On Tue, Dec 1, 2009 at 11:12 AM, Ivan Gromov <[email protected]> wrote:
> Hello,
>
> I have a situation in which a hb agent failed in the stop method. In
> this case heartbeat calls STONITH reset. How frequently will heartbeat
> call STONITH if the first calling takes too long?

Once per timeout plus a bit of processing time.

> Where may this time be configured?

If you have a recent version, you can use the stonith-timeout option
in the crm_config section.
Run
   /usr/lib/heartbeat/pengine metadata
to see if your version supports the option.

> For instance, I have configuration like this:
>
> <clone id="DoFencing">
>  <instance_attributes>
>    <attributes>
>      <nvpair name="clone_max" value="2"/>
>      <nvpair name="clone_node_max" value="1"/>
>    </attributes>
>  </instance_attributes>
>  <primitive id="child_DoFencing" class="stonith"
> type="external/yourplugin" provider="heartbeat">
>    <operations>
>      <op name="monitor" interval="20s" timeout="60s"
> id="DoFencing-monitor"/>
>      <op name="start" timeout="20s" prereq="nothing"/>
>    </operations>
>    <instance_attributes>
>      <attributes>
>        <nvpair name="parm1-name" value="parm1-value"/>
>        <nvpair name="parm2-name" value="parm2-value"/>
>        <!-- ... -->
>      </attributes>
>    </instance_attributes>
>  </primitive>
> </clone>
>
> Monitor have interval=20s and timeout=60s. Do these parametrs affect the
> STONITH behaviour. Does it mean that STONITH agent is called with reset
> every 20 seconds?
>
> Best wishes,
> Ivan
>
> _______________________________________________
> Linux-HA mailing list
> [email protected]
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems
>
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to