Date: Sun, 22 Jan 2006 15:26:01 -0800 From: "Kevin Oberman" <oberman@es.net> To: Kris Kennaway <kris@obsecurity.org> Cc: stable@freebsd.org, Tobias Roth <roth@iam.unibe.ch> Subject: Re: system laggy under load, SCHED_BSD Message-ID: <20060122232601.9386E4505B@ptavv.es.net> In-Reply-To: Your message of "Sun, 22 Jan 2006 17:25:33 EST." <20060122222532.GA62073@xor.obsecurity.org>
next in thread | previous in thread | raw e-mail | index | archive | help
> > --pWyiEgJYm5f9v55/ > Content-Type: text/plain; charset=us-ascii > Content-Disposition: inline > Content-Transfer-Encoding: quoted-printable > > On Sun, Jan 22, 2006 at 11:00:28PM +0100, Tobias Roth wrote: > > Hi > >=20 > > Since some time, I experience extremely bad system behaviour under > > load. For instance when compiling something, or unpacking a large > > archive, the system is almost unusable, even the mouse pointer > > reacts sloppy. > >=20 > > My system: > >=20 > > 6.0-STABLE FreeBSD 6.0-STABLE #7: Sun Jan 22 15:53:39 CET 2006 i386 > >=20 > > I use SCHED_BSD. > >=20 > > How can I diagnose the problem, or provide more information? What > > factors might influence what I experience? I cannot give a more precise > > point in time as when this started, since I switched versions and > > systems a few times recently. > > This has been discussed a number of times: the short answer is to look > for interrupt storms (vmstat -i), interrupt sharing (also vmstat -i), > and in particular shared interrupts involving usb or other drivers > still under Giant. It has been, but I never received any response to my last message. I eliminated shared IRQs and have only the mouse and keyboard under giant. No improvement. I enabled mutes profiling and posted the results. I'll admit not have much expertise, but nothing obvious. FWIW, GIANT was not an issue according to the profile data. I'm afraid that my chance to work on this is probably at an end as I received a replacement for my trusty old Dell (a dual core AMD 64 4400+), and I doubt that the Dell will be around for more than a week or so. If there is something I can try to look at before then, I'd be happy to, but I am at a total loss as to what the problem might be. -- R. Kevin Oberman, Network Engineer Energy Sciences Network (ESnet) Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab) E-mail: oberman@es.net Phone: +1 510 486-8634
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060122232601.9386E4505B>