[Linux-cluster] Fencing: Prevent rebooting halted node

[Kaloyan Kovachev]

Hi,
On Wed, 21 Mar 2012 11:43:26 +0100, Nicolas Ecarnot <nicolas at ecarnot.net>
wrote:
> Hi,
> 
> We are setting up a new cluster and we still have tests and questions.
> At present, our cluster is two nodes only, with a very simple setup.
> fencing is done with fence_ipmilan, and the only action we do is
rebooting.
> Today, I tried to completely switch both nodes off, then boot up node 1.
> It perfectly boots up and serves as it should.
> But detecting the missing one, fencing is ran on node 2 and boots it up.
> 
> I would like to avoid that, and keep the stopped nodes stopped.
> 

if method=cycle does not help, you may create a wrapper script that will
first check the status and if it is OFF simply return success or if it is
ON to call fence_ipmilan with reboot

> I don't know if there's a way I could improve my cluster.conf to do
that?
> Either improve my fencedevice command, but I did not find many more 
> option in the fence_ipmilan man page...
> Either there's a way to first do a test (?) before doing any further
> action?
> 
> I'd be glad to read your advice.
> 
> My setup :
> 
> [...]
> 
> <clusternodes>
>   <clusternode name="node1" nodeid="1">
>    <fence>
>     <method name="1">
>      <device name="node1_ipmi" action="reboot"/>
>     </method>
>    </fence>
>   </clusternode>
>   <clusternode name="node2" nodeid="2">
>    <fence>
>     <method name="1">
>      <device name="node2_ipmi" action="reboot"/>
>     </method>
>    </fence>
>   </clusternode>
> </clusternodes>
> 
> <fencedevices>
>   <fencedevice name="node1_ipmi" agent="fence_ipmilan" ipaddr="c-node1" 
> login="alogin" passwd="apwd"/>
>   <fencedevice name="node2_ipmi" agent="fence_ipmilan" ipaddr="c-node2" 
> login="alogin" passwd="apwd"/>
> </fencedevices>
> 
> [...]
> 
> Regards,