[test-results] [Bamboo] No agents to build plan Asterisk SCF - Temp - sip - Ubuntu Maverick - amd64

Bamboo bamboo at asterisk.org
Wed Aug 15 14:40:48 CDT 2012


-------------------------------------------------------------------------------
ASCFTEMP-SIP-UBUNTUMAVERICKAMD64-15 has been queued, but there's no agent capable of building it.
-------------------------------------------------------------------------------

http://bamboo.asterisk.org/browse/ASCFTEMP-SIP-UBUNTUMAVERICKAMD64/log

--------------
Code Changes
--------------
Brent Eagles <beagles at digium.com> (42310a3a78e9b4eb3372c48ebb36c2698b40b906):

>FINALLY uncovered a huge source of memory consumption. Once things got to a
>certain point in the call setup sequence a lot of timers started to be created.
>Once 500 living calls (1000 session objects) memory would fly through the roof.
>

Brent Eagles <beagles at digium.com> (d070d485e3030bcd5e4c1a6820e44ea0223cd754):

>Reverting change made in commit 3dd0ab1b6ac94f6e8cc41a70e568ab9fbcad17ff.  This
>causes race conditions on pjsip memory pools when the invite session is
>destroyed. It appears that if everything else is working properly this code
>should not be necessary anyways.
>
>The other change introduces a tiny sleep the pjsip polling thread to encourage
>yields in a more timely manner. Otherwise, if a system manages to load up pjsip
>with a lot of waiting work, this thread will dominate things. Something to look
>at: even if this thread is running crazy, on a multi-core system thread pool
>worker threads should make some headway, but this simply isn't happening. There
>is some kind of contention occurring.
>

Brent Eagles <beagles at digium.com> (df0edc0b3f42960b1fc2fd65ab19fb8b1f120406):

>Fix a race condition where the dialog may disappear while we are using it.
>


--
This message is automatically generated by Atlassian Bamboo
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/test-results/attachments/20120815/f9f62138/attachment-0001.htm>


More information about the Test-results mailing list