[226] in arla-drinkers
linux/SMP/arla difficulties
daemon@ATHENA.MIT.EDU (Dave Morrison)
Sun Aug 23 21:57:09 1998
From owner-arla-drinkers@stacken.kth.se Mon Aug 24 01:57:08 1998
Return-Path: <owner-arla-drinkers@stacken.kth.se>
Delivered-To: arla-drinkers-mtg@bloom-picayune.mit.edu
Received: (qmail 29013 invoked from network); 24 Aug 1998 01:57:07 -0000
Received: from unknown (HELO sundance.stacken.kth.se) (130.237.234.41)
by bloom-picayune.mit.edu with SMTP; 24 Aug 1998 01:57:07 -0000
Received: (from majordom@localhost)
by sundance.stacken.kth.se (8.8.8/8.8.8) id DAA04011
for arla-drinkers-list; Mon, 24 Aug 1998 03:51:23 +0200 (MET DST)
Received: from assaris.pdc.kth.se (assaris.pdc.kth.se [193.10.159.45])
by sundance.stacken.kth.se (8.8.8/8.8.8) with ESMTP id DAA04007
for <arla-drinkers@stacken.kth.se>; Mon, 24 Aug 1998 03:51:20 +0200 (MET DST)
Received: (from assar@localhost) by assaris.pdc.kth.se (8.8.5/8.7.3) id DAA04343; Mon, 24 Aug 1998 03:53:28 +0200 (MET DST)
Original-Sender: dave@bnl.gov
Message-ID: <35DC5C49.79E6D3C0@bnl.gov>
Date: Thu, 20 Aug 1998 13:26:33 -0400
From: Dave Morrison <dave@bnl.gov>
X-Mailer: Mozilla 4.5b1 [en] (X11; I; Linux 2.1.116 i686)
X-Accept-Language: en
MIME-Version: 1.0
To: arla-drinkers <arla-drinkers@stacken.kth.se>
Subject: linux/SMP/arla difficulties
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Lines: 39
Sender: owner-arla-drinkers@stacken.kth.se
Precedence: bulk
Dear arla-drinkers,
We have a bunch on Dell dual PII boxes that are running linux-2.1.116 (yes, I
know, yesterday's technology) and arla-0.9. Well, they're sort of running -
they're a bit unstable and the instability seems to be correlated with arla's
presence. Here are a few symptoms that may or may not be related:
o once arla is up and running, the system crashes after anywhere from a few
minutes to a few hours - no obvious correlation with AFS activity.
o before the machine finally hangs, tons of messages appear of the form "sending
wakeup: ..." (which are apparently generated in xfs).
o during this same time, arla goes nonlinear and chews up all the CPU
o when arlad is first started (using the startarla script), the sysname can't be
changed successfully for at least a minute.
o this same machine appears to be stable if arla is not started
Other oddments:
o arla was built with -D__SMP__ (same behavior when built without this)
o this version of kernel and arla runs happily together on a bunch of quad PPro
boxes we also have
Is anyone else experiencing similar problems? Does anyone have suggestions
that'd help with this situation? Is there some additional info that I could
gather that could help diagnose things?
Thanks,
Dave
--
David Morrison Brookhaven National Laboratory phone: 516-344-5840
Physics Department, Bldg 510 C fax: 516-344-3253
Upton, NY 11973-5000 email: dave@bnl.gov
