IT37500: FDCs with ProbeID 'CO609020' generated by queue manager for a SVRCONN channel with SHARECNV(1)

APAR status

• Closed as program error.

Error description

• Numerous failure data capture (FDC) records with Probe Id
  CO609020 from Component
  cciHeartbeatMonitorThread and error 'MQRC_UNEXPECTED_ERROR'
  are generated when an MQ client connects to the queue manager
  with its shared conversation attribute of the server connection
  channel is set to 1: SHARECNV(1).
  

Local fix

• Change the server connection channel's SHARECNV(1) to
  SHARECNV(2) or higher.
  

Problem summary

• ****************************************************************
  USERS AFFECTED:
  MQ clients, especially MQ JMS clients, using SHARECNV(1)
  channels to connect to queue managers on platforms other than
  z/OS.
  
  
  Platforms affected:
  AIX, IBM iSeries, Linux on Power, Linux on x86-64, Linux on
  zSeries, Windows
  
  ****************************************************************
  PROBLEM DESCRIPTION:
  IBM MQ clients which connect over SHARECNV(1) channels use a
  faster synchronous form of MQGET to retrieve messages from the
  queue manager.  However, no other MQ calls are allowed while a
  synchronous MQGET is in progress.
  
  IBM has determined that MQ clients can sometimes violate this
  restriction by sending additional MQ requests to the queue
  manager before an outstanding synchronous MQGET calls has
  completed.  When this occurs, the queue manager can generate an
  FFST with Probe Id CO609020 from Component cciProcessOne.  In
  some cases, the queue manager can also suffer a memory fault
  (SIGSEGV) after reporting the CO609020 FFST.
  

Problem conclusion

• This APAR avoids the CO609020 FFST and memory fault by making
  sure the queue manager does not attempt to process any MQ
  requests received during a synchronous MQGET.  Instead, the
  queue manager generates a new FFST for this specific case with
  Probe Id CO547103 from Component cciProcessOne from the
  HeartbeatMonitor thread.  For example:
  
    Probe Id          :- CO547103
    Component         :- cciProcessOne
    Program Name      :- amqrmppa
    Thread            :- 5    HeartbeatMonitor
    Major Errorcode   :- rrcE_PROTOCOL_ERROR
    Minor Errorcode   :- OK
    Probe Type        :- MSGAMQ9504
    Probe Description :- AMQ9504E: A protocol error was detected
  for channel 'TEST.1'.
    Comment1          :- TEST.1
    Comment2          :- MQJM09030000
    Comment3          :- 192.168.1.2
  
  The comments in the FFST identify the client channel name, the
  client type and version (refer to the RPRODUCT and RVERSION
  fields in the DISPLAY CHSTATUS documentation:
  https://www.ibm.com/docs/en/ibm-mq/9.3?topic=reference-display-c
  hstatus-display-channel-status), and the client network address.
   This information makes it much easier to identify the client
  which caused the problem in order to address the fault in the MQ
  client.
  
  After generating the FFST, the queue manager writes message
  AMQ9504 ("A protocol error was detected") to its error logs and
  the client channel connection ends.  The client application
  receives MQCC_FAILED, MQRC_CONNECTION_BROKEN from MQGET.
  
  This APAR also provides a way to avoid the FFSTs and keep the
  client from having its connection broken.  For each MQ client
  RPRODUCT value, such as "MQJM" for the IBM MQ classes for JMS in
  normal mode, or "MQCC" for the IBM MQ C client, you can set an
  environment variable of the form AMQ_xxxx_USE_SYNC_GET=0, where
  "xxxx" must be replaced with the four-character client RPRODUCT
  value.
  
  For example, if the CO547103 FFST from cciProcessOne shows
  "MQJM" in the Comment2 field indicating a problem with an MQ JMS
  client, you can set the environment variable
  AMQ_MQJM_USE_SYNC_GET=0 before starting the queue manager.  The
  queue manager will then direct MQ JMS clients to use
  asynchronous MQGET, avoiding the FFSTs and memory fault.  More
  than one such environment variable may be set if required.
  
  ---------------------------------------------------------------
  The fix is targeted for delivery in the following PTFs:
  
  Version    Maintenance Level
  v9.2 LTS   9.2.0.11
  v9.3 LTS   9.3.0.5
  v9.x CD    9.3.3
  
  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:

Fix information

• Fixed component name

  MQ BASE V9.2

• Fixed component ID

  5724H7281

Applicable component levels