[796] in arla-drinkers

home help back first fref pref prev next nref lref last post

RE: inode clashes (linux 2.2.x, arla 0.2x)

daemon@ATHENA.MIT.EDU (Lyle Seaman)
Fri Apr 30 19:37:36 1999

From owner-arla-drinkers@stacken.kth.se Fri Apr 30 23:37:36 1999
Return-Path: <owner-arla-drinkers@stacken.kth.se>
Delivered-To: arla-drinkers-mtg@bloom-picayune.mit.edu
Received: (qmail 8096 invoked from network); 30 Apr 1999 23:37:35 -0000
Received: from unknown (HELO sundance.stacken.kth.se) (130.237.234.41)
  by bloom-picayune.mit.edu with SMTP; 30 Apr 1999 23:37:35 -0000
Received: (from majordom@localhost)
	by sundance.stacken.kth.se (8.8.8/8.8.8) id BAA17476
	for arla-drinkers-list; Sat, 1 May 1999 01:31:39 +0200 (MET DST)
Received: from gw.stormsystems.com (stormsystems.com [209.49.190.30])
	by sundance.stacken.kth.se (8.8.8/8.8.8) with ESMTP id BAA17472
	for <arla-drinkers@stacken.kth.se>; Sat, 1 May 1999 01:31:34 +0200 (MET DST)
Received: from fs1.office.stormsystems.com (exchange.office.stormsystems.com [172.29.1.5])
	by gw.stormsystems.com (8.9.1a/8.9.1) with ESMTP id TAA08386
	for <arla-drinkers@stacken.kth.se>; Fri, 30 Apr 1999 19:31:02 -0400
Received: by fs1.stormsystems.com with Internet Mail Service (5.5.2448.0)
	id <J5B3YPTG>; Fri, 30 Apr 1999 19:31:01 -0400
Message-ID: <B5C3AE5DDF77D211BD2500A0C9D3A6A10C0041@fs1.stormsystems.com>
From: Lyle Seaman <LSeaman@stormsystems.com>
To: arla-drinkers@stacken.kth.se
Subject: RE: inode clashes (linux 2.2.x, arla 0.2x)
Date: Fri, 30 Apr 1999 19:30:53 -0400
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2448.0)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-arla-drinkers@stacken.kth.se
Precedence: bulk

The mapping in "stock" AFS uses the volume and vnode numbers, so this should
happen most often when dealing with two "cellular" mount points.  It can
also happen with two mount points whose mounted volume IDs are the same in
the lower 16 bits.

The best solution is probably to use a better mixing function to generate
inode numbers.  vnode and volume numbers really aren't very random, and the
simple function that AFS uses doesn't adequately compensate for their
predictability.


home help back first fref pref prev next nref lref last post