APAR: IY03669 COMPID: 5765B7300 REL: 500
ABSTRACT: INFINITE LOOP OF XC174006 FDCS DURING AGENT SHUTDOWN

PROBLEM DESCRIPTION:
  During abnormal shutdown situations, it is possible for
MQSeries agents to hang indefinitely while terminating the
internal health thread. While hanging, the agents produce
a stream of FDCs with probe XC174006 from xllSemSetVal.

LOCAL FIX:
  Ensure that SysV semaphore sets are not asynchronously
removed while MQSeries is running. If agents do enter this
hang state, kill them.

PROBLEM SUMMARY:
This problem has been fixed for MQSeries V5.0 and MQSeries V5.1.

PROBLEM CONCLUSION:
This problem has been fixed in MQSeries V5.0 and MQSeries V5.1
and the fix will be shipped in the following PTFs:
    A) MQSeries for V5.0 CSD08
          OS/2 U200116
          Windows NT U200117
          AIX U469658
          HP-UX U469659
          Sun Solaris U469671
    B) MQSeries for V5.1 CSD06
          OS/2 U200121
          Windows NT U200122
          AIX U471218
          HP-UX (V10) U471240
          HP-UX (V11) U471444
          Sun Solaris U471246

------

APAR: IY15169 COMPID: 5765B7300 REL: 520
ABSTRACT: CUMULATIVE MAINTENANCE #01 (CSD01) FOR MQSERIES FOR AIX V5.2

PROBLEM DESCRIPTION:
Service offering #01 (CSD01) for MQSeries for AIX V5.2 provides
fixes for the problems described in this APAR.

