[14378] in Athena Bugs
sun4 7.7V: Solaris
daemon@ATHENA.MIT.EDU (Kevin L. Mitchell)
Wed Jun 5 16:03:07 1996
To: bugs@MIT.EDU
Date: Wed, 05 Jun 1996 16:03:00 EDT
From: "Kevin L. Mitchell" <klmitch@MIT.EDU>
System name: starkiller
Type and version: SPARC/5 7.7V (4 update(s) to same version)
Display type: cgthree
What were you trying to do?
I was running several things at once, including one instance of emacs19
that had several windows open, vtwm.gamma, xload, xclock, and several xterms.
In two of the xterms I had gdb running on a client/server networking
application.
What's wrong:
The machine became heavily loaded; focus response was an observable
time behind mouse movements (as much as 2 or 3 seconds). The machine finally
just hung for about 90 seconds. It wouldn't respond to stop-A, it wouldn't
respond to pings. It finally unwedged, but this has happened twice before on
this machine (in one case, it had to be power-cycled because it wouldn't
listen to stop-A after a "sync" command was issued; in both cases, the only
way to get the machine's attention was to unplug the keyboard and plug it back
in), and once on another machine (that was running only MAE, as far as I could
tell). The first time this happened on starkiller, after getting the machine's
attention (by unplugging the keyboard and plugging it back in), we issued a
"sync" command, which proceeded to flood the screen with "interrupt 14 not
processed" or something similiar; nothing could seem to break the machine out
of the loop back to an OK prompt, so it was power-cycled. In this case, and
in the case of the machine running MAE, a message appeared in the console that
looked like this:
15:50 NOTICE: zs3:ring buffer overflow
What should have happened:
Well, I suppose that a machine should never hang this hard.
Please describe any relevant documentation references:
Well, common sense dictates that a machine should not just hang up and
sulk like a child...
