[11989] in Athena Bugs
decmips 7.6G: mwm
daemon@ATHENA.MIT.EDU (Mary E. Gibson)
Wed Apr 27 11:11:27 1994
To: bugs@MIT.EDU
Date: Wed, 27 Apr 94 11:11:21 EDT
From: Mary E. Gibson <mgibson@MIT.EDU>
System name: monisha
Type and version: KN02ca 7.6G (1 update(s) to same version)
Display type: PMAG-DV
What were you trying to do?
Infrequently I will find that I have multiple mwm's running
at the same time. This has been going on for a couple of years (see
old email on this attached below), and doesn't happen very often.
This is not real serious, since I can just kill the extra mwm's, but
I wanted y'all to know it still happens.
What's wrong:
Workstation gwets bogged down, and is very slow.
What should have happened:
I suppose mwm should have been terminated in my previous session.
Please describe any relevant documentation references:
Old email:
Received: from ATHENA.MIT.EDU by po6.MIT.EDU (5.61/4.7) id AA09563; Tue, 22 Sep 92 23:17:46 EDT
Received: from MIT.MIT.EDU by Athena.MIT.EDU with SMTP
id AA29495; Tue, 22 Sep 92 23:17:42 EDT
Received: from BAD-LOUD-MUSIC.MIT.EDU by MIT.EDU with SMTP
id AA21659; Tue, 22 Sep 92 23:17:34 EDT
Received: by bad-loud-music (4.1/4.7) id AA23232; Tue, 22 Sep 92 23:17:30 EDT
Message-Id: <9209230317.AA23232@bad-loud-music>
To: Mary E. Gibson <mgibson@Athena.MIT.EDU>
From: Chris VanHaren <vanharen@MIT.EDU>
Subject: Re: Two mwm's running again
Date: Tue, 22 Sep 92 23:17:30 BST
Sender: vanharen@MIT.EDU
> Hi Chris,
>
> Here we go again! This happened Thursday or Friday, but hadn't
> happened before since the new release. Maybe it's because of
> something I did recently.....
> Should I just kill one and go on? Do you see anything else that looks odd - I
> don't know what everything that runs does. Thanks,
Nothing else looks odd to me. Everything seems fine, except for that
stray mwm. Yeah, you can just go ahead and kill it. Obviously, the one
that you want to kill is the one that is just spinning, eating up CPU
time and performance. In this case, that is obviously:
> PID USERNAME PRI NICE SIZE RES STATE TIME WCPU CPU COMMAND
> 23862 mgibson 96 4 1704K 28K run 8156:39 83.59% 83.59% mwm
As you can see, it has racked up a considerable amount of CPU time,
which is not normal, of course. So, if you haven't already killed it,
you can go ahead and do so.
I doubt that there is anything that you did specifically to cause this
-- it seems to be a race-condition (a fancy term for a timing problem,
if you aren't familiar with the term, though I suspect you've heard it
before). As such, it's not something that is necessarily easily
reproduced, nor is there something which will always cause it... It's
just sort-of luck, or bad luck, depending on how you look at it.
Anyway, I'm glad it's not happening so much anymore. It seems to occur
less frequently on faster machines than on slower ones, so your upgrade
from the RT (was it? or VAX 3100?) that you had to the DECstation 5000
probably also lessened the chance that you will run across it.
Anyway, thanks for reporting it. I think I may have a fix for it
finally, which hopefully will appear in the next release (whenever that
might be).
-C.
