[Linux-HA] 'monitor' action/operation on the resource

[Dejan Muhamedagic]

Hi,

On Wed, Oct 03, 2007 at 05:21:49PM +0300, Andrew W. Nosenko wrote:
> Some questions about 'monitor' operation on the resource (heartbeat-2.1.2).
> 
> 1. Resource is controlled by OCF-style Resource Agent.  Meta-data have
> 'monitor' action.  Whether I should to explicitly add 'monitor'
> operation to the CIB in some way (e.g. using hb_gui) for allow
> heartbeat to restart resource if it failed (e.g. resource-process is
> killed by signal)?

Yes, there's no monitoring by default.

> 2. Assuming that 'monitior' operation exists in CIB, but 'on_fail'
> attribute there is undefined (just absent in the "<op .../>" node.
> What things heartbeat will do in this case (restart resource? reboot
> node?  migrate resource to another node?  Something else?)

Either restart or move to another node, depending on other
settings.

Thanks,

Dejan

> -- 
> Andrew W. Nosenko <andrew.w.nosenko at gmail.com>
> _______________________________________________
> Linux-HA mailing list
> Linux-HA at lists.linux-ha.org
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems