From owner-freebsd-scsi Mon Mar 27 14:24:27 2000 Delivered-To: freebsd-scsi@freebsd.org Received: from panzer.kdm.org (panzer.kdm.org [216.160.178.169]) by hub.freebsd.org (Postfix) with ESMTP id 9368837B5F0 for ; Mon, 27 Mar 2000 14:24:23 -0800 (PST) (envelope-from ken@panzer.kdm.org) Received: (from ken@localhost) by panzer.kdm.org (8.9.3/8.9.1) id PAA07150; Mon, 27 Mar 2000 15:23:50 -0700 (MST) (envelope-from ken) Date: Mon, 27 Mar 2000 15:23:50 -0700 From: "Kenneth D. Merry" To: Matthew Hanselman Cc: freebsd-scsi@FreeBSD.ORG Subject: Re: Terrible SCSI performance with 4.0 -- please help Message-ID: <20000327152350.A7100@panzer.kdm.org> References: <200003272149.PAA14773@alumni.cs.wisc.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i In-Reply-To: <200003272149.PAA14773@alumni.cs.wisc.edu>; from mjhans@cs.wisc.edu on Mon, Mar 27, 2000 at 03:49:33PM -0600 Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org On Mon, Mar 27, 2000 at 15:49:33 -0600, Matthew Hanselman wrote: > I'm running 4.0-RELEASE on a K6-2 with an Adaptec 29160 and some IBM > DMVS18M drives (nice 10k disks), and am getting absolutely terrible > performance. My setup has a Quantum IDE drive in it as well, and I get > about 2X better read/write performance on the IDE drive than the SCSI > (the SCSI should be a bit faster than the IDE...) > > I've tried swapping in an Adaptec 2940 to no avail, so it's not the 29160 > specifically; I've tried moving the disks to another (Linux) box and the > drives work perfectly fine, so it's not the drives/cables. I don't know > what else to try, even what other info to send out as diagnostics. > > Apologies if this is not the right list; I'm running out of ideas, please > point me to the right list if nothing else. Thanks a lot. Well, it'll take a little more information to try to track down the problem. How about dmesg output, and benchmark numbers from bonnie and iozone. Also, is write caching enabled or disabled on those drives? You can find out with camcontrol, like this: camcontrol modepage da0 -m 8 will print out mode page 8 for da0. While you're doing the benchmarks, run top(1) to see if you're CPU bound. Also, are you getting any error messages? Ken -- Kenneth Merry ken@kdm.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message