OM list:

If you search for an existing meeting when creating a new meeting, it
can cause a race condition that takes down the webserver... At least in
my case.

Humm.. that's not right. SugarCRM seems to run fine with everything
else. I don't use a php opcache, because I know it doesn't play nice
with the threaded php model I'm using (at least in the short term - and
besides it performs well).

"GET
/sugar/cache/jsLanguage/openmeetings/en_us.js?v=iMjYhjAUH0YotcqG1cZeKw
HTTP/1.1" 200 711

Result: Webserver only goes out to lunch. Session dead.

PID USER          NR  NI  VIRT   RES   SHR  S CPU   MEM TIME COMMAND 
3217 apache    20   0  783m 156m 8912 S  99.7 10.5   8:31.31 apache2

What do you think the culprit could be. My webserver or some kind of php
endless loop? Scripts can run for up to 5-10 mins in my configuration
(due to SugarCRM upgrades / installs / large queries / batch jobs etc),
so I'm expecting it to unlock itself in another 10 mins or so.

Cheers,

Luke

The new openmeetings nightly build works great however. I just would
like to know how to dial-down the debugging a little bit and then we are
ready for production ;)

Reply via email to