Re: Two node storage failover with lvm and ISCSI
Justin Jereza put forth on 2/18/2011 6:37 AM:
>> we have two nodes connected to one big SAS storage (LSI 630j Jbod) with
>> SAS HBAs and they can see all disks at same time.
>> Now we want build a failover construct for lvm with ISCSI:
>>
>> LSI Jbod -> node* | raid | lvm | ISCSI -> Global IP ->> Client
>>
>> If the primary node fails, start raid on node two, activate lvm, export
>> them via ISCSI, take over the IP address.
>>
>> There are many layers, that can fails, like raid, lvm, timing ....
>>
>> any suggestions?
>
> Bugger. I always fail at replying to mailing lists using GMail. Repost.
>
> I'd consider running clvm + gfs2 instead. That way, both nodes can
> stay up and connected to the same filesystem at the same time. The
> only decision left would be which node to use. OTOH, you can have an
> HA configuration as well.
I recommend drbd + gfs2 instead of clvm.
http://www.drbd.org
--
Stan
Reply to: