734v32a-87056 queue all) Opened DataEquator session log file /u10/julia30014/var7/ log/o734v32a_declt1228-01.log You can search the log file names for the server process that died, and look for the client process associated with that log file to determine the correct PID to kill.
On Windows systems, the logs also record the startup of the associated sp_cop process.
To kill a server process If you need to kill a sp_desvr server process when a sp_declt client process dies, look in the Event Log to find out which log the sp_declt client process was writing to. The Event Log records the startup of each client process and its PID. A subsequent entry in the log records the compare log file to which the process is writing. Within the compare log file's name in that entry is the PID of the server process. For example, in the following sample entry, the sp_declt process PID is 2450. The process writes to log ../o734v32a_declt-1228-01.log. The 1228 is the PID of the server process, and that is the process to kill. 05/04/01 17:01 Process launched: sp_declt (for o.o734v32a-o.o734v32a- 87056 queue all) [pid = 2450] 05/04/01 17:01 Notice: sp_declt(deq) (for o.o734v32a-o.o734v32a-87056 queue all) Opened DataEquator session log file /u10/julia30014/var7/ log/o734v32a_declt-122801.log On Windows systems, the logs also record the startup of the associated sp_cop process.
Solve other replication problems
This section reviews solutions to other replication problems. If the issue you are experiencing is not listed in this documentation, search the SharePlex Knowledge Base at: https://support.quest.com. The Knowledge Base provides filtering options and links to other resources that can help you use and troubleshoot SharePlex.
SharePlex does not run on a Windows system
SharePlex uses the MKS Toolkit® (also known as NuTCRACKER) operating environment from Parametric Technology Corporation (PTC) to run on Windows systems. If the NuTCRACKER service is stopped or disabled, or if the NuTCRACKER files have been removed or relocated, there will be errors when you try to run SharePlex. Solution:
SharePlex 9.2.8 Administrator Guide 258

1. Run the Windows Task Manager, then click the Processes tab. 2. Check to see if the NuTCRACKER service is running. The process name is nutsrvx, where x is the
version of the NuTCRACKER software. 3. If the process is not running, start the NuTCRACKER service by using the Services panel of the
Administrative Tools Control Panel. l If you do not see the NuTCRACKER service, re-install SharePlex. l If the NuTCRACKER service started, but SharePlex still returns errors, check to see if the NuTCRACKER files were relocated. To determine the correct installed location, look at the following in the Windows Registry: HKEY_LOCAL_MACHINE\SOFTWARE\Data Focus\Runtime l The InstallDir string in the right pane of the Runtime node shows the correct location for the files. Search for the MKS Toolkit folder and restore the files to the location specified in the InstallDir entry.
If you cannot locate the files or cannot restore them to the correct location, do the following: 1. Stop the SharePlex and NuTCRACKER services, if running. 2. Run regedit to open the Registry Editor. 3. Delete the Data Focus and Mortice Kern Systems registry folders under HKEY_LOCAL_ MACHINE\SOFTWARE\. 4. Close the Registry Editor. 5. Restart the system. 6. Reinstall SharePlex. Make certain to reinstall SharePlex in the same location, and make certain to install the NuTCRACKER component when prompted. 7. Restart the system to make the new NuTCRACKER environment effective.
A "Can't unlink file" error occurs on Windows systems
A nuisance error similar to the ones below sometimes occur on Windows systems. The files eventually unlink. Text file busy Unlinking file: 'r:\splex2102/rim/o.SERV+C+0.0000000
Or... System call error: sp_ordr.exe(osp) (for o.SERV queue o.SERV) Text file busy 17003 - Can't unlink file R:\Splex2100/state/o.SERVlog_ sp_ordr.30
Common connection errors
The following are solutions to common errors when starting sp_ctrl, or with forming a connection with the host, port or [on host] commands in sp_ctrl.
SharePlex 9.2.8 Administrator Guide 259

Explanation of connection error messages

Error

Cause

Solution

Host unknown: cannot form connection

Appears when either the host command or [on host] option is issued.

Network unreachable

The network is down.

Verify that the system to which you want to connect is running and that you are using the correct system name.
Find out how long the network administrator expects it to last. If the downtime could cause the SharePlex queues to exceed their disk space, take measures to avoid having to resynchronize the data. For more information, see How to resolve disk space shortage on page 268.

Export cannot connect to import on hostname: timeout waiting for ack

Export cannot connect to the target because its connection was timed out by the network configuration. This can occur when there is little replication activity and the network has a timeout setting.

Set the SP_XPT_KEEPALIVE parameter to 1. This setting tells the Export process to send a "hello" message to Import at regular intervals to prevent the TCP timeouts.

User is not authorized as a SharePlex user -- check /etc/group

You do not have user permissions to execute the operation.

SharePlex users must be listed in the /etc/group file (Unix and Linux) or in the Users list (Windows) under one of the SharePlex user groups: SharePlex Admin, spoper, spview.

unauthorized connection attempt from host hostname.
net

A connection from a remote machine was denied because its name is not listed in the auth_ hosts file.

See the error message for the name of the system. To allow that system to connect to the sp_cop on the local system, add its name to the auth_hosts file.

Common command errors

Error Deactivate/flush a nonactive datasource Bad routing specification
Status db file is corrupt.

Cause
You are attempting to flush a configuration that is not active. The syntax in the routing map is incorrect.
The Status Database has been damaged.

Solution
None required.
For more information, see Routing specifications in a configuration file on page 66.
Shut down SharePlex and remove the statusdb file, which resides in the data subdirectory of the SharePlex variable-data directory. SharePlex will create

SharePlex 9.2.8 Administrator Guide 260

Error
Parameter does not exist in database.

Cause

Solution

another one when you start sp_ cop again.

You tried to set a parameter, and you entered the wrong name or the parameter is deprecated for your SharePlex version.

Use the list param command to view the SharePlex parameters for your version and to verify the spelling.

Parameter type checking failed - look in param defaults file.
Unknown service specified. or... No such module. or... Service may be only one of: post, read, import, export, capture, all.

You might have entered the wrong data type for the parameter.

Use the list param command to determine the valid data type.

Valid service (process) names are Issue the command again with capture, read, export, import, the correct name. post.

Command was called with an invalid argument.

The command contains invalid input.

Issue the help command to view valid input for the command.

or...

Unknown keyword used in command.

Permission denied for command You are not a member of the

- check your authorization

user group that can issue this

level.

command.

Default host is not defined: SharePlex cannot to determine

use the `host' command or [on which system you want the

host] option.

command to affect.

Issue the authlevel command to view your authorization level.
Either establish a default host with the host command or use the [on host] option with the command that you want to issue (if available).

How to resynchronize source and target tables
The following instructions help you decide how to resynchronize out-of-sync tables.
l For Oracle tables, if only a few tables are out of synchronization, and they are not large, you can use the compare command in sp_ctrl to see how many rows are out-of-sync in each one. If the number of outof-sync rows is small, you can run the repair command to resynchronize them. If the number of out-ofsync tables is large, it might take less time to re-synchronize the databases than it will to use the compare and repair commands to repair them. For more information, see Overview of Compare and Repair on page 272.

SharePlex 9.2.8 Administrator Guide 261

l For both Oracle and Open Target databases, you can use the SQL statement(s) in the ID_errlog.sql file to patch the tables manually after you correct the problem. For more information, see Manually patch outof-sync tables on page 262.
l You can resynchronize the data in a number of ways. See the following topics: l Manually patch out-of-sync tables on page 262 l Resynchronize by copying the source tables on page 262 l Resynchronize with Oracle transportable tablespace on page 263 l Resynchronize with an Oracle hot backup on an active database on page 264
Manually patch out-of-sync tables
Valid for: All database types If the number of synchronization errors is small, you can try to repair out-of-sync tables manually. When the Post process detects an out-of-sync condition, it ignores the error and continues to apply the next operations in the post queue. However, Post logs source SQL statements that cause out-of-sync errors to an error file calleID_ errlog.sql. (ID is the identifier that SharePlex uses for the target instance, such as the ORACLE_SID or the database name.) You can apply those SQL statements to a target table through the native SQL interface of the database. Because this procedure bypasses the comparison made by Post, the operations should succeed assuming the structure of the target table did not change. SharePlex stores ID_errlog.sql in the data sub-directory of the variable-data directory on the target system. The entries in the file are similar to the following example:
-- Host (irvlabua) Sid (al920u64) -- session 2, 1 error ---- [1] Tue Dec 11 13:31:32 2007 -- redolog seq#/offset 26622/26980368 -- redolog timestamp 641050290 (12/11/15 13:31:30) -- original rowid AAE0m8AAWAAAAFEAAA -- -- NOT FOUND delete from "SP_5"."QA_LOB_DISABLE_INROW" t where rownum = 1 and "KEY"='01';
To apply the SQL manually
1. Stop user access to the affected source table. 2. On the target system, open the ID_errlog.sql file. 3. Apply the SQL statement(s) to the target table. 4. Reactivate the configuration if you had to make any changes to it.
sp_ctrl> activate config filename 5. Allow user access to the source table.
Resynchronize by copying the source tables
Valid for: All database types This procedure restores synchronization to out-of-sync target tables by applying a copy of the source tables. You only need to resynchronize the tables that are out of synchronization, so users can continue accessing all other tables.
SharePlex 9.2.8 Administrator Guide 262

IMPORTANT! Before you start, review this procedure and see the SharePlex Reference Guide for more information about the commands that are used.
1. On the source and target systems, make sure sp_cop is running. 2. On the target system, run sp_ctrl. 3. [If necessary] On the target system, issue the show sync command to identify the tables that are out of
synchronization. sp_ctrl> show sync
4. On the source system, stop activity for the out-of-sync tables. 5. On the source system, run sp_ctrl. 6. On the source system, issue the flush command. NOTE: This command has additional options for use
with named queues or multiple targets. See the SharePlex Reference Guide for more information about this command.
sp_ctrl> flush datasource 7. On the source system, copy the tables. 8. On the source system, reactivate the configuration file if you had to make any changes.
sp_ctrl> activate config filename 9. On the source system, allow users back onto the source tables. 10. On the target system, issue the status command until it shows that Post stopped.
sp_ctrl> status 11. On the target system, restore the tables. 12. On the target system, disable or modify triggers, referential integrity constraints and check constraints
according to the requirements of your replication strategy. 13. On the target system, determine the status ID of each message by viewing the Status Database.
sp_ctrl> show statusdb detail 14. On the target system, clear each message with the following command.
sp_ctrl> clear status statusID 15. On the target system, start the Post process.
sp_ctrl> start post
Resynchronize with Oracle transportable tablespace
Valid for: Oracle database The transportable tablespace feature enables you to resynchronize numerous out-of-sync tables quickly and with minimal downtime. To use the transportable tablespace feature, follow the instructions in the Oracle documentation for generating a tablespace set, moving the tablespace set to the target database, and plugging the set into the database. The following instructions contain steps only for using this feature to resynchronize data. It assumes familiarity with using the transportable tablespace feature. IMPORTANT! Before you start, review this procedure and see the SharePlex Reference Guide for more information about the commands that are used.
SharePlex 9.2.8 Administrator Guide 263

1. On the source system, set the source tablespace to READ ONLY. SQL> ALTER TABLESPACE name READ ONLY;
2. On the source system, run sp_ctrl. 3. On the source system, issue the flush command in sp_ctrl. NOTE: This command has additional
options for use with named queues or multiple targets. See the SharePlex Reference Guide for more information.
sp_ctrl> flush datasource 4. Export the metadata to an export file according to the Oracle documentation. 5. When the export is finished, copy the datafiles to a secondary location on the source system. This
minimizes the impact on the source database of copying the files to the target system. 6. On the source system, set the source tablespace(s) to READ WRITE mode.
SQL> ALTER TABLESPACE name READ WRITE; 7. On the target system, drop the existing datafiles and tablespaces from the target database so that the
copied files can be applied. 8. Copy the files from the secondary location on the source system to the target system. 9. On the target system, use the Oracle import utility to import the metadata and the tablespace definitions. 10. On the target system, set the tablespace(s) to READ WRITE mode.
SQL> ALTER TABLESPACE name READ WRITE; NOTE: SharePlex must be the only user permitted to have write access to the target tables, unless you are using peer-to-peer replication. 11. On the source system, reactivate the configuration file if you had to make any changes to it.
sp_ctrl> activate config filename 12. On the target system, run sp_ctrl. 13. On the target system, start the Post process.
sp_ctrl> start post
Resynchronize with an Oracle hot backup on an active database
Valid for: Oracle database When you use an Oracle hot backup and the reconcile command to resynchronize a target instance, users can continue to access the production data while the backup is made and applied.
IMPORTANT: l To resynchronize centralized reporting, such as a data warehouse, you cannot use a hot backup from all source systems. One backup would override the data from the previous one. You can use a hot backup of one of the source instances to establish the target instance, and then use another method such as export/import or transportable tablespaces to copy the tables from the other instances. l To resynchronize peer-to-peer replication, you must quiet all of the secondary source systems for the duration of this procedure. Move all users to the primary system, and then follow the procedure. After the procedure has been performed on all of the secondary systems, users can resume activity on them. l Before you start, review this procedure and see the SharePlex Reference Guide for more information about the commands that are used.
SharePlex 9.2.8 Administrator Guide 264

