A fix is available
APAR status
Closed as program error.
Error description
Capture always checks the Db2 catalog to determine if the table is altered and needs to be REORG when activating a subscription.
Local fix
Problem summary
**************************************************************** * USERS AFFECTED: 1- All Q and SQL Capture users * * 2- users using Partition sub and Proxy mode * * feature * * 3- Q capture server alert latency warning * * 4- All SQL Apply users * * 5- All Qrep EP users * * 6- All Capture users * * 7- Q Apply with CCD target CONDENSED=N * **************************************************************** * PROBLEM DESCRIPTION: 1- Q and SQL Capture needs to support * * Db2 ALTER TABLESPACE MOVE TABLE * * 2- warmstart proxyMode Q capture * * failed if remote Db is not available * * and ON_SOURCE_DB_UNAVAIL is R * * 3- When Capture starts it might take * * couple of minutes to initialize and * * the CURRENT_LOG_TIME will display * * 1900-01-01-00.00.00.000000 (zero), * * which can not be fixed in engine. * * ASNMON generates alerts when it reads * * this value. What this customer wants * * is for ASNMON to realize this issue * * and not send the alerts until * * CURRENT_LOG_TIME goes to a non zero * * value. * * 4- SQL Apply high CPU consumption in * * ASNAISO after migration to DB2 V12 * * 5- inconsistent timestamp for Qrep EP * * 6- Capture always checks the Db2 * * catalog to determine if the table is * * altered and needs to be REORG when * * activating a subscription. * * 7- Q Apply may fail to resolve * * duplicate -803 errors for * * non-condensed CCD target after Q * * Capture is restarted from older LSN * * point * **************************************************************** * RECOMMENDATION: * **************************************************************** 1- Q and SQL Capture will detect the SYSIBM.SYSTABLE update log record for ALTER TABLESPACE MOVE TABLE operation and update the IFI log filter array with the new PSID for the source table. 2- Q capture queries system table in remote db to get partition key information. Q capture goes down if the query fails because remote DB is unavailable. 3- The requirement is to ignore the alert that qcap latency warning when the current log time is 1900- 4- The CPU costs increased from 3 b , to 300 b , per day in our apply-job after migration to DB2 V12. 5- Due to a logical operation error, the MQ header column b For most of the messages, the timestamp is 26 bytes. 6- It is not always possible to determine if REORG is required by querying the Db2 catalog. Only the user knows if the change was safe. 7- Q Apply may reports ASN0569E failures when encountering duplicates for non-condensed CCD after Q Capture is restarted from older LSN point
Problem conclusion
1- Q and SQL Capture should continue to work even after a source table is moved to a different tablespace. 2- With this fix, capture stores partition key information into two new columns, PART_KEY_COLNO and PART_KEY_ORDERING of IBMQREP_SRC_COLS, if these two columns exist. When capture warmstart, if remote DB is not available and ON_SOURCE_DB_UNAVAIL is R, capture queries IBMQREP_SRC_COLS to get partition key information. 3- Ignore the alert that qcap latency warning when the current log time is 1900- 4- Rewrote some subqueries to improve query performance. 5- Correct the code defect so that EP always generates a correct 'commit_time' for the MQ header column 6- Introduce a new Q Capture startup parameter CAPSTART_REORGCHECK (default is Y for Db2 V11 and N for Db2 V12) to turn on/off checking for REORG. 7- Q Apply is fixed to correctly handle duplicates for non-condensed CCD targets after Q Capture is restarted from older LSN point ASNCLP was not updated. Function Level: 1140.103
Temporary fix
Comments
APAR Information
APAR number
PH29264
Reported component name
WS REPLICATION
Reported component ID
5655L8800
Reported release
B33
Status
CLOSED PER
PE
NoPE
HIPER
NoHIPER
Special Attention
NoSpecatt / Xsystem
Submitted date
2020-09-08
Closed date
2020-10-28
Last modified date
2020-12-01
APAR is sysrouted FROM one or more of the following:
APAR is sysrouted TO one or more of the following:
UI72310 UI72311 UI72312 UI72313
Modules/Macros
ASNAAPP ASNACMD ASNACMP ASNADMSP ASNAFET ASNAISO ASNAMAN ASNAPPLY ASNAPRS ASNAWPN ASNCAP ASNCATM ASNCCMD ASNCDINS ASNCLPB ASNCTLZD ASNMCMD ASNMON ASNMONIT ASNQACMD ASNQAPP ASNQAROW ASNQCAP ASNQCCMD ASNQCTLZ ASNQDEP ASNQEXRP ASNQMFMT ASNQXFMT ASNQ1140 ASNRBASE ASNTDIFF ASNTDSP ASNTRC ASNVSQL ASN2BASE
Fix information
Fixed component name
WS REPLICATION
Fixed component ID
5655L8800
Applicable component levels
RB33 PSY UI72310
UP20/11/07 P F011
RB34 PSY UI72311
UP20/11/10 P F011
RB35 PSY UI72312
UP20/11/10 P F011
RB36 PSY UI72313
UP20/11/07 P F011
Fix is available
Select the PTF appropriate for your component level. You will be required to sign in. Distribution on physical media is not available in all countries.
[{"Line of Business":{"code":"LOB10","label":"Data and AI"},"Business Unit":{"code":"BU059","label":"IBM Software w\/o TPS"},"Product":{"code":"SSDP5R","label":"InfoSphere Replication Server"},"Platform":[{"code":"PF025","label":"Platform Independent"}],"Version":"B33"}]
Document Information
Modified date:
02 December 2020