PH50187: CDC-REPLICATED TABLES OUT OF A SUDDEN GET SUSPENDED FROM REPLICATION ("TABLE ... WAS PARKED BECAUSE OF AN ERROR...")

APAR status

• Closed as documentation error.

Error description

• During CDC-based replication the apply of changes on an
  accelerator-shadow table suddenly fails. The issue becomes
  obvious by series of DXNX881I messages in DB2MSTR which include
  combinations of message 1 and message 2a or message 1 and
  message 2b described below:
  
  Message 1:
  DSNX881I Subscription: (subscription name)
  Message: Table (table name> in subscription (subscription name)
  was parked because an error occurred during replication.
  Originator:
  com.datamirror.ts.engine.component.target.TargetSubscriptionBulk
  ApplyController
  Message: Mirror bulk apply operation failed.
  Issue: Cannot apply changes to table
  
  Message 2a:
  DSNX881I ... A SQL exception has occurred. The SQL error code is
  '0'. The SQL state is: Ýnull¨.
  The error message is:
  Mirror bulk apply external table insert error occurred.
  Originator:
  com.datamirror.ts.target.apply.bulkapply.ExecuteBulkApplyStateme
  ntJob
  Actual number of affected rows (0) are not equal to expected
  number of affected rows (nnnn)."
  (Note: nnnn is a number larger than 0)
  
  Message 2b:
  DSNX881I ... A SQL exception has occurred. The SQL error code is
  '0'. The SQL state is: Ýnull¨.
  The error message is:
  Mirror bulk apply external table insert error occurred.
  Originator:
  com.datamirror.ts.target.apply.bulkapply.ExecuteBulkApplyStateme
  ntJob
  Actual number of affected rows (xxxx) are not equal to expected
  number of affected rows (yyyy)."
  (Note:
    - xxxx and yyyy are numbers larger than 0)
    - xxxx is less than yyyy)
  In addition message CHC0742I may be seen.
  Accelerator maintenance level 7.5.7 includes a new edition of
  the replication engine (RE) component that fixes the underlying
  issue resulting in the occurrence of messages 1 and 2a.
  Accelerator maintenance level 7.5.10 will include a new edition
  of the replication engine (RE) component that fixes the
  underlying issue resulting in the occurrence of messages 1 and
  2b.
  Additional keywords:
  TS007194163 TS006194434 TS006739372 TS010921821
  DSNX881I-20
  GH/.../Customer-Cases/issues/280
  

Local fix

• Customers have to reload the tables that have been affected by
  the issue and were parked.
  

Problem summary

• Problem Summary:
  CDC-replicated tables out of a sudden get suspended from
  replication. The event is accompanied by a series of DSNX881I
  messages in DB2MSTR. One of these messages denotes an affected
  table as "has been parked". Furthermore the information "Mirror
  bulk apply external table insert error occurred. Originator:
  com.datamirror.ts.target.apply.bulkapply.ExecuteBulkApplyStateme
  ntJob" is contained in an DSNX881I message.
  
  Users Affected:
  Users of CDC-based replication
  
  Problem Scenario:
  See APAR Error description.
  
  Problem Symptoms:
  See APAR Error description.
  

Problem conclusion

• Accelerator maintenance level 7.5.7 includes a new edition of
  the replication engine (RE) component that fixes the underlying
  issue resulting in the occurrence of the DSNX881I message with
  the text
  "Actual number of affected rows (0) are not equal to expected
  number of affected rows (nnnn)."
  where nnnn is a number larger than zero.
  
  Accelerator maintenance level 7.5.10 will include a new edition
  of the replication engine (RE) component that fixes the
  underlying issue resulting in the occurrence of the DSNX881I
  message with the text
  "Actual number of affected rows (xxxx) are not equal to expected
  number of affected rows (yyyy)."
  where xxxx and yyyy are numbers larger than 0 and xxxx is less
  than yyyy.
  
  Upgrade your Accelerator maintenance level accordingly.
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

• APAR is sysrouted TO one or more of the following:

Fix information

Applicable component levels