[795] in arla-drinkers
Re: inode clashes (linux 2.2.x, arla 0.2x)
daemon@ATHENA.MIT.EDU (Brandon S. Allbery KF8NH)
Fri Apr 30 18:33:50 1999
From owner-arla-drinkers@stacken.kth.se Fri Apr 30 22:33:49 1999
Return-Path: <owner-arla-drinkers@stacken.kth.se>
Delivered-To: arla-drinkers-mtg@bloom-picayune.mit.edu
Received: (qmail 7225 invoked from network); 30 Apr 1999 22:33:48 -0000
Received: from unknown (HELO sundance.stacken.kth.se) (130.237.234.41)
by bloom-picayune.mit.edu with SMTP; 30 Apr 1999 22:33:48 -0000
Received: (from majordom@localhost)
by sundance.stacken.kth.se (8.8.8/8.8.8) id AAA17027
for arla-drinkers-list; Sat, 1 May 1999 00:27:54 +0200 (MET DST)
Received: from ECE.CMU.EDU (ECE.CMU.EDU [128.2.236.200])
by sundance.stacken.kth.se (8.8.8/8.8.8) with ESMTP id AAA17023
for <arla-drinkers@stacken.kth.se>; Sat, 1 May 1999 00:27:38 +0200 (MET DST)
Received: from hilfy.ece.cmu.edu (root@HILFY.ECE.CMU.EDU [128.2.253.106])
by ECE.CMU.EDU (8.9.2/8.8.8) with ESMTP id SAA18997;
Fri, 30 Apr 1999 18:27:10 -0400 (EDT)
Received: from speaker.kf8nh.apk.net (root@ANNEX-3.ECE.CMU.EDU [128.2.236.3])
by hilfy.ece.cmu.edu (8.9.2/8.8.8) with ESMTP id SAA01486;
Fri, 30 Apr 1999 18:27:34 -0400 (EDT)
Received: from rushlight.kf8nh.apk.net (allbery@rushlight.kf8nh.apk.net [10.9.204.1])
by speaker.kf8nh.apk.net (8.8.7/8.8.7) with ESMTP id SAA11721;
Fri, 30 Apr 1999 18:27:32 -0400
Received: (from allbery@localhost)
by rushlight.kf8nh.apk.net (8.8.7/8.8.7) id SAA13628;
Fri, 30 Apr 1999 18:27:56 -0400
Message-Id: <199904302227.SAA13628@rushlight.kf8nh.apk.net>
To: Simon Josefsson <jas@pdc.kth.se>
Cc: arla-drinkers@stacken.kth.se
Subject: Re: inode clashes (linux 2.2.x, arla 0.2x)
In-reply-to: Your message of "30 Apr 1999 23:03:35 +0200."
<ilur9p1u8go.fsf@xiphias.pdc.kth.se>
Date: Fri, 30 Apr 1999 18:27:56 -0400
From: "Brandon S. Allbery KF8NH" <allbery@kf8nh.apk.net>
Sender: owner-arla-drinkers@stacken.kth.se
Precedence: bulk
In message <ilur9p1u8go.fsf@xiphias.pdc.kth.se>, Simon Josefsson writes:
+-----
| Since I've never seen this when using a AFS client from another vendor
| I wonder whether this is something that can be (or is being) improved
| in arla?
+--->8
I've seen it once or twice.
The problem is that AFS clients must map the AFS file ID, which is 4 32-bit
values, into a 32-bit inode number. It's quite impossible to completely
avoid collisions, although you might be able to minimize the chances of it
for certain access patterns.
--
brandon s. allbery [os/2][linux][solaris][japh] allbery@kf8nh.apk.net
system administrator [WAY too many hats] allbery@ece.cmu.edu
carnegie mellon / electrical and computer engineering KF8NH
We are Linux. Resistance is an indication that you missed the point.
