[prev in list] [next in list] [prev in thread] [next in thread] 

List:       veritas-vx
Subject:    Re: [Veritas-vx] filesystem corruption when migrating RAID5 subdisks
From:       Doug Hughes <doug () eng ! auburn ! edu>
Date:       2004-02-27 1:41:08
Message-ID: Pine.GSO.4.44.0402261937390.19645-100000 () goodall ! eng ! auburn ! edu
[Download RAW message or body]

On Thu, 26 Feb 2004, Mark Sandrock wrote:

> On Thu, Feb 26, 2004 at 01:54:42PM -0800, Darren Dunham wrote:
> > Mark Sandrock wrote:
> > > I don't see anything that I did was wrong, and since this is
> > > the second instance of corruption on this filesystem within
> > > the past 6 months, (the other time I did a relayout of the
> > > same RAID5 f/s), I'm suspecting a problem with the Veritas
> > > installation on that host.
> > >
> > > Is there any reason not to remove and reinstall VxVM then?
> > > It's running 3.1.1 --
> >
> > you should certainly make sure you have any available patches installed,
> > but I don't know that you've said anything that would make me think that
> > corruption should have occurred.
>
> We install patches every 90 days. But I no longer trust this
> VxVM installation. Imo a relayout operation should _never_
> cause corruption on a filesystem--and that's what happened
> the last time.

do you specifically have the vxvm patches? They (3.1.1 p4 is
downloadable from veritas.com)
That's, I think, the most important thing to note. You are right,
a relayout should never cause corruption. Did you try a parity
verify?

_______________________________________________
Veritas-vx maillist  -  Veritas-vx@mailman.eng.auburn.edu
http://mailman.eng.auburn.edu/mailman/listinfo/veritas-vx
[prev in list] [next in list] [prev in thread] [next in thread]