[1014] in arla-drinkers
Re: proposed PAG handling changes for Arla
daemon@ATHENA.MIT.EDU (Chris Wing)
Sat Jul 24 14:27:55 1999
From owner-arla-drinkers@stacken.kth.se Sat Jul 24 18:27:54 1999
Return-Path: <owner-arla-drinkers@stacken.kth.se>
Delivered-To: arla-drinkers-mtg@bloom-picayune.mit.edu
Received: (qmail 8814 invoked from network); 24 Jul 1999 18:27:53 -0000
Received: from unknown (HELO sundance.stacken.kth.se) (130.237.234.41)
by bloom-picayune.mit.edu with SMTP; 24 Jul 1999 18:27:53 -0000
Received: (from majordom@localhost)
by sundance.stacken.kth.se (8.8.8/8.8.8) id UAA16034
for arla-drinkers-list; Sat, 24 Jul 1999 20:23:30 +0200 (MET DST)
Received: from shaft.engin.umich.edu (wingc@shaft.engin.umich.edu [141.213.33.85])
by sundance.stacken.kth.se (8.8.8/8.8.8) with ESMTP id UAA16030;
Sat, 24 Jul 1999 20:23:24 +0200 (MET DST)
Received: from localhost (wingc@localhost)
by shaft.engin.umich.edu (8.9.3/8.9.3) with ESMTP id OAA01304;
Sat, 24 Jul 1999 14:23:22 -0400
X-Authentication-Warning: shaft.engin.umich.edu: wingc owned process doing -bs
Date: Sat, 24 Jul 1999 14:23:22 -0400 (EDT)
From: Chris Wing <wingc@engin.umich.edu>
To: Assar Westerlund <assar@stacken.kth.se>
cc: arla-drinkers@stacken.kth.se
Subject: Re: proposed PAG handling changes for Arla
In-Reply-To: <5lu2qu8ylv.fsf@assaris.sics.se>
Message-ID: <Pine.LNX.4.10.9907241412380.1275-100000@shaft.engin.umich.edu>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-arla-drinkers@stacken.kth.se
Precedence: bulk
Assar:
> Yes, `fs gcpags' will do a pioctl VIOC_GCPAGS. (Which is a nop in
> Arla :-)
Oops, I guess I should have read more of the code before commenting :)
>
> > In Arla, it's less of an issue since the tokens are stored in user
> > space, not kernel space. (arlad does all the real work)
>
> Yes. Expanding on this a little bit, arlad stores the credentials of
> the actual users in a hash table from which credentials get deleted
> when you do `unlog' or when the credential expires.
I've seen arlad messages in the syslog when tokens expire, and I've
noticed the credentials cache in the code.
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.
-Chris
wingc@engin.umich.edu
