From owner-freebsd-scsi Wed Jul 17 11:36:39 2002 Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E163D37B407 for ; Wed, 17 Jul 2002 11:36:33 -0700 (PDT) Received: from odin.ac.hmc.edu (Odin.AC.HMC.Edu [134.173.32.75]) by mx1.FreeBSD.org (Postfix) with ESMTP id 61D9543E42 for ; Wed, 17 Jul 2002 11:36:33 -0700 (PDT) (envelope-from brdavis@odin.ac.hmc.edu) Received: from odin.ac.hmc.edu (IDENT:brdavis@localhost.localdomain [127.0.0.1]) by odin.ac.hmc.edu (8.12.3/8.12.3) with ESMTP id g6HIaVri032325; Wed, 17 Jul 2002 11:36:31 -0700 Received: (from brdavis@localhost) by odin.ac.hmc.edu (8.12.3/8.12.3/Submit) id g6HIaVBQ032323; Wed, 17 Jul 2002 11:36:31 -0700 Date: Wed, 17 Jul 2002 11:36:31 -0700 From: Brooks Davis To: Dave Hayes Cc: Andrew Heybey , freebsd-scsi@FreeBSD.ORG Subject: Re: Random hangs on (at least) SCSI reads Message-ID: <20020717113631.A30914@Odin.AC.HMC.Edu> References: <200207171826.g6HIQr049869@hokkshideh2.jetcafe.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-md5; protocol="application/pgp-signature"; boundary="sm4nu43k4a2Rpi4c" Content-Disposition: inline User-Agent: Mutt/1.2.5.1i In-Reply-To: <200207171826.g6HIQr049869@hokkshideh2.jetcafe.org>; from dave@jetcafe.org on Wed, Jul 17, 2002 at 11:26:48AM -0700 X-Virus-Scanned: by amavisd-milter (http://amavis.org/) on odin.ac.hmc.edu Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org --sm4nu43k4a2Rpi4c Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Jul 17, 2002 at 11:26:48AM -0700, Dave Hayes wrote: > Well we've isolated the problem (we think). These machines are built > using a 2U case. This requires a PCI riser card that plugs into the > motherboard PCI slot and provides horizontal (rather than vertical) > card mounting options. >=20 > With that riser card out, I've run the same "crashme" script which > puts a moderate to heavy load on the disk subsystem, and the machine > is still up (16 hours and counting). >=20 > Electronically speaking that riser card adds a extra capacitance and > resistance that is probably not in the design spec for PCI. That's > the current theory anyway. I don't know enough about PCI to know > if that makes sense or not. I do know we are going to try slowing > down the SCSI transfer speed to see if that helps any.=20 I'd bet a lot more of a bum riser card then on a generic problem with riser cards. We've got got quite a few 2U systems running various cards with few problems. The only problem we've seen with a SCSI card was a RAID control that got fried when we power cycled a hung system. One other possiablity to consider is lack of decent air flow causing heating problems. If you run without a riser card, the system has to be open which would improve air flow around passivly cooled components. -- Brooks --=20 Any statement of the form "X is the one, true Y" is FALSE. PGP fingerprint 655D 519C 26A7 82E7 2529 9BF0 5D8E 8BE9 F238 1AD4 --sm4nu43k4a2Rpi4c Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE9NbkuXY6L6fI4GtQRAuJXAJkBQALt/A3OD/xzdzPfri/OZH0xZACeKt9v 2JtTB/N6V6iltc7MlNMi0ts= =QQVi -----END PGP SIGNATURE----- --sm4nu43k4a2Rpi4c-- To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message