[427] in Info-AFS_Redistribution
Re: AFS and NeXTs
daemon@ATHENA.MIT.EDU (reuss@ni.umd.edu)
Fri Nov 15 12:48:05 1991
To: Craig_Everhart@transarc.com, Info-AFS@transarc.com, reuss@ni.umd.edu
Cc: Cal_Thixton@next.com
In-Reply-To: Your message of "Fri, 15 Nov 91 09:24:52 EST."
Date: Fri, 15 Nov 91 11:20:22 -0500
From: reuss@ni.umd.edu
> One possible, if questionable, strategy can emerge from all this, and it
> goes something like the following. Leave root.afs alone for non-NeXT
> machines. Make a special root.afs.next volume that's just like root.afs
> except all the mount points have the sticky bit on. Use this as the
> root volume for all NeXT machines in a cell. Alter the CM for the NeXT
> platform so that if it gets an lstat() for a file that's really an AFS
> mount point with the sticky bit on, it doesn't follow the mount point,
> but instead returns information that's made up to suit the case, so that
> 'ls -F' and the Workspace will both simply print something like
> ``(automount)''.
> I don't know what the most reasonable answer is, but I'm rather leery of
> the consequences of some AFS mount points suddenly really looking like
> symbolic links whose readlink(2) contents are meaningless.
One problem here is the the automounter turns the sticky bit off after it
does the mount so ls and friends can look through the link and see the newly
mounted filesystem. The CM would probably have to only do the lstat tricks
when the stat information for the mount point wasn't in cache...??.. Sounds
a little messy, to say the least.
I really didn't think the symlink/stickybit could be exploited to solve the
problem; I just found it intertesting that the Workspace was doing some
tricks with automounts. I really hope these dicussions lead to some
solution. We've got a cell consisting mostly of NextStations. Currently
the cell doesn't mount remote cells because of this very problem (ok, they
actually mount them as /afs/remote/..., but I don't concider that a
solutuion). Unfortunatly I want to add some non-NeXT machines to the cell,
and they expect to have the standard AFS naming conventions and remote cell
access.
-Karl
