Re: Errors and later panics in 2.6.0-test11.
From: Nathan Scott (nathans_at_sgi.com)
Date: Fri, 5 Dec 2003 14:20:23 +1100 To: "Kevin P. Fleming" <email@example.com>
On Wed, Dec 03, 2003 at 10:25:27AM -0700, Kevin P. Fleming wrote:
> Linus Torvalds wrote:
> >The ones I've seen seem to be raid-0, but Nathan (firstname.lastname@example.org)
> >reported problems in RAID-5 under load. I didn't decode the full oops on
> >that one, but it really looked like a stale "bi" bio that trapped on the
> >PAGE_ALLOC debug code.
> The problem I reported was also with RAID-5, and I have also found a
> problem similar to Nathan's (probably the same one) by just trying to
> run bonnie++ on an XFS filesystem on DM over RAID-5, even after
> formatting the XFS filesystem to forcibly align everything to RAID-5
> stripes (64K units).
FWIW, this doesn't align _everything_ (space allocations done
through the XFS allocator are influenced, which means "most")
-- log IO is still going to be sector aligned, as are any IOs
to the four XFS allocation group header metadata structures.
-- Nathan - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to email@example.com More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/