IT26163: IN DPF , DB2 MAY CRASH WHEN TRYING TO UNLATCH A LATCH THAT IS NOT HELD

APAR status

• Closed as program error.

Error description

• db2diag.log reports following error:
  
  2018-07-31-16.43.57.183345+480 I333558187A369    LEVEL: Severe
  PID    : 15139118            TID : 1543       PROC : db2sysc 60
  INSTANCE: db2inst1            NODE : 060
  EDUID  : 1543                EDUNAME: db2fcmr 60
  FUNCTION: DB2 UDB, fast comm manager,
  sqkfChannel::DeliverInboundBuffer, probe:15
  MESSAGE : Invalid Sequence No. Detected = 1. Expected No. = 2
  
  2018-07-31-16.43.57.183328+480 I333558557A1565   LEVEL: Severe
  PID    : 42009450            TID : 1543       PROC : db2sysc 59
  INSTANCE: db2inst1            NODE : 059
  EDUID  : 1543                EDUNAME: db2fcmr 59
  FUNCTION: DB2 UDB, SQO Latch Tracing,
  sqlo_xlatch::releaseConflict, probe:10
  DATA #?1 : String, 27 bytes
  unlocking an unlatched lock
  DATA #?2 : Pointer, 8 bytes
  0x07800000b0dd4ca0
  DATA #?3 : String, 117 bytes
  {
    lock         = { 0x00000000 [ unlocked ] }
    identity     = sqkfRecvLockManager::sqkfRecvLockManager (103)
  }
  DATA #?4 : Hexdump, 8 bytes
  0x07800000B0DD4CA0 : 0000 0000 0067 0000
  .....g..
  CALLSTCK: (Static functions may not be resolved correctly, as
  they are resolved to the nearest symbol)
   [0] 0x090000001EB05C6C pdLog + 0xF8
   [1] 0x090000001B0A8904 pdLog@glue415 + 0x12C
   [2] 0x090000001B05321C sqloSpinLockReleaseConflict + 0x5C
   [3] 0x090000001E9162AC sqloSpinLockReleaseConflict@glue73 +
  0x78
   [4] 0x090000001BE64F34
  DeliverInboundBuffer__11sqkfChannelFP10sqkfBufferP17SQLKF_SESSIO
  N_HDLP18SQLZ_PDB_UNIQUE_IDP15sql_static_data + 0x4A4
   [5] 0x090000001B638BEC
  DeliverBufferToTargetChannel__19sqkfFastCommManagerFP10sqkfBuffe
  riN2217SQLKF_CHANNEL_PRIP17SQLKF_SESSION_HDLP15sqkfSendConduit +
  0x84
   [6] 0x090000001AA59B78
  RouteInboundBuffer__19sqkfFastCommManagerFRP10sqkfBufferP17SQLKF
  _SESSION_HDLiT3 + 0x6A0
   [7] 0x090000001BAC2D98 HandleDataEvent__15sqkfRecvConduitFUl +
  0xCA0
   [8] 0x090000001BAC1BC0 RunEDU__15sqkfRecvConduitFv + 0xEC0
   [9] 0x090000001E82D9BC EDUDriver__9sqzEDUObjFv + 0xE4
  
  In certain case, a FastCommManager claimed the channel and
  put it in its receive table and at same time the channel is
  pointing to another FastCommManager, a possible reason here is
  that somehow the channel is reused before it was actually
  closed. This may also come with double free of a channel.
  

Local fix

• Disable communication between MLN nodes via shared memory. In
  this case, they should set the reg var DB2_FORCE_FCM_BP to NO.
  When this is set, FCM resources are created per logical node and
  are not shared among other nodes on the same host.
  db2set DB2_FORCE_FCM_BP=NO
  recycle the instance
  

Problem summary

• ****************************************************************
  * USERS AFFECTED:                                              *
  * all                                                          *
  ****************************************************************
  * PROBLEM DESCRIPTION:                                         *
  * See Error Description                                        *
  ****************************************************************
  * RECOMMENDATION:                                              *
  * Upgrade to db2_v111m4fp6 or later                            *
  ****************************************************************
  

Problem conclusion

• Upgrade to db2_v111m4fp6 or later
  

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

  DB2 FOR LUW

• Fixed component ID

  DB2FORLUW

Applicable component levels