[asterisk-dev] [Code Review] 2796: Testsuite: sip_hold_direct_media adaptations for Asterisk 12

[Mark Michelson]

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/2796/#review9526
-----------------------------------------------------------


The scenarios used for versions <= 11  (assuming they're the same ones that are in the current trunk of the testsuite) don't actually appear to use direct media. For instance, in phone_A.xml, after completing the initial INVITE transaction with Asterisk, the next expected event is the BYE to end the call. I would expect the same three re-invite transactions that phone_A_12.xml has (first to begin direct media, second to reinvite back to Asterisk for MOH, third to reinstate direct media). In fact, I wouldn't be surprised to see a fourth reinvite to direct the media back to Asterisk again when phone B sends its BYE to Asterisk.

Similarly the non-12 phone B scenarios don't appear to have the reinvites I would expect to see in order to initiate direct media.

I know the testsuite has required many tweaks to account for differences in Asterisk 12, but I would expect that the SIPp scenarios used for this test should be the same between different Asterisk versions. Here is what I suggest doing:

1) Using the phone_A.xml and any of the various non-12 phone B scenarios, manually run against Asterisk 11 and try to debug why it is that direct media reinvites do not get sent to the scenarios. If the reason is legitimate, then we likely should also not be offering direct media in 12 given the same setup. If Asterisk 11 *should* be offering direct media, then Astrisk 11 (and possibly 1.8) will need to be adjusted.

Based on the result of this, the path can branch a couple of different ways:

2a) If Asterisk 12 should not be offering direct media in this scenario, then the test will need to be adjusted in a way that will allow for direct media to be used. Once the test is altered properly, then the scenarios between Asterisk 11 and Asterisk 12 should be identical. If we find that the same scenarios do not pass in both 11 and 12, then we can look at the differences and try to rectify them.

2b) If Asterisk 11 should be offering direct media in this scenario, then Asterisk 11 should be altered to properly send the necessary reinvites. Like with 2a, I would expect that the same scenarios should then work with both Asterisk 11 and Asterisk 12. If not, then we need to find what is different and figure out how to resolve the issue.

- Mark Michelson


On Aug. 26, 2013, 8:18 p.m., jrose wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/2796/
> -----------------------------------------------------------
> 
> (Updated Aug. 26, 2013, 8:18 p.m.)
> 
> 
> Review request for Asterisk Developers, Joshua Colp, kmoore, Matt Jordan, and Mark Michelson.
> 
> 
> Bugs: https://issues.asterisk.org/jira/browse/ASTERISK-22217
>     https://issues.asterisk.org/jira/browse/https://issues.asterisk.org/jira/browse/ASTERISK-22217
> 
> 
> Repository: testsuite
> 
> 
> Description
> -------
> 
> On Friday I committed a patch which addressed some bugs with holding in Asterisk 12 while using native RTP bridges and directmedia. As part of that effort, the SIP hold tests in the testsuite were split up and divided into tests which used direct media and tests which didn't use direct media. At the time, Asterisk 12 failed the tests which used direct media. This patch fixes those tests by making the test use Asterisk 12 specific sipp scenarios (which were based on the existing scenarios). The main difference between the Asterisk 12 scenarios and their older counterparts was always just the addition of more expected invites and responses to those invites on account of how directmedia is established in Asterisk 12 both during the initial setup.
> 
> 
> Diffs
> -----
> 
>   /asterisk/trunk/tests/channels/SIP/sip_hold_direct_media/sipp/phone_B_12_media_restrict.xml PRE-CREATION 
>   /asterisk/trunk/tests/channels/SIP/sip_hold_direct_media/sipp/phone_B_12_IP_restrict.xml PRE-CREATION 
>   /asterisk/trunk/tests/channels/SIP/sip_hold_direct_media/sipp/phone_B_