[Linux-HA] status check in resource script for vmware

[Jure Pečar]

Hi all,

I'm using drbd and heartbeat to create a simple two-node failover cluster that runs a vmware virtual machine. 

I have a question about the resource script.

It's simple to create one to start and stop a virtual machine, however I'd like the status check to test status of the guest OS, not just the state of virtual machine. "vmware-cmd" offers a few tests, such as getstate, getuptime, getheartbeat and gettoolslastactive, which all return the same regardless if the linux guest is running or if I force it to panic.

I know this is more a vmware question ... but has anyone come up with a more elegant way than setting up a simple shell script "daemon" under xinetd in guest that would return something that host would periodically check?

-- 

Jure Pečar
http://jure.pecar.org