[11244] in Athena Bugs
Re: sun4 7.6L: dm
daemon@ATHENA.MIT.EDU (Scott E Kullberg)
Thu Oct 21 15:42:39 1993
To: mar@MIT.EDU
Cc: sekullbe@MIT.EDU, bugs@MIT.EDU, marc@MIT.EDU
In-Reply-To: Your message of Thu, 21 Oct 93 15:29:30 -0400.
Date: Thu, 21 Oct 93 15:42:18 EDT
From: Scott E Kullberg <sekullbe@MIT.EDU>
> A while back you reported a bug where logging out of a sun failed for
> you. You reported this on deathtongue, a machine known to have a very
> non-standard configuration. Have you seen this on any other machines,
> particularly any configured as public workstations? A quick look at
> your world-readable dotfiles shows that you also have a very
> non-standard session configuration. How did you attempt to logout
> when it failed? And what was the symptom?
>
> Your bug report only shows the partial output of a ps, which is not
> enough information to determine anything. If you supply us with more
> information, we may be able to find out what happened.
> -Mark Rosenstein
> DCNS Development
I should also note that I often have the same problem logging out of
RS/6000s.
It's not just deathtongue, that was the machine I happened to have been
on when Marc reported the bug for me. I've had similar problems on
cluster Suns and RS6000s and on air-traffic-controller, the RS6000 in
the OLC office.
I don't think I made any changes to the session in my dotfiles; I do
have a lot of configurations and run extra stuff but the basic session
stuff oughtn't be any different than normal.
What I think happens is that my session_gate doesn't finish dying; it
becomes a zombie and can't be killed. Usually I fix this by su-ing and
killing the X server.
The next time I have the problem I will record all the information I can
think of as being possibly relevant and report it. It hasn't happened
for a couple of days.
-scott
