PH29166: MESSAGE ENGINE DEADLOCK PROBLEM

APAR status

• Closed as program error.

Error description

• At 12:30:09:355 on June 23rd, the background processing that
  was being executed on the application side failed and the
  following error was output.
  ----------
  [20/06/23 12:30:09:355
  JST] 0000009b SibMessage E
  [intjmsbus:CLUSTERNME.002-intjmsbus] CWSIS1501E:
  ??????????????? com.ibm.db2.jcc.am.BatchUpdateException:
  [jcc][t4][102][10040][3.69.76] ???????????
  ?????????????????????????????? 1
  ????????????
  getNextException()
  ??????????????????????????????????? ERRORCODE=-4229,
  SQLSTATE=null ??????????
  ----------
  
  The message store is a
  data store, not a file store. The error output to the Db2 side
  at the same time is as
  follows.
  ----------
  2020-06-23-12.30.09.343842+540
  I30299588E869 LEVEL: Error
  PID : 11220 TID : 47037747816192
  PROC : db2sysc 0
  INSTANCE: db2mam NODE : 000 DB : MAMDB
  APPHDL
  : 0-11627 APPID: 184.120.2.1.51500.200503103434
  AUTHID : DB2MAM
  HOSTNAME: HOSTNAME
  EDUID : 18483 EDUNAME: db2agent (MAMDB)
  0
  FUNCTION: DB2 UDB, relation data serv, sqlrr_array_input,
  probe:210
  DATA #1 : SQLCA, PD_DB2_TYPE_SQLCA, 136 bytes
  
  sqlcaid : SQLCA sqlcabc: 136 sqlcode: -911 sqlerrml: 1
  
  sqlerrmc: 2
  sqlerrp : SQLRI2CC
  sqlerrd : (1) 0x80100002 (2)
  0x00000002 (3) 0x00000000
  (4) 0x00000000 (5) 0xFFFFFE0C (6)
  0x00000000
  sqlwarn : (1) (2) (3) (4) (5) (6)
  (7) (8) (9)
  (10) (11)
  ----------
  
  After that, we received the following
  report from the Db2 team investigation.
  ----------
  The
  deadloack happens like as followings:
  Application A holding a
  row lock on IBMWASSIB000.SIB001 table, then request another row
  locks on the same table, which already held by B.
  Application B
  also holding a row lock on IBMWASSIB000.SIB001 table, then
  request another row locks on the same table, which already held
  by A.
  i.e, this deadlock is caused by a defect on the
  application A/B. So, the application have to be
  fixed.
  ----------
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED:  Users of IBM WebSphere Application Server   *
  *                  who are running a SIB messaging engine      *
  *                  using a database as a message store         *
  ****************************************************************
  * PROBLEM DESCRIPTION: Message processing can stop and the     *
  *                      database can report a deadlock within   *
  *                      the SIB database tables.                *
  ****************************************************************
  * RECOMMENDATION:                                              *
  ****************************************************************
  When multiple threads are attempting to update message records
  in the datastore, a lock is held on to an updated record while
  an attempt is made to lock the next record to be updated. If
  multiple threads are updating records in different orders it
  is possible for a deadlock to occur.
  

Problem conclusion

• The fix for this APAR is targeted for inclusion in fix pack
  9.0.5.9.
  For more information, see 'Recommended Updates for WebSphere
  Application Server':
  https://www.ibm.com/support/pages/node/715553
  

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

  WEBS APP SERV N

• Fixed component ID

  5724H8800

Applicable component levels

• R850 PSY

     UP

• R900 PSY

     UP