 |
Thinjon100
Apprentice
Joined: 12 Jul 2004
Posts: 190
Location: Canada
|
 Posted: Fri Jul 13, 2007 8:14 pm
[1.34] Having trouble with processor lock-ups
|
Unfortunately I can't use the built-in bug report for this one. I'm boggled by this problem, because it only started 2 days ago.
My CMud will, at "seemingly" random points, lock up the processor at 99% usage and stop responding.
Now, admittedly, I am putting an immense strain on Cmud... I normally have 4-5 sessions connected simultaneously, one of which maintains a mapper, and another of which has 5 capture windows. The one running the mapper is very script-minimal. The one with capture windows has admittedly more scripting, having capture triggers for 4 windows and a mini-database feeding the 5th. The other sessions have almost no triggers at all.
It seems that the processor lock-ups typically occur during a window-switching operation. One occasion I was switching from one session window to another, another when I attempted to switch from the Package Editor window to one of the session windows, and another instance when I was in another program entirely and attempted to switch back to CMud.
I've tried "waiting it out" to no avail, and end up having to kill the process directly from Task Manager.
Any tips for tracking this down? Any more information I can supply to help debug? |
|
_________________
If you're ever around Aardwolf, I'm that invisible guy you can never see. Wizi ftw! :) |
 |
|
|
Zugg
MASTER
Joined: 25 Sep 2000
Posts: 23379
Location: Colorado, USA
|
Posted: Sun Jul 15, 2007 7:14 pm
|
It's hard when it's random like that. If it was something you could cause to happen when offline, then you could send me all of your *.PKG files along with the instructions on getting it to hang and I could run it in the Delphi IDE and see where it is looping.
If it's using 99% of the processor, then it's infinite looping somewhere. Check to be sure you don't have a #WHILE or #UNTIL loop that might be infinite looping. Also check for #WAIT command usage that might be causing trouble. |
|
|
|
|
Thinjon100
Apprentice
Joined: 12 Jul 2004
Posts: 190
Location: Canada
|
Posted: Sun Jul 15, 2007 11:34 pm
|
After some testing, I narrowed it down a bit.
I don't use ANY #UNTIL, #WHLIE, or #WAIT anywhere in my scripts, so I know it's not that.
Oddly enough, the lockups only occur when this trigger is enabled:
| Code: |
#TR {^-->} {
#PSUB {%time("hh:mm:ss")} 0 1
}
|
As you can see, I just wanted it to change the --> arrow in the front of stock wiznet lines to a hh:mm:ss> timestamp.
After disabling this trigger, I've had zero lockups. With it re-enabled, it locks up again.
If it means anything, the wiznet lines are being #CAPped to the window where this trigger is. |
|
_________________
If you're ever around Aardwolf, I'm that invisible guy you can never see. Wizi ftw! :) |
|
|
|
Zugg
MASTER
Joined: 25 Sep 2000
Posts: 23379
Location: Colorado, USA
|
Posted: Mon Jul 16, 2007 5:18 pm
|
Strange...nothing wrong with that trigger. Try changing the trigger script to just #NOOP and see if it hangs then. That will tell me if it's the trigger processor, or the #PSUB command.
|
|
|
|
|
Thinjon100
Apprentice
Joined: 12 Jul 2004
Posts: 190
Location: Canada
|
Posted: Mon Jul 16, 2007 8:45 pm
|
Changed to #TR {^-->} {#NOOP} and reenabled. Will let you know if it locks up.
Thanks, Zugg :) |
|
_________________
If you're ever around Aardwolf, I'm that invisible guy you can never see. Wizi ftw! :) |
|
|
|
Thinjon100
Apprentice
Joined: 12 Jul 2004
Posts: 190
Location: Canada
|
Posted: Tue Jul 17, 2007 5:48 pm
|
OK, I've run for a day roughly now, with the above #NOOP trigger... no lockups or anything. I think it's #PSUB.
|
|
_________________
If you're ever around Aardwolf, I'm that invisible guy you can never see. Wizi ftw! :) |
|
|
|
|
|
|
