IT00089: ALL MIGRATED VIRTUAL MACHINES MIGRATED TO THE SAME NEW NODE AFTER COMPUTE NODE POWER OFF.

IBM PureApplication System Version 1.1.0.2

APAR status

• Closed as program error.

Error description

• When a compute node is powered off the virtual machines on that
  node should be migrated across the remaining nodes in the cloud
  group to evenly balance the total load on the remaining compute
  nodes. In some cases this all virtual machines can be placed on
  the same compute node even when more than one compute node
  remains active in the cloud group.
  

Local fix

Problem summary

• When a compute node is evacuated, the management node would
  find the first available compute node left in the cloud group
  and place all evacuated virtual machines on that compute node,
  as long as it did not violate certain colocation constraints.
  Only after that compute node is fully utilized did it move on
  to another compute node. The system will rebalance the virtual
  machines across all remaing compute nodes in the cloud group
  after the initial evacuation.
  

Problem conclusion

• The placement is modified to improve colocation constraint
  processing, allowing virtual machines to spread more evenly
  across available compute nodes in the cloud group.
  

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

  PUREAPP SYS W15

• Fixed component ID

  5725G3200

Applicable component levels

• R110 PSY

     UP