To resynchronize with a hot backup
1. On the source and target systems, run sp_ctrl. 2. On the target system, stop the Post process. This allows the replicated data to accumulate in the post
queue until the target instance has been recovered and reconciled. sp_ctrl> stop post
3. On the source system, run the Oracle hot backup. 4. On the source and target systems, verify that sp_cop, sp_ctrl and all SharePlex processes (Capture,
Read, Export, Import, Post) are running. sp_ctrl> status
5. Switch log files on the source system.
l To recover the database to a sequence number, make a note of the highest archive-log sequence number.
l To recover the database to an Oracle System Change Number (SCN), pick an SCN to recover to on the target database.
6. Recover the target database from the hot backup:
l If recovering to a sequence number, recover the database from the hot backup using the UNTIL CANCEL option in the RECOVER clause, and cancel the recovery after Oracle has fully applied the log from the previous step.
l If recovering to a SCN, recover the database from the hot backup using the UNTIL CHANGE SCN option in the RECOVER clause, and cancel the recovery after Oracle has applied the logs matching the SCN from the previous step.
7. Open the database with the RESETLOGS option.
8. On the target system, issue the reconcile command. If you are using named post queues, issue the command for each one. Issue the qstatus command if you are unsure of the queue name.
l If recovering to a sequence number, substitute the sequence number of the log that you noted in step 5. sp_ctrl> reconcile queue queuename for datasource-datadest seq sequence_ number Example: reconcile queue SysA for o.oraA-o.oraA seq 1234
l If recovering to a SCN, substitute the SCN that you noted in step 5. sp_ctrl> reconcile queue queuename for datasource-datadest scn scn_number Example: reconcile queue SysA for o.oraA-o.oraA scn 0123456789
The reconcile process retains control of sp_ctrl until it is finished, and then the sp_ctrl prompt returns.
9. On the target system, log onto SQL*Plus as the Oracle user for SharePlex, and run the cleanup.sql script located in the bin sub-directory of the SharePlex product directory. This script truncates and updates the SharePlex tables, which are owned by the SharePlex user. If you are running multiple instances of sp_cop with multiple variable-data directories, there is a SharePlex Oracle user for each one. Make sure you run this script as the SharePlex user that owns the tables you want to restore. The script prompts you for the SharePlex user name and password. SQL> @/productdir/bin/cleanup.sql
SharePlex 9.2.8 Administrator Guide 265

10. On the target system, disable or modify the following according to your replication strategy:
l triggers l foreign key constraints l cascading delete constraints (or configure SharePlex to ignore them) l check constraints l scheduled jobs that perform DML 11. On the source system, reactivate the configuration file if you had to make any changes to it.
sp_ctrl> activate config filename
12. On the target system, start the Post process. The two instances are now in synchronization, and SharePlex will continue replicating. sp_ctrl> start post

How to restore Oracle archive logs

If you decide to restore the archive logs to enable SharePlex to resume capture and replication, use the following procedure to determine the required archive logs.
1. Determine the sequence number that Capture needs to resume processing from. Capture stops when it encounters a log wrap and prints a message to the Event Log (event_log) containing the redo log sequence number it needs. You also can find out this number by querying the SHAREPLEX_ACTID table and looking at the SEQNO column, as shown in the following example: SQL> select * from splex.shareplex_actid;

ACTID SEQNO OFFSET

AB_ FLAG

QUE_SEQ_NO_1 QUE_SEQ_ COMMAND NO_2

----- ------ -------- ------ ------------ --------- ------------

-----

114

9757200 0

672101000

2. Query the Oracle V$LOG_HISTORY table to find out when that sequence number was archived, then copy the logs from that point forward to the source system. SQL> select * from V$LOG_HISTORY;

RECID STAMP ----- ------
111 402941650 112 402941737 113 402941930 114 402942019 115 402942106

THREAD# ------1 1 1 1 1

SEQUENCE# --------
111 112 113 114 115

FIRST_CHANGE# FIRST_TIM NEXT_CHANGE# ------------- --------------

2729501 2729548 2729633 2781791 2836155

14-JUL-00 2729548 14-JUL-00 2729633 14-JUL-00 2781791 14-JUL-00 2836155 14-JUL-00 2890539

SharePlex 9.2.8 Administrator Guide 266

How to release semaphores after process failure
If database corruption or other system problem forced you to shut down SharePlex, verify that SharePlex released the semaphores and shared memory that it was using.
To verify and release semaphores 1. Look for any SharePlex processes that did not shut down, and kill them. $ ps -ef | grep sp_ $ kill -9 PID 2. Change directories to the rim sub-directory of the SharePlex variable-data directory, then issue the od -x command for the shmaddr.loc and the shstinfo.ipc files. # od -x shmaddr.loc 0000000 0000 00e1 ed40 0000 4400 9328 0080 0000 0000020 0002 0021 0000024 # od -x shstinfo.ipc 0000000 0000 00e0 ee90 0000 4100 9328 0010 0000 0000020 0002 0020 0000024 3. Make a note of the following values: l The first 32-bit word of each of the files above reveals the hexadecimal equivalent of the ID of the shared memory segment. Convert this value to decimal. For example, in the shmaddr.loc file shown in step 2, the first word is 0000 00e1, which equates to a decimal value of 225. In the shstinfo.ipc file, the first word is 0000 00e0, which equates to a decimal value of 224. l The third word of the shmaddr.loc and the shstinfo.ipc files reveals the hexadecimal equivalent of the KEY of the shared memory segment and the semaphore. (Each set has the same key value.) Do not convert this value to decimal. For example, in the shmaddr.loc file, the third word is 4400 9328. In the shstinfo.ipc file, the third word is 4100 9328. l The fifth word of each file is the SEMAPHORE ID. Convert this value to decimal. The semaphore IDs in the examples are hex 0002 0021 and 0020 0020, which in decimal are 131105 and 131104, respectively.
SharePlex 9.2.8 Administrator Guide 267

4. Issue the ipcs -smaa command to view all of the shared memory segments and semaphores. (Shared memory segments are listed first and are denoted with an "m." Semaphores are denoted with an "s.") The display looks similar to the following, but will be more extensive.
5. Verify that the shared memory IDs from the shmaddr.loc and shstinfo.ipc are in the list and that the keys match.
6. For each shared memory segment, verify that the value in the NATTCH column is 0. This ensures that the SharePlex processes that you killed released their memory segments.
7. For the semaphores, verify that the semaphore IDs and keys match the file values. 8. As root, issue the ipcrm -m command for the ID values (224 and 225 in the examples) to remove the
memory segments. # ipcrm -m 224 # ipcrm -m 225
9. As root, issue the ipcrm -s command for the key values (131104 and 131105 in the examples) to remove the semaphores. # ipcrm -s 131104 # ipcrm -s 131105
How to resolve disk space shortage
This topic helps you resolve disk space issues that can occur when something interferes with replication. See Solve replication problems on page 234 for possible causes.
SharePlex 9.2.8 Administrator Guide 268

How to conserve disk space on the target
SharePlex captures and processes data much faster than it posts it with SQL statements on the target system, so the target is where most disk problems can occur, assuming the network is operational and data is being sent from the source. If you think the post queue may exceed its disk space, there may be enough free space on the source system to store the data temporarily until the Post queue clears out.
1. Stop the Import process. 2. Let the data accumulate on the source system until Post processes enough messages to clear
the post queue. 3. Start Import. 4. Continue to stop and start Import until the amount of data accumulating in the post queue levels out. When you implement this method, monitor the replication services and disk usage on the source system. On Unix and Linux systems, you can use the sp_ps script to monitor processes and the sp_qstatmon monitoring script to monitor the queues. On Windows systems, you can use the Sp_Nt_Mon utility to monitor those components. For more information, see Monitor SharePlex on page 208.
How to restore disk space
If a queue disk runs out of disk space, you may see messages similar to this in the Event Log: 11/22/07 14:14 System call error: No space left on device bu_wt.write [sp_mport (que)/1937472] 11/22/07 14:14 System call error: No space left on device bu_rls.bu_wt [sp_mport (que)/1937472] 11/22/07 14:14 Error: que_BUFWRTERR: Error writing buffer to file que_writecommit (irvspxuz+P+o.a920a64z-o.a102a64z) [sp_mport(rim)/1937472] 11/22/07 14:14 Error: sp_mport: rim_writecommit failed 30 - exiting [sp_mport/ 1937472] 11/22/07 14:14 Process exited sp_mport (from irvspxuz.domain.com queue irvspxuz) [pid = 1937472] - exit(1)
If a queue disk is almost out of free space, you might be able to add disk space without the need to resynchronize the data.
To restore disk space 1. Stop SharePlex on the affected system. 2. Add more disk space. 3. Start SharePlex.
SharePlex 9.2.8 Administrator Guide 269

4. View the Event Log and look for the messages "queue recovery started" and "queue recovery complete." l If both messages are there, SharePlex resumes processing where it stopped and the recovery succeeded. If your applications generate high volumes of transactions, there may be numerous backlogged messages in the queues. Depending on the nature of the transactions, how well the target database and the Post process are tuned, and your tolerance for latency, it might be more practical to resynchronize the data instead of waiting for replication to regain parity with transactional activity. l If one or more queues is corrupted, the Event Log records a message like this: Bad header magic... or peekahead failure. Or, you will see the message queue recovery started, but you will not see the queue recovery complete message that signifies successful queue recovery. In this case, you must restore replication an initial state.
To restore replication to an initial state
1. Run db_cleansp to restore the variable-data directory and SharePlex tables. It must be run on all systems in the affected replicationconfiguration. See the utilities documentation in the SharePlex Reference Guide.
2. Synchronize the data using your method of choice, then reactivate the configuration. For more information, see Start replication on your production systems on page 186.
3. You can prevent this problem from occurring again by using the SharePlex monitoring utilities to start unattended monitoring of key replication events, including queue volume alerts. For more information, see Monitor SharePlex on page 208.
How to find the ORACLE_SID and ORACLE_HOME
When setting up SharePlex to work with an Oracle database, you provide the ORACLE_SID and then SharePlex gets the ORACLE_HOME from the Windows Registry or theoratab file on Unix/Linux. Both values are stored in the SharePlex environment. SharePlex uses the Oracle libraries that are in the location specified with ORACLE_HOME.
To determine the ORACLE_SID and ORACLE_HOME being used by SharePlex Issue the orainfo command in sp_ctrl.
sp_ctrl (mysysl11:2101)> orainfo Oracle instance #1:
Oracle SID ora12 Oracle HOME /oracle/products/12 Oracle Version 12 Oracle instance #2: Oracle SID ora12 Oracle HOME /oracle/products/12 Oracle Version 12
SharePlex 9.2.8 Administrator Guide 270

To determine the default ORACLE_SID and ORACLE_HOME on Windows View the Oracle entry in the Registry at \HKEY_LOCAL_MACHINE\SOFTWARE\ORACLE. To determine the default ORACLE_SID and ORACLE_HOME on UNIX and Linux On most Unix and Linux systems the oratab file is under /etc/oratab. On Oracle Solaris systems, it is under /var/opt/oracle, but sometimes there is an oratab file in the /etc directory as well. The entry in the file looks like the following example:
qa12:/qa/oracle/ora12/app/oracle/product/12.0 In the example, qa12 is the ORACLE_SID and /qa/oracle/ora12/app/oracle/product/12.0 is the ORACLE_HOME.
SharePlex 9.2.8 Administrator Guide 271

19
Repair out-of-sync data
This chapter contains an overview of the SharePlex Compare and Repair feature. SharePlex provides this feature as built-in support for Oracle tables to help you maintain data that is synchronized between the source and target systems.
Contents
Overview of Compare and Repair Before you use Compare and Repair How to use the repair and compare commands
Overview of Compare and Repair
In addition to regularly monitoring the health and performance of replication, it is good practice to compare the source and target data on a regular schedule to ensure that all of the data is still synchronized. Post detects outof-sync conditions for the rows that it is processing, but there can be hidden out-of-sync conditions. Examples of these are DML applied on the target or an incomplete backup restore. These conditions can go undetected until Post applies an operation that affects the out-of-sync row. The SharePlex Compare and Repair feature enables you to detect hidden out-of-sync conditions and then repair them.
NOTE: To understand how hidden out-of-sync conditions can occur, see Understand the concept of synchronization on page 30. SharePlex provides the following commands for comparing and repairing out-of-sync data: l compare: Compares an individual source table to its target table or compares a wildcarded set of tables in the same schema. l compare using: Takes input from a file to compare some or all of the tables in the active replication configuration. l repair: Repairs an individual target table or a wildcarded set of tables in the same schema. l repair using: Takes input from a file to repair some or all of the tables in the active replication configuration.
SharePlex 9.2.8 Administrator Guide 272

Supported sources
Oracle
Supported targets
Oracle
Overview of the server and client processes
The compare and repair commands are always issued on the source system. The command spawns a server process on the source system and then sends a message through the SharePlex queues to spawn a client process on the target system.
The server and client processes then begin communication with each other. Depending on the syntax options included in the command, the processes may be multithreaded on the target. The two processes compare the source and target tables and then write the results to a log file.
How locks are managed
During a comparison, SharePlex obtains a brief exclusive lock on the source table to get read consistency for its row selection. On the target system, SharePlex obtains an exclusive lock on the target table and retains the lock for the duration of the comparison of that table. This prevents the table from being modified while SharePlex is processing it. After locking the tables, the rows are read and sorted in identical fashion on both source and target. Next, a batch of rows is read and a check sum is performed. If the check sums match, another batch of rows is processed the same way. If any check sums do not match, the processes determine which rows are out-ofsynchronization and then they create the SQL statements to repair them. If a repair command is issued, SharePlex repairs the rows.
Before you use Compare and Repair
Before running the compare or repair commands, review these guidelines. l All of the SharePlex processes (Capture, Read, Export, Import, Post) must be running when you run a comparison or repair command. l The tables that you want to compare or repair must be part of an active configuration file.
SharePlex 9.2.8 Administrator Guide 273

l Uncommitted transactions on a source table prevent the compare and repair processes from obtaining the brief locks they need to obtain read consistency. Make certain that all transactions are committed before you run a comparison or repair.
l If a table is large, it will probably need to be sorted in the TEMP tablespace. Before running the compare or repair commands, the TEMP tablespace may need to be made larger. The size depends on the setting of the SP_DEQ_THREADS parameter or the threads option within the command syntax, both of which controls the number of processing threads used by SharePlex on the target. Each thread processes a table. At the default of two threads, the size of the tablespace should be larger than the sum of the sizes of the two largest tables. If you set the number of threads higher, then increase the size of the tablespace to accommodate a proportionate number of the largest tables. However,
l The UNDO tablespace may also need to be increased. Based on transaction volume and the length of time it takes to compare the largest table, increase the size of the UNDO tablespace and increase the undo_retention database parameter to avoid an ORA-1555 Snapshot too old error. Tables with LOBs take much longer to compare or repair than tables without them.
The following are commonly modified compare and repair parameters. Do not increase the values unless necessary. For details about these parameters, see their documentation in the SharePlex Reference Guide.

