On 2005-11-22T12:32:14, Andrew Beekhof <[EMAIL PROTECTED]> wrote:
> The only sane interpretation is therefore to treat the resource failed.
Right. However, we need to recover from as many things as possible,
gracefully.
> Also, for the record, option A is not allowed as it would
> bypass/conflict with the "multiple active" recovery policies
> configured in the CIB and carried out in the PE.
BTW, this is not true. "A" in particular referred to the case where,
after a successful start, a "monitor" returned an unexpected "stopped"
status code. Treating this as a failed resource (and thus doing a "stop"
first) is absolutely harmless and might help catch corner cases.
Sincerely,
Lars Marowsky-Brée <[EMAIL PROTECTED]>
--
High Availability & Clustering
SUSE Labs, Research and Development
SUSE LINUX Products GmbH - A Novell Business -- Charles Darwin
"Ignorance more frequently begets confidence than does knowledge"
_______________________________________________________
Linux-HA-Dev: [email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev
Home Page: http://linux-ha.org/
