From owner-freebsd-fs@FreeBSD.ORG Sat Aug 20 05:24:16 2005 Return-Path: X-Original-To: freebsd-fs@freebsd.org Delivered-To: freebsd-fs@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3C5EC16A41F for ; Sat, 20 Aug 2005 05:24:16 +0000 (GMT) (envelope-from andrew@starfishzone.com) Received: from mail.starfishzone.com (mail.starfishzone.com [81.187.184.167]) by mx1.FreeBSD.org (Postfix) with ESMTP id 4EA6D43D45 for ; Sat, 20 Aug 2005 05:24:12 +0000 (GMT) (envelope-from andrew@starfishzone.com) Received: (qmail 29752 invoked by uid 1013); 20 Aug 2005 05:24:09 -0000 Received: from 81.2.81.105 by linckia.starfishzone.com (envelope-from , uid 88) with qmail-scanner-1.25 (clamdscan: 0.86.2/1034. Clear:RC:1(81.2.81.105):. Processed in 0.229287 secs); 20 Aug 2005 05:24:09 -0000 Received: from unknown (HELO moray.starfishzone.com) (andrew.liles@starfishzone.com@81.2.81.105) by 0 with (RC4-SHA encrypted) SMTP; 20 Aug 2005 05:24:08 -0000 Message-Id: <6.0.3.0.2.20050820061711.03b530d0@mail.starfishzone.com> X-Sender: andrew.liles@starfishzone.com@mail.starfishzone.com X-Mailer: QUALCOMM Windows Eudora Version 6.0.3.0 Date: Sat, 20 Aug 2005 06:23:51 +0100 To: freebsd-fs@freebsd.org From: Andrew Liles Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Subject: Panic "softdep_move_dependencies: need merge code" X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 20 Aug 2005 05:24:16 -0000 I have a Highpoint RocketRAID 454 card configured for RAID5 with 4 discs (driver is hpt374.ko) on FreeBSD 5.2.1-RELEASE. I have had a history of panics that I think emanate from the filesystem and I have far more physical failures of the discs than one might expect (I suspect heat damage due to poor ventilation). Having built myself a debug kernel I was able to pinpoint the following in 3 recent panics that occurred within a few days of each other. Around this time I found that another disc has died on me. In all cases I believe the panics occur under periods of intense disc activity; e.g. copy many files from one area to another, or running big meta-data updates like chmod -R and chgrp -R. This the instruction that calls the panic: #14 0xc0776904 in softdep_move_dependencies (oldbp=0xcd0d9c20, newbp=0xcd0c0838) at ../../../ufs/ffs/ffs_softdep.c:800 800 panic("softdep_move_dependencies: need merge code"); I have read this post: http://tinyurl.com/c3fpj a) Is it reasonable to assume that my panics are exactly because I have suffered physical disc problems/ b) shouldn't my external RAID5 have insulated me from such issues? c) might there be some incompatibility with Soft Updates and this particular RAID5 implementation? I am happy to forego the benefits of SoftUpdates if I can then have a stable machine. What would you recommend? Andrew Liles