IT48682: IBM STORAGE PROTECT SERVER TARGET SERVER CRASHES DURING STGRULE FOR REPLICATION, DUE TO GSK ISSUE

8.2.1.000-IBM-SPSRV-WindowsX64
8.2.1.000-IBM-SPSRV-Linuxx86_64
8.2.1.000-IBM-SPSRV-Linuxs390x
8.2.1.000-IBM-SPSRV-Linuxppc64le
8.2.1.000-IBM-SPSRV-AIX

APAR status

• Closed as program error.

Error description

• IBM Storage Protect Server target server crashes during stgrule
  for replication, stack points to GSK issue :
  
  Segmentation fault in . at 0x900000011f9cc5c ($t606)
  0x900000011f9cc5c (???) e8c60048 ld r6,0x48(r6)
  
  (dbx) where
  _dt_19GSKBufferAttributesFv(??, ??) at 0x900000011f9cc5c
  _as_13GSKFastBufferFRC13GSKFastBuffer(??, ??) at
  0x900000011fa0410
  Receive__14TLSV13ProtocolFPvib(??, ??, ??, ??) at
  0x9000000112b7228
  Receive__14TLSV13ProtocolFPvi(??, ??, ??) at 0x9000000112b7d10
  Receive__18SSLProtocolManagerFPvi(??, ??, ??) at
  0x9000000112bb6c8
  gsk_secure_soc_read(??, ??, ??, ??) at 0x9000000114dad64
  tlsRecv(??, ??, ??) at 0x1010ffa08
  IPRA.$ReceiveVerb(??, ??) at 0x1005962b0
  SmRecvVerbX(??) at 0x10058d1f8
  SmReplServerSession(??) at 0x100c1ed38
  IPRA.$DoReplServer(??, ??) at 0x1009845ec
  smExecuteSession(??, ??, ??, ??, ??, ??, ??, ??) at 0x100973480
  psSessionThread(??) at 0x10110542c
  StartThread(0x0) at 0x100011a6c
  
  (dbx) registers
  $r0:0x0000000000000001 $stkp:0x000000013a966ff0
  $toc:0x09001000a1206000
  $r3:0x000000012fcda3d0 $r4:0x0000000000000003
  $r5:0x0000000000000000
  $r6:0x0900100000000040 $r7:0x0000000000000000
  $r8:0x000000012fc713d8
  $r9:0x0000000012ce0877 $r10:0x0000000000000020
  $r11:0x0000000000000000
  $r12:0x0900000011fa03ac $r13:0x000000013a973800
  $r14:0x00000001025543e8
  $r15:0x0000000000000003 $r16:0x0000000000000000
  $r17:0x0000000101f51ea4
  $r18:0x00000001025643e8 $r19:0x0000000000000005
  $r20:0x0000000000000004
  $r21:0x0000000000000007 $r22:0x000000016394d9d0
  $r23:0x0000000000000004
  $r24:0x0000000000000001 $r25:0x0900000011a83278
  $r26:0x0000000000000000
  $r27:0x000000013a967bd8 $r28:0x0000000000000004
  $r29:0x0000000000000001
  $r30:0x0000000000000003 $r31:0x000000012fc713d0
  $iar:0x0900000011f9cc5c $msr:0x800000000200d032 $cr:0x44204244
  $link:0x0900000011fa0414 $ctr:0x090000000043be20 $xer:0x2004001a
  
  
  Condition status = 0:g 1:g 2:e 4:g 5:e 6:g 7:g
  [unset $noflregs to view floating point registers]
  [unset $novregs to view vector registers]
  [unset $novsregs to view vector scalar registers]
  in . at 0x900000011f9cc5c ($t606)
  0x900000011f9cc5c (???) e8c60048 ld r6,0x48(r6)
  
  Crash does not occur when running REPLICATE NODE instead of
  stgrule for replication
  
  IBM Storage Protect Server version Affected: IBM Storage Protect
  Server 8.1.25.x and later on AIX platforms.
  

Local fix

• Make these changes at the bottom of the rc.dsmserv file by
  adding new line for MALLOCBUCKETS
  and change multiheap. Multiheap should be set to optimal value
  of 16 instead of default 8 to avoid crash
  (Replace existing block with the following mentioned block) :
  
  if [ "$OS" = "AIX" ]; then
  If MALLOCOPTIONS isn't set by invoker, set it now
  if [ "$MALLOCOPTIONS" = "" ]; then
  MALLOCBUCKETS=number_of_buckets:128,bucket_sizing_factor:64,blo
  cks_per_bucket:1024
  MALLOCOPTIONS=multiheap:16
  fi
  LDR_CNTRL is 1 environment variable. This isn't 5 environment
  variables
  envVars="LDR_CNTRL=TEXTPSIZE=64K@DATAPSIZE=64K@STACKPSIZE=64K@S
  HMPSIZE=64K"
  envVars="${envVars} MALLOCBUCKETS=$MALLOCBUCKETS"
  envVars="${envVars} MALLOCOPTIONS=$MALLOCOPTIONS"
  if [ "$instanceUser" != "" ]; then
  exec su - $instanceUser $bashFlag "-c $envVars
  $serverBinDir/dsmserv $cmdParms"
  else
  exec /usr/bin/env $envVars $serverBinDir/dsmserv $cmdParms
  fi
  else
  if [ "$instanceUser" != "" ]; then
  exec su - $instanceUser -c "nohup $serverBinDir/dsmserv
  $cmdParms"
  else
  exec nohup $serverBinDir/dsmserv $cmdParms
  fi
  fi
  
  You will then have to halt and restart the SP server to pick up
  these changes.
  Please refer https://www.ibm.com/docs/en/aix/7.2.0?topic=subsys
  tem-malloc-multiheap
  

Problem summary

• ****************************************************************
  * USERS AFFECTED:                                              *
  * All IBM Storage Protect server users                         *
  ****************************************************************
  * PROBLEM DESCRIPTION:                                         *
  * See error description.                                       *
  ****************************************************************
  * RECOMMENDATION:                                              *
  * Apply fixing level when available. This problem is currently *
  * projected to be fixed in levels 8.2.1. Note that this is     *
  * subject to change at the discretion of IBM.                  *
  ****************************************************************
  

Problem conclusion

• This problem was fixed.
  Platforms fixed:  AIX
  

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