IT45131: Probe Id XC130004 SIGSEGV in rrmRecoClqMgrDelRemote function in amqrrmfa causes queue manager to fail

APAR status

• Closed as program error.

Error description

• Diagnostic outputs similar to those below are written in FDC
  files in /var/mqm/errors.  The queue manager ends immediately.
  
  Probe Id :- XC130004
  Component :- xehExceptionHandler
  Program Name :- amqrrmfa
  Arith1 :- 11 (0xb)
  Comment1 :- SIGSEGV: address not mapped(0xc)
  
  O/S Call Stack for current thread includes
  /opt/mqm/lib64/libmqml_r.so(rrmRecoClqMgrDelRemote+0x2f1)[0x14d8
  
  MQM Function Stack
  amqrrfma_main
  rrmMain
  rrmRepository
  rrmProcessMsg
  rrmChangeClqMgr
  rrmRecoClqMgr
  rrmRecoClqMgrDelRemote
  xcsFFST
  

Local fix

Problem summary

• ****************************************************************
  USERS AFFECTED:
  Any users of MQ clustering, who have installed one of the
  affected fix packs, are potentially affected by this issue.
  These include the IBM MQ versions:
  
    9.2.0.20
    9.2.0.21
    9.2.0.22
  
    9.3.0.10
    9.3.0.15
    9.3.0.16
  
    9.3.3 CD
    9.3.4 CD
  
  The defective code path is executed on the local queue manager
  only when it is running one of the versions above.  The version
  of IBM MQ on the remote queue manager is not significant.
  
  The defective code path is executed on the local queue manager
  when a remote queue manager is removed from a cluster (for
  example, by setting its CLUSRCVR channel's CLUSTER attribute to
  blank), and the local queue manager knows about that remote
  queue manager in that cluster.
  
  The defective code paths are executed on the local queue
  manager, only when the updates on the remote queue manager
  arrive out of order at the local queue manager.
  Two known examples of when this might happen are:
  - The remote queue manager was restored from a backup, within in
  the last few months.
  - The remote queue manager was removed, then re-added to a
  cluster within the last few months, and the resulting update
  messages were delayed in transmission to the Full Repositories,
  and/or to the local queue manager.
  
  
  Platforms affected:
  MultiPlatform
  
  ****************************************************************
  PROBLEM DESCRIPTION:
  A new code branch was added via APAR IT43646 into the affected
  levels of MQ code.
  
  The new branch allowed a Delete of a CLUSRCVR channel to be
  processed, when this was not correct as per the design. This new
  branch did not initialise a pointer variable needed later in the
  code, and therefore it invalidly had a null (zero) value. When
  de-referenced in the later code, this gave a SIGSEGV.
  

Problem conclusion

• The code branch has been corrected to avoid this issue
  occurring.
  
  ---------------------------------------------------------------
  The fix is targeted for delivery in the following PTFs:
  
  Version    Maintenance Level
  v9.2 LTS   9.2.0.25
  v9.3 LTS   9.3.0.17
  v9.x CD    9.3.5
  
  The latest available maintenance can be obtained from
  'IBM MQ Recommended Fixes'
  https://www.ibm.com/support/pages/recommended-fixes-ibm-mq
  
  If the maintenance level is not yet available information on
  its planned availability can be found in 'IBM MQ
  Planned Maintenance Release Dates'
  https://ibm.biz/mqplannedmaintenance
  
  ---------------------------------------------------------------
  

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

  MQ BASE V9.2

• Fixed component ID

  5724H7281

Applicable component levels