[Pacemaker] automatic reset of failcount?

[Andrew Beekhof]

On Thu, Jan 21, 2010 at 1:17 PM, Schaefer, Diane E
<diane.schaefer at unisys.com> wrote:
> Hi,
>
>
>
> We are using Pacemaker 1.0.6 and heartbeat 3.0.  We’d like to know if there
> is a way to configure the failcount for a resource to automatically reset.
> We have several resources that have a migration-threshold in a two node
> cluster.  Once the threshold is met, the resource migrates to the other node
> as expected.  We have configured the failure-timeout and
> start-failure-is-fatal to false.  After the failure-timeout expires the
> resource can now go back to the other node.  If the threshold on the second
> node is reached, the resource migrates back.  After this point, since the
> failcount > migration-threshold any single failure causes the resource to
> switch.

There's no built-in way to do this, we haven't found a nice solution.
cron + script is probably your best bet for now