[Top] [All Lists]

Re: 2.6.9: unkillable processes during heavy IO

To: Brad Fitzpatrick <brad@xxxxxxxxx>
Subject: Re: 2.6.9: unkillable processes during heavy IO
From: Nathan Scott <nathans@xxxxxxx>
Date: Thu, 18 Nov 2004 16:05:01 +1100
Cc: Andrew Morton <akpm@xxxxxxxx>, linux-kernel@xxxxxxxxxxxxxxx, Lisa Phillips <lisa@xxxxxxxxx>, Mark Smith <marksmith@xxxxxxxxx>, linux-xfs@xxxxxxxxxxx
In-reply-to: <Pine.LNX.4.58.0411162251170.7904@xxxxxxxxx>
References: <Pine.LNX.4.58.0411141403040.22805@xxxxxxxxx> <Pine.LNX.4.58.0411160549070.7904@xxxxxxxxx> <20041116200156.2b2526e5.akpm@xxxxxxxx> <20041117045506.GA1802@frodo> <Pine.LNX.4.58.0411162251170.7904@xxxxxxxxx>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Mutt/1.5.3i
On Tue, Nov 16, 2004 at 11:00:41PM -0800, Brad Fitzpatrick wrote:
> > Brad, could you send me details on how you've setup mysqld
> > and how to generate a load similar to yours, so that I can
> > reproduce the hang locally?
> The MySQL people made a tool to reproduce MySQL-like load for the specific
> purpose of not putting you through the database setup pain:

Ah, fantastic.

I've reproduced it now with this tool and a simplified version
of your recipe - I'll have a fix for you to try before too long.

> Well, sysbench should help you find the problem.

Yep, it sure did.



<Prev in Thread] Current Thread [Next in Thread>