Wed, 17 Oct 2001 10:29:47 +0200
Bernard Frit wrote:
> Newby question.
> I'm currently running a test cluster as described in
> the Getting Started Document and everything is working
> Let's say :
> o smb is running on machine_A
> o http is running on machine_B
> o smb tree is mirrored via scsi on machine_B
> o http tree is mirrored via scsi on machine_A
> (cross mirroring)
> What I want to do :
> 1 if machine_A detects any problem on mirrored
> disk running on machine_B
> 2 then machine_A will send a message to machine_B
> "telling" her to stop heartbeat
> 3 machine_B will stop heartbeat and unmount
> remote mirrored volumes (http)
> 4 machine_A will mount localy the mirrored http
> and unmount remote mirrored volumes (smb)
> 5 machine_A will activate http service
> I don't know how to do #2
> Is there any simple solution ?
What about STONITH? This is absolutely necessary if you use shared disks
(which I assume you do - or what else do you mean with 'mirrored via
Also I would do the monitoring on machine_B via mon and tell mon to stop
the local Heartbeat if it detects problems with the disk.
If each of us have one object, and we exchange them,
then each of us still has one object.
If each of us have one idea, and we exchange them,
then each of us now has two ideas.