PROBLEM SUMMARY:
Service offering #01 (CSD01) includes fixes to the following
problems:
IC20274 - The backup and recovery facilities of MQSeries on
          Windows NT did not work correctly when following an
          RCRMQOBJ command with an RCDMQIMG command. The
          RCDMQIMG command generated an FDC. The problem was
          caused because RCRMQOBJ does not organize the queue
          file in the way that is expected by RCDMQIMG.
          However, if the recreated queue was accessed (by an
          application) then the queue file was reorganized again
          and RCDMQIMG worked correctly again.
IC20277 - The queue manager failed after producing some FDCs.
          The first FDC contained probe id AD027000 and referred
          to RC32 from CreateFile (ERROR_SHARING_VIOLATION).
          This error caused a second FDC with probe id ZX005000,
          because the error with CreateFile had ended the
          process. The problem was caused because the queue
          file had been opened by another application and the
          queue manager did not expect this situation. The
          queue manager will now retry a number of times to open
          the file before returning with an error.
IY03547 - After a successful browsing MQGET call, followed by a
          removing MQGET call to read an available message
          consisting of several segments, a subsequent browsing
          MQGET call with the LOCK option led to a failure with
          reason code 2195 (MQRC_UNEXPECTED_ERROR), completion
          code 2 (MQCC_FAILED) and to a broken connection to the
          queue manager. An FDC with probe id AQ105001 was
          produced from component aqhLogicalMsgLock.
IY03669 - During abnormal shutdown situations, it was possible
          for MQSeries agents to hang indefinitely while
          terminating the internal health thread. While
          hanging, the agents produced a stream of FDCs with
          probe id XC174006 from component xllSemSetVal.
IY08799 - MQSeries tracing on AIX could have made an impact on
          performance on a stressed system. MQSeries
          performance on AIX may have been slowed somewhat even
          when tracing was not active. On systems which were
          extremely heavily loaded, this overhead could slow
          MQSeries processing noticeably, although it should not
          have caused any hard errors.
IY12240 - This APAR has replaced the system() call in runmqtrm
          with a call to fork() and exec() in order to start
          triggered programs. This was necessary in order to
          avoid potential hangs due to the non-threadsafe nature
          of system().
IY14251 - The MQSeries trigger monitor (runmqtrm) places trigger
          messages to the dead letter queue with a feedback code
          of 265 (MQFB_APPL_CANNOT_BE_STARTED) if it cannot
          start the triggered application, which is correct
          behaviour. However, runmqtrm would do the same if the
          triggered program terminated with anything other than
          a zero exit status.
IY14563 - A SIGSEGV failure occurred in the amqrrfma process.
IY15320 - DB2 will not participate in MQSeries-coordinated
          global units of work on Sun Solaris Version 2.6 or Sun
          Solaris 7, complaining that it cannot access the
          MQSeries AX interface. The same failure is not seen
          on Sun Solaris 8.
IY15801 - When an attempt was made to shutdown MQSeries for Sun
          Solaris V5.2 using the command endmqm -i, an FDC with
          probe id ZL000028 from component zlaMain occurred and
          endmqm -i just blocked, waiting forever. A brutal
          shutdown (kill -9) had to be performed.
IY15839 - CRTMQM failed with FDC probe ids ZS142006 or ZS142005.
47379 - CCSID and encoding fields were not updated in a
          chained MQDLH header conversion.
50368 - Put disable of a clustered QMGRALIAS was ignored.
52172 - When creating a namelist in the MQExplorer, entering a
          long list of long names in capital separated by
          commas, and then clicking on OK, error message AMQ4082
          'Unexpected error (4008)' would appear.
52385 - MQPUT did NOT fail when CCSID was equal to 0 in
          certain structures.
52485 - Trying to run the internet gateway sample dmqsamp1
          gave a segmentation fault(core dump) on MQSeries for
          Sun Solaris V5.2.
52885 - Errors occurred in an MQRMH header conversion with
          expanding/contracting data.
52928 - In the "Parameter" tab of the Alert Moniter Properties
          panel, if the mnemonic of "Hide" (Ctrl+H) was pressed
          as soon as the panel appeared then it did not work.
53072 - Exception from amqcrsta and runmqchl on endmqm -p on
          MQSeries for AIX V5.2 accompanied by an FDC with probe
          id XC130003 from component xehExceptionHandler.
53156 - Application trapped if Oracle returned -3 from
          xa_close.
53234 - An FDC with probe id AD030001 from component
          adiReadFile (RC=0 from read) was created by rcdmqimg
          when disk was full.
53235 - Component xcsGetGrnam returned error number 2 when
          running under stress on MQSeries for Linux V5.2.
53244 - A SEGV could occur processing an FDC that was detected
          in the early stages of initialising a process.
53329 - Exception from component xtmTimerThread during endmqm
          -p on MQSeries for Linux V5.2 accompanied by an FDC
          with probe id XC130003 from component
          xehExceptionHandler.
53339.1 - The return code xecL_W_SEM_OWNER_DIED correctly was
          not always handled correctly on MQSeries for Windows
          NT V5.2.
53351 - The command to kill an idle queue manager failed to
          stop the queue manager on MQSeries for AIX V5.2.
53368 - An "endmqm -p" command failed to stop a queue manager
          on MQSeries for Linux V5.2.
53477 - The French translation for error message AMQ8048 was
          incorrect.
53479 - The magnifier was getting lost while tabbing because
          it was tabbing to hidden controls.
53621 - Error meesage AMQ9495 was incorrectly reported when
          the CLWL exit was correct.
54142 - Poor performance used to occur on preloaded queues.
54673 - Attempting to create a Queue Manager with a very long
          directory path failed with error message AMQ7064 (Log
          path not valid or inaccessable).
54852 - A queue manager would fail to restart after a
          fastbound application terminated abruptly and an FDC
          with probe id AT067013 would occur.
54919 - A SIGSEGV would occur after an "endmqm -i" command if
          a previous fastbound application terminated abruptly.
54962 - Archives on MQSeries for AIX V5.2 were 64 bit even on
          32 bit machines.

PROBLEM CONCLUSION:
Cumulative maintenance #01.

------

APAR: IY18751 COMPID: 5765B9500 REL: 120
ABSTRACT: RECEIVE ERRONEOUS MESSAGE "6027-1127 INSUFFICIENT SPACE

PROBLEM DESCRIPTION:
Customer is receiving the error message "6027-1127
Insufficient space for metadata" when adding disks to
a filesystem. The code that is producing this message
is no longer needed and should be removed.

PROBLEM SUMMARY:
Remove incorrect message 6027-1127 when adding
disks to a file system

PROBLEM CONCLUSION:
remove incorrect message from mmaddisk

------

APAR: IY19102 COMPID: 5765B8100 REL: 220
ABSTRACT: ERROR_ID 1 GENERATED IF CALLER HANGS UP BEFORE DT ANSWERS

PROBLEM DESCRIPTION:
Sometimes an error_id 1 is produced if a caller hangs up before
DirectTalk answers the call.

