[26782] in Athena Bugs

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

Re: 9.4.26 breaks AFS on earlgrey.mit.edu

daemon@ATHENA.MIT.EDU (Garry Zacheiss)
Tue Mar 28 16:37:18 2006

Message-Id: <200603282137.k2SLb6nb015855@brad-majors.mit.edu>
To: Daniel Jared Dominguez <danjared@mit.edu>
In-Reply-To: Message from Daniel Jared Dominguez <danjared@MIT.EDU> of "Tue,
	28 Mar 2006 15:22:32 EST." <20060328202232.GB440@gaston.mit.edu> 
Date: Tue, 28 Mar 2006 16:37:06 -0500
From: Garry Zacheiss <zacheiss@mit.edu>
X-Spam-Score: 1.217
X-Spam-Level: * (1.217)
X-Spam-Flag: NO
Cc: bugs@mit.edu
Errors-To: bugs-bounces@mit.edu

For the record, the problem with turns out to be that the machine in
question has a 20GB AFS cache partition, and was using 7GB of it, which
apparently caused afsd to segfault partway through the cache sweep.

The machine is now configured to use a smaller amount of the partition
(with AFSADJUST=false in rc.conf to keep the init script from stomping
on it) and is running AFS happily.

This is a bug and a new one, and I'll look into it and report it
upstream, but for now, the answer to having especially large cache
partitions is "don't do that".

Garry

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