Parameter SP_DEQ_MALLOC
SP_DEQ_PARRALLISM

Description
This parameter controls the fetch batch size. The batch size controls the number of rows that SharePlex selects at once for comparison. Larger batch sizes increase processing speed but require more memory. The value is divided equally by the number of compare threads to be used, and then the batch size is recalculated based on all column sizes added together.
This parameter manages the select statement Degree of Parallelism hint. The parallelism option of the command overrides this setting.

SP_DEQ_PART_TABLE_UPDATE
SP_DEQ_READ_BUFFER_SIZE
SP_DEQ_ROW_LOCK_ THRESHOLD SP_DEQ_SKIP_LOB

This parameter controls how the repair commands work on Oracle partitioned tables, depending on whether row movement is possible.
This parameter controls the size of the buffer that holds fetched LONG and LOB data and can be adjusted based on available system memory.
This parameter sets a threshold that controls whether SharePlex uses row-level or table-level locking when a where option is used.
This parameter determines whether or not LOBs are included in the compare/repair processing.
l When the parameter is set to the default of 0, the compare processes include LOBs in their processing.
l When the parameter is set to 1, only non-LOB columns are compared and repaired. If LOBs are not modified once inserted, you can speed up processing by setting this parameter to 1.
Set this parameter on the source system.

SharePlex 9.2.8 Administrator Guide 274

Parameter SP_DEQ_TIMEOUT

Description
This parameter sets a queue backlog threshold. High backlogs delay the establishment of a connection between the source and target compare/repair processes. If the backlog meets or exceeds this value, any compare or repair command that is issued on the source will exit and return an error. If this happens, consider running the compare or repair when the system is less busy.

How to use the repair and compare commands
The recommended procedure for maintaining synchronized data through the comparison and repair commands is to run the compare or compare using command first, then view the results with the repair status command. This command shows any rows that are out-of-sync and the possible cause. Unless the cause of the out-of-sync condition is corrected, replication will go out of synchronization again, even if you repair the rows this time. After the problem is fixed, issue the repair or repair using command. You can run the repair or repair using command without doing a preliminary comparison. The command performs a comparison first, to identify the out-of-sync rows, and then it repairs those rows. However, the underlying cause of the out-of-sync condition must be corrected to prevent future out-of-sync conditions. See for causes and solutions for out-of-sync conditions. To view the status or results of a comparison, use the compare status command in sp_ctrl. To view the status or results of a repair, use the repair status command in sp_ctrl.
When to run a repair
The best time to repair a target table depends on its size, the cause of the problem, the extent of out-of-sync rows, and how long you are willing to tolerate users being locked out. Before you initiate a repair, consider the following:
l Although the users of the tables are not usually affected by the brief locks that are applied when tables are compared, they are locked out of the target table for the duration of the repair process. For a small table, this might not be disruptive, but for a large table needing extensive repairs, the wait can be significant.
l Locks on a target table can reduce posting performance if Post must wait for the repair to finish before it can apply changes to that table and move on to other tables. This increases the latency of the target data and causes operations to accumulate in the post queue. If the objects that Post needs to change are different from those being repaired, the two processes run simultaneously.
l If you must repair a table immediately, but cannot tolerate locks or replication latency, you can use the where option to limit the repair to certain rows. An alternative is to use the key option, but this option may cause the repair to miss some out-of-sync rows.
l If the repair can wait, correct the cause of the problem immediately and then do the repair during nonpeak hours.
l Replication latency can slow down the compare and repair processing. The message sent from the source to spawn the command processes on the target is sent through the queues along with regular

SharePlex 9.2.8 Administrator Guide 275

replicated data. Delays caused by a data backlog will delay the spawn message and cause the process on the source to lose its read consistency, which results in errors. If possible, perform comparisons and repairs during off-peak hours.
How to run the compare and repair commands
To get additional information and syntax for the compare and repair commands, see the command documentation in the SharePlex Reference Guide.
SharePlex 9.2.8 Administrator Guide 276

20
Tune the Capture process
This chapter contains instructions for improving the performance of the Capture process to prevent Capture from losing pace with the volume of redo that an Oracle source database generates.
Contents
Disable LOB Mapping Tune Capture on Exadata Tune checkpointing Add a second thread
Disable LOB Mapping
If you have PK/UK logging enabled on the source database (recommended to support more SharePlex features and faster processing), check the setting of the SP_OCT_ENABLE_LOBMAP parameter. This parameter controls whether or not SharePlex uses a LOB map when replicating tables that contain out-of-row LOB columns. The LOB map is used by the Capture process to map LOBIDs and rows when PK/UK logging is not enabled. LOB mapping is enabled by default. The SHAREPLEX_LOBMAP table stores these mappings.Transactions with numerous LOB operations can slow down Capture because it needs to maintain and refer to the mappings. If PK/UK logging is enabled on the database, you can disable LOB mapping by setting this parameter to 0. To disable LOB mapping during active replication
1. Run sp_ctrl on the source system. 2. Set SP_OCT_ENABLE_LOBMAP to 0.
sp_ctrl> set param SP_OCT_ENABLE_LOBMAP 0 3. Stop Capture.
sp_ctrl> stop capture 4. Truncate the SHAREPLEX_LOBMAP table.
SharePlex 9.2.8 Administrator Guide 277

5. Restart Capture. sp_ctrl> start capture
Tune Capture on Exadata
The Capture process can be configured to use multiple capture threads for faster performance on an Exadata system. Capture reads directly from the logs on the Exadata ASM disks. The SP_OCT_ASM_MULTI_OCI parameter controls the number of threads that Capture uses to read the redo logs. The value for this parameter must be set to at least 2 but no more than the number of disks in the redo log disk group. A large number of threads is not required, and performance actually diminishes with too many threads. The more threads, the more memory Capture requires. Start with a small number of threads and monitor performance, then add threads if needed until you obtain an ideal balance between performance gain and memory usage.
To configure SharePlex for multi-threaded capture on Exadata 1. Run sp_ctrl. 2. Set the SP_OCT_ASM_MULTI_OCI parameter to the number of threads that you want Capture to use. sp_ctrl> set param SP_OCT_ASM_MULTI_OCI 3 3. Restart Capture.
NOTE: Capture automatically adjusts its buffer size to the value of the AU_SIZE parameter that is set for the disk group where the logs reside. This is the recommended buffer size for best performance and should not be changed. The SP_OCT_ASM_MULTI_OCI_BLOCK_SIZE parameter can override the default behavior if necessary.
Tune checkpointing
Capture checkpoints it state to disk on a regular basis to support recovery. This information includes the log and location within that log of the most recently processes data. In a database environment where there are frequent log switches, a switch can occur before SharePlex writes its checkpoint. You can use the SP_OCT_ CHECKPOINT_LOG parameter to ensure that Capture issues a checkpoint before a log switch. The checkpoint is triggered when Capture lags a specified number of logs behind Oracle. For example, with the default of 2, Capture does a checkpoint when it falls 2 or more logs behind Oracle. The range of permissible values for this parameter is from 2 (the default) to a value equal to the number of logs you are using. A value of 0 disables this feature.
Add a second thread
You can set the SP_OCT_OLOG_RDS_MINER parameter to 1 to add a second thread to Capture. This thread can be used to address performance issues when Capture is lagging behind Oracle on a very busy system.
SharePlex 9.2.8 Administrator Guide 278

Due to the processing load incurred by using this thread, it is disabled by default. To enable it, set this parameter to 1. NOTE: Enabling the SP_OCT_OLOG_RDS_MINER parameter is deprecated and no longer supported starting with Oracle 19c.
SharePlex 9.2.8 Administrator Guide 279

21
Tune the Post process
This chapter contains instructions for improving the performance of the Post process. Because replicated data is applied through standard SQL mechanisms, the Post process provides the most potential for performance tuning.
Contents
Use Oracle INDEX hints Tune SQL Caching Adjust open cursors Skip large maintenance transactions Make small transactions faster Split a large transaction into a smaller one Tune queue performance Tune hash-based horizontally partitioned replication
Use Oracle INDEX hints
Valid for: Oracle targets
When SharePlex performs UPDATEs and DELETEs on a target table, Oracle sometimes does not pick the most efficient index for SharePlex. Without the right index, the Post process slows down when multiple UPDATEs and DELETEs are performed. SharePlex enables you to make use of Oracle's INDEX hints to enforce the use of the correct index on target objects. To use INDEX hints, use the hints.SID file, where SID is the ORACLE_SID of the target instance. When Post applies a SQL statement, it reads the hints file. If the file contains entries, Post reads the data into memory and then checks each UPDATE and DELETE statement that it processes. If any of those operations involve tables listed in the hints file, Post sends the hints to Oracle. Use hints only for tables that need them. For example, if Post is doing full-table scans on tables where there are defined indexes, use hints only for those tables. The use of hints causes Post to read the hints.SID file for each operation on tables listed in the file. This can slow down processing if numerous tables are listed. The default maximum number of hints (table/index pairs) is 100. You can adjust this value with the SP_OPO_ HINTS_LIMIT parameter. See the SharePlex Reference Guide for more information.
SharePlex 9.2.8 Administrator Guide 280

Make certain all indexes are valid. Although SharePlex will use an invalid index as a hint, Oracle ignores invalid hints and returns no errors. SharePlex writes the following information to the event_Log if it detects abnormal conditions relating to the specified hints. 15050 hint file not found 17000 error opening hint file 15051 missing column in the hint file (either table or index name) 15052 syntax error for tablename 15053 syntax error for indexname 15054 source table's object_id not found in object cache 15055 more than 20 valid entries were entered into the hints file

To use the hints file
There is a blank hints.SID file in the SharePlex variable-data directory on each system. Use the hints.SID file that resides on the target system. If a hints file does not exist, create one in this directory and make certain to use the hints.SID naming format.

1. Stop Post if it is running.
2. Open the file.
3. You can add comment lines anywhere in the file. Start a comment line with a pound symbol (#).
4. On a non-commented line, use the following template to specify a source table and the index that you want to use for that table. Put at least one space between the table name and the index name. Place each specification on a separate line.

src_owner.table

tgt_owner.index

Example scott.emp

scott.emp_index

Tune SQL Caching
SharePlex caches frequently-used SQL statements for reuse so that they do not have to be parsed and bound every time they recur. This is an adjustable feature of SharePlex that is named SQL Cache. You can tune this feature to maximize its advantages based on the amount of repetitive statements your application generates. SQL Cache improves the performance of Post only if the same SQL statements are issued over and over again, with no variation except the data values. If that is not true of your environment, then SQL Cache adds unnecessary overhead to the Post process, and you should disable it.
Supported targets
All

SharePlex 9.2.8 Administrator Guide 281

Enable or disable SQL Cache
Control SQL Cache as follows:

Oracle

Parameter Description

SP_OPO_ SQL_ CACHE_ DISABLE

Enables or disables SQL Cache. Enabled by default with a setting of 0. To disable SQL Cache set the parameter to 1. To disable SQL Cache only for batch operations set the parameter to 3, which reduces the amount of memory that Post uses.

SP_OPO_ MAX_CDA

Determines the number of active statements to cache per Post session. Post opens 50 cursors per session by default. You can increase or decrease this setting if needed. For more information, see Adjust open cursors on page 283.

Open Target

Parameter

Description

SP_OPX_SQL_ CACHE_DISABLE

Enables or disables SQL Cache. Enabled by default with a setting of 0. To disable SQL Cache set the parameter to 1.

Use the target command:
target r.database [queue queuename] set resources max_ active_ statements=number_ of_active_statements

Determines the number of active statements to cache per Post session. For Open Target databases, Post gets the number of allowed active statements from the ODBC driver. If that value is lower than the setting for max_active_statements, Post stops and returns an error. You can either disable the SQL Cache feature or reduce the value of max_active_statements.

Tune SQL Cache for best performance
Follow these steps to make certain that the number of active statements is optimal for the operations that are replicated.
1. Determine the hit ratio for cached statements by running sp_ctrl and issuing the show post detail command.
2. Look for the SQL cache hit count field. It shows the ratio of the total number of messages that are executed without parsing and binding divided by the total number of INSERT, UPDATE and DELETE operations. For example, a hit ratio of 36% indicates that Post is using cached statements 36 percent of the time.
3. View the hit ratio after several days of typical replication activity to gauge the ideal setting for the number of active statements. If the hit ratio is under 50 percent, increase the parameter value in a small increment of about 5 statements.

