IJ32159: LONG WAITER ON FILEALLOCATORMUTEX DUE TO SLOW BLOCK ALLOCATION OF METADATA.

APAR status

• Closed as program error.

Error description

• Full block allocation of metadata could become
  slower when block allocation
  regions become unbalanced over time and more
  regions contain only free space for fragments.   This
  in turn could cause long waiters on fileAllocatorMutex.
  

Local fix

• Add more disks to the system pool.
  

Problem summary

• Full block allocation of metadata could become
  slower when block allocation
  regions become unbalanced over time and more
  regions contain only free space for fragments.   This
  in turn could cause long waiters on fileAllocatorMutex.
  

Problem conclusion

• Benefits of the solution:
  Avoid slow down of block allocation
  for metadata block when all disks in
  system pool can contain both data and metadata.
  
  Work Around:
  Add more disks to the system pool.
  
  Problem trigger:
  Operations requiring allocation of full metadata blocks.
  Examples:
  Expand number of allocated inode
  Create new independent fileset
  
  Symptom:
  Performance Impact/Degradation
  
  Platforms affected:
  ALL Operating System 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:

  IJ32796

Fix information

• Fixed component name

  SPEC SCALE STD

• Fixed component ID

  5737F33AP

Applicable component levels