From owner-freebsd-current Wed Jan 29 16:51: 0 2003 Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id A921637B401; Wed, 29 Jan 2003 16:50:58 -0800 (PST) Received: from mail.rpi.edu (mail.rpi.edu [128.113.2.7]) by mx1.FreeBSD.org (Postfix) with ESMTP id ACB7443F85; Wed, 29 Jan 2003 16:50:56 -0800 (PST) (envelope-from drosih@rpi.edu) Received: from [128.113.24.47] (gilead.netel.rpi.edu [128.113.24.47]) by mail.rpi.edu (8.12.1/8.12.1) with ESMTP id h0U0olsR239576; Wed, 29 Jan 2003 19:50:47 -0500 Mime-Version: 1.0 X-Sender: drosih@mail.rpi.edu Message-Id: In-Reply-To: References: Date: Wed, 29 Jan 2003 19:50:46 -0500 To: current@FreeBSD.ORG From: Garance A Drosihn Subject: Re: Panic#2 copying many files from UFS to UFS2 partition Cc: kirk@mckusick.com, Poul-Henning Kamp Content-Type: text/plain; charset="us-ascii" ; format="flowed" X-Scanned-By: MIMEDefang 2.3 (www dot roaringpenguin dot com slash mimedefang) Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG At 9:49 PM -0500 1/28/03, Garance A Drosihn wrote: >Hmm, well, I finally got my first actual system panic which wasn't >obviously caused by my own screwing around. On the console I have: > > free inode /usr/cvs/net/64 had 0 blocks > panic: Negative bio_offset (-19038208) on bio 0xce51be28 > cpuid = 0; lapic.id = 01000000 > Debugger("panic") > Stopped at Debugger+0x55: xchgl %ebx,in_Debugger.0 It looks like the sequence which generated that panic can be repeated to generate other panics. My sequence was: use sysinstall to remove and then recreate the new UFS2 partition, 3100 meg in size, mounted on /usr/cvs/net pushd /usr/cvs/net-old tar cf - ./ | (popd; cd /usr/cvs/net; tar xpf -) and again I got a panic, although not quite the same panic. It also seemed like it happened much sooner in the process. This time it says: free inode /usr/cvs/net/64 had 0 blocks mode = 0177777, inum = 78, fs = /usr/cvs/net panic: ffs_valloc: dup alloc cpuid = 0; lapic.id = 01000000 Debugger("panic") Stopped at Debugger+0x55: xchgl %ebx,in_Debugger.0 The first panic was while doing a 'mkdir'. There's no mkdir in the traceback this time. It looks like the system was in the middle of an open() call (syscall 5). It's interesting that all three panics mentioned /usr/cvs/net/64, even though the source directory has no file-or-directory named '64' in it. I expect that some system dumps would be more useful than me typing in backtraces, so I set up dumpdev and dumpdir in rc.conf, rebooted, and tried again. The system panic'ed again, looking about the same as the second panic. I typed 'panic' in the debugger, it says panic: from debugger cpuid = 1; lapic.id = 00000000 boot() called on cpu#1 Uptime: 6m8s Dumping 511 MB ata0: resetting devices I let it sit there, well, for about 20 minutes, before deciding that nothing is happening and I powered the machine off and on to get it back. There was no dump in the dump-dir. I've tried this a number of times now, and it seems I never get a dump. My dumpdir is on a UFS1 partition, I'm using it instead of /var because my /var is 256meg and this partition is over 1 gig. So, what should I do at this point to help pin this down? -- Garance Alistair Drosehn = gad@gilead.netel.rpi.edu Senior Systems Programmer or gad@freebsd.org Rensselaer Polytechnic Institute or drosih@rpi.edu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message