JR63828: LOCK TIMEOUT ON TS_BOOKMARK TABLE WHEN TARGET IS KAFKA AND METADATA IS BEING UPDATED AND SYSTEM IS BUSY

APAR status

• Closed as program error.

Error description

• Every few days mirroring fails with the following error:
  Exception occurred during replication.
  Unable to load bookmark data from database. The SQL error code
  is "84,000". The SQL state is "ZZC01". The SQL message is "Lock
  time out; try later.
  
  This indicates a lock timeout occurred on the TS_BOOKMARK table.
  For a Kafka target, the TS_BOOKMARK table is in the Pointbase md
  (metadata) database. When a Pointbase lock timeout occurs,
  Pointbase dumps the lock table to a file called md_Debug.log.
  This shows there is no database deadlock. The lock timeout
  occurs for a select from TS_BOOKMARK table to get the bookmark .
  The select wants to get an S lock on the row, but it is blocked
  by an X lock being held by an update of that bookmark row. The
  txn that performed the update should immediately be committed,
  so the X lock should only be held a very short time before it is
  released. But for some reason it is held long enough for the
  select to wait longer than the default Pointbase lock timeout (1
  minute).
  

Local fix

• Increase the Pointbase lock timeout. The default is 60 seconds.
  
  It can be increased by placing a file called "pointbase.ini" in
  the CDC install directory.
  
  To set the lock timeout to 120 seconds, the file should contain
  the following line:
  locks.timeout=120
  

Problem summary

• ****************************************************************
  * USERS AFFECTED:                                              *
  * This issue may affect you if the host running CDC for Kafka  *
  * is busy.                                                     *
  ****************************************************************
  * PROBLEM DESCRIPTION:                                         *
  * Mirroring ends on error.                                     *
  ****************************************************************
  * RECOMMENDATION:                                              *
  * Upgrade to IDR 11.4.0.4-5609 or newer for Kafka.             *
  ****************************************************************
  

Problem conclusion

• Upgrading to IDR 11.4.0.4-5609 or newer for Kafka resolves the
  issue.
  

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

• Fixed component name

  IDR KAFKA

• Fixed component ID

  5725E30KA

Applicable component levels