Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 7 May 2003 13:51:52 -0500
From:      Glenn Johnson <gjohnson@srrc.ars.usda.gov>
To:        John Baldwin <jhb@FreeBSD.org>
Cc:        Doug Barton <DougB@FreeBSD.org>
Subject:   Re: recent performance problems?
Message-ID:  <20030507185152.GA39911@node1.cluster.srrc.usda.gov>
In-Reply-To: <XFMail.20030507140236.jhb@FreeBSD.org>
References:  <20030507173548.GA24421@node1.cluster.srrc.usda.gov> <XFMail.20030507140236.jhb@FreeBSD.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, May 07, 2003 at 02:02:36PM -0400, John Baldwin wrote:

> On 07-May-2003 Glenn Johnson wrote:
>
> > On Tue, May 06, 2003 at 10:22:20PM -0700, Doug Barton wrote:
> >
> >> On Wed, 7 May 2003, W. Josephson wrote:
> >>
> >> > On Tue, May 06, 2003 at 09:40:19PM -0700, Doug Barton wrote:
> >> >
> >> > > On Wed, 7 May 2003, W. Josephson wrote:
> >> > >
> >> > > > On Tue, May 06, 2003 at 08:51:09PM -0700, Doug Barton wrote:
> >> > > >
> >> > > > > Do the recommendations in UPDATING help you guys at all?
> >> > > >
> >> > > > I'm not quite sure what you are after
> >> > >
> >> > > Did you actually read /usr/src/UPDATING?
> >> >
> >> > yes.  Do you actually care about having feedback on -CURRENT?
> >>
> >> Yes we do. That's why I asked. Which of the options suggested in
> >> UPDATING did you try, and what were the results?
> >
> > In my case I have all of the debugging options off as per UPDATING.
> > I think I have a system with a different twist though.  Here is the
> > uname output:
> >
> > FreeBSD 5.1-BETA #2: Tue May 6 10:38:29 CDT 2003                    
> > root@node1.cluster.srrc.usda.gov:/usr/obj/usr/src/sys/CLUSTER-FW    
> >
> > When I first boot the system up it is fine.  However, it slows
> > down after a period of time.  Within a few hours it is essentially
> > unusable.  It does not seem to be a gradual slowdown but I do
> > not know how to quantify that.  I did find the following in
> > /var/log/messages after the most recent slowdown:
> >
> > May 6 19:52:21 node1 kernel: swap_pager: indefinite wait buffer:
> > device: twed0s1b, blkno: 408, size: 4096
> >
> > The system had been up for about four and a half hours and the time
> > stamp above corresponds to when the slowdown occurred.  Now, I could
> > be having problems with the 3-Ware RAID card or this could be the
> > same thing that others are reporting.  Anyway, I thought I should
> > share that as it might provide a clue to someone.
>
> It sounds like perhaps a lock is being held across sleep or some such.
> Have you tried running the system with WITNESS and INVARIANTS turned
> on?

No I have not but I will.

-- 
Glenn Johnson
USDA, ARS, SRRC			 Phone: (504) 286-4252
New Orleans, LA 70124		e-mail: gjohnson@srrc.ars.usda.gov



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20030507185152.GA39911>