Red Hat Bugzilla – Bug 472918
[RFE] - Documentation - A service is disabled if it runs out of nodes in a cluster / failover domain with which it can fail to.
Last modified: 2016-01-05 14:01:01 EST
Presently, the recovery policy for service failover is user selectable: relocate, restart, or disable. Should one of the first two options be selected, the service has the potential to move to another available node in the cluster (or within its failover domain if it is restricted to it).
However, should a service run out of nodes within a cluster (or failover domain) to fail to, the service is then disabled regardless of the recovery policy.
Upon investigation internally with several different sources, search engines, etc, this fact is not stated anywhere within our current documentation.
Can we have this policy documented in our cluster suite documentation?
Thank you. Yes. I will follow up and verify.
Actually, it's marked "stopped" if it runs out of nodes.
Cluster Suite is no longer supported.