> Hi Eric,
>
> I took out one RAID 5 array, remains 894GB. mongo runs w/o any problems
> so far. That >1TB is causing the problem seem to be more and more
> likely.
Since Eric is sat 2 feet from me at the moment, I will reply for him...
This is good (from the point of view of xfs, not from the point of view
of the 3ware drivers). Although come to think of it, the only thing which
is bigger than 1 Tbyte in your configuration is the md device - perhaps
md actually has some issues with 1 Tbyte. In any case, it may be worth
describing the scenario on linux kernel and seeing what people think.
>
> Once it's finished, I'll run mongo with ext2 instead of xfs over the
> 1.2TB md0 device. If the 3ware driver has a problem with file systems
> over 1TB, it *should* show faulty behavior with ext2fs, too. Or am I
> wrong?
Yes, but the detection process may need some work - ext2 will probably
have radically different behavior on an error. It may throw up on the
filesystem, or you may have to run fsck afterwards to see what happened
to the fs.
Steve
>
> Greetz
> Marc
>
> Eric Sandeen wrote:
> >
> > Hi again Marc -
> >
> > Can you try chopping this down to under 1TB? Martin suggested that
> > there may be a sign problem in the 3ware driver; perhaps >1TB is causing
> > the trouble...
> >
> > -Eric
> >
> > On Mon, 2001-10-22 at 07:39, Marc Schmitt wrote:
> > > Hello all,
> > >
> > > I`m currently setting up a 1.2TB file server and consider using XFS. The
> > > configuration is:
> >
> > --
> > Eric Sandeen XFS for Linux http://oss.sgi.com/projects/xfs
> > sandeen@xxxxxxx SGI, Inc.
|