Originally posted by: vermor

This is PowerHA 7.1.2.  I need to update the EMC and I need to varyoff all the volume groups, including the caa disk.

I asked IBM for a process because I could not find anything related in the literature.     The process involves removing and restoring the Cluster from a snapshot.  I think that is a very drastic and time consuming process.

Is there anyone that has varyoff the respository disk for an update?

Originally posted by: cla

Hello,

call IBM Support and refer to

  IV54199 POWERHA: ADD NEW CLMGR FLAG TO ENABLE STOP/START CAA CLUSTER

Thanks, Claudio.

Originally posted by: POWERHAguy

Yes there is a way to do it. I believe there is also an EMC doc that states how exactly including the above referenced APAR to stop the CAA cluster. Overall IBM, AIX, HA support SHOULD have this information. This roughly the same state as IF the cluster repos disk was gone. The cluster can/should continue to operate normally however don't manage/make any cluster changes during this time period.

Originally posted by: RussellAdams

See the PowerHA Cookbook, sg247739.pdf page 225, or search for "STOP_CAA".

Essentially the new clmgr command now has the ability to stop a node and take down CAA, but it must be restarted via the clmgr command. Smit clstart, sync & verify, reboot, nothing else will configure CAA again.

Originally posted by: POWERHAguy

Thanks for the plug Russell, I put that in the book. Actually I used this thread as a reminder to make sure we documented it... : )

Shawn Bodily
Senior Consultant - Clear Technologies

PowerHAguy on Youtube and Twitter