Re: Oops on 2.5.15 with ext3 & AIM 7 shared workload

Andrew Morton (akpm@zip.com.au)
Fri, 17 May 2002 12:52:03 -0700


Peter Rival wrote:
>
> I've recently built a 2x300MHz P2 system with 512MB of memory for a simple
> x86 testbed. To start with, I ran the AIM VII shared workload on the
> system, using the 18 SCSI disks that I have available with ext2
> filesystems. I got the following oops when I rebooted and started the run
> with ext3. This was only at the simulated '100 users' load point; the
> system is capable of ~1100 simulated users (i.e. the system was hardly
> heavily loaded). Does this look familiar to anyone, or is there
> somehting anyone needs me to do? Thanks!
>
> - Pete
>
> Assertion failure in journal_commit_transaction() at commit.c:523:

(When reporting ext3 problems, please specify the journalling
mode - it makes a rather large difference).

I'll assume it's data=ordered. If you're using data=journal
then please don't - it's not working very well in 2.5.15.

These patches:

http://www.zip.com.au/~akpm/linux/patches/2.5/2.5.15/ext3-no-steal.patch
http://www.zip.com.au/~akpm/linux/patches/2.5/2.5.15/sct-jbddirty.patch

Will likely fix it. Please let me know...

-
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/