From owner-freebsd-current  Thu Oct 17 09:36:30 1996
Return-Path: owner-current
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id JAA10774
          for current-outgoing; Thu, 17 Oct 1996 09:36:30 -0700 (PDT)
Received: from tatooine.vader.org (tatooine.vader.org [194.159.100.200])
          by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id JAA10767
          for <current@FreeBSD.org>; Thu, 17 Oct 1996 09:36:26 -0700 (PDT)
Received: (from chris@localhost)
          by tatooine.vader.org (8.8.0/8.8.0) id RAA01857;
          Thu, 17 Oct 1996 17:35:57 +0100 (BST)
From: Chris Dabrowski <chris@vader.org>
Message-Id: <199610171635.RAA01857@tatooine.vader.org>
Subject: Re: system hangs with current kernel...?
In-Reply-To: <Pine.NEB.3.95.961016195948.372A-100000@quagmire.ki.net> from "Marc G. Fournier" at "Oct 16, 96 08:03:02 pm"
To: scrappy@ki.net (Marc G. Fournier)
Date: Thu, 17 Oct 1996 17:35:57 +0100 (BST)
Cc: current@FreeBSD.org
Reply-To: chris@vader.org
X-Mailer: ELM [version 2.4ME+ PL28 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-current@FreeBSD.org
X-Loop: FreeBSD.org
Precedence: bulk

In a previous message, Marc G. Fournier wrote:
> 
> 	This is getting slightly annoying, so I figured I'd bring it to
> the list...
> 
> 	Has anyone noticed a recent problem with the -current kernel
> that is causing system hangs?  I've seen the reports going through
> about panics in the NFS area and whatnot, but nothing in particular
> dealing with actually hangs...
> 
> 	I'm running -current as of about Sunday night, and this morning
> the system locked up completely as if it was a SCSI bus hang...something
> I haven't experienced in *many* months on that machine.  It did it last
> night too, and I recompiled a kernel with debugging enabled, but, of course,
> if it doesn't panic, it doesn't dump core :(
> 

Yes, it happens to me too. However, I've noticed one strange phenomenon.
Before the system hangs completely, connections to port 119 (INN 1.4unoff4)
timeout whilst other services (21, 25, etc) are unaffected. Maybe it's
because INN uses the disk intensively and hence hammers the SCSI code
more.

I'm not much of a hacker myself but are there any tests or info that I
can provide to help?


chris@tatooine.vader.org:/home/chris $ uname -a
FreeBSD tatooine.vader.org 2.2-CURRENT FreeBSD 2.2-CURRENT #0: Wed Oct 16 23:57:37 BST 1996     chris@tatooine.vader.org:/usr/src/sys/compile/TATOOINE  i386

Pentium 100
48Mb RAM
Adaptec 2940 PCI
HP 2Gb Hard Drive
4.4x CD-ROM

Chris
-- 
Chris Dabrowski
chris@vader.org