xfs
[Top] [All Lists]

Re: nfs/local performance with software raid5 and xfs and SMP

To: "linux-xfs@xxxxxxxxxxx" <linux-xfs@xxxxxxxxxxx>
Subject: Re: nfs/local performance with software raid5 and xfs and SMP
From: Nathan Straz <nstraz@xxxxxxx>
Date: Fri, 20 Jul 2001 07:33:45 -0500
In-reply-to: <3B57CCB9.4386DD23@xxxxxxxxxxxxxxxx>
Mail-followup-to: "linux-xfs@xxxxxxxxxxx" <linux-xfs@xxxxxxxxxxx>
References: <200107191812.f6JICsI32762@xxxxxxxxxxxxxxxxxxxx> <3B57CCB9.4386DD23@xxxxxxxxxxxxxxxx>
Sender: owner-linux-xfs@xxxxxxxxxxx
User-agent: Mutt/1.3.18i
On Fri, Jul 20, 2001 at 08:16:25AM +0200, Simon Matter wrote:
> Steve Lord schrieb:
> > Jani, this may also be the source of your problems, we did experience
> > an almost complete lockup here with the 1.0.1 rpm kernel. If you get a
> 
> I'd like to reproduce this 'almost complete lockup' here, how can I do
> this. I mean do I need SMP to reproduce this and what kind operation
> did you do on the filesystem to force the lookup? Are we safe on UP?

The system was a 2p PIII 500 running RH2.4.3-XFS.  I had 7 disks in a
RAID5 configuration using left-symmetric for the parity algorithm.  I
believe the RAID was completely synced.  The RAID was exported over NFS.

I was running the `rm -r` stage of the untar-remove test over NFS.
Steve started the untar stage locally in a separate directory on the
RAID.  

-- 
Nate Straz                                              nstraz@xxxxxxx
sgi, inc                                           http://www.sgi.com/
Linux Test Project                                  http://ltp.sf.net/


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