IT20685: Repeated FDCs showing Probe Id RM702021 component rrmCheckSubsForQMgr

APAR status

• Closed as program error.

Error description

• The following FDC is getting generated continuously.  There is
  no harm to product functionality, but the record uses unwanted
  quantities of disk space under /var/mqm/errors.
  
  Probe Id          :- RM702021
  Component         :- rrmCheckSubsForQMgr
  Program Name      :- amqrrmfa
  Comment1          :- Missing subscription
  
  MQM Function Stack ends with:
    rrmMaintainClqMgr
    rrmCheckSubsForQMgr
    xcsFFST
  

Local fix

Problem summary

• ****************************************************************
  USERS AFFECTED:
  Only users who see this FDC should be concerned.  No functional
  problem arises from this situation.  Applications continue to
  work ok.  But the FDCs are quite large and will be written every
  hour, eventually filling the file system in which
  /var/mqm/errors is placed.
  
  
  Platforms affected:
  MultiPlatform
  
  ****************************************************************
  PROBLEM DESCRIPTION:
  A certain sequence of creation of manually-defined CLUSSDR
  channels in a multi-cluster scenario, causes the immediate
  appearance of one of these FDCs, then a repeated appearance
  every hour.
  
  The necessary conditions are:
  - a CLUSRCVR must be defined on a Partial Repository queue
  manager, which then flows to a Full Repository queue manager.
  - the CLUSRCVR channel is in multiple clusters.
  - for at least one of these clusters, the recipient is a Full
  Repository for that cluster.
  - for at least one of these clusters, the recipient is not a
  Full Repository for that cluster.
  - the recipient queue manager has at least one manually-defined
  CLUSSDR that points to the Partial Repository queue manager.
  
  In this scenario MQ has failed to store a necessary flag, and
  the state is incorrect.  The FDC was correctly reporting the
  incorrect state.
  

Problem conclusion

• The FDC was correctly reporting an invalid state stored some
  short time beforehand.
  
  The MQ code has been corrected so that the correct state is
  stored with the object on the recipient queue manager in the
  above scenario.
  
  If you are already seeing these FDCs, note that, after applying
  a fix for this issue, the FDCs will not immediately cease to be
  written.
  
  To stop them being written, you will need to install a fix for
  this issue on the local queue manager, then visit the remote
  queue manager, and issue a REFRESH CLUSTER for just the one
  cluster named in the FDC.
  
  The remote queue manager name and channel name appear in the
  first dump area, which is beneath the text: ""Failing Qmgr
  record""
  
  The cluster name appears in the second dump area, which is
  beneath the text: "Failing Clus record".
  
  ---------------------------------------------------------------
  The fix is targeted for delivery in the following PTFs:
  
  Version    Maintenance Level
  v8.0       8.0.0.9
  v9.0 CD    9.0.5
  v9.0 LTS   9.0.0.4
  
  The latest available maintenance can be obtained from
  'WebSphere MQ Recommended Fixes'
  http://www-1.ibm.com/support/docview.wss?rs=171&uid=swg27006037
  
  If the maintenance level is not yet available information on
  its planned availability can be found in 'WebSphere MQ
  Planned Maintenance Release Dates'
  http://www-1.ibm.com/support/docview.wss?rs=171&uid=swg27006309
  ---------------------------------------------------------------
  

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:

  PI92380 412AC3Ÿ IT36360

Fix information

• Fixed component name

  WMQ BASE MULTIP

• Fixed component ID

  5724H7251

Applicable component levels