IJ33000: AFM: DEADLOCK AFTER ABNORMAL EXIT OF AFM EVICT

APAR status

• Closed as program error.

Error description

• After an abnormal exit (crash or killed) of AFM evict,
  Spectrum Scale might detect deadlock as below:
  
  Waiting 41491.0316 sec since 23:19:16, monitored, thread
  954231 dmRequestRightHandlerThread: on ThCond
  0x183214B2370 (LkObjCondvar), reason 'waiting for XW
  lock'
  
  Reported In:
  Spectrum Scale 5.1.0.1
  

Local fix

• Restart Spectrum Scale on gateway node
  

Problem summary

• In the current implementation of Eviction on a file, the
   eviction program acquires a DMAPI lock first on the file
   first and punches a hole on it.
  The program can be terminated at any point without the
   DMAPI lock to be released - causing a lock leak and hence
   later DMAPI lock acquire on the file can deadlock and the
   only way to come out of this is to bounce the mmfsd.
  

Problem conclusion

• The problem is fixed in 5.0.5 PTF8.
  Benefits of the solution:
  No more deadlock
  
  Work around:
  None
  
  Problem trigger:
  Try to evict a file or list of files, and the eviction
   getting killed midway through.
  
  Symptom:
  Deadlock
  
  Platforms affected:
  ALL Linux and AIX OS environments
  
  Functional Area affected:
  AFM
  
  Customer Impact:
  High Importance
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

  IJ32512

• APAR is sysrouted TO one or more of the following:

Fix information

• Fixed component name

  SPEC SCALE STD

• Fixed component ID

  5737F33AP

Applicable component levels