From: Sten Fredriksson (no email)
Date: Mon Apr 04 2005 - 04:48:09 EDT
On Apr 4, 2005 10:34 AM, Phil Brutsche <> wrote:
> Sten Fredriksson wrote:
> > Hi,
> >
> > I know that this has been up before but after searching I found a fix
> > that maybe have changed the thought on NFS as back end storage [1]
>
> [...]
>
> > If NFS Is a big no no (as it's almost always are by default) how
> > would I build a back end that is redundant/fail-over?
>
> While NFS may work under RHEL, there's still no guarantee that it will
> work correctly under other operating systems, or even other Linux
> distributions. Therefore I doubt the maintainers will update the FAQ.
>
> What some people do for fail-over is use some sort of heartbeat
> mechanism that will detect when the "master" is unavailable and cause
> the "slave" to take over the IP address (if one isn't using the MURDER
> aggregator), mount the volumes, etc.
>
> The volumes would be shared between multible machines using:
>
> a) a shared SCSI bus
> b) fiber channel SAN
> c) DRBD (http://www.drbd.org/)
>
> This will give you active/passive failover.
>
> While you could theoretically share the volumes between 2 (or more)
> computers directly for active/active failover, you run into many of the
> same problems as with NFS (mmap not working right over the cluster file
> system, etc). It would also require the use of the pre-alpha Cyrus IMAP
> 2.3 code.
Would it still be "big no no" if back ends store their mail on NFS mounted
storage but not sharing and use some sort of heartbeat (keepalived /
heatbeat etc) to take over the ip and mount up the storage. Or is NFS
even if not sharing mail storage is not supported and/or recommended at all?
DRBD (http://www.drbd.org/) looks interesting. Do anyone of you use it
and how does it work for you?
// Sten
--- Cyrus Home Page: http://asg.web.cmu.edu/cyrus Cyrus Wiki/FAQ: http://cyruswiki.andrew.cmu.edu List Archives/Info: http://asg.web.cmu.edu/cyrus/mailing-list.html
|
|
|