Okay everyone, I'm sure you are all TERRIBLY anxious to hear how the bulk
call testing went....
We fixed a bug where we were sending fast busy when the phones were
on-hook (not like you'd ever notice, but we still shouldn't do it, so we
got that fixed).
Next, we discovered that the Tormenta card was having trouble timing off
the bulk call tester directly. Curiously, we haven't had any trouble
timing off anything else. We placed an Atlas 550 in the middle and DAX'd
across it, and having done so we were able to place all 24 calls
simultaneously (or staggered) (and answer them simultaneously) repeatedly
for hours at a time with no failures.
Ah, and one more thing that was strange... We would get deadlocks with
the mutexes while the tests were going on, yet when I put in some
debugging code to see where the mutexes were getting frozen up they failed
to get frozen up. I suspect that there may be some issues with mutexes
under glibc 2.1 (RedHat 6.2).
I would REALLY like for someone to try to find out if there really are
issues with mutexes in RedHat 6.2. In any case the problem went away with
the addition of some (what should be unnecessary) checks in the mutexes.
Mark
-------------------------------------------------------------------------
Mark Spencer, President Free and Commercial Linux Support
Linux Support Services, Inc. On the web or on the phone
6700 Odyssey Drive Suite 101-ABF
Huntsville, AL 35806 http://www.linux-support.net
(256) 428-6000 staff@linux-support.net
Toll free: (877) LINUX-ME
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum