[2932] in Athena Bugs
6.3B: xlogin spin-dry loop
daemon@ATHENA.MIT.EDU (probe@ATHENA.MIT.EDU)
Sun Aug 20 13:46:55 1989
From: <probe@ATHENA.MIT.EDU>
Date: Sun, 20 Aug 89 13:46:43 -0400
To: bugs@ATHENA.MIT.EDU
Reply-To: Richard Basch <probe@ATHENA.MIT.EDU>
I hit it again (running the latest xlogin). Apparently, my last bug
report about xlogin not being up-to-date on the AFS cell was incorrect;
everything has been reconciled.
Anyway, since I am definitely running the latest version (as in, I
logged out and back in), including a deactivation, I can insure this
xlogin "spin-dry" problem still persists.
System:
CRONOS.MIT.EDU VAX 6.3B AFS
Symptoms:
I shrank my xlogin window to be tiny (only enough for the Hide
button to remain), and hid the window under my Logout button.
I had left for a while and was going to re-kinit... My AFS
tokens had expired and kinit is on AFS land, so I would expect
the message "AFS tokens have expired for uid 14640", or
something like that... Anyway, I was not going to be bothered
by the console window since I dealt with it appropriately :)
Well, just at about the time that those messages should have
appeared, my system got very slow; xlogin had run-away again.
This time, I gcore'd the xlogin process:
__XtTextGetText(8ba00,9701c) from _DisplayTextWindow+bf
_DisplayTextWindow(8ba00) from _XtTextReplace+99
_XtTextReplace(8ba00,2b,2b,345a0) from _WriteToConsole+5b
_WriteToConsole(89b40,4bba8) from _ReadSession+b2
_ReadSession(4b8c0,939b4,7fffe5b4) from 12678
__XtwaitForSomething(939a0) from 12a97
_XtRemoveInput(60800) from _XtAppNextEvent+30
_XtAppNextEvent(60800,7fffe61c) from 536
_main(3,7fffe794,7fffe7b4) from start+3d
The core-dump is sitting in /mit/bitbucket/core.xlogin.spin.vax
Suspicion: (I am not going to test it right now)
Attempts to hide the console window by shrinking it and moving
it out of the way, may cause the xlogin process to run-away
when the next console message appears. Unfortunately, since
the console window can get annoying, I expect many people will
try this...
-Richard
