IT07002: CDRSPOOL4ALT THREAD CAN DEADLOCK WITH CDRGFAN ON CDR_GC LOCK, BLOCKING ALTER TABLE AS WELL AS WHOLE ER

APAR status

• Closed as program error.

Error description

• An ALTER TABLE on a replicated table internally would create a
  CDRSpool4Alt thread for advancing the ER replay position to the
  position of the current transaction.  That thread would spool
  the send queue and/or sit waiting on the replay position to
  advance  -  all while the ALTER TABLE holds the CDR_GC lock.
  
  ALTER session:
  
  Stack for thread: 4444659 sqlexec
   base: 0x0000000177034000
    len:   135168
     pc: 0x000000010122fb70
    tos: 0x00000001770517f1
  state: join wait
     vp: 21
  
   yield_processor_mvp
   mt_join
   cdrInvokeForceSpool4AlterMode
   dac_altmode_on
   SetCDRAltMode          << has gcLock()'ed the GC (CDR_GC)
   cdrDDLSetAltMode
   cdrDDLAlterTablePreDDL
   cdrPreDDLCallBack
   ddalter
   aud_ddalter
   excommand
   sq_execute
   sqmain
   listen_verify
  
  Stack for thread: 4446885 CDRSpool4Alt
   base: 0x000000016620d000
    len:   69632
     pc: 0x000000010122fb70
    tos: 0x000000016621d2f1
  state: sleeping
     vp: 22
  
   yield_processor_mvp
   cdrSpoolSendQ4AlterMode
   th_init_initgls
   startup
  
  Now one of the threads involved in getting the replay position
  moved, the CDRGfan thread, can end up in the following activity
  -  having to wait on that very CDR_GC lock:
  
  Stack for thread: 145 CDRGfan
   base: 0x0000000112872000
    len:   69632
     pc: 0x000000010122fb70
    tos: 0x0000000112880571
  state: cond wait
     vp: 16
  
   yield_processor_mvp
   mt_wait
   gcLock
   grid_lockGC
   gauto_processRemasterRepl
   grid_processEROp
   grprF_processGridOps
   grouperFanOutThread
   cdrTrampolineThread
   th_init_initgls
   startup
  

Local fix

Problem summary

• ****************************************************************
  * USERS AFFECTED:                                              *
  * All users                                                    *
  ****************************************************************
  * PROBLEM DESCRIPTION:                                         *
  * See Error Description                                        *
  ****************************************************************
  * RECOMMENDATION:                                              *
  * Update to IBM Informix Server 12.10                          *
  ****************************************************************
  

Problem conclusion

• Problem Fixed In IBM Informix Server 12.10
  

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

  INFORMIX SERVER

• Fixed component ID

  5725A3900

Applicable component levels

• RA10 PSN

     UP

• RA10 PSY

     UP

• RB10 PSN

     UP

• RB10 PSY

     UP

• RB50 PSN

     UP

• RB50 PSY

     UP

• RB70 PSN

     UP

• RB70 PSY

     UP

• RC10 PSN

     UP

• RC10 PSY

     UP