SharePlex 9.2.8 Administrator Guide 282

4. Monitor the hit ratio over the next few days. If the hit ratio increases, it means your applications are using all of the cursors allowed for active statements. Continue to increase the parameter value in small increments until the hit ratio remains constant.
Adjust open cursors
Valid for: Oracle targets
The value of the Oracle parameter OPEN_CURSORS needs to be set high enough to support the level of performance expected of the Post process. This parameter defines the maximum number of cursors that a process (such as Post) can open. Internally, Post establishes its maximum total number of open cursors from the value of OPEN_CURSORS, minus the 10 required for routine calls. You view this value in the event_log. For the following example, OPEN_ CURSORS is set to 512.
Notice: sp_opst_mt (for o.oracle-o.oracle queue oracle) Post will not open more than 502 cursors (OPEN_CURSORS 10). Post maintains a record of the number of cursors it has open. If Post detects that it will exceed the maximum number of cursors, it closes the least-recently used cursor in the least-recently used session. To avoid running out of cursors, the Post process queries the OPEN_CURSORS value when it starts. If the value is not high enough, Post writes the following warning to the event_log:
Warning: (sp_opst_mt for o.oracle-o.oracle queue oracle)Oracle parameter 'OPEN_ CURSORS' is < number. Check 'OPEN_CURSORS' setting. The OPEN_CURSORS value can be modified or added if absent. To view the OPEN_CURSORS value, query the database using the following SQL statement:
select value from v$parameter where name = 'open_cursors';
To estimate a value for OPEN_CURSORS that is high enough for the Post process
1. Estimate the peak number of concurrent transactions (sessions) that will be expected for the target instance. Post opens a session on the target system for each one on the source system. You can get a good estimate of the number of transactions by issuing the show post detail command in sp_ctrl when production is at its maximum level. The Number of Open Transactions field in the display shows the number of concurrent transactions.
2. Use the following formulas to determine the correct setting for OPEN_CURSORS to support SharePlex (and other applications that may be accessing the target data). SQL Cache enabled (default): By default, Post needs to reserve 10 cursors for routine calls that are closed once they finish, plus a minimum of 7 cursors per transaction (the base minimum of 2 plus an additional 5). The formula is: 10 + (peak number of concurrent transactions x 7) = minimum open cursors needed SQL Cache disabled: The Post process needs to reserve 10 cursors for routine calls that are closed once they finish, plus a minimum of 2 cursors per transaction. The formula is: 10 + (peak number of concurrent transactions x 2) = minimum open cursors needed
SharePlex 9.2.8 Administrator Guide 283

Skip large maintenance transactions
Valid for: Oracle targets
Large transactions that are applied by application patches or other internal Oracle operations can be omitted from replication if they are not relevant to the data needed by user applications. These operations can translate into thousands or millions of individual UPDATE or DELETE statements for SharePlex, all to be applied by Post. Such transactions can adversely affect Post performance and increase the latency between the source and target data that user applications need to perform their work. There may be reasons to prevent other DML operations from being posted to a target database. There are two ways you can handle such transactions:
l Assuming there are no referential relationships between those operations and the user data, configure those operations to process through a dedicated named post queue. For more information, see Configure named post queues on page 89.
l Configure Post to skip the operations, and then apply the SQL statement directly through Oracle. See the following instructions.
To skip maintenance DML 1. On the source system, run the create_ignore.sql script from the util sub-directory in the SharePlex product directory. This script creates the SHAREPLEX_IGNORE_TRANS public procedure in the database. When executed at the start of the transaction, the procedure directs the Capture process to ignore the DML operations that occur until the transaction is committed or rolled back. Thus, the affected operations are not replicated. For more information about the script, its limitations, and how to run it, see create_ignore.sql in the SharePlex Reference Guide. 2. Edit your patch script to call SHAREPLEX_IGNORE_TRANS before UPDATE or DELETE operations. This allows SharePlex to ignore the transaction and not send it to the target. The script will also have to be run on the target to bring the database back into sync.
NOTE: Only DML operations are affected by the SHAREPLEX_IGNORE_TRANS procedure. It does not cause SharePlex to skip DDL operations, including TRUNCATE. DDL operations are implicitly committed by Oracle, so they render the procedure invalid.
Make small transactions faster
Valid for: Oracle and Open Target (as indicated per feature) You can improve the speed of Post when it is processing mostly small transactions, such as those most commonly found in OLTP. There are two features you can use, depending on the supported database:
l Increase the level of concurrency l Reduce the number of commits Together these features are called Post Enhanced Performance, or PEP.
SharePlex 9.2.8 Administrator Guide 284

Increase the level of concurrency
Valid for Oracle, SQL Server, and PostgreSQL targets The Transaction Concurrency feature configures a Post process to apply transactions in parallel to increase overall throughput. To use this feature, supplemental logging for primary and unique keys must be enabled on the source.
To enable Transaction Concurrency l For an Oracle target database, set the SP_OPO_DEPENDENCY_CHECK parameter to 1. l For SQL Server and PostgreSQL, set the SP_OPX_THREADS parameter to 2 or greater.
NOTE: The use of Transaction Concurrency may reduce or eliminate the need to run multiple Post processes, but you can still benefit from that configuration because it eliminates a single point of failure. If a Post process fails, the other Post processes can continue, resulting in less recovery time, after the problem is resolved. The Transaction Concurrency feature can be used in a multi-Post configuration, so long as the rules for using multiple Post processes are followed (such as including all tables with referential integrity in the same process stream). For more information, see Configure named post queues on page 89.
Reduce the number of commits
Valid for Oracle and all Open Target databases The Commit Reduction feature of Post combines batches of small transactions into larger ones. One large transaction runs faster than multiple smaller ones by having fewer commits and acknowledgments to process. Post skips the commits of small transactions until their combined size reaches the threshold specified by one of the following parameters:
l SP_OPO_COMMIT_REDUCE_MSGS (Oracle targets) l SP_OPX_COMMIT_REDUCE_MSGS (Open Target) The default batch transaction size is 100 messages. This value is an approximation. If the size of the last transaction in the batch exceeds the specified threshold, SharePlex waits for the remaining messages and the commit before applying the batch transaction to the target. Commit reduction is enabled by default. To disable commit reduction, set this parameter to a value of 1.
Split a large transaction into a smaller one
Valid for: Currently supported for Teradata You can configure Post to split a large transaction into a series of smaller ones. This option can work around resource limits that affect large transactions, such as the number of row locks permitted per transaction.
To split a large transaction into smaller ones Use the target command to set the commit_frequency parameter.
target r.database [queue queuename] set resources commit_frequency=number_of_operations
SharePlex 9.2.8 Administrator Guide 285

This parameter specifies a maximum number of operations after which Post issues a commit. It can be any integer greater than 1.
Example: target r.mydb queue q1 set resources commit_frequency=10000
Tune queue performance
You can tune the performance of Post by tuning the performance of the post queue.
Reduce queue contention
You can use the SharePlex queue contention reduction feature to ensure that shared memory is not swapped to disk when the post queue is becoming full. This feature is enabled by the SP_IMP_QUEUE_PAUSE parameter. This parameter pauses the writing of data to the post queue when that queue contains the specified number of messages. Post stores queue messages in shared memory until it issues a checkpoint, after which it releases the data from memory. If the post queue runs out of shared memory, the read and write functions will start incurring file IO to free up the memory buffers. By pausing the queue writing, this parameter helps Post maintain its performance by avoiding the need for disk storage and the resultant slowdown in IO. Use the SP_IMP_QUEUE_RESUME parameter to set the number of messages at which Import resumes writing to the post queue. This parameter works in conjunction with SP_IMP_QUEUE_PAUSE. If the number of messages in the post queue is lower or equal to the value set with this parameter, Import resumes writing to the post queue. To use this feature, both SP_IMP_QUEUE_PAUSE and SP_IMP_QUEUE_RESUME must be greater than zero, and SP_IMP_QUEUE_PAUSE must be greater than SP_IMP_QUEUE_RESUME.
Tune subqueue indexing
You can improve Post queue performance by enabling subqueue indexing to access the subqueue structures that represent a transaction session. A message "Subqueue index enabled queuename" is written to the Event Log for every Post queue for which this parameter is enabled. To enable this feature, set the SP_QUE_USE_SUBQUE_INDEX parameter to 1. This parameter does not support VARRAYs. If you are replicating VARRAYs and this parameter is enabled, the parameter is ignored.
Tune hash-based horizontally partitioned replication
Hash-based horizontally partitioned replication uses a hash algorithm that is based on the rowid by default. You may be able to improve the processing of tables that use hash-based horizontally partitioned replication by switching the hash algorithm to one that is based on the block where the row resides.
SharePlex 9.2.8 Administrator Guide 286

Because changing the algorithm has the same effect as a routing change (the potential to switch partitions), you must reactivate the configuration file. The activation locks the tables that are affected by this change so that the hash change is applied when there are no open transactions. The locking eliminates the potential for out-ofsync conditions by preventing data that is processed under the new hashing algorithm from being posted before in-flight data that was processed under the old algorithm. To switch to a block-based hash
1. Set the SP_OCF_HASH_BY_BLOCK parameter to 1. 2. Reactivate the configuration file.
SharePlex 9.2.8 Administrator Guide 287

22
Recover replication after Oracle failover
This chapter contains instructions for recovering Oracle replication along with the database and applications during a failover in a high-availability environment. To support these procedures, SharePlex must be properly configured to support high availability. See Configure replication to maintain high availability.
Contents
Recover replication if the primary system fails Recover replication if the secondary Oracle instance fails Move replication during planned failover and failback Resume replication after failure and recovery
Recover replication if the primary system fails
In an unplanned failure of the primary (source) machine, replicated data remaining in the SharePlex queues on that system will be unrecoverable as a result of buffering and the likelihood that the queues are corrupted. In a high-availability environment, you can move replication to the secondary (target) machine along with the database users to maintain data availability. When the primary system is restored, you can move users and replication back to that system with minimal downtime using a hot backup of the secondary instance. For this procedure, you will activate a configuration file and then use the reconcile command to synchronize the results of the backup with ongoing replicated user changes after the copied instance is recovered.
Supported databases
Oracle database on Unix or Linux
SharePlex 9.2.8 Administrator Guide 288

Requirements
l SharePlex must be configured correctly to support high availability. For more information, see Configure replication to maintain high availability on page 148.
l You must have a backup of the SharePlex replication environment. l You must know how to run SharePlex. For more information, see Run SharePlex on page 36. l You must understand the activate config, reconcile, and delete queue commands. See the SharePlex
Reference Guide.
Procedure 1: Move replication to the secondary system
This procedure moves replication to the secondary system after an unplanned failure of the primary system. 1. Verify that Export is stopped on the secondary system. sp_ctrl> stop export 2. Use the qstatus command to view the post queue, and keep issuing this command until the number of backlogged messages is 0. NOTE: Do not wait for the actual numberof messages to be 0. If the primary system failed before the commit for some transactions were received, messages for these partial transactions will remain in the queue until cleared later in this procedure. sp_ctrl> qstatus 3. Run the script that grants INSERT, UPDATE and DELETE access to all users on the secondary system. 4. Run the script that enables triggers and constraints on the secondary system when users begin using this system. 5. Implement the failover procedure for relocating users to the secondary system, including starting the applications. 6. Move users to the secondary system and let them resume working, but do not start Export. Their transactions will now be accumulating in the export queue awaiting restoration of the source database. NOTE: If started, Export will repeatedly try to connect to the primary system, wasting system resources. 7. Go to Procedure 2: Move replication to the restored primary system.
Procedure 2: Move replication to the restored primary system
This procedure moves users back to the primary machine after it is recovered from an unplanned failure. Follow each segment in the order presented.
Restore the replication environment on the primary system
1. On the primary system, recover the SharePlex directories, system files and Oracle files from the backups and archives.
SharePlex 9.2.8 Administrator Guide 289

2. On the primary system, start sp_cop using the -s option to prevent the SharePlex processes (Capture, Read, Export, Import, Post) from starting. $ /productdir/bin/sp_cop -s &
3. On the primary system, start sp_ctrl. 4. On the primary system, deactivate the configuration file. When you copied the archived SharePlex
variable-data directory to the primary system, you copied the configuration that was active before the system failed. This causes the Capture process to set the transaction number to "1" when replication from the primary system resumes.
sp_ctrl> deactivate config filename
Purge the queues
1. Run sp_ctrl on the primary and secondary systems. 2. On the primary system, delete the capture queue.
sp_ctrl> delete capture queue for datasrc [on host] Example: sp_ctrl> delete capture queue for o.oraA 3. On the primary system, delete the export queue. sp_ctrl> delete export queue quename [on host] Example: sp_ctrl> delete export queue sysA 4. On the secondary system, delete the post queue. sp_ctrl> delete post queue quename for datasrc-datadst [cleartrans] [on host] Example: sp_ctrl> delete post queue sysA for o.oraA-o.oraB NOTE: l You are issuing the delete queue command on the primary system because you restored the old capture and export queues when you restored the archived SharePlex directories. l You are issuing the delete queue command on the secondary system because data remaining in the post queue cannot be posted. The primary system failed before Post received a COMMIT for remaining transactions. SharePlex will rebuild the queues when you reactivate the configuration and the two systems are reconciled.
Start replication from secondary to primary system
1. On the primary system, verify that all SharePlex processes are stopped. sp_ctrl> status
2. On the primary system, stop Post. sp_ctrl> stop post
3. On the secondary system, start Export. This establishes communication between the primary and secondary systems.: sp_ctrl> start export
SharePlex 9.2.8 Administrator Guide 290

Synchronize the source and target data
1. On the secondary system, run an Oracle hot backup. 2. When the hot backup is finished, switch log files on the secondary system and make a note of the
highest archive-log sequence number. 3. On the primary system, recover the primary database from the backup by using the UNTIL CANCEL
option in the RECOVER clause, and cancel the recovery after the log with the number you recorded has been fully applied. 4. Open the database with the RESETLOGS option. NOTE: This resets the sequences on the primary system to the top of the cache upon startup. 5. On the primary system, issue the reconcile command using the sequence number of the log that you noted previously. If you are using named post queues, issue the command for each one. If you do not know the queue names, issue the qstatus command first.
sp_ctrl> qstatus sp_ctrl> reconcile queue queuename for datasource-datadest seq sequence_number Example: reconcile queue SysB for o.oraA-o.oraA seq 1234 6. On the primary system, disable triggers on the tables, or run the sp_add_trigger.sql utility script so that the triggers ignore the SharePlex user. 7. On the primary system, disable check constraints and scheduled jobs that perform DML. 8. On the primary system, log onto SQL*Plus as the SharePlex Oracle user and run the cleanup.sql utility from the bin sub-directory of the SharePlex product directory. This truncates the SharePlex tables and updates the SHAREPLEX_ACTID table. 9. On the secondary system, truncate the SHAREPLEX_TRANS table. This table contains transaction information that the Post process was using before the primary system failed, and therefore that information is obsolete. Truncating the table restores transaction consistency between the two systems.
Activate replication on the primary system
1. On the primary system, activate the configuration file. sp_ctrl> activate config filename
2. On the primary system, start Post. sp_ctrl> start post
Restore the object cache
1. On the primary system, view the status of the SharePlex processes. sp_ctrl> status
NOTE: This command shows that Post stopped due to an error, which occurred because the object cache is missing. 2. On the primary system, issue the show log command with the filter option, and filter on the keyword "objcache." sp_ctrl> show log filter=objcache
SharePlex 9.2.8 Administrator Guide 291