PROBLEM SUMMARY:
ERROR_ID 1 GENERATED IF CALLER HANGS UP BEFORE
DT ANSWERS

PROBLEM CONCLUSION:
In CHP, upon failure of a device driver
ioctl issued to answer an incoming phone call, if errno is
EACCES and Clr_hup is set, we now recognise the fact that the
caller hung up early and eventually return an
EDGE_AC_NOT_RINGING, which makes execution resume normally
without raising any errors.

------

APAR: IY19123 COMPID: 5765B9501 REL: 310
ABSTRACT: USER NOT IN GROUP ABLE TO TOUCH A FILE IN DIRECTORY WHERE GROUP

PROBLEM DESCRIPTION:
USER NOT IN GROUP ABLE TO TOUCH A FILE IN DIRECTORY WHERE GROUP
STICKY BIT IS SET.

PROBLEM SUMMARY:
A user not in the group can touch a file in a directory
where the group sticky bit is on.

PROBLEM CONCLUSION:
If the SGID bit is set in the parent directory, do not
modify the
credentials group id while picking the group id to set on a
file.

------

APAR: IY19801 COMPID: 5765B8100 REL: 220
ABSTRACT: ISDN TRANSFER FAILS ON MULTIPL TRUNKS ISDN COM

PROBLEM DESCRIPTION:
ISDN transfer fails across trunks on non NFAS protocol.
Works okay within the same trunk.

PROBLEM SUMMARY:
ISDN TRANSFER FAILS ON MULTIPL TRUNKS

PROBLEM CONCLUSION:
Fixed RLT function for multiple trunk
support. The Rose call id had to be passed and stored in
ISDNDL4 rather than ISDNDL3, so that call A and call B and
be cross referenced.

------

APAR: IY20035 COMPID: 5765B8100 REL: 220
ABSTRACT: ISDN TRANSFER FAILS ON MULTIPL TRUNKS ISDN ATT

PROBLEM DESCRIPTION:
ISDN transfer fails across trunks on non NFAS protocol.
Works okay within the same trunk.

PROBLEM SUMMARY:
ISDN TRANSFER FAILS ON MULTIPL TRUNKS

PROBLEM CONCLUSION:
Fixed RLT function for multiple trunk
support. The Rose call id had to be passed and stored in
ISDNDL4 rather than ISDNDL3, so that call A and call B and
be cross referenced.

------

APAR: IY20040 COMPID: 5765B8100 REL: 220
ABSTRACT: ISDN TRANSFER FAILS ON MULTIPL TRUNKS ISDN DMS100

PROBLEM DESCRIPTION:
ISDN transfer fails across trunks on non NFAS protocol.
Works okay within the same trunk.

PROBLEM SUMMARY:
ISDN TRANSFER FAILS ON MULTIPL TRUNKS

PROBLEM CONCLUSION:
Fixed RLT function for multiple trunk
support. The Rose call id had to be passed and stored in
ISDNDL4 rather than ISDNDL3, so that call A and call B and
be cross referenced.

------

APAR: IY20044 COMPID: 5765D5100 REL: 311
ABSTRACT: LATEST PSSP 3.1.1 FIXES AS OF JUNE 2001

PROBLEM DESCRIPTION:
This is the lastest PSSP ptf as of June 2001.
Order this apar to get all of the ptfs as of June 2001.

PROBLEM SUMMARY:
This is a packaging apar for PSSP 3.1.1 fixes
as of June 2001.

PROBLEM CONCLUSION:
This is a packaging apar for PSSP 3.1.1
fixes as of June 2001.

------

APAR: IY20066 COMPID: 5765B8100 REL: 220
ABSTRACT: ISDN TRANSFER FAILS ON MULTIPL TRUNKS ISDN EURO

PROBLEM DESCRIPTION:
ISDN transfer fails across trunks on non NFAS protocol.
Works okay within the same trunk.

PROBLEM SUMMARY:
ISDN TRANSFER FAILS ON MULTIPL TRUNKS

PROBLEM CONCLUSION:
Fixed RLT function for multiple trunk
support. The Rose call id had to be passed and stored in
ISDNDL4 rather than ISDNDL3, so that call A and call B and
be cross referenced.

------

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]