From owner-freebsd-scsi Sat Apr 10 19:19:46 1999 Delivered-To: freebsd-scsi@freebsd.org Received: from panzer.plutotech.com (panzer.plutotech.com [206.168.67.125]) by hub.freebsd.org (Postfix) with ESMTP id 3CE9F153FC for ; Sat, 10 Apr 1999 19:18:58 -0700 (PDT) (envelope-from ken@panzer.plutotech.com) Received: (from ken@localhost) by panzer.plutotech.com (8.9.3/8.8.5) id UAA34825; Sat, 10 Apr 1999 20:16:28 -0600 (MDT) From: "Kenneth D. Merry" Message-Id: <199904110216.UAA34825@panzer.plutotech.com> Subject: Re: Strange SCSI panic - devstat_end_transaction In-Reply-To: <370DAC97.2B48323E@tdx.co.uk> from Karl Pielorz at "Apr 9, 1999 8:30:31 am" To: kpielorz@tdx.co.uk (Karl Pielorz) Date: Sat, 10 Apr 1999 20:16:28 -0600 (MDT) Cc: freebsd-scsi@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL43 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Karl Pielorz wrote... > Hi All, > > My box did this last night, and I was wondering if anyone could shed any light > on it? - Unfortunately the machine was locked solid, and didn't manage to > dump... > > I was newfs'ing a Vinum volume at the time - da2 is dedicated solely to /usr2, > and is on a separate controller to all the drives used for Vinum - so I don't > think Vinum is related... > > /usr2 has user home directories on it, a lightly used MySQL database - and the > directories for the machines lightly used Apache server (i.e. it was probably > being used at the time, but it's never thrashed). > > The machine is a dual P-Pro200, w/256Mb RAM running on a SuperMicro > P6DNF/440FX motherboard. System is running -current as of 22nd March. > > DA2 is: > > da2 at ahc0 bus 0 target 5 lun 0 > da2: Fixed Direct Access SCSI-2 device > da2: 10.000MB/s transfers (10.000MHz, offset 15), Tagged Queueing Enabled > da2: 1021MB (2091144 512 byte sectors: 64H 32S/T 1021C) > > And is running softupdates... The volume got 'mildly' trashed after this - > fsck found ~2 errors but does seem to have recovered it without incident. > > I've quickly checked termination etc. for that controller - and it all appears > fine... The contoller (ahc0) is: > > ahc0: rev 0x01 int a irq 17 on pci0.18.0 > ahc0: aic7880 Wide Channel A, SCSI Id=7, 16/255 SCBs > > > Any info / rough ideas would be greatly appreciated... The devstat warning is just a symptom of the problem, and not the real cause of the problem. It basically indicates that one of a couple of things have happened: - a transaction has been completed twice - the buffer queue has been corrupted somehow (Justin said that duplicate buffers might do it) and we ended up getting more transaction completions than transaction starts. The devstat message is informational, and won't cause any system problems, but it does indicate some other problem. > devstat_end_transaction: HELP!! busy_count for da2 is < 0 (-1)! > panic: vwakeup: neg numoutput > mp_lock = 01000001; cpuid = 1; lapic.id = 01000000 > boot() called on cpu#1 I'm not sure what the problem is, and I haven't seen this one recently. Have you ever had the problem before? Can you reproduce it? i.e., if you newfs the Vinum volume again, does it happen again? Ken -- Kenneth Merry ken@plutotech.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message