Re: Bugs with mixed Zimbra 7/Zimbra 8 cluster

[Vladimir Gabrielescu <vgabriel@oit.rutgers.edu>]

Sure, we actually use NFS and that is part of our failover plan, mount the file systems of the failed host on a different host, change IPs, come right back up. We've done most of our upgrades this way over the years, bring up new server, turn off zimbra on the old one, mount file systems, restart on the new one.

On Jun 13, 2014, at 10:34 PM, Rich Graves <rgraves@carleton.edu> wrote:

>> Have you considered doing an temp in house install and migration to centos 6, then upgrade to v8, then make the move to outside Ubuntu ?
> 
> Yes, but that's a lot of work and risk I'm trying to avoid.
> 
> Also, our VMWare admin refuses to consider connecting the existing volumes as raw devices. Can I report that some of you have done that? (See: zero help from other local staff.)