[SSI-users] Re: Problem with implementing Root disk failover using DRBD

[Andreas <ro...@co...>]

Kilian CAVALOTTI wrote:
> On Tuesday 06 July 2004 21:35, Jaideep Dharap wrote:
> 
>>Have you tried just normal drbd on the cluster yet. I would suggest to
>>try that before moving
>>to root-failover. It is fairly easy and quick to try it out (if you have
>>an partition that you can
>>use to mirror). All you need is the drbd module (which you should
>>already have) and a drbd
>>conf file in /etc/drbd.conf.
>>
>>I personally would feel much more comfortable to try and debug the root
>>failover knowing
>>that a non-root drbd failover is working on ES3.
> 
> 
> Hi all,
> 
> Still having the same problem as described here, I also tried to test 
> failover on a non-root disk. Drbd is working fine, but I wondered how to 
> test failover with such a non-root disk. Do I have to use the SSIfailover 
> script?
> What the process to simulate and test the failover in such a case?
> 
> Thanks in advance,
> Regards,
> 

Hello.

I think the easiest way would be if you use your non-Init-Node as 
primary and some other node (can be the Init-Node) as secondary. After 
shutting down the second node, the other should take over.

Andreas