IT11581: STORING DATA TO DIRECTORY-CONTAINER STORAGE POOL CAN HANG CAUSING HIGH SYSTEM CPU/MEMORY USAGE WITH DSMSERV

7.1.3.100-TIV-TSMSRV-WIN
7.1.3.100-TIV-TSMSRV-SolarisSPARC
7.1.3.100-TIV-TSMSRV-Linuxx86_64
7.1.3.100-TIV-TSMSRV-Linuxs390x
7.1.3.100-TIV-TSMSRV-Linuxppc64
7.1.3.100-TIV-TSMSRV-HP-UX
7.1.3.100-TIV-TSMSRV-AIX
IBM Spectrum Protect Server V7.1.3.X interim fix downloads

APAR status

• Closed as program error.

Error description

• When backing up or archiving data to a Tivoli Storage Manager
  7.1.3 directory-container storage pool, or using the REPLICATE
  NODE command to replicate source data to a 7.1.3 target system's
  directory-container storage pool, the process can appear to be
  hung.  While the client activity or replication process appears
  to be hung, the server is actually processing in a tight loop,
  leading to excessive CPU/memory usage by the dsmserv process on
  the system.  In some cases, the system may run out of memory.
  
  Cancelling the client activity or replication process will not
  clear the condition, and the Tivoli Storage Manager server must
  be recycled to recover from the hang.
  
  
  
  
  
  Customer/L2 Diagnostics:
  Messages similar to the following *might* be recorded in the
  system's FFDC log to indicate a match for this problem.  In the
  case of replication, the messages will appear on the target
  server and look like the following:
  
  [10-01-2015 13:26:21.435][ FFDC_GENERAL_SERVER_ERROR ]:
  (sdcreate.c:2039) Encountered batchP 000000F02CB4A000 with
  0 chunks. batch is global, controlP 0000000000000000,
  controlP overallReturnCode 0, sessP 0000000000000000, poolId
  0, pool (can't eval) cloud, obj rptSize 0, sinkRc 0
  [10-01-2015 13:26:21.435][ FFDC_GENERAL_SERVER_ERROR ]:
  (sdcreate.c:2039) Encountered batchP 000000F042D48000 with
  0 chunks. batch is global, controlP 000000F045522000,
  controlP overallReturnCode 0, sessP 000000F0342FFC90, poolId
  4, pool is not cloud, obj rptSize 206832191, sinkRc 0
  
  Additionally, a message similar to the following *might* be
  recorded in the system's activity log:
  
  09/30/2015 20:14:31 ANR9999D_2225955985 outFFDCLog
  (sdcreate.c:2039) Thread<242>:
  Encountered batchP 00000074D2D4A000
  with 0 chunks. batch is global,
  controlP 00000074F1B90000, controlP
  overallReturnCode 0, sessP
  00000074D95BE640, poolId 4, pool is
  not cloud, obj rptSize 206832191,
  sinkRc 0(SESSION: 1234)
  
  In the case of client backup/archive activity, the following
  *might* be recorded in the system's activity log:
  
  2015-10-11 23:14:37 ANR9999D_3047338377 outFFDCLog(sdcreate.c:
                      2039) Thread<602538>: Encountered batchP
                      0x7ffabe8ad000 with 0 chunks. batch is
                      global, controlP (nil), controlP
                      overallReturnCode 0, sessP (nil), poolId 0,
                      pool (can't eval) cloud, obj rptSize 0, sink
                      Rc 0(SESSION: 41149)
  Initial Impact:
  High
  
  Tivoli Storage Manager Versions Affected:
  All 7.1.3 Tivoli Storage Manager servers.
  
  Additional Keywords:
  TSM 104926 FFDC_GENERAL_SERVER_ERROR SDCREATE REPL NODE
  REPLICATE HANG HUNG CPU LOOP MEMORY FFDC ANR9999D_2225955985
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED:                                              *
  * Users affected: All Tivoli Storage Manager server users of   *
  * container storage pools.                                     *
  ****************************************************************
  * PROBLEM DESCRIPTION:                                         *
  * See error description.                                       *
  ****************************************************************
  * RECOMMENDATION:                                              *
  * Apply fixing level when available. This problem is projected *
  * to be fixed in levels 7.1.3.100 and 7.1.4. Note that this is *
  * subject to change at the discretion of IBM.                  *
  ****************************************************************
  

Problem conclusion

• This problem was fixed.
  Affected platforms: AIX, Solaris, 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

• R71W PSY

     UP

• R71S PSY

     UP

• R71A PSY

     UP

• R71L PSY

     UP