HA-LVM vs CLVM

[Adam Miller]

If you are going to have to mount it manually on the passive node when
a failure occurs, then why even worry about clustering? Why not just
have an ext3 volume on some shared storage?

You won't have to worry about locking and/or LWW because only one
machine will have it mounted at a time.

(This does sound like it needs to be a full blown RHCS deployment to
provide any realistic HA, but this ext3 and manual fail over should
"work")

-AdamM


-- 
http://maxamillion.googlepages.com
---------------------------------------------------------
()  ascii ribbon campaign - against html e-mail
/\  www.asciiribbon.org   - against proprietary attachments