IT28428: AUTOMATIC CONTAINER DEFRAG PROCESS FAILS MARKING CONTAINER UNAVAILABLE WITH ERROR ANR3660E.

APAR status

• Closed as program error.

Error description

• In version 8.1.5, new functionality was added to automatically
  defragment containers using MOVE CONTAINER. During the auto
  defrag process the operation can fail with the following
  messages:
  .
  ANR0204I The container state for /xxx/yyy/000000000000123.dcf is
           updated from AVAILABLE to UNAVAILABLE.
  .
  ANR3660E An unexpected error occured while opening or writing to
           the container. Container /xxx/yyy/000000000000123.dcf i
           stgpool CNTRPOOL has been marked as UNAVAILABLE and
           should be audited to validate accessibility and content
  .
  AUDIT CONTAINER using option ACTION=SCANALL or SCANDAMAGED
  will mark the container as available, but the next time the
  automatic defrag container runs the volume is marked unavailable
  again.
  .
  DIAGNOSTICS:
  .
  The problem can be verified by checking for the return code 4306
  EOF error in the FFDC LOG, the message would be similar to the
  following:
  .
  [06-24-2018 20:17:36.357][ FFDC_GENERAL_SERVER_ERROR ]:
      (sdcntr.c:10010) Unable to find valid range: rc 4306, list
      count 2276
  [06-24-2018 20:17:37.034][ FFDC_GENERAL_SERVER_ERROR ]:
      (sdio.c:2290) Unexpected rc=4306
  .
  The problem is caused by the container size recorded in the
  database not matching the container's actual size.
  .
  To verify the container size recorded in the database for
  unavailable containers use the following db2 select to get
  the size from the sd_containers table:
    db2 connect to TSMDB1
    db2 set schema TSMDB1
    db2 select CNTRID,POOLID,substr\(cntrname,1,50\) as CNTRNAME,
        MAXSIZE,FILESIZE,FREESPACE,LARGEST_FREE,STATE  from
        sd_containers where state=1
  .
  Then use the OS command to list the actual file size reported:
     ls -l /xxx/yyy/000000000000123.dcf
  .
  If the database size does not match the size listed by the
  OS this APAR applies.
  .
  Note that after this fix is applied there may still be existing
  containers where the internal size does not match the OS size.
  TO make sure all containers have matching sizes run the
  following command after applying the fix:
     RESET CONTAINERSIZE <stgpool_name>
  .
  Tivoli Storage Manager Versions Affected:
  All supported server versions 8.1.5 and higher.
  .
  Initial Impact:
  medium
  .
  Additional Keywords:
  DEFRAG MOVE AUDIT CONTAINER ANR0204I ANR3660E TS001071749
  

Local fix

• 1. Use AUDIT CONTAINER with ACTION=SCANALL to make the volume
     available.
  2. Run this command the RESET CONTAINERSIZE to resolve the
     mismatch size:
       RESET CONTAINERSIZE <stgpool_name>
  3. Manually run  MOVE CONTAINER.
  

Problem summary

• ****************************************************************
  * USERS AFFECTED:                                              *
  * All IBM Spectrum Protect server users.                       *
  ****************************************************************
  * PROBLEM DESCRIPTION:                                         *
  * See error description.                                       *
  ****************************************************************
  * RECOMMENDATION:                                              *
  * Apply fixing level when available. This problem is currently *
  * projected to be fixed in level 8.1.7. Note that this is      *
  * subject to change at the discretion of IBM.                  *
  ****************************************************************
  

Problem conclusion

• This problem was fixed.
  Affected platforms for reported release:  AIX, Linux, and
  Windows.
  Platforms fixed:  AIX, Linux, and Windows.
  

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

  TSM SERVER

• Fixed component ID

  5698ISMSV

Applicable component levels