3. Look for a Post error message that refers to a file with a name similar to the following example: 0x0a0100c5+PP+sys4+sp_opst_mt+o.quest-o.ov-objcache_sp_opst_mt.18
You are looking for a string that contains the string "objcache_sp_opst_mt," followed by a number. This is the object-cache file that the Post process needs. If you are using named post queues, there will be more than one error message, each referring to a different object-cache file but ending with the same number, such as the number .18 in the example.
4. Make a note of the full pathname of the Post object-cache file(s) named in the error message. The path will be the state directory in the SharePlex variable-data directory, for example: splex_vardir/state/0x0a0100c5+PP+sys4+sp_opst_mt+o.quest-o.ov-objcache_sp_ opst_mt.18
5. On the secondary system, shut down SharePlex. sp_ctrl> shutdown
6. On the secondary system, change directories to the state sub-directory of the SharePlex variable-data directory and locate the Capture object-cache file. This file will have a name similar to the one in the following example. o.quest-objcache_sp_ocap.18 IMPORTANT! If there are more than one of these files, use the one with the most recent number at the end of it -- this number should match the number at the end of the Post object-cache file, such as .18 in the example.
7. Copy the Capture object-cache file to the primary system and rename it to the full pathname of the Post object-cache file that you noted previously.
8. On the primary system, start Post. sp_ctrl> start post
Switch users back to the primary system
1. On the secondary system, stop user access to the database. 2. On the primary system, view the post queue and continue to issue the command until the number of
messages is 0. sp_ctrl> qstatus
3. On the secondary system, shut down the Oracle instance. svrmgr1> shutdown
4. On the secondary system, start the Oracle instance. svrmgr1> startup
NOTE: This resets the sequence on the secondary system to the top of the cache to synchronize with the primary system.
5. On the primary system, enable database objects that were disabled. 6. On the secondary system, start SharePlex. 7. On the secondary system, stop Post.
sp_ctrl> stop post
8. On the secondary system, disable triggers on the tables, or run the sp_add_trigger.sql utility script so that the triggers ignore the SharePlex user.
SharePlex 9.2.8 Administrator Guide 292

9. On the secondary system, disable check constraints and scheduled jobs that perform DML. 10. On the secondary system, start Post.
sp_ctrl> start post 11. On the primary system, view the number of messages in the post queue, and keep checking until the
number of messages is 0. sp_ctrl> qstatus
12. When the number of messages is 0, switch users back to the primary system. 13. On the secondary system, stop Export to prevent accidental changes made on that system from being
replicated to the primary system. sp_ctrl> stop export
NOTE: The secondary system is now in a failover-ready state again, with: l no users l an active configuration l disabled or modified triggers, constraints, and scheduled jobs l a stopped Export process
Recover replication if the secondary Oracle instance fails
In an unplanned failure of the secondary (target) Oracle instance, the replication environment from that system to the primary system is corrupted. This procedure enables you to restore the replication configuration without affecting the database users on the primary system, and without the need to reactivate the configuration file on the primary system. Only the secondary configuration is affected. This procedure cleans out the SharePlex queues and restores the target instance by means of a hot backup from the source system. You will use the reconcile command to synchronize the results of the backup with ongoing replicated user changes after the copied instance is recovered.
Supported databases
Oracle database on Unix or Linux
Requirements
l SharePlex must be configured correctly to support high availability. For more information, see Configure replication to maintain high availability on page 148.
l You must know how to run SharePlex. For more information, see Run SharePlex on page 36. l You must understand the activate config, reconcile, and delete queue commands. See the SharePlex
Reference Guide.
SharePlex 9.2.8 Administrator Guide 293

l This procedure assumes that the secondary system itself is operational so that you can interact with SharePlex on the system.
Procedure
This procedure is divided into logical segments. Follow them in the order presented.
Purge the queues
1. On the secondary system, stop Post. sp_ctrl> stop post
2. On the secondary system, deactivate the configuration file. sp_ctrl> deactivate config filename
NOTE: The deactivation causes the error "Error in sp_cnc." in the Event Log. You may ignore it and continue with the procedure. 3. On the primary system, run sp_ctrl. 4. On the primary system, delete the post queue.
sp_ctrl> delete post queue quename for datasrc-datadst [cleartrans] [on host] Example: sp_ctrl> delete queue sysB:P for o.oraA-o.oraB NOTE: You are deleting the queues on the primary system because there could be messages remaining from uncommitted transactions sent from the secondary instance before it failed. 5. On the primary system, truncate the SHAREPLEX_TRANS internal table in the SharePlex schema. This table contains transaction information that the Post process on that system was using before the secondary instance failed, and therefore the information is obsolete. Truncating the tables restores transaction consistency. 6. On the secondary system, run sp_ctrl. 7. On the secondary system, delete the capture queue. sp_ctrl> delete capture queue for datasrc [on host] Example: sp_ctrl> delete queue o.oraB:C 8. On the secondary system, delete the export queue. sp_ctrl> delete export queue quename [on host] Example: sp_ctrl> delete queue sysB:X NOTE: You are deleting the queues on the secondary system because the capture and export queues on that system still retain a record of the transactions that have already been processed.
SharePlex 9.2.8 Administrator Guide 294

Synchronize the data
1. On the primary system, begin a hot backup of the primary Oracle instance. 2. On the primary system, switch log files.
On-premises database: svrmgr1> alter system switch logfile;
Amazon RDS database: Use Amazon RDS procedure rdsadmin.rdsadmin_util.switch_logfile.
3. Make a note of the highest sequence number of the archive logs. 4. On the secondary system, recover the secondary database from the hot backup using the UNTIL
CANCEL option in the RECOVER clause. Cancel the recovery after Oracle has fully applied the log from the previous step. 5. On the secondary system, open the secondary database with the RESETLOGS option. This resets the sequence on the secondary system to the top of the cache upon startup 6. On the secondary system, run SQL*Plus as the SharePlex database user. 7. In SQL*Plus, run the cleanup.sql script from the bin subdirectory of the SharePlex product directory. 8. On the secondary system, issue the reconcile command using the sequence number of the log that you noted previously. If you are using named post queues, issue the command for each one. If you do not know the queue names, issue the qstatus command first.
sp_ctrl> qstatus sp_ctrl> reconcile queue queuename for datasource-datadest seq sequence_number Example: reconcile queue SysA for o.oraA-o.oraA seq 1234 9. On the secondary system, disable triggers on the tables, or run the sp_add_trigger.sql utility script so that the triggers ignore the SharePlex user. 10. On the secondary system, disable check constraints and scheduled jobs that perform DML. 11. On the secondary system, after the sp_ctrl prompt returns, stop Export. This ensures that nothing accidentally gets replicated to the primary system when you activate the configuration on the secondary system. sp_ctrl> stop export
Start replication on the secondary system
1. On the secondary system, activate the configuration file. sp_ctrl> activate config filename
2. On the secondary system, start Post. sp_ctrl> start post
3. Use the status command to determine if any other SharePlex processes have a status of Stopped due to Error and start them. sp_ctrl> status sp_ctrl> start process
NOTE: The secondary system is now prepared for future failover.
SharePlex 9.2.8 Administrator Guide 295

Move replication during planned failover and failback
In a planned failover of database activity to a secondary Oracle instance, you can quickly move SharePlex to the secondary system. While users continue their transactions on that system, SharePlex captures their changes and stores them until the primary system is back online and activity is moved back to that system.
Supported databases
Oracle database on Unix or Linux
Requirements
l SharePlex must be configured correctly to support high availability. For more information, see Configure replication to maintain high availability on page 148.
l There must be enough disk space where the queues reside on the secondary system to contain the data that accumulates while users are working on that system.
l You must know how to run SharePlex. For more information, see Run SharePlex on page 36. l You must understand the SharePlex flush command. For more information, see the SharePlex
Reference Guide.
Procedure
This procedure is divided into logical segments. Follow them in the order presented. Do not shut down the primary instance until prompted in the procedure.
Switch users to the secondary system
1. On the primary system, stop user access to the primary instance. 2. On the primary system, flush the data in the queues to the secondary system. This command stops Post
on the secondary system and places a marker in the data stream to establish a synchronization point between the primary and secondary data.
sp_ctrl> flush datasource where: datasource is the datasource specification of the primary Oracle instance, for example o.OraA. 3. On the secondary system, verify that Post stopped. (Continue to issue this command until it shows that Post stopped.)
sp_ctrl> status 4. On the primary system, verify that there are no messages in the capture and export queues. The Number
of Messages and the Backlog (messages) fields must be 0. sp_ctrl> qstatus
SharePlex 9.2.8 Administrator Guide 296

5. On the secondary system, verify that there are no messages in the post queue. The Number of Messages and the Backlog (messages) fields must be 0. sp_ctrl> qstatus
6. On the primary system, shut down SharePlex. 7. On the primary system, shut down the Oracle instance with the abort option. Do not use the
immediate option. svrmgr1> shutdown abort
NOTE: This resets the sequence on the primary system to the top of the cache when the database starts. 8. On the secondary system, verify that Export is stopped. This prevents user changes from being
replicated to the primary system until it is back online and SharePlex is ready to receive them. Stop Export if needed.
sp_ctrl> status sp_ctrl> stop export 9. On the secondary system, run the script that grants INSERT, UPDATE and DELETE access to all users. 10. On the secondary system, run the script that enables triggers and constraints on the secondary instance. 11. Run your failover procedure for relocating users to the secondary system, including starting the applications and starting jobs that were running on the primary system. 12. Move the users to the secondary system to resume working, but do not start the Export process.
Switch users back to the primary system
1. On the primary system, open the Oracle instance. The sequence on this system should now be at the top of the cache.
2. On the primary system, disable triggers on the tables, or run the sp_add_trigger.sql utility script so that the triggers ignore the SharePlex user.
3. On the primary system, disable check constraints and scheduled jobs that perform DML. 4. On the primary system, start SharePlex. 5. On the secondary system, start Export so that SharePlex sends the accumulated replicated data to the
primary system. sp_ctrl> start export
NOTE: SharePlex passes any sequence updates from the secondary system back to the primary system when Export starts. 6. On the primary system, stop Export.
sp_ctrl> stop export 7. On the primary system, allow Post to process the message backlog that was sent from the
primary system. 8. On the secondary system, stop user access to the Oracle instance. 9. On the secondary system, flush the data in the queues to the primary system.
sp_ctrl> flush datasource where: datasource is the datasource specification of the secondary Oracle instance, for example o.OraB.
SharePlex 9.2.8 Administrator Guide 297

10. On the primary system, verify that Post stopped. (Continue to issue this command until it shows that Post stopped.) sp_ctrl> status
11. On the secondary system, verify that there are no messages in the capture and export queues. The Number of Messages and the Backlog (messages) fields must be 0. sp_ctrl> qstatus
12. On the primary system, verify that there are no messages in the post queue. The Number of Messages and the Backlog (messages) fields must be 0. sp_ctrl> qstatus
13. On the secondary system, shut down SharePlex. sp_ctrl> shutdown
14. On the secondary system, shut down the Oracle instance with the abort option. Do not use the immediate option. svrmgr1> shutdown abort NOTE: This resets the sequence on the secondary system to the top of the cache when the database starts.
15. On the secondary system, start the Oracle instance. svrmgr1> startup
NOTE: The sequence on the secondary system is now at the top of the cache. When the next value is selected on the primary system, a new cache is acquired and is replicated to the secondary system. Now, the primary system is at the start of a cache, and the secondary system is at the top of a cache. 16. On the primary system, run the script that grants INSERT, UPDATE and DELETE access to all users. 17. On the primary system, run the script that enables triggers and constraints on the primary system when users begin using this system. 18. Run your failover procedure for moving users back to the primary system, including starting the applications and starting jobs that were running on the secondary system. 19. Switch the users to the primary system to resume working, but do not start the Export process. This prevents replicated data from being sent to the secondary system until SharePlex is ready to receive it there.
Resume replication to maintain the secondary instance
1. On the secondary system, disable triggers on the tables, or run the sp_add_trigger.sql utility script so that the triggers ignore the SharePlex user.
2. On the secondary system, disable check constraints and scheduled jobs that perform DML. 3. On the secondary system, start SharePlex. 4. On the secondary system, stop Export. This prevents any accidental DML on that system from being
replicated to the primary system. sp_ctrl> stop export
5. On the primary system, start export. sp_ctrl> start export
SharePlex 9.2.8 Administrator Guide 298

6. On the secondary system, start Post. sp_ctrl> start post
Replication from the primary instance to the secondary instance is now active to keep the two databases synchronized and ready for future failover when needed.
Resume replication after failure and recovery
The procedure is typically used in these situations: l When the source system fails and replication must be switched to a standby database system l When replication must be positioned back in time to re-read old archive logs.
Requirements to support SharePlex replication recovery
To resume replication when the source, target or both have failed, there must be the following in place at the onset of replication:
l A disaster recovery (DR) solution that provides a physically identical copy of the production source instance and another physical copy of the production target instance. Methods such as Oracle Data Guard or disk mirroring, tape backups and other methods support this requirement.
l The SP_OPO_UPDATE_SCN parameter must be set to a value of 1. This parameter directs SharePlex to keep a record of the SCNs of the transactions that it processes. When you set this parameter to 1, it also disables the Post Enhanced Performance feature.
Overview of initial setup
The following diagram depicts a DR configuration at the onset of replication. There is a source production instance and a mirrored source DR instance that is kept current by Oracle Data Guard. Similarly, there is a production target instance and a mirrored DR target instance that is kept current by Oracle Data Guard.
l The solid (blue) lines represent the Oracle Data Guard DR deployment. l The dotted (bright green) line between the production source instance and the production target
instance represent SharePlex replication under normal operating circumstances. l The dashed lines (red, orange or aqua) show possible replication recovery paths if the source, target or
both fail.
SharePlex 9.2.8 Administrator Guide 299

Figure 2: DR configuration at the onset of replication
Example failure/recovery scenario
This example illustrates one of the potential failure/recovery scenarios, in this case where the production target instance fails. The recovery path is shown as the diagonal, orange dotted line in the DR configuration at the onset of replication diagram.
Normal replication
The following diagram illustrates the configuration and the names that are used in this example. l The production source is named Source PROD and the DR source is named Source DR. l The production target is named Target PROD and the DR target is named Target DR. l SharePlex (SPO in the diagram) replicates from Source PROD to Target PROD.
SharePlex 9.2.8 Administrator Guide 300

