IJ30673: LOGASSERTFAILED: ISSGPANICKED

APAR status

• Closed as program error.

Error description

• logAssertFailed: isSGPanicked
  return code 0, reason code 0, log record tag 0
   *** Assert exp(isSGPanicked) in line 837 of file
  /project/spreltac503/build/rtac503s003a/src/avs/fs/mmfs/t
  s/fs/bufdesc.C
   *** Traceback:
  ?2:0x555555D99AF8 logAssertFailed + 0x418 at ??:0
  ?3:0x55555597C4AA BufferDesc::cleanBuffer() + 0xA9A at
  ??:0
  ?4:0x555555993A64 BufferDesc::recycleBufferDesc(int,
  stealState**) + 0x574 at ??:0
  ?5:0x555555994098 StealBufferDesc(StripeGroup*, int) +
  0x248 at ??:0
  ?6:0x555555B36C2D OpenFile::acquireBuffer(long long,
  BufferDesc**, LkObj::LockModeEnum*, int) + 0xBED at ??:0
  ?7:0x555555A7C9FF FSOperation::acquireBuffer(long long,
  LkObj::LockModeEnum, OpenFile*, int) + 0x2AF at ??:0
  ?8:0x555555A05B42 OpenFile::fetchRandomBuffer(long long,
  LkObj::LockModeEnum, long long, int, int, FSOperation*,
  LkObj::LockModeEnum,BufferDesc**) + 0x1D2 at ??:0
  ?9:0x555555A0AFAA BufferHashTab::fetch(CacheObj*,
  unsigned short, LkObj::LockModeEnum, int, void*, int) +
  0x5AA at ??:0
  ?10:0x5555558B8D43 HandleMBHashFetch(MBHashFetchParms*) +
  0x153 at ??:0
  ?11:0x5555558B4F63 Mailbox::msgHandlerBody(void*) + 0x363
  at ??:0
  ?12:0x555555898733 Thread::callBody(Thread*) + 0x63 at
  ??:0
  ?13:0x555555885AE2 Thread::callBodyWrapper(Thread*) +
  0xA2 at ??:0
  ?14:0x7FFFF79C4DD5 start_thread + 0xC5 at ??:0
  ?15:0x7FFFF6AC902D __clone + 0x6D at ??:0
  
  Reported In:
  Spectrum Scale 5.0.3.3
  

Local fix

Problem summary

• When aioSyncDelay config is enabled,
  the buffer steal and the aio writes that need
  to be done as buffered I/O
  may race with each other and causes
  log assert  isSGPanicked in clearBuffer
  

Problem conclusion

• Benefits of the solution:
  No more assert
  
  Work Around:
  mmchconfig aioSyncDelay=0 to disable it
  
  Problem trigger:
  enable aioSyncDelay config and do aio writes
  
  Symptom:
  Abend/Crash
  
  Platforms affected:
  ALL Linux OS environments
  
  Functional Area affected:
  All Scale Users
  
  Customer Impact:
  High Importance
  

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:

  IJ31062

Fix information

• Fixed component name

  SPEC SCALE ADV

• Fixed component ID

  5737F35AP

Applicable component levels