Bug#295657: EXT3 on RAID problems in all 2.6-smp
Hi all,
I am mailing to close this bug. Some time ago I found out what the real
cause of it all was: a faulty memory chip. At boot time it checked out
fine, and even during operations it faulted only in 5% of the cases or
so. Intense runs of memtest86+ revealed the problem.
I removed the chip and no more problems since.
Thanks to all that have tried to help out!
Cheers,
Jeroen
On Tuesday 31 May 2005 13:52, Jeroen van Disseldorp wrote:
> Well it happened again last weekend, with 2.6.11-1-686-smp this time:
>
> May 26 22:39:11 einstein kernel: EXT3-fs error (device dm-2):
> ext3_readdir: bad entry in directory #82935810: rec_len %% 4 != 0 -
> offset=0, inode=23147 May 26 22:39:11 einstein kernel: Aborting
> journal on device dm-2. May 26 22:39:12 einstein kernel: EXT3-fs
> error (device dm-2) in ext3_reserve_inode_write: Journal has aborted
> May 26 22:39:12 einstein kernel: EXT3-fs error (device dm-2) in
> ext3_dirty_inode: Journal has aborted May 26 22:39:12 einstein
> kernel: ext3_abort called.
> May 26 22:39:12 einstein kernel: EXT3-fs error (device dm-2):
> ext3_journal_start_sb: Detected aborted journal May 26 22:39:12
> einstein kernel: Remounting filesystem read-only
>
> If I can provide any more info, please let me know.
>
> Jeroen
>
> On Thursday 19 May 2005 17:04, maximilian attems wrote:
> > severity 295657 important
> > stop
> >
> > On Thu, 19 May 2005, Erik Forsberg wrote:
> > > On Wed, 2005-05-18 at 22:44 +0200, maximilian attems wrote:
> > > > can you try to reproduce it with more recent 2.6.11 from
> > > > unstable?
> > >
> > > In my case, the machine in question is unfortunately running as a
> > > medium-sized mailserver. It's a rather bad idea to use it for
> > > file system trashing tests.
> >
> > agreed.
> >
> > > I will receive a P4 (HyperThreaded) machine at work in a few
> > > weeks. I might be able to do some quick tests on that hardware.
> > >
> > > Btw,
> > > http://marc.theaimsgroup.com/?l=linux-kernel&m=107913912311421&w=
> > >2 seems to be a relevant thread on Linux-kernel.
> >
> > thanks for the pointer,
> > unfortunately the thread ends without conclusion.
> >
> > so i can't help you atm.
> > good luck.
> >
> > --
> > maks
Reply to: