From owner-freebsd-scsi Wed Dec 15 16:58:39 1999 Delivered-To: freebsd-scsi@freebsd.org Received: from narnia.plutotech.com (narnia.plutotech.com [206.168.67.130]) by hub.freebsd.org (Postfix) with ESMTP id CB01915650; Wed, 15 Dec 1999 16:58:33 -0800 (PST) (envelope-from gibbs@narnia.plutotech.com) Received: (from gibbs@localhost) by narnia.plutotech.com (8.9.3/8.7.3) id RAA30807; Wed, 15 Dec 1999 17:48:28 -0700 (MST) Date: Wed, 15 Dec 1999 17:48:28 -0700 (MST) From: "Justin T. Gibbs" Message-Id: <199912160048.RAA30807@narnia.plutotech.com> To: Mike Smith Cc: scsi@FreeBSD.ORG Subject: Re: Unexpected bus free & swap_pager: indefinite wait buffer X-Newsgroups: pluto.freebsd.scsi In-Reply-To: <199912142011.MAA00665@mass.cdrom.com> User-Agent: tin/pre-1.4-980818 ("Laura") (UNIX) (FreeBSD/4.0-CURRENT (i386)) Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org In article <199912142011.MAA00665@mass.cdrom.com> you wrote: >> We're running 3.3-stable from about 10/21/1999 on a machine that we >> haven't physically touched for about four months. It has a (supposedly >> good) LVD cable with a terminator at the end of the cable, that came >> with the Asus motherboard. Last night, the kernel started printing the >> errors below and the machine became unusable: couldn't login, >> unresponsive shells, though it did still respond to pings. >> >> The blknos reported in the swap_pager message repeated in a cycle (e.g., >> 8944, 328, 8944, 328...), and every now and then another blkno would be >> added to the cycle, until at the end (just before we hit the reset >> button) the cycle was >> 8944,328,2640,174968,44560,42848,40720,3208,512,3160. >> >> Are we losing a disk, or is it some kind of bug, or...? >> >> Thanks in advance for any help you can offer. >> >> Matt >> >> Dec 13 17:35:00 merry /kernel: Unexpected busfree. LASTPHASE == 0x40 > > This is indicative of a disk going away, either from a firmware bug or > overheating. Everything goes downhill from there (the command has been > lost, and we don't recover well from that). Hmm. We should have retried the command. Of course, this might have failed if the drive decided to stop responding. Does the swap-pager code in 3.3 talk about "indefinite wait buffers" if the I/O system returns an error? I would expect this only if the command never returns, but I'll have to poke around in the code to find out. -- Justin To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message