From owner-freebsd-bugs Wed Oct 29 11:15:53 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id LAA06241 for bugs-outgoing; Wed, 29 Oct 1997 11:15:53 -0800 (PST) (envelope-from owner-freebsd-bugs) Received: from freefall.freebsd.org (freefall.FreeBSD.ORG [204.216.27.21]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id LAA06232 for ; Wed, 29 Oct 1997 11:15:50 -0800 (PST) (envelope-from billh@microware.com) Received: from gate.microware.com (firewall-user@gate.microware.com [198.17.151.51]) by freefall.freebsd.org (8.8.6/8.8.5) with SMTP id LAA15492 for ; Wed, 29 Oct 1997 11:15:24 -0800 (PST) Received: by gate.microware.com; id AA25705; Wed, 29 Oct 97 13:19:10 CST Received: from mcrware.microware.com(172.16.1.32) by gate.microware.com via smap (V3.1.1) id xma025671; Wed, 29 Oct 97 13:18:48 -0600 Received: from microware.com (orac.microware.com) by mcrware.microware.com with SMTP id AA12384 (5.67a8/IDA-1.5); Wed, 29 Oct 1997 13:12:08 -0600 Message-Id: <34578A51.C27984AD@microware.com> Date: Wed, 29 Oct 1997 13:11:13 -0600 From: Bill Hutchison Organization: Microware Systems Corp. X-Mailer: Mozilla 4.03 [en] (X11; I; SunOS 5.5.1 sun4c) Mime-Version: 1.0 To: gnat@frii.com Cc: "Jordan K. Hubbard" , freebsd-bugs@freebsd.com Subject: Re: 2.2.5-RELEASE reboots References: <199710290028.RAA00828@elara.frii.com> <22314.878091829@time.cdrom.com> <199710291534.IAA24297@elara.frii.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-freebsd-bugs@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk gnat@frii.com wrote: > > Jordan K. Hubbard writes: > > > We're seeing reboots with no panic in the log file or on the screen. > > > It then hangs when trying to synch disks, and the machine must be > > > Ctrl-Alt-Del rebooted or reset. This is very frustrating. > > > > Hmmm. Are the reboots coincident with any other activity, say > > heavy NFS usage or something? > > It's possible, but we can't know one way or the other. We suspect a > cron process interacting badly with another running process, but it's > impossible to prove. The machine reboots before the offending process > can exit, so process accounting isn't giving us clues. > > The machine does run a daemon process for a user, which prepares HTML > in response to requests from a CGI script (it's the back end to his > web site). This is moving to another machine today, and I'll look > eagerly to see if the reboots follow it. > > Is there any way to get more data on what's going on when the machine > reboots? Any clues about the synching disks situation? > > Nat Nat, FYI, I recently set up a 2.2.2 system for primilary running Samba 1.9.17p4. It's been sitting idle (except for testing) without any problems. As I've moved users over to it, I've had exact same crashes occur as you describe. I managed to catch it once before it reboot and saw that it was a "fatal trap 12: page fault while in kernel" and the current process was nfsiod. So, I'm guessing my crashes are related to increased NFS activity... I see there is already an outstanding open problem report describing this crash (kern/4186). Does FreeBSD 2.2.2 and 2.2.5 have known problem (and workarounds?) for high NFS traffic servers. -Bill