OA40172: THE REPLY QUEUE IS FULL BECAUSE OF TOO MANY MONITORING EVENTS

Obtain the fix for this APAR.

APAR status

• Closed as program error.

Error description

• Since agent subscribes one general topic, there are too many
  message received when the monitoring events is enabled.
  

Local fix

• The reason to make high curdepths is because customer enabled
  the broker
  monitoring events(business-level monitor), the details, please
  refer to:
  
  http://publib.boulder.ibm.com/infocenter/wmbhelp/v6r1m0/topic/co
  m.ibm.et
  ools.mft.doc/ac37850_.htm?resultof=%22%62%75%73%69%6e%65%73%73%2
  2%20%22%
  62%75%73%69%22%20%22%70%72%6f%63%65%73%73%22%20%22%6d%6f%6e%69%7
  4%6f%72%
  69%6e%67%22%20%22%6d%6f%6e%69%74%6f%72%22%20
  
  You can issue the broker command "mqsireportflowmonitoring
  <BrokerName>
  -g -j" to check if the business-level monitoring is enabled. And
  you can
  issue "mqsichangeflowmonitoring <BrokerName> -c inactive -g -j"
  command
  to disable the business-level monitoring or disable parts of
  monitoring
  by the parameter -e executionGroupName and -f messageFlow.
  

Problem summary

• ****************************************************************
  * USERS AFFECTED: All IBM Tivoli OMEGAMON XE for Messaging:    *
  *                 WebSphere Message Broker Monitoring 7.0.0    *
  *                 users.                                       *
  ****************************************************************
  * PROBLEM DESCRIPTION: The following symptom may be discovered *
  *                      when the Message Broker monitoring      *
  *                      events are enabled:                     *
  *                                                              *
  *                      . The temporary dynamic queue,          *
  *                        KQI.AGENT.REPLY.QUEUE*, used by the   *
  *                        WebSphere Message Broker monitoring   *
  *                        agent has a high current depth or     *
  *                        regularly becomes full.               *
  *                                                              *
  *                      . The WebSphere Message Broker          *
  *                        monitoring agent process, KQIAGENT,   *
  *                        has the high CPU usage.               *
  ****************************************************************
  * RECOMMENDATION: Apply this PTF.                              *
  ****************************************************************
  The Message Broker may emit a large amount of event messages
  if the business-level monitoring is enabled. Since the
  WebSphere Message Broker monitoring agent subscribes one
  general topic, all of these events will arrive at the queue
  KQI.AGENT.REPLY.QUEUE*, which is used by the agent to receive
  publications and reply messages from the broker. The
  processing of so many events may result in the high CPU usage
  by the agent. And a high current depth (or even queue full)
  will be observed on the queue KQI.AGENT.REPLY.QUEUE* if the
  agent is not able to process these events in a timely manner.
  

Problem conclusion

• The agent code is updated not to subscribe the high level
  general topic, but only those low level topics required by
  the agent.
  

Temporary fix

Comments

APAR Information

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

  OA39415

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

Modules/Macros

•    KQIAGENT
  

Fix information

• Fixed component name

  OMXE MSG BROKER

• Fixed component ID

  5698A87MB

Applicable component levels

• R700 PSY UA67235

     UP12/12/20 P F212

Fix is available

• Select the PTF appropriate for your component level. You will be required to sign in. Distribution on physical media is not available in all countries.