[Lustre-discuss] manual ost failover problems

[Andreas Dilger]

On Aug 21, 2008  18:29 +1000, Marcus Schull wrote:
> We are currently testing lustre 1.6.5.1 on RHEL 5 (64bit) with 3 OSTs  
> for a 'data' filesystem running on server1 and 4 OSTs for a 'common'  
> filesytem running on server2.  Each OST is a 1TB SAN LUN that can be  
> seen from either server.  The idea was to run the servers as an  
> active/active failover pair, being able to mount the 'other' LUNs on  
> the remaining server if one server failed.   Also, we could have the  
> flexibility of striping (between the 2 nodes initially -->  more in  
> the future), if the OSTs of each fs  were spread out amongst the  
> servers.
> 
> At present, this works well if all LUNs are only mounted on the  
> initial server they are mounted on after creation.
> 
> I had assumed that OSTs could be unmounted from server1 and then  
> remounted on then remounted on server2 (never simultaneously  
> mounted), but this does not seem to work whether or not clients are  
> using (have mounted) the file system, or even whether the servers are  
> rebooted in between the change.
> 
> Even though the LUNs will mount on the other server, any clients that  
> access the filesytem will 'hang' until the LUN is mounted back in its  
> initial location.
> 
> The filesystems were created using the --failnode option.

Odd, this is exactly what should work.  Do the clients report trying
to contact the backup server?

> Is there a command to 'update' the ?MGS/MDT's information regarding  
> this, and so communicate this to the clients?

No, the clients should know this from the configuration they got at
mount time, and try automatically with the backup server if the primary
is down.

Cheers, Andreas
--
Andreas Dilger
Sr. Staff Engineer, Lustre Group
Sun Microsystems of Canada, Inc.