Would ZFS and gmirror work well together in a two-node failover cluster?

[Maurice Volaski]

>My findings have been that ZFS and ggate[cd] do *not* play nicely

I am not that surprised. Gmirror doesn't look like it was designed to 
be as integrated as dedicated remote mirroring solutions such as 
Linux drbd or AVS (Sun's "drbd") are. And it's still not clear 
whether gmirror could be configured to stay out of way should disk 
errors occur.

I've been setting up a virtual system in OpenSolaris and it's 
starting to look like my Linux-based drbd system.

I do hope FreeBSD fixes the outstanding issues and also gives these 
the scheme more functionality so it works like drbd. FreeBSD ought to 
have something like that. :-)

>As was suggested in other followups, it would seem that zfs send/recv
>may be a viable option depending on whethere it is granular enough
>(timewise) to be practical.
>

The post at 
http://mail.opensolaris.org/pipermail/zfs-discuss/2008-January/045168.html 
highlights an important difference between zfs send/recv and an 
integrated remote mirroring solution like AVS.

The more I read about AVS, the more it sounds like drbd. For example, 
I can monitor it as it keeps things in sync. I guess I've been 
spoiled by drbd for all these years and that includes a fair share of 
unplanned failovers.

Thanks for the advice.
-- 

Maurice Volaski, mvolaski at aecom.yu.edu
Computing Support, Rose F. Kennedy Center
Albert Einstein College of Medicine of Yeshiva University