[1176] in Athena Bugs
Re: A bug with the workstation environment
daemon@ATHENA.MIT.EDU (John T Kohl)
Fri Oct 14 13:26:42 1988
Date: Fri, 14 Oct 88 13:25:55 EDT
From: John T Kohl <jtkohl@ATHENA.MIT.EDU>
To: jnrees@ATHENA.MIT.EDU, bugs@ATHENA.MIT.EDU, consultants@ATHENA.MIT.EDU
In-Reply-To: bugs:[1175]
The symptoms you describe sound like what happens when a disk goes
off-line (or, equivalently, an NFS server you use is unreachable).
One process attempts to access the unavailable file, directory, or
whatever. It waits forever for the missing filesystem to come back.
But it locks some resources on the workstation. Eventually the lock
cascades such that all processes get wedged up.
(The C shell will often wedge itself by searching the user's path for an
executable.)
On the VAXes, the mouse will probably keep tracking forever. On the RT
it may at some point stop tracking (if the X server gets wedged by the
missing filesystem).
On either machine, if the X server gets wedged, the mouse will cease
changing shape between emacs, the root window, and xterms.
So the problem you describe is most likely related to the unreliability
of our NFS servers, and Athena has adopted a plan to replace these
servers with new (and much more reliable) equipment.
John
