[1016] in arla-drinkers
Re: proposed PAG handling changes for Arla
daemon@ATHENA.MIT.EDU (Jeffrey Hutzelman)
Sat Jul 24 17:04:58 1999
From owner-arla-drinkers@stacken.kth.se Sat Jul 24 21:04:57 1999
Return-Path: <owner-arla-drinkers@stacken.kth.se>
Delivered-To: arla-drinkers-mtg@bloom-picayune.mit.edu
Received: (qmail 9562 invoked from network); 24 Jul 1999 21:04:57 -0000
Received: from unknown (HELO sundance.stacken.kth.se) (130.237.234.41)
by bloom-picayune.mit.edu with SMTP; 24 Jul 1999 21:04:57 -0000
Received: (from majordom@localhost)
by sundance.stacken.kth.se (8.8.8/8.8.8) id XAA19350
for arla-drinkers-list; Sat, 24 Jul 1999 23:00:27 +0200 (MET DST)
Received: from minbar.fac.cs.cmu.edu (MINBAR.FAC.CS.CMU.EDU [128.2.185.161])
by sundance.stacken.kth.se (8.8.8/8.8.8) with SMTP id XAA19333
for <arla-drinkers@stacken.kth.se>; Sat, 24 Jul 1999 23:00:02 +0200 (MET DST)
Received: from AFSTEST-2.FAC.CS.CMU.EDU by minbar.fac.cs.cmu.edu id aa16922;
24 Jul 99 16:59 EDT
Date: Sat, 24 Jul 1999 16:58:29 -0400
From: Jeffrey Hutzelman <jhutz+@cmu.edu>
To: Chris Wing <wingc@engin.umich.edu>
cc: arla-drinkers@stacken.kth.se
Subject: Re: proposed PAG handling changes for Arla
Message-ID: <827606258.932835509@AFSTEST-2.FAC.CS.CMU.EDU>
In-Reply-To: <Pine.LNX.4.10.9907241412380.1275-100000@shaft.engin.umich.edu>
Originator-Info: login-token=Mulberry:01pKJDmQ4KuA3w7cIc6SS0+y8Hr8bmG8sa0KYMpBC/IqQVqQo=;
token_authority=postmaster@andrew.cmu.edu
X-Mailer: Mulberry (Win32) [1.4.4, s/n S-100002]
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Disposition: inline
Sender: owner-arla-drinkers@stacken.kth.se
Precedence: bulk
> So does the official AFS never get rid of tokens or PAG data structures
> unless there is always an explicit unlog? We run some pretty big login
> servers around here that use the latest 'official' AFS on Solaris or
> HP-UX, and I haven't noticed them being particularly slow to setpag(),
> despite long uptimes and the fact that most of the users don't unlog
> before logging out.
A PAG isn't a data structure; it's just a number, like a UID. The mere
existence of a PAG doesn't occupy any resources -- the resources are
occupied by the structure that holds a user's credentials. Such structures
are created as needed (when you set tokens), and are automatically
garbage-collected from time to time, if they contain no current tokens.
The problem is that if a user doesn't unlog, the data structure holding his
credentials is not reclaimed until up to two hours after those credentials
expire.
The net effect is that you don't see problems on machines with long uptimes
but relatively few login sessions, but you _will_ see problems on machines
which have lots of users logging in and out all day. Also, the performance
hit isn't on setpag(), which is essentially free; it's on looking up
credentials for users.
