Attached is a resource agent that I call exportfs.

Rather than starting/stopping NFS, it uses exportfs to add/remove
individual exports.

It also takes care to use cluster-wide unique fsid parameters for each
export. It ensures that this fsid is migrated with the resource.

Also, it continuously backs up the /var/lib/nfs/rmtab file to be
transferred to the remote node on failover.

As my application does not use locking, I have not implemented or
tested anything to do with sm-notify and /var/lib/nfs/sm(.bak).
However, If someone can give me information on this, I would be happy
to implement it. Also, this is probably very CentOS specific as that
is what I implemented it for, I did not really put much thought into
portability as I wrote it for my specific use (servers going into
production soon).

More information is on my blog where I recorded the details as a
memory aid for myself :-).

http://ben.timby.com/?p=109

Feedback is encouraged, as I would love to see this RA get some use
beyond my own!

Also, I would like to tell the developers of Heartbeat/Pacemaker that
I am VERY impressed with the new versions of these projects. I have
been using Heartbeat for a very long time. I only had the pleasure of
using the 3 series of Heartbeat this week. Also, I have never used
Pacemaker before this week.

Great work, I had my cluster up and running in no time and the
problems I have had with Heartbeat 2 series are apparently gone!

Thanks.

Attachment: heartbeat-exportfs.sh
Description: Bourne shell script

_______________________________________________________
Linux-HA-Dev: [email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev
Home Page: http://linux-ha.org/

Reply via email to