[139] in The Cryptographic File System users list
Re: CFS goes unresponsive under NetBSD
daemon@ATHENA.MIT.EDU (Alex Raftis)
Tue Aug 3 13:46:43 1999
From owner-cfs-users@nsa.research.att.com Tue Aug 03 17:46:43 1999
Return-Path: <owner-cfs-users@nsa.research.att.com>
Delivered-To: cfs-mtg@bloom-picayune.mit.edu
Received: (qmail 2818 invoked from network); 3 Aug 1999 17:46:42 -0000
Received: from unknown (HELO mail-green.research.att.com) (135.207.30.103)
by bloom-picayune.mit.edu with SMTP; 3 Aug 1999 17:46:42 -0000
Received: from amontillado.research.att.com (amontillado.research.att.com [135.207.24.32])
by mail-green.research.att.com (Postfix) with ESMTP
id 8E1791E02A; Tue, 3 Aug 1999 13:46:41 -0400 (EDT)
Received: from nsa.research.att.com (majordomo@nsa.research.att.com [135.207.24.155])
by amontillado.research.att.com (8.8.7/8.8.7) with ESMTP id NAA24499;
Tue, 3 Aug 1999 13:46:34 -0400 (EDT)
Received: (from majordomo@localhost) by nsa.research.att.com (8.7.3/8.7.3) id NAA07478 for cfs-users-list; Tue, 3 Aug 1999 13:44:09 -0400 (EDT)
X-Authentication-Warning: nsa.research.att.com: majordomo set sender to owner-cfs-users@nsa.research.att.com using -f
Received: from mail-green.research.att.com (mail-green.research.att.com [135.207.30.103]) by nsa.research.att.com (8.7.3/8.7.3) with ESMTP id NAA07474 for <cfs-users@nsa.research.att.com>; Tue, 3 Aug 1999 13:44:06 -0400 (EDT)
Received: by mail-green.research.att.com (Postfix)
id B04F61E02B; Tue, 3 Aug 1999 13:44:59 -0400 (EDT)
Delivered-To: cfs-users@research.att.com
Received: from echidna.doverpacific.com (unknown [209.132.92.4])
by mail-green.research.att.com (Postfix) with ESMTP id 0A2251E02A
for <cfs-users@research.att.com>; Tue, 3 Aug 1999 13:44:59 -0400 (EDT)
Received: from timewarp.doverpacific.com (timewarp.doverpacific.com [209.132.92.3])
by echidna.doverpacific.com (8.8.8/8.8.8) with SMTP id KAA08088
for <cfs-users@research.att.com>; Tue, 3 Aug 1999 10:44:58 -0700 (PDT)
Message-Id: <199908031744.KAA08088@echidna.doverpacific.com>
Received: by timewarp.doverpacific.com (NX5.67g/NX3.0X)
id AA08683; Tue, 3 Aug 99 10:44:57 -0700
From: "Alex Raftis" <alex@doverpacific.com>
Date: Tue, 3 Aug 1999 10:44:56 -0700
To: cfs-users@research.att.com
Subject: Re: CFS goes unresponsive under NetBSD
Reply-To: "Alex Raftis" <alex@doverpacific.com>
Received: by NLS.Mailer (0.9)
Mime-Version: 1.0 (NLS Mail v1)
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
Sender: owner-cfs-users@research.att.com
Precedence: bulk
There's a similar problem under OpenStep/Mach 4.2. I haven't spent a whole
lot of time tracing it down, since I've generally found how to avoid lock up.
The problem seems to be a race condition in the nfsd caused by heavy access
on a cfs mounted disk. Basically, I can pretty consistently lock my nfs (both
cfs and other network access), by copying or moving a large set of files
between two cfs disks or a non-cfs disk to a cfs disk. I find that if I slow
down the access to the disks, I don't have the problem.
For example, I needed to move about a hundred files from a non-cfs disk to a
cfs disk. To do so, rather than issuing a mv * /crypt/..., I did:
foreach file (*)
echo $file
mv $file /crypt/...
end
Also, while doing this, I refrain from doing other cfs access, and have been
able to avoid further lock ups.
Alex Raftis
---
alex@doverpacific.com
