[8803] in Info-AFS_Redistribution
Re: NeXTStep AFS client
daemon@ATHENA.MIT.EDU (michael shiplett)
Mon Jan 28 22:35:00 2002
To: info-afs@transarc.com
In-reply-to: Your message of Mon, 28 Jan 2002 18:46:33 -0800.
<ylofjdx4ae.fsf@windlord.stanford.edu>
From: michael shiplett <walrus+info-afs@monkey.org>
Date: Mon, 28 Jan 2002 22:31:40 -0500
Message-Id: <20020129033140.51AD23D4A0C@naughty.monkey.org>
> > The one problem I remember is that every NeXT did an fsck every time it
> > was booted. They didn't find anything, but it certainly took a while.
> Oh, yeah, that's right, there was that. The NeXT boxes in the
> residences tended to take something like a day to boot, in part
> because of AFS. (But the cluster machines weren't as bad.)
unfortunately i don't remember why the systems didn't shutdown
cleanly, but running w/a 12-16M memory cache was a decent way to
access afs w/o having to wait a day for the system boot.
there was also trying to get an afs-authenticating loginwindow and
having to create a special root.afs w/the sticky bit set to keep
them from stat'ng all the mountpoints. the other problem we had
w/them at umich was having to deal with netinfo instead of a global
password. it's too bad mac osx has resurrected it.
michael
