[Pacemaker] [DRBD-user] Performing crm failover using crm node standby

[Vladislav Bogdanov]

Hi!

22.06.2011 14:44, Florian Haas wrote:
...
> 
> This is something that has been discussed here previously. A resource
> agent could report to Pacemaker during monitor (via an exit code named,
> say, OCF_ERR_DEGRADED) that a resource, or a resource instance in the
> case of a clone or master/slave set, is in a degraded state.

At least one use case which comes to mind immediately is a kind of
multipath RA, which may use OCF_ERR_DEGRADED to notify upper layers that
not all paths are currently running. This is not a resource failure,
because paths are configured, and degradation may be caused by a
communication layer or storage partial failure. Resource restart may or
may not help, so it is safer not to do anything with degraded resource,
just mark resource as degraded and wait. Dependent resources may be
migrated to healthy nodes if any.

Best,
Vladislav