[asterisk-dev] [Code Review] 3562: chan_sip: Start session timer at 200, not at INVITE.
Matt Jordan
reviewboard at asterisk.org
Tue May 27 09:26:16 CDT 2014
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/3562/#review11972
-----------------------------------------------------------
Ship it!
Would it be worthwhile to incorporate the SIPp scenario into the Test Suite?
- Matt Jordan
On May 23, 2014, 5:03 a.m., wdoekes wrote:
>
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/3562/
> -----------------------------------------------------------
>
> (Updated May 23, 2014, 5:03 a.m.)
>
>
> Review request for Asterisk Developers.
>
>
> Bugs: ASTERISK-22551
> https://issues.asterisk.org/jira/browse/ASTERISK-22551
>
>
> Repository: Asterisk
>
>
> Description
> -------
>
> RFC says Asterisk should start counting session timers at 200. Asterisk starts at INVITE.
>
> For short intervals (e.g. 90 secs) and long ringing times (e.g. 30 secs), this means that
> a caller with refresher=uac will get disconnected by Asterisk before it has a chance to
> send a refreshing reINVITE.
>
> Reproduced using:
> https://github.com/ossobv/sipp-scenarios/blob/master/INVITE-test-session-refresher-uac.xml
>
>
> Diffs
> -----
>
> /branches/1.8/channels/chan_sip.c 414344
>
> Diff: https://reviewboard.asterisk.org/r/3562/diff/
>
>
> Testing
> -------
>
> Tested against the scenario mentioned above.
>
> Before the patch, Asterisk aborts the call during <pause milliseconds="45000"/>
> After the patch, the scenario completes succesfully.
>
> The scenario tests both:
> - asterisk not killing the dialog too early, and
> - killing it when expected
>
> After the scenario and a reasonable time, no excess 'sip show objects' were seen.
>
>
> Thanks,
>
> wdoekes
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20140527/d861d7a3/attachment.html>
More information about the asterisk-dev
mailing list