Figure 3: Normal replication and mirroring configuration
Production target fails
The Target PROD target fails, as represented by the red X across it in the following diagram. SharePlex can no longer replicate to Target PROD, as represented by the red X over the original replication data stream. Because Target PROD is offline, Oracle Data Guard can no longer keep Target DR up to date. However, SharePlex can. SharePlex (SPO in the diagram) is able to resume replication from Source PROD to Target DR, thus resuming data availability. Figure 4: Failure and recovery by SharePlex (SPO)
Resume replication after failover
In this procedure you will do the following to direct SharePlex to recover replication: l Direct SharePlex to capture the correct Oracle SCN of the last committed transaction processed by each post queue. l Direct SharePlex, through the reconcile command, to discard all transactions that were committed to the target before the failure, so that SharePlex resumes replication at the correct point in the data stream.
NOTE: This procedure requires the following:
SharePlex 9.2.8 Administrator Guide 301

l The source instance is recovered to a later point in time than the target instance; otherwise, this method will not work.
l The SP_OPO_UPDATE_SCN parameter is set to 1.
To resume replication NOTE: In these instructions, the source and target systems are whichever source and target are operational after the failover.
1. Shut down SharePlex on the source system, if it is still running. sp_ctrl> shutdown
2. On the target, start sp_cop if it is not running already. $ /productdir/bin/sp_cop &
3. On the target, use the qstatus command to make certain that all of the message in the queues are posted to the target database. The command output should show 0 backlog in the post queue. sp_ctrl> qstatus
4. From the command line of the target, run the show_scn utility from the bin subdirectory of the SharePlex product directory. For ORACLE_SID use the ORACLE_SID of the target database. $ /productdir/bin/show_scn ORACLE_SID
5. Keep the output of the show_scn utility open. The output displays the complete reconcile command that you will use for each of your post queues to reposition Post to the correct transaction for recovery. It also shows the SCN to which you will activate the configuration later in these steps.
6. Shut down sp_cop on the source and target. sp_ctrl> shutdown
7. Run ora_cleansp on the source and target to clean out the queues. $ /productdir/bin/ora_cleansp
8. Start sp_cop on the source and target. $ /productdir/bin/sp_cop &
9. On the target, stop Post. sp_ctrl> stop post
10. On the source, issue the activate config command with the scn option to activate the configuration. For scn_value, use the value that is shown in the output of the show_scn utility on the line that states On source activate to scn=nnnnnnn. sp_ctrl> activate config configname scn=scn_value Example: sp_ctrl> activate config myconfig scn=510012416
SharePlex 9.2.8 Administrator Guide 302

11. On the target, copy the first reconcile command from the show_scn output and then execute it in sp_ ctrl. Then do the same for the second reconcile command, and work your way down the list. Example: sp_ctrl> reconcile queue spx11 for o.ora112-o.ora112 scn 235690 sp_ctrl> reconcile queue pq1 for o.ora112-o.ora112 scn 132436 sp_ctrl> reconcile queue pq2 for o.ora112-o.ora112 scn 246843 sp_ctrl> reconcile queue pq3 for o.ora112-o.ora112 scn 123457 The reconcile command may seem stalled until new data comes in. However, the command is working.
12. On the target, start Post. sp_ctrl> start post
SharePlex 9.2.8 Administrator Guide 303

23
Make changes to an active replication environment
This chapter contains instructions for making datbase changes, or performing system and software maintenance, on systems where SharePlex replication is active.
Contents
Change an active configuration file Add or change objects in an active configuration Change Partitioned Replication Add Oracle sequences to an active replication configuration Remove source objects from replication Make DDL changes in an active replication configuration Make Oracle changes that affect replication Change the SharePlex database account Change the name or IP address of a replication host Set the SharePlex port number
Change an active configuration file
Many procedures that change an element of the replication environment will also involve changing the SharePlex configuration file. The recommended method to modify an active configuration file is to make a copy of the file first using a new file name. For example, append the date to the file name. By copying the file first, you preserve the original file in case it is needed again or you encounter a problem with the new file. By copying the file first, you also can control when to activate the new configuration file. In most cases, activation of an edited configuration is less time-consuming than the original activation if you do not deactivate the original configuration. The activation of the new configuration automatically deactivates the original configuration, and SharePlex only needs to analyze the new, changed, and removed objects. If you deactivate the original configuration before you activate a new one, SharePlex re-analyzes all of the objects. To change a configuration file, see Add or change objects in an active configuration on page 305.
SharePlex 9.2.8 Administrator Guide 304

Add or change objects in an active configuration
This section provides instructions for adding a supported object to replication, or changing the specifications of an object, while replication is active.
Supported databases
Oracle source All targets
Oracle procedure
NOTE: To add sequences to replication, see Add Oracle sequences to an active replication configuration on page 307. If you are using wildcards and an object that you are adding satisfies the wildcard specification, it is not necessary to add the object to the configuration file if the source is Oracle. Any new objects that match the wildcard criteria are automatically added into replication. Only add objects that must be explicitly stated by name. IMPORTANT! Do not deactivate the original configuration.
1. If adding new tables, add them to the source and target (populated in both places, if applicable) to establish a synchronized initial state. Do not allow transactional access to the source table yet.
2. In sp_ctrl, issue the copy config command to make a copy of the active configuration file. sp_ctrl> copy config filename to newname
Where: filename is the name of the active file and newname is the name of the new one. 3. Issue the edit config command to open the new configuration file in the default text editor.
sp_ctrl> edit config newname 4. Add the entries for the new tables or change existing entries.
NOTE: To change partitioned replication, see Change Partitioned Replication on page 305. 5. Save the configuration file. 6. Activate the new configuration. This deactivates the original configuration. Only the new or changed
tables are activated, so the activation should not be as long as the initial activation. sp_ctrl> activate config newname
7. Allow access to the newly added tables.
Change Partitioned Replication
You can change a horizontally partitioned or vertically partitioned replication configuration while replication is active. Both of these procedures require the reactivation of the active replication configuration, but SharePlex only locks tables that are associated with those changes.
SharePlex 9.2.8 Administrator Guide 305

NOTE: To learn more about changing a configuration file, review Change an active configuration file. For more information about partitioned replication, see Configure partitioned replication.

Supported databases
Oracle All targets

To change horizontally partitioned replication
1. Run sp_ctrl. 2. Issue one of the following commands to change the partition or partition scheme. For syntax and other
information, see the alphabetical command listings in the SharePlex Reference Guide.

Command add partition drop partition drop partition scheme modify partition

Auth. level 2 2 2
2

Description Creates partition schemes and row partitions. Removes a row partition from a partition scheme. Removes a partition scheme.
Modifies a row partition of a partition scheme.

3. If you dropped a partition scheme:
a. Copy (but do not deactivate) the active configuration file to a new file. sp_ctrl> copy config filename to newname
b. Edit the copy to remove or change the routing map where the partition scheme was specified. sp_ctrl> edit config filename
4. Activate the new configuration file. sp_ctrl> activate config filename

To change vertically partitioned replication
1. Make a copy of (but do not deactivate) the active configuration file. sp_ctrl> copy config filename to newname
2. Edit the copy to change the appropriate column partition. sp_ctrl> edit config filename
3. Activate the new configuration file. sp_ctrl> activate config filename

SharePlex 9.2.8 Administrator Guide 306

Add Oracle sequences to an active replication configuration
The procedure to use to add a sequence to an active configuration file depends on whether or not you can stop user access to the objects that use the sequence. If the sequences are used to populate a column in a table, you may not be able to stop user access. Review the following procedures to determine which one will work best in your environment: Enable auto-add of sequences Add sequences if auto-add is not enabled
Supported databases
Oracle source and target
Enable auto-add of sequences
You can configure SharePlex to add sequences to replication automatically if their names satisfy a wildcard in the configuration file. For more information, see Control Oracle DDL replication on page 152.
Add sequences if auto-add is not enabled
The following procedures apply if the auto-add feature for sequences is not enabled.
Add a sequence if the sequence does not populate a column
1. Stop user activity to the objects on the source system. 2. In sp_ctrl, issue the copy config command to make a copy of the active configuration file.
sp_ctrl> copy config filename to newname Where: filename is the name of the active file and newname is the name of the new one. 3. Issue the edit config command to open the new configuration file in the default text editor.
sp_ctrl> edit config newname 4. Add the new sequences to the configuration file. 5. Save and close the file. 6. Create the target sequence on the target system. To ensure uniqueness on the target system, the start
value of the target sequence must be larger than the start value of the source sequence. Use the following formula to determine the target START_WITH value: source_INCREMENT_BY_value = START_WITH_value
SharePlex 9.2.8 Administrator Guide 307

7. Activate the new configuration. This deactivates the original configuration. sp_ctrl> activate config newname
8. Allow users to access the objects.
Add a sequence if the sequence populates a column
1. In sp_ctrl, issue the copy config command to make a copy of the active configuration file. sp_ctrl> copy config filename to newname
Where: filename is the name of the active file and newname is the name of the new one. 2. Issue the edit config command to open the new configuration file in the default text editor.
sp_ctrl> edit config newname 3. Add the new sequences to the configuration file. 4. Save and close the file. 5. Activate the new configuration. This deactivates the original configuration.
sp_ctrl> activate config newname 6. On the source system, flush the data from source system to the target system. This command stops Post
and places a marker in the data stream that establishes a synchronization point between source and target data. sp_ctrl> flush datasource Where: datasource is o.ORACLE_SID of the source instance -- for example o.oraA. 7. After Post stops, issue the following Oracle command on the target system to find the last known value of the sequence. Make a record of this value.
select max(column_name) = last known value 8. Determine the value of the following equation.
source_INCREMENT_BY_value x source_CACHE_value For example, if the source sequence is incremented by 2 and the cache size is 10, the value would be 20. 9. Starting with the value that you recorded for the select max (column_name) command, determine the next highest multiple of (source_INCREMENT_BY_value x source_CACHE_value). Example: INCREMENT_BY = 2 CACHE = 10 select max(column_name) = 24 Next highest multiple of (2 x 10) after 24 = 40. 10. To the value obtained in the previous step, add another multiple of (source_INCREMENT_BY_value x source_CACHE_value). The result determines the START WITH value of the target sequence. For example, in the previous equation the START WITH value would be: 40 + (2 x 10) = 60. 11. Create the target sequence with the START WITH value that you calculated.
SharePlex 9.2.8 Administrator Guide 308

12. On the target, start Post. sp_ctrl> start post
SharePlex will continue replicating the data, while keeping the target sequence at least one multiple of (source_INCREMENT_BY_value x source_CACHE_value) ahead of the source sequence.
IMPORTANT! Sequences continue to be incremented even when a transaction is rolled back. If numerous rollbacks are issued for a source table that uses a replicated sequence, it causes the sequence values to increase without actually being used in columns in the table. As a result, when Post applies the next valid operation, the sequence value on the target system could be less than the value in the replicated row. When there are numerous rollbacks, view the target table regularly to ensure that the current value of the target sequence remains greater than the maximum value in the table. If the current value of the target sequence is less than the maximum value in the table, repeat the preceding procedure to re-establish the sequence relationships.
Remove source objects from replication
To remove source objects from replication, the configuration must be reactivated. NOTE: (Oracle only) Objects being removed are locked when the configuration is activated, but only those objects are locked, so the activation is less time-consuming that the original activation. You can prevent posting to a table without removing it from the configuration file. You may need to do this if, for example, there is data corruption and you do not want DML or DDL operations to be applied to that table. To prevent posting to a table, use the SP_OPO_DISABLE_OBJECT_NUM parameter. For more information about this parameter, see the SharePlex Reference Guide.
Supported databases
All databases supported by SharePlex
Procedure
1. In sp_ctrl, issue the copy config command to make a copy of the active configuration file. sp_ctrl> copy config filename to newname
Where: filename is the name of the active file and newname is the name of the new one. 2. Issue the edit config command to open the new configuration file in the default text editor.
sp_ctrl> edit config newname 3. In the new configuration file, delete the entries for the objects that you want to remove from replication. If
the object that you want to remove from replication satisfies a wildcard, use the not notation to exclude the object. For more information, see Use Wildcards to specify multiple objects on page 72. 4. Save and close the file.
SharePlex 9.2.8 Administrator Guide 309

5. Activate the new configuration. This deactivates the original configuration. sp_ctrl> activate config newname
6. Allow users to access the removed objects.
Make DDL changes in an active replication configuration
This procedure applies to DDL changes that are not of a type that is supported by SharePlex. DDL that is supported by SharePlex can be applied to the source database without reactivating the configuration file or stopping user access to objects, assuming the applicable SharePlex parameters are set correctly. Supported DDL is replicated by SharePlex to the target, where it is applied by Post. For a list of supported DDL operations and required parameters, see the SharePlex Release Notes. Use this procedure to apply DDL that is not of a type that is supported by SharePlex. The DDL must be applied outside SharePlex on both the source and target systems. This procedure requires stopping access to the objects in the configuration file and a reactivation of the configuration file to update the internal tables. However, only the changed objects are analyzed, so the activation time will be shorter than the time required for a full activation.
Supported databases
Oracle
Requirements
l You must know how to run SharePlex. For more information, see Run SharePlex on page 36. l You must understand how to activate a configuration file with the activate config command. l You must understand the SharePlex flush command. For more information, see the SharePlex
Reference Guide.
Procedure
1. On the source system, stop access to the source objects (on all systems if using peer-to-peer replication).
2. On the source system (trusted source in peer-to-peer), flush the data from the source system to the target systems. This command stops the Post process and places a marker in the data stream that establishes a synchronization point between the source and target data. sp_ctrl> flush datasource where: datasource is the database specification of the source instance, for example o.oraA.
SharePlex 9.2.8 Administrator Guide 310

3. On the target system (all secondary systems in peer-to-peer) verify that the number of messages in the post queue is 0 on each system and that Post stopped. sp_ctrl> lstatus
4. On the source system, make the DDL changes. 5. On the source system, reactivate the configuration file.
sp_ctrl> activate config filename 6. On the source system, allow user activity to resume. Their replicated changes will accumulate in
the post queue. 7. On the target system, make the corresponding DDL changes. 8. [High availability and peer-to-peer replication only] On the secondary systems, reactivate the
configuration file. sp_ctrl> activate config filename
9. On the target systems, start Post. sp_ctrl> start post
SharePlex resumes replication from the last stop point and the data remains synchronized.
Make Oracle changes that affect replication
This topic helps you make common changes to the Oracle environment while replication is active.
Supported databases
Oracle on Linux and UNIX
Move the location of ORACLE_HOME
If you change the ORACLE_HOME, you need to relink SharePlex to the Oracle libraries. 1. Shut down SharePlex. sp_ctrl> shutdown 2. Move the ORACLE_HOME. 3. Edit the oratab file to point to the new ORACLE_HOME. 4. Edit the defaults.yaml file to point to the new ORACLE_HOME. This file is in the data subdirectory of the SharePlex product directory. 5. Start SharePlex.
SharePlex 9.2.8 Administrator Guide 311

