IV95162: HA:ROHA YANKING MAY FAIL WITH ERRORS DURING RG FAILOVER

PowerHA-7.2.1 SP3
PowerHA-7.2.1 SP1

APAR status

• Closed as program error.

Error description

• ROHA Yanking may fail with errors during RG failover
  
  Issuing halt -q command to trigger HA failover from
  active
  to standby nodes, but the standby node failed to acquire
  EPCoD resource to start resource group. The ROHALOG
  indicated that there was not enough Mobile CoD resource
  for the standby nodes when the standby node made a
  request
  to obtain 8 cores, there are 2 core available in PEP, and
  the rest of the resources (6 cores) should be yanked from
  the halted primary node. but PowerHA7.2.1 is not behaving
  like that.
  As my understanding all the iFix provided for
  PowerHA7.2.0
  is imported into PowerHA 7.2.1, so this must not be the
  same issue what found in PowerHA 7.2.0
  LPAR name and PowerHA node name are identical.
  
  In order to narrow down the issue: I asked the
  customer/partner to run the following test, but it didn't
   make any change:
  
      Changed the defaut SPP (default) use to dedicated
  core use.
      Change the resource allocatioi policy from
      "Enterprise Pool First" to "Free Pool First"
  
  Of cource, if there is enough Mobile resources which has
  not been assigned to any nodes, HA failover works as
  expected.
  

Local fix

• n/a
  

Problem summary

• When EPCoD resources are not available, Yanking may
  fail with errors during failover
  

Problem conclusion

• Modified code to resolve the failures during the
  RG failover.
  

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

  POWERHA SYSMIR

• Fixed component ID

  5765H3900

Applicable component levels