IC89767: SERVER MAY HANG WHEN SHOW DBCONN COMMAND IS ISSUED

IBM Tivoli Storage Manager (TSM) server V6.3.3.x interim fix downloads

APAR status

• Closed as program error.

Error description

• There is a small timing window in which the SHOW DBCONN command
  may result in a hang condition on the Tivoli Storage Manager
  Server.
  Platforms affected:
  All Tivoli Storage Manager 6.3.x Servers
  Customer/L2 Diagnostics:
  The output from the pstack/procstack command issued against the
  dsmserv process id will show a call stack similar to the
  following for the thread associated with the SHOW DBCONN
  command:
  pth_cond._cond_wait_global(??, ??, ??) at 0x9000000006fd4d0
  pth_cond._cond_wait(??, ??, ??) at 0x9000000006fe058
  pth_cond.pthread_cond_wait(??, ??) at 0x9000000006fed48
  pkmon.pkWaitConditionTracked(??, ??, ??, ??, ??) at 0x100007870
  output.OutQueueData(??, ??, ??) at 0x10001f7c0
  output.StdPutText(??, ??, ??, ??) at 0x10001c994
  outvarg.outPrintf(0x710112023b220000, 0x100e95b70, 0x167d336b0,
  0x167d336e0, 0x167d33700, 0x101402a70, 0x66db, 0x167d3277c) at
  0x10000a6b4
  tmshow.ShowTxnDesc(??, ??, ??) at 0x1000c778c
  tmshow.tmShowTxnId(??, ??, ??) at 0x1000c7d08
  dbiconn.dbiShowConnPool@AF26_7(??) at 0x1000b77d8
  admshow.AdmShow(??) at 0x100cbc50c
  admcmd.AdmCommandLocal(??, ??, ??, ??, ??) at 0x1005a0348
  admcmd.admCommand(??, ??, ??, ??, ??) at 0x10059ecb8
  smadmin.SmAdminCommandThread(??) at 0x100a46530
  pkthread.StartThread(??) at 0x10000c0dc
  The dbiShowConnPool() function is holding the DBV mutex.  The
  amount of data generated by the SHOW DBCONN command exceeded the
  size of the output buffer and, thus, the thread had to signal
  the admin thread to send data back to the client.  The call
  stack associated with the admin thread shows the following:
  pth_spinlock._global_lock_common(??, ??, ??) at
  0x9000000006db7b4
  pth_mutex._mutex_lock(??, ??, ??) at 0x9000000006e8bd8
  pkmon.pkAcquireMutexTracked(??, ??, ??) at 0x100006648
  smtrans.SmSendVerbX(??, ??) at 0x100333158
  smadmin.SmAdminOutput(??) at 0x100a4687c
  smadmin.SmAdminSession(??) at 0x100a45124
  smexec.DoAdminGeneral(??) at 0x1001aa168
  smexec.smExecuteSession(??, ??, ??, ??, ??, ??, ??, ??) at
  0x1001a3774
  tcpcomm.psSessionThread(??) at 0x100063858
  pkthread.StartThread(??) at 0x10000c0dc
  The admin thread is attempting to acquire the SMV mutex, but the
  mutex is currently held by another thread (restore/retrieve
  operation), as seen below
  
  pth_spinlock._global_lock_common(??, ??, ??) at
  0x9000000006db7b4
  pth_mutex._mutex_lock(??, ??, ??) at 0x9000000006e8bd8
  pkmon.pkAcquireMutexTracked(??, ??, ??) at 0x100006648
  dbiinit.dbIsRestoreInFlight@AF28_20(??, ??) at 0x1002335ac
  smtrans.SmSendData(??, ??, ??, ??) at 0x10033931c
  sstrans.RtrvFramed(??, ??, ??, ??, ??, ??, ??, ??) at
  0x100659dd0
  sstrans.ssRtrv(??, ??, ??, ??, ??, ??, ??, ??) at 0x1006589d0
  afrtrv.AfRtrv(??, ??, ??, ??, ??, ??, ??, ??) at 0x1008f5bdc
  bfrtrv.RtrvOne(??, ??, ??, ??, ??, ??, ??, ??) at 0x100864e84
  bfrtrv.bfRtrvExt(??, ??, ??, ??, ??, ??, ??, ??) at 0x100860c5c
  smnqr.SmRetrieveBitfile(??, ??, ??, ??, ??, ??, ??) at
  0x10094c8c0
  smnode.SmDoObjRtrv(??, ??, ??, ??, ??, ??) at 0x100794834
  smnode.SmNodeSession(??, ??) at 0x10077ee54
  smexec.HandleNodeSession(??, ??, ??) at 0x1001b0db8
  smexec.DoNodeGeneral(??, ??) at 0x1001a8730
  smexec.smExecuteSession(??, ??, ??, ??, ??, ??, ??, ??) at
  0x1001a374c
  tcpcomm.psSessionThread(??) at 0x100063858
  pkthread.StartThread(??) at 0x10000c0dc
  The restore/retrieve thread is holding the SMV mutex while
  attempting to acquire the DBV mutex held by the SHOW DBCONN
  command.  This results in the deadlock condition that ultimately
  causes the Tivoli Storage Manager Server to hang.
  Initial Impact:
  Medium
  Additional Keywords:
  wait lock unresponsive txnt dbtxnt deadlocks
  

Local fix

• Do not issue SHOW DBCONN command.
  

Problem summary

• ****************************************************************
  * USERS AFFECTED: All Tivoli Storage Manager server users.     *
  ****************************************************************
  * PROBLEM DESCRIPTION: See error description.                  *
  ****************************************************************
  * RECOMMENDATION: Apply fixing level when available. This      *
  *                 problem is currently projected to be fixed   *
  *                 in levels 6.3.3.100 and 6.3.4. Note that     *
  *                 this is subject to change at the             *
  *                 discretion of IBM.                           *
  ****************************************************************
  *
  

Problem conclusion

• This problem was fixed.
  Affected platforms:  AIX, HP-UX, 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

• R63A PSY

     UP

• R63H PSY

     UP

• R63L PSY

     UP

• R63S PSY

     UP

• R63W PSY

     UP