IT24748: CONTAINER-COPY STORAGE POOL VOLUMES CONTAINING 0 REFCOUNT EXTENTS MAY NOT BE RETURNED TO SCRATCH

8.1.7.000-IBM-SPSRV-Linuxppc64le
8.1.7.000-IBM-SPCMS-WindowsX64
8.1.7.000-IBM-SPCMS-WindowsI32
8.1.7.000-IBM-SPCMS-Linuxx86_64
8.1.7.000-IBM-SPOC-WindowsX64
8.1.7.000-IBM-SPOC-Linuxx86_64
8.1.7.000-IBM-SPOC-Linuxs390x
8.1.7.000-IBM-SPOC-LinuxPPC64le
8.1.7.000-IBM-SPOC-AIX
8.1.7.000-IBM-SPSRV-WindowsX64
8.1.7.000-IBM-SPSRV-Linuxx86_64
8.1.7.000-IBM-SPSRV-Linuxs390x
8.1.7.000-IBM-SPSRV-AIX
IBM Spectrum Protect Server V8.1 Fix Pack 7 (V8.1.7) Downloads

APAR status

• Closed as program error.

Error description

• It is possible that during the reclaim phase of local protect,
  volumes may not be returned to scratch. This is caused when
  reclamation leaves 0 refcount extents on the reclaimable volume.
   When the 0 refcount extent meets the reusedelay value of the
  container pool, it is then marked for deletion. The next run of
  local protect performs the deletion which will then decrement
  the occupancy on the volume and the volume is set to pending
  assuming all extents chunks with refcount>0 have been moved.
  When the volume's reusedelay is met, the volume is changed from
  pending to scratch. This can also cause a decrease in
  performance during the reclamation phase of local protect.
  
  Tivoli Storage Manager Versions Affected: 7.1.X and 8.1.X on all
  versions and supported platforms
  
  Customer/L2 Diagnostics: In addition to seeing a decrease in
  performance of local protect, the following can be run  from a
  DB2 command line to verify if there are a large number of 0
  refcount extents:
  
  1) Pick a volume that should have been reclaimed after local
  protect reclamation but was not.
  
  2) Identify the volume ID
  
  db2 connect to TSMDB1
  db2 set schema TSMDB1
  
  db2 "select volid,volname,poolid from ss_volume_ids where
  VOLNAME='VOLUME_NAME_IN_CAPS' "
  
  3) Use the volume ID in the following select.
  
  db2 "select count_big(cc.chunkid) as tot_count,
  sum(cast(cc.length as bigint)) as tot_length, cc.poolid,
  cc.copy_cntrid, cc.flags, ac.refcount from sd_chunk_copies cc
  left join sd_all_chunks ac on (cc.chunkid=ac.chunkid) where
  cc.copy_cntrid=<volume ID> group by cc.poolid, cc.copy_cntrid,
  cc.flags, ac.refcount for read only with ur"
  
  If you see a non-zero tot_count with a refcount of 0 and no
  results with a refcount > 0  returned, then this applies.
  
  Initial Impact: Medium
  
  Additional Keywords: perf performance recl reclaim reclaimed
  reclamation local protect stgpool stg
  

Local fix

• Running the protect and reclaim phases separately will help
  improve this. To do this, use RECLAIM=NO during local protect to
   run only the protect phase and RECLAIM=ONLY to run only the
  reclaim phase.
  

Problem summary

• ****************************************************************
  * USERS AFFECTED:                                              *
  * All IBM Spectrum Protect server users of PROTECT STGPOOL     *
  * with TYPE=LOCAL                                              *
  ****************************************************************
  * PROBLEM DESCRIPTION:                                         *
  * See ERROR DESCRIPTION.                                       *
  ****************************************************************
  * RECOMMENDATION:                                              *
  * Apply fixing level when available. This problem is currently *
  * projected to be                                              *
  * fixed in level 8.1.7.0. Note that this is subject to change  *
  * at the discretion of IBM.                                    *
  ****************************************************************
  

Problem conclusion

• This problem was fixed.  However, the fix only applies if the
  REUSEDELAY of the container-copy pool is greater than or equal
  to the REUSEDELAY of the primary container pool.
  Affected platforms:  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