On Thu, Oct 15, 2009 at 9:22 AM, Ronny Tjoa <[email protected]> wrote:
> Hello,
>
> I am new to Pacemaker and clustering. I manage to setup 3 nodes (VMWare
> ESXi) monitoring VIP, Asterisk and SIP messages. I would like to reboot the
> node with fail resource after switching to another node.
>
> Stonith seems to be the one that I need. However, there is no
> information/example about external/vmware on the documentation and forum. I
> found a recent post by Thomas Glanzmaan and his vsphere script for VMWARE
> ESX. I cannot find any further information/example on this script. Has
> anyone use this before? Can you please share information on how to use it?
> Are there any changes required in the VMWare ESXi server?

There is also external/vmware which I wrote a long time ago but
haven't used much recently.

>From memory you needed to define
   hostmap="name1=cfgfile1 name2=cfgfile2..."
and
   device_host="host where the vmware server is running"

also guests needed to be able to (non-interactively) ssh (as root) to
the vmware host.

Not totally ideal, but it worked and didnt require modification of the
vmware server (just make sure the scripting tools are installed)


Hope that helps.

_______________________________________________
Pacemaker mailing list
[email protected]
http://oss.clusterlabs.org/mailman/listinfo/pacemaker

Reply via email to