[8423] in Athena Bugs

home help back first fref pref prev next nref lref last post

(VS2000) Version 7.3M: ez spins...

daemon@ATHENA.MIT.EDU ("Mark W. Eichin")
Thu Oct 17 15:09:12 1991

Date: Thu, 17 Oct 91 15:10:01 -0400
From: eichin@Athena.MIT.EDU ("Mark W. Eichin")
To: bugs@Athena.MIT.EDU
Cc: Bill Cattey <wdc@Athena.MIT.EDU>

System type, version:	(VS2000) Version 7.3M
System name:		w20-575-something
What's wrong:

Last night, a user came in to the SIPB office, having had dreadful
trouble with ez. She just wanted to get her paper done... anyway, she
had a file of about 5K. 
USER       PID %CPU %MEM   SZ  RSS TT STAT  TIME COMMAND
ackbaree  6633105.7  7.8  901  209 p0 R N  10:43  (runapp)
ackbaree  6685  8.0  3.6  136   95 p0 R     0:00  (ps)
ackbaree  6618  3.4  6.0  548  160 ?  S     0:15 xterm -geometry 80x48+0+30
ackbaree  6610  0.8  6.3  922  167 ?  I     0:19 mwm
ackbaree  6620  0.4  4.3  245  116 p0 S     0:08  (tcsh)
ackbaree  6616  0.3  2.7  409   68 ?  S     0:08 dash -default -logout
ackbaree  6607  0.0  0.4    7    5 ?  I     0:08 session_gate -logout
ackbaree  6602  0.0  0.0   36   18 ?  IW    0:13 sh /etc/athena/login/Xsession 
ackbaree  6626  0.0  0.0  293    0 ?  IW    0:00 zwgc

That 105.7% is a drop from around 150%. Ez was not responding to any
input. This was reproducible - if she started up ez on her file, then
used the arrow keys to scroll down, it would start out slow and
eventually stop reacting altogether. It was handling expose events, it
seems (unless perhaps backing store was set - in any case, hiding it
and exposing it didn't leave holes).
	I sigSTOP'ed it and gcore'd it; eichin:cores/core.ez.eichin
has a copy. I didn't copy the user's file, however if you get in touch
with her, the file "test" in her homedir is a copy of the original
"memory" from the time it happenned.
	The second RT she tried had no such problems. (the first one
powered itself off - giving her no additional confidence :-)
	She had no styles set at all; the paper was mostly long
paragraphs (probably over 1K characters each.) 

What should have happened:

Ez shouldn't spin like that. If you can analyze the core dump, please
let me know what you find.
				_Mark_ <eichin@athena.mit.edu>
				MIT Student Information Processing Board

home help back first fref pref prev next nref lref last post