Re: Suspected bug - System slowdown under unexplained excessive disk I/O - 2.4.13

Nikita Danilov (Nikita@Namesys.COM)
Thu, 8 Nov 2001 15:15:43 +0300


Oktay Akbal writes:
> I just did a quick retest with sql-bench on my machine and found that
> reiserfs seems to be a source for that problem.
> I did not reproduce the whole problem, since i stopped
> test-create before the system was to unresponsive.
>
> But I started test-create on a partition with
> reiserfs. After much files were created the system to very
> long for bash command-completion or doing simple thing
> free, top etc. this was even after test-create was stopped.
> It took about 30 seconds till the system responded normally.
>
> Same test with ext2 on the same partition showed no problems.
>
> Reran the test with reiserfs and the system got unresponsive
> again.

It doesn't seem that you have the same problem as Sasha. Reiserfs has
known problem with dealing with large holes, so may be this explains
situation. We'll try to reproduce your results.

>
> Did you use reiserfs ?
>
> Oktay
>

Nikita.

>
> On Thu, 8 Nov 2001, Oktay Akbal wrote:
>
> > On Wed, 7 Nov 2001, Sasha Pachev wrote:
> >
> > > Summary:
> > >
> > > System slowdown under unexplained excessive disk I/O
> > >
> > > Full description:
> > >
> > > While running X, KDE, having a few windows open, I ran make -j4 on MySQL
> > > source tree. I do this all the time and it usually works just fine - the
> > > system is a little bit unresponsive. However, occasionally the system becomes
> > > completely unresponsive - the disk goes crazy, the machine pings but neither
> > > ssh or telnet work - connection to the port is established, but nothing
> > > further than that. It does respond to magic SysRQ. I was able to get a memory
> > > info dump + stack traces into syslog, included below. The filesystem is
> > > ReiserFS.
>
-
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/