[prev in list] [next in list] [prev in thread] [next in thread] 

List:       opensolaris-nfs-discuss
Subject:    [nfs-discuss] Consistent minor IDs of MPxIO-based LUNs for HA-NFS
From:       Thomas Blank <thomas.blank () tds ! de>
Date:       2009-12-24 2:48:14
Message-ID: 1191226922.21261651724289.JavaMail.Twebapp () sf-app1
[Download RAW message or body]

Hi all,
we have an issue with MPxIO-based LUNs used for HA-NFS with Veritas Cluster.

To enable clients mounting an NFS-shared filesystem a transparent failover \
(reconnect), it's necessary to have consistent major- and minor-IDs on all nodes in \
the cluster which are sharing the filesystem. It's easy to ensure consistent \
major-IDs of the ssd driver, but it seems there's no way to affect the way the \
minor-IDs are created.

For example, we have a two-node cluster with a service group sharing a filesystem \
which resides on SAN. Both nodes have access to this device and are connected to it \
via MPxIO. We neither use veritas nor solaris volume manager - and there's no way to \
use it!

If this LUN is the first and only LUN, it is highly propable that the device on both \
nodes have the same minor ID. But what if one node has more LUNs attached than the \
other? Minor IDs will differ and in case of failover, clients to not reconnect and \
report a "stale nfs handle".

I know, using veritas or solaris volume manager would solve this issue. But I do not \
want to use this - additionally, sds does not work with vcs...

Does anyone have a solution for this?
Any comments or suggestions are welcome!

Merry christmas to all!
Thanks, Thomas
-- 
This message posted from opensolaris.org
_______________________________________________
nfs-discuss mailing list
nfs-discuss@opensolaris.org


[prev in list] [next in list] [prev in thread] [next in thread]