[Linux-cluster] GFS

[Patrick]

Basically, Guys... has anyone done these types of setups with
100k+mailboxes / etc.

P


On Tue, 26 Oct 2004 16:41:29 -0400, Dascalu Dragos
<dascalu_dragos at bah.com> wrote:
> We are working on a similar scenario but adding mailman into the mix.
> The ideal outcome would be for multiple mailman/postfix servers to
> write archives, etc to the same centralized location on a SAN. After
> doing some tests this setup does not appear to be trivial. We ran into
> a similar problem when using NFS; if multiple machines write to the
> same file at the same time the file gets mangled as the machines cut
> each other off. With GFS we noticed that each machine has a 4k buffer
> window in which it writes its data. If a second process decides to
> start writing to the same file we noticed alternating writes to the
> file after 4k of data.
> 
> We would like multiple mailman machines to write their archives to a
> centralized space on a SAN. If a mail list message gets sent
> concurrently from both machines to the same list, the archives may get
> mangled. I would also like to know of any alternatives to make this
> work.
> 
> Dede.
> 
> 
> On Oct 26, 2004, at 4:14 PM, Patrick wrote:
> 
> > Hi,
> >
> > If i was going to use GFS on a mail system for a LOT of users, and i
> > was going to have multiple machines writing to the mailstore. I was
> > thinking to decrease the ammount of possible problems by splitting the
> > device up into multiple smaller fs's to lower the chance of the same
> > directory being hit / drive / lun etc.
> >
> > Anyone else have any ideas on what would be good things to take into
> > account ? It would be on FC - > SAN ( fast T600 ) and maildir with
> > postfix 2.x
> >
> > Comments anyone ?
> >
> > P
> 
> --
> Linux-cluster mailing list
> Linux-cluster at redhat.com
> http://www.redhat.com/mailman/listinfo/linux-cluster
> 
> 
> 


-- 
</N>