[OpenAFS] Solaris servers and >1TB disks

[Alf Wachsmann]

On Tue, 30 Mar 2004, Stephen Joyce wrote:
> On Mon, 29 Mar 2004, Jeffrey Hutzelman wrote:
>
> > Not quite.  A >1TB filesystem as a vice partition won't work, because vfsck
> > doesn't understand the modified filesystem structure required to support
> > this.  We're working with Sun on getting this situation resolved.
> >
> > But what Stephen is asking about is several smaller filesystems on a >1TB
> > _disk_ (i.e. a RAID array).  I see no reason why that shouldn't work; none
> > of our code (including vfsck) knows anything about disk labels.
>
> Jeffrey & Alf, thanks for the replies.
>
> Just to recap: do I understand correctly that a >1TB disk is OK as long
> as the separate /vice partitions are reasonable (read < 1TB)?

That is correct. I misread your original request.


> I notice that solaris, by default, turns logging on for filesystems
> (partitions) larger than 1TB.  If that is the only problem, would it
> suffice to simply specify "nologging" when mounting the /vice
> partitions?...  Or is there some other issue with the filesystem on such
> large partitions?  (note, I'm not saying it's wise to potentially need to
> fsck such a large partition; namei on ufs with logging is probably wisest).

There shouldn't be any other problem with these large partitions.
(Disclaimer: We haven't tried this at SLAC yet - our largest partitions
are ~810GB).

-- Alf.

-----------------------------------------------------------------------
  Alf Wachsmann                       | e-mail: alfw@slac.stanford.edu
  SLAC Computing Service              | Phone:  +1-650-926-4802
  2575 Sand Hill Road, M/S 97         | FAX:    +1-650-926-3329
  Menlo Park, CA 94025, USA           | Office: Bldg. 50/323
-----------------------------------------------------------------------
                http://www.slac.stanford.edu/~alfw (PGP)
-----------------------------------------------------------------------