Ciao,

On Thu, Jun 18, 2009 at 10:33:32AM +0200, Cristina Bulfon wrote:
> Ciao,
>
> I am still setting the HA configurazion, we have 2 node cluster in 
> active/passive mode.
>
> - to check the SAN storage device we use an external stonith device and in 
> case of SAN's  failure
>  on the active machine, it  should do a shutdown instead it makes a reboot.
>
>  In the cluster property section of the cib.xml I have
>
>          <nvpair id="cib-bootstrap-options-stonith-enabled" 
> name="stonith-enabled" value="true"/>
>            <nvpair name="stonith-action" 
> id="cib-bootstrap-options-stonith-action" value="poweroff"/>

The sbd does reboot for both "reset" and "off" requests. You can
open a bugzilla for this issue. But why don't you just prevent
heartbeat from starting at boot, shouldn't that be the same?

> - regarding the "failback OFF"
>
>    In the cluster property section of the cib.xml file I have
>
>            <nvpair name="default-resource-stickiness" 
> id="cib-bootstrap-options-default-resource-stickiness" value="INFINITY"/>
>            <nvpair 
> id="cib-bootstrap-options-default-resource-failure-stickiness" 
> name="default-resource-failure-stickiness" value="0"/>
>
>  The failback is working , when the active node is coming back the resource 
> still remains on the passive node , for migrating the resource
> I have to execute the following command
>
>               crm_resource -M -H <active_node> -r <group> -f  (it doesn't 
> migrate 
> without -f )
>                 crm_resource -U  -H <passive_node>
>
> If I repeat the exercise: simulate failure on the active node a couple of 
> times it happens that the passive_node don't take the resource.

Uh, that really shouldn't happen. Are you sure that there are no
location constraints for the resource with -INFINITY? Also, check
the fail counts. If you can reproduce it, please create hb_report
for the period of the incident.

> In attachment you will find the ouput of cibadmin -Q
>
> Thanks in advance for any help

Thanks,

Dejan

>
> cristina
>
>


>
>
>
>
>
>




> _______________________________________________
> 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