Change the target ORACLE_SID
1. On the source system, run sp_ctrl. 2. On the source system, copy the active configuration file to a new name, but do not deactivate it.
sp_ctrl> copy config filename to newname 3. On the source system, open the new configuration file.
sp_ctrl> edit config filename 4. Change the ORACLE_SID to the new one in all of the routing maps that include this target database and
target system. 5. Save and close the configuration file, but do not activate it. 6. On the source system, stop user access to the objects involved in replication. 7. On the source system, flush the data in the queues to the target. This stops the Post process and
establishes a synchronization point between the source and target databases. sp_ctrl> flush datasource
where: datasource is the database indicator of the source instance, for example o.oraA. 8. On the source system, activate the new configuration file. This will deactivate the original
configuration file. sp_ctrl> activate config filename
NOTE: The activation will be brief because SharePlex does not need to analyze the tables. 9. On the source system, allow users to access the objects involved in replication. 10. On the target system, verify that Post stopped. If Post is not stopped, continue to issue the command until
it shows that Post stopped. sp_ctrl> status
11. On the target system, shut down the database and then rename the ORACLE_SID. 12. On the target system, start Post.
sp_ctrl> start post
Change the SharePlex database account
You can change the user name (schema or database), the password, or both in the SharePlex database account. The database account was established during the installation of SharePlex. These procedures guide you through the process in the correct order to maintain an active replication configuration.
Supported databases
All SharePlex-supported databases
SharePlex 9.2.8 Administrator Guide 312

Procedure
This procedure changes the user account name and/or password of the SharePlex user account in a database. This user account is the one that the SharePlex processes use to connect to the database when performing replication tasks. IMPORTANT! If using multiple variable-data directories, you must run this procedure for each one that you want to change.
1. (Unix and Linux only) If you are using multiple variable-data directories, export the environment variable that points to the variable-data directory for the SharePlex instance for which you are changing the account name or password. ksh shell: export SP_SYS_VARDIR=/full_path_of_variable-data_directory csh shell: setenv SP_SYS_VARDIR=/full_path_of_variable-data_directory
2. Run sp_ctrl on the system where you are changing the password. 3. Stop the SharePlex processes on the system where you are changing the account information. For
example, if you are changing the SharePlex account in the source database, stop Capture and Read. If you are changing the account in the target database, stop Post.
sp_ctrl> stop service
4. Verify that all SharePlex replication processes for this instance of SharePlex are stopped. sp_ctrl> status
5. Log in to the database as a DBA user and change the SharePlex account name and/or password to the new ones. Important! Do not delete the SharePlex objects!
6. If you changed the account name, copy all of the SharePlex database objects from the old account to the new one. NOTE: Keep the old account and SharePlex objects as backup until you are certain replication resumes properly.
7. In sp_crtl, issue the following command to change the account name and/or password in the SharePlex internal records. To change the user account: sp_ctrl> connection {o.SID | r.database} set user=username To change the password: sp_ctrl> connection {o.SID | r.database} set password=password where:
l SID is the ORACLE_SID of the database, if the database is Oracle. l database is the name (not the DSN) of the database, if the database is non-Oracle. l username is the new account name.
l password is the new password.
8. Start the SharePlex processes. sp_ctrl> start service
SharePlex 9.2.8 Administrator Guide 313

Change the name or IP address of a replication host
Description
Use the provision utility to change a host name or IP address in the SharePlex configuration. The SharePlex processes rely on the host names or IP addresses of the source and target machines to route data properly. The provision utility enables you to change host names or IP addresses within an active SharePlex instance, without reactivating a new configuration. NOTE: The provision utility does not change anything in the database. It only affects SharePlex internal objects.
Supported databases
All databases supported by SharePlex on all supported platforms
Guidelines for using provision
l provision is only valid for SharePlex version 6.1.1 and above. l (SharePlex 7.0.2 only) If provision is being used because the host name of a machine changed, contact
Quest Support to request a new license. Otherwise SharePlex will not start after the change. l (SharePlex 7.0.2 only) Add the following parameter to the param-defaults file before you run provision:
SP_CFP_DEBUG I live int() 0x00000000 l If running SharePlex on an AIX machine, set EXTSHM before running provision.
export EXTSHM=ON l Run provision on all of the machines in the SharePlex configuration. Each machine can reference the
IP addresses of all the other machines.
Run provision
1. Stop sp_cop. If sp_cop is running, provision will fail. NOTE: provision prevents sp_cop from being started while it is running.
SharePlex 9.2.8 Administrator Guide 314

2. Using the command line of the operating system, run provision from the SharePlex util sub-directory of productdir with the following syntax: provision -f old_name[:old_ipaddress] -t new_name[:new_ipaddress] [-p port] [-n]

Argument

Input

-f old_hostname[:old_ ipaddress]

l -f is required and represents "from." l old_hostname is the old (current) host name. l old_IPaddress is the old IP address. Use if the IP address cannot
be obtained from the network.

-t new_hostname[:new_ ipaddress]

l -t is required and represents "to." l new_hostname is the new host name. l new_IPaddress is the new IP address. Use if the IP address cannot
be obtained from the network.

-p port -n

For Windows systems, specifies the port of the SharePlex instance for which provision is being run.
Runs provision without actually making any changes. Generates a report on the changes that provision will make. IMPORTANT! The best practice is to run provision with -n first, to make certain you agree with the potential changes, then run it without -n to make the changes.

Example:
provision -f oldname -t newname -n
3. View the event log to view every change that was made. If the provision run fails or you do not agree with the changes that were made, you can undo them by running the undo_provision script. See Undo changes made by provision .

Undo changes made by provision
The provision utility creates an undo_provision script that can be used to restore the host names and IP addresses to their previous state. Run the undo_provision script from the util subdirectory of the SharePlex product directory. There are no input arguments to this script.

Known issues
The following may occur but do not affect the integrity of the replication environment:
l The provision utility does not change the active configuration file. This means that the configuration file no longer represents the current state of replication after provision is run. If you need to run the compare config command, or if you decide to reactivate the configuration, update the host name or IP address in the configuration file first.

SharePlex 9.2.8 Administrator Guide 315

l If an Export or Import error occurred when SharePlex connected to a machine before the name or address was changed, the error status persists and cannot be cleared.
l If the new or changed machine is a source machine, provision generates new routing information, but the Read process may still have the old routing in its cache. When you start sp_cop, Read might generate a warning that the stored IP address does not match the one for the machine. You can ignore this error.
l After provision is run for a source host, it might not update the "hostname" column in SHAREPLEX_ ACTID table with the new host name details. If that column is not correctly updated, you must update the SHAREPLEX_ACTID table manually to specify the new host name. This is only required if the name change affected a source machine.
Set the SharePlex port number
The SharePlex processes use TCP and UDP ports to communicate with each other on different systems. The default TCP and UDP port numbers for SharePlex are both set to 2100 at the time of installation. For some deployments of SharePlex you may need to change the port number or add additional ones. Before selecting a port number, review the following points:
l If your replication strategy requires multiple instances of sp_cop on a system, you must set a unique port number for each one. For more information, see Run multiple instances of SharePlex on page 43.
l When an non-default port is required, the same number must be used for both the TCP and UDP ports, and it must be used for the TCP and UDP ports of all other instances of sp_cop that are involved in the same replication configuration. If the ports are different, sp_cop on one system cannot connect to the sp_cop on another system to send or receive messages and data.
l On the Windows platform, SharePlex permits a maximum of 64 port numbers (SharePlex instances) on one system.
Supported databases
All databases supported by SharePlex on all supported platforms
Set the SharePlex port on Unix and Linux systems
To set the port number on Unix and Linux systems, a SharePlex Administrator must set both the TCP and UDP port parameters in the SharePlex environment. If there is an active configuration, you will be instructed to stop access to the source objects and shut down sp_cop.
1. (If using multiple variable-data directories] Export the SP_SYS_VARDIR environment variable to point to the correct variable-data directory for the port you are setting. ksh shell: export SP_SYS_VARDIR=/full_path_of_variable-data_directory csh shell:
SharePlex 9.2.8 Administrator Guide 316

setenv SP_SYS_VARDIR=/full_path_of_variable-data_directory 2. Export the following environment variables.
ksh shell: export SP_COP_TPORT=port export SP_COP_UPORT=port
csh shell: setenv SP_COP_TPORT port setenv SP_COP_UPORT port
where: port is the new port number 3. Change directories to the SharePlex product directory. 4. Start sp_cop and sp_ctrl. NOTE:If you are using multiple variable-data directories, start sp_cop with
the -uport option, where port is the port number that you have chosen for the variable-data directory that you exported. ./sp_cop [-uport] & 5. Run sp_ctrl. ./sp_ctrl 6. In sp_ctrl, set the following SharePlex parameters. sp_ctrl> set param SP_COP_TPORT port sp_ctrl> set param SP_COP_UPORT port 7. Do one of two things:
l If there is not an active configuration, use the shutdown command in sp_ctrl to stop sp_cop. The next time you start sp_cop, the new port number takes effect. NOTE: If you do not have an active configuration, you are finished setting the port number.
l If there is an active configuration, continue to the next step.
To finish setting the port in an active configuration 8. Stop access to the replicating objects on the source system, then issue the flush command in sp_ctrl on
the source system to clear the queues. sp_ctrl> flush o.database_identifier where: database_identifier is o. followed by the Oracle SID, TNS alias, or PDB name that SharePlex uses to connect to the source database (depends on whether the database is a regular Oracle instance, RAC, or PDB in a container database).
9. On the source system, issue the qstatus command to verify that all of the messages reached the target system. sp_ctrl> qstatus Continue to issue the command until the export queue is empty.
10. On the target system, issue the qstatus command to verify that all of the messages were posted to the database. Continue to issue the command until the post queue is empty.
SharePlex 9.2.8 Administrator Guide 317

11. On the target system, issue the status command to verify that Post stopped. sp_ctrl> status
12. Shut down SharePlex on the source and target systems. sp_ctrl> shutdown
13. Start sp_cop on the source and target systems. NOTE:If you are using multiple variable-data directories, start sp_cop with the -uport option, where port is the port number that you have chosen for the variabledata directory that you exported. ./sp_cop [-uport] &
14. Run sp_ctrl on the target system. 15. Start the Post process.
sp_ctrl> start post 16. Allow users to access the replicating objects. 17. Use the status command on the source and target systems to verify that all SharePlex processes
are running. sp_ctrl> status
Set the SharePlex port on Windows systems
On Windows systems, SharePlex port numbers are stored in the Windows Registry. If it becomes necessary to change a port number, this is the SharePlex entry location:
\HKEY_LOCAL_MACHINE\Software\Wow6432node\Quest Software\SharePlex.
1. Log onto Windows as a SharePlex Administrator using your system password and user name. Your user name must be assigned to the SharePlex Admin group.
2. Run the SpUtils utility. 3. Select the SharePlex Services tab. 4. Under Port, select the port number for the instance of SharePlex whose port you want to change. 5. Under SharePlex Service Status, click Stop. 6. Click Close. 7. From the Start menu, select Run. 8. In the Run dialog box, type regedit to open the Registry Editor. 9. In the Registry Editor, under \HKEY_LOCAL_MACHINE\Software\Wow6432node\Quest
Software\SharePlex, right click the current port number, then select Rename. 10. Replace the highlighted port number with the new number. 11. Click the new port number to highlight it. 12. In the Name column of the right-hand pane, right-click the SP_SYS_VARDIR entry that is associated
with the new port number, then select Modify. 13. In the Edit String dialog, change the port number (only the port number) in the path to the new port
number, then click OK. 14. Exit the Registry Editor.
SharePlex 9.2.8 Administrator Guide 318

15. In the Windows file explorer, navigate to the variable-data directory for which you changed the port number (the one whose name still shows the old port number).
16. Right click the name, and then use the Rename command to change only the port number in the name to the new port number.
17. Run the SpUtils utility, then select the SharePlex Services tab. 18. Under Port, select the new port number. 19. Under SharePlex Service Status, click Install to establish the SharePlex service under the new port. 20. After Current State shows that the service is installed, click Start to start the service.
SharePlex 9.2.8 Administrator Guide 319

Apply an Oracle application patch or upgrade

This chapter contains procedures to follow when you need to apply an application patch or upgrade and there is an active replication configuration. These procedures apply to Oracle databases.
Contents
Before you patch or upgrade an application Apply patch/upgrade to source then copy it to target Apply patch/upgrade to source and target Apply patch to source and replicate it to the target

Before you patch or upgrade an application
Review the following topics before you patch or upgrade an application on a system where SharePlex replication is active.

Which procedure to use?
There are different procedures for applying an application patch or upgrade to an Oracle database while replication is in process. Which one to choose depends on what changes the patch or upgrade makes.

Changes made by the patch/upgrade
If the patch/upgrade applies DDL that is not supported by SharePlex. For details on the DDL that SharePlex supports, see the SharePlexSharePlex Release Notes.

Steps to take
Manually apply the patch/upgrade to the source and target by following either of these procedures:

SharePlex 9.2.8 Administrator Guide 320

Changes made by the patch/upgrade
If the patch/upgrade does any of the following: l Performs DML changes. l Performs supported DDL on the source system. For details on the DDL that SharePlex supports, see the SharePlexSharePlex Release Notes. l Changes users and security on source system (other than SharePlex)

Steps to take
Apply patch/upgrade to source then copy it to target on page 322 Apply patch/upgrade to source and target on page 324
Manually apply the patch/upgrade to the source, then allow SharePlex to replicate the changes to the target. Follow this procedure: Apply patch to source and replicate it to the target on page 325 NOTE: Because this procedure assumes that SharePlex can replicate all of the changes that the patch or upgrade applies, the patch/upgrade is not applied to the target.

The effect of patches and upgrades on partitioned replication
A patch or upgrade can make changes that affect the column partitions of vertically partitioned replication in your configuration file. Take the following into account when you perform this procedure.

If the patch or upgrade does this to a table: Do this:

Adds columns that do not satisfy the column partition of the table
Adds columns that need to be in the column partition of the table
Drops columns that are part of the column partition of the table

(Optional) Drop the columns from the target table after the patch or upgrade is applied.
Add those columns to the source and target column partition lists in the configuration file.
Remove those columns from the source and target column partition lists in the configuration file.

Changes the name of a column that is in the Change the column name in the source and target column

column partition of a table

partition lists in the configuration file.

For more information, see Configure vertically partitioned replication on page 104.

Naming conventions used
In these procedures, the "source" system is one of the following: l The source system of a single-direction replication configuration, including cascading replication. l All source systems of a consolidated replication configuration. l The trusted source system in a peer-to-peer replication configuration.
In these procedures, the "target" system is one of the following:

SharePlex 9.2.8 Administrator Guide 321

l The target system of a single-direction replication configuration, including cascading and consolidated replication.
l The secondary systems in a peer-to-peer replication configuration. In this procedure, the SharePlex commands in the procedure apply to all sp_cop instances that apply to the replication strategy you are using (for example, all sp_cop processes on a target in consolidated replication).
Apply patch/upgrade to source then copy it to target
Supported databases
Oracle on all supported platforms
When to use this procedure
Use this procedure if the patch or upgrade makes DDL changes of a type not replicated by SharePlex. For a list of objects for which DDL is supported, see the SharePlex Release Notes.
Overview of the procedure
Use this procedure to run an Oracle hot backup to copy patches or upgrades from the source system to the target system, instead of applying the patch or upgrade directly on the target system. This is useful if the patch or upgrade makes extensive changes that are of the type(s) not supported by SharePlex replication, or if you are unsure of what it does. With this procedure, you can keep the configuration file active on the source system. You use the reconcile command to identify and eliminate the following:
l Duplicate DML and supported DDL from the patch or upgrade operations that were replicated but also applied by the backup.
l Production transactions that were replicated but also applied by the backup.
Apply the patch/upgrade
1. Stop user access to the instances involved in replication on the source and target systems, but do not shut down SharePlex.
2. On the source system, run sp_ctrl.
SharePlex 9.2.8 Administrator Guide 322

3. On the source system, flush the data to the target system. This command stops Post and places a marker in the data stream that establishes a synchronization point between source and target data. sp_ctrl> flush datasource where: datasource is the datasource of the source instance, for example o.oraA.
4. On the source system, apply the patch or upgrade. 5. On the source system, restore user access to the source instance. 6. [If the patch/upgrade adds objects that must be replicated] Edit the configuration file as follows (do not
deactivate it). The patch or upgrade may have affected column partitions or column conditions in partitioned replication. For more information, see Change an active configuration file on page 304.
l Copy the configuration file. sp_ctrl> copy config filename to newname
l Edit the copy. sp_ctrl> edit config newname
Save the file. 7. Do one of the following:
l If you added objects in the previous step, activate the new configuration file. sp_ctrl> activate config newname
l If you did not make any changes to the original configuration file, activate that one. sp_ctrl> activate config filename
8. On the source, run the Oracle hot backup. 9. On the source, switch log files and make a note of the highest archive-log sequence number.
On-premises database: svrmgr1> alter system switch logfile;
Amazon RDS database: Use Amazon RDS procedure rdsadmin.rdsadmin_util.switch_logfile.
10. On the target, recover the target database from the hot backup using the UNTIL CANCEL option in the RECOVER clause, and cancel the recovery after Oracle fully applies the log that you recorded in the previous step.
11. On the target system, open the database with the RESETLOGS option. 12. Run the Database Setup utility on the target instance, but do not create a new user. Choose the existing
SharePlex user and password (copied in the backup). For more information, see Database Setup Utilities in the SharePlex Reference Guide. 13. On the target system, issue the reconcile command using the sequence number of the log that you noted previously. If you are using named post queues, issue the command for each one. If you do not know the queue names, issue the qstatus command first.
sp_ctrl> qstatus sp_ctrl> reconcile queue queuename for datasource-datadest seq sequence_number Example: reconcile queue SysA for o.oraA-o.oraA seq 1234 NOTE: The reconcile process retains control of sp_ctrl until it is finished.
SharePlex 9.2.8 Administrator Guide 323

14. On the target system, if the patch or upgrade installed triggers on the tables in replication, disable them or run the sp_add_trigger.sql utility script so that the triggers ignore the SharePlex user.
15. On the target system, if the patch or upgrade added check constraints or scheduled jobs that perform DML, disable them.
16. On the target system, perform any cleanup required by partitioned replication. 17. On the target system, start Post.
sp_ctrl> start post The two instances are now in synchronization, and SharePlex resumes replication.
Apply patch/upgrade to source and target
Supported databases
Oracle on all supported platforms
When to use this procedure
Use this procedure if the patch or upgrade makes DDL changes of a type not replicated by SharePlex. For a list of objects for which DDL is supported, see the SharePlex Release Notes.
Overview of the procedure
Use this procedure to apply an application patch or upgrade if it includes changes to the database that are not replicated by SharePlex and you can stop user access to the source database to deactivate and reactivate the configuration file. It requires deactivation of the configuration file so that SharePlex can rebuild its object information to incorporate the changes that the patch or upgrade applied. When you reactivate the configuration, SharePlex will re-analyze all of the objects again. You can allow users to access the source data while the patch or upgrade is applied to the target system.
Apply the patch/upgrade
1. Stop user access to the instances involved in replication on the source and target systems, but do not shut down SharePlex.
2. On the source system, flush the data to the target system. This command stops Post and places a marker in the data stream that establishes a synchronization point between source and target data. sp_ctrl> flush datasource where: datasource is the datasource of the source instance, for example o.oraA.
SharePlex 9.2.8 Administrator Guide 324

3. On the source system, deactivate the configuration. sp_ctrl> deactivate config filename
4. On the source system, apply the patch or upgrade. 5. [If the patch/upgrade adds objects that must be replicated] On the source system, edit the configuration
file, including making any changes to column partitions or column conditions if using partitioned replication. For more information, see Change an active configuration file on page 304.
sp_ctrl> edit config filename 6. On the source system, activate the configuration file.
sp_ctrl> activate config filename 7. On the source system, restore user access to the source instance. 8. On the target system, apply the patch or upgrade. 9. On the target system, if the patch or upgrade installed triggers on the tables in replication, disable them
or run the sp_add_trigger.sql utility script so that the triggers ignore the SharePlex user. 10. On the target system, if the patch or upgrade added check constraints or scheduled jobs that perform
DML, disable them. 11. On the target system, perform any cleanup required by partitioned replication. 12. On the target system, start Post.
sp_ctrl> start post The two instances are now in synchronization, and SharePlex resumes replication.
Apply patch to source and replicate it to the target
Supported databases
Oracle on all supported platforms
When to use this procedure
Use this procedure if all of the operations applied by a patch or upgrade are supported by SharePlex and can be replicated to the target. This includes DML changes and DDL that is supported by SharePlex. If you are not sure whether the patch or upgrade performs operations that are not supported by SharePlex, use the procedure Apply patch/upgrade to source then copy it to target on page 322. Note: For a list of operations that SharePlex supports, see the SharePlex Release Notes.
SharePlex 9.2.8 Administrator Guide 325

Apply the patch/upgrade
1. Stop user access to the Oracle instances on the source and target systems. 2. On the source system, flush the data to the target system. This command stops Post and places a marker
in the data stream that establishes a synchronization point between source and target data. sp_ctrl> flush datasource where: datasource is the datasource of the source instance, for example o.oraA.
3. On the source system, apply the patch or upgrade. 4. On the source system, restore user access to the source instance. 5. On the target system, if the patch or upgrade created or modified triggers, disable them or run the sp_
add_trigger.sql utility script so that the triggers ignore the SharePlex user. 6. On the target system, restore user access to the target instance.
SharePlex 9.2.8 Administrator Guide 326

25
Back up data on the source or target
This topic contains procedures for making backups of source and target data while replication is active.
Contents
Perform a partial backup of the source data Perform a full backup of the source system
Perform a partial backup of the source data
To perform a partial back up of a source system (for example, to extract data or populate a data warehouse) while data is being replicated, you can perform the partial backup on the target system instead and copy the same data as you would on the source system. This procedure does not interrupt user access to the source data and does not deactivate or reactivate the configuration file.
Supported databases
Oracle All targets
SharePlex 9.2.8 Administrator Guide 327

Procedure
1. On the source system, start sp_ctrl. 2. On the source system, flush the data to the target system. This command stops the Post process and
places a marker in the data stream that establishes a synchronization point between source and target data.
sp_ctrl> flush datasource where: datasource is the datasource specification of the source database in the configuration file, for example o.ora1. 3. On the target system, back up the data. NOTE: The data now matches the source data at the flush marker. 4. On the target system, start the Post process. sp_ctrl> start post
Perform a full backup of the source system
To back up the entire source system, including SharePlex, you must shut down SharePlex replication while the backup is performed. This procedure stops user access to the source data but does not deactivate or reactivate the configuration file. Replication resumes when started after the backup.
Supported databases
Oracle All targets
Procedure
Perform these steps on the source system: 1. Stop all system activity. 2. Start sp_ctrl. 3. Flush the data to the target system. This command stops the Post process and places a marker in the data stream that establishes a synchronization point between source and target data. sp_ctrl> flush datasource where: datasource is the datasource specification of the source database in the configuration file, for example o.ora1. 4. Shut down SharePlex. This command shuts down SharePlex. sp_ctrl> shutdown
SharePlex 9.2.8 Administrator Guide 328

5. Shut down the database. 6. Perform the backup. 7. Start the database. 8. Start sp_cop (Unix and Linux) or the SharePlex service (Windows). 9. Start sp_ctrl. 10. Allow users to access the database. 11. Verify that the SharePlex Capture, Read, and Export processes started.
sp_ctrl> status Perform these steps on the target system:
1. Start Post. sp_ctrl> start post
2. Verify that Post started. sp_ctrl> status
SharePlex 9.2.8 Administrator Guide 329

26
Troubleshooting Tips
Use the following resources from the Quest Support Portal to help you troubleshoot SharePlex. l SharePlex Knowledge Base articles l How to do basic troubleshooting with SharePlex (video)
See also, Prevent and solve replication problems on page 227.
SharePlex 9.2.8 Administrator Guide 330

A
Appendix A: Peer-To-Peer Diagram
This diagram visually explains the concept of peer-to-peer replication. For more information, see Configure peer-to-peer replication on page 123.
SharePlex 9.2.8 Administrator Guide 331

SharePlex 9.2.8 Administrator Guide 332

Appendix B: SharePlex variables

SharePlex uses the following environment variables, which you may need to set in certain situations. Usually you must perform additional steps before or after setting a variable, so refer to the recommended instructions before setting a SharePlex variable.

Environment Variable EDITOR HOST SP_COP_TPORT SP_SYS_HOST_NAME
SP_SYS_VARDIR

Description
Sets the default ASCII text editor for sp_ctrl commands that use one, for example the create config command.
Sets a host name for all locally run sessions of sp_ctrl.
Sets a non-default port number for an instance of SharePlex. The default port number is 2100. You may need to set a different port number if one of the following is true:
l You are setting up additional instances of sp_cop.
l A different port number than 2100 must be used.
Sets the host name that SharePlex binds to during configuration activation. This variable is used for the following:
l Sets the virtual IP address (also known as the global cluster package name) on a clustered system, such as Oracle RAC. This variable must be set on all cluster nodes.
l In the case where there are multiple network cards on Windows, it must be set to the network card or IP address that you want SharePlex to use as the local host. Otherwise, if the server reboots after the SharePlex configuration is activated, the IP address may bind to a different network card from the one that is bound in the replication configuration.
l If SP_SYS_HOST_NAME is set to an IPV6 address on the source system, SharePlex on the target system must be version 9.0 or later.
Sets the full path to the SharePlex variable-data directory so that sp_cop

SharePlex 9.2.8 Administrator Guide 333

Environment Variable SP_SYS_SECURE_MODE

Description
can locate the configuration data, queues, logs and other information. If there is only one instance of sp_cop on the local system, this variable is set by default*. If there are multiple instances of sp_cop on the local system, always set this variable to point to the correct variable-data directory of an instance before setting any other SharePlex variables for that instance.
Suppresses the output of the compare and repair SQL log file for all compare and repair runs while the current instance of SharePlex is running. This variable must be set before starting SharePlex, so if the sp_cop process is running it must be restarted after setting this variable. When sp_cop is run with this environment variable, the compare and repair commands will not put data into SQL files and the Post process will not put data into the SharePlex error log.

* On Unix and Linux, the variable-data directory is set in the proddir/data/defaults.yaml file. On Windows, it is set in the Windows Registry.
To set an environment variable in Unix or Linux ksh shell:
export variable_name=value csh shell:
setenv variable_name value

ksh shell: export SP_SYS_VARDIR=full_path_of_variable-data_directory
csh shell: setenv SP_SYS_VARDIR full_path_of_variable-data_directory
To set an environment variable on Windows Set the environment variable by adding it to the SharePlex node of the Windows Registry. IMPORTANT! Incorrect Registry settings can adversely affect the operation of SharePlex. Consult a System Administrator if you need help with this procedure.
1. Shut down the SharePlex service. 2. Open the Run dialog. The location varies with the Windows version. 3. In the Run dialog, type regedit to run the Registry Editor. 4. Expand the SharePlex node:
\HKEY_LOCAL_MACHINE\Software\Wow6432node\Quest Software\SharePlex
5. Right click the port number of the SharePlex instance to which you want to add a variable, then select New, then String Value.
6. Under the Name column, right click the new variable, then select Rename. 7. Type the correct name. 8. Double click the new variable.

SharePlex 9.2.8 Administrator Guide 334

9. Under Value Data, enter the string for the new variable and then click OK. 10. Exit the Registry.
SharePlex 9.2.8 Administrator Guide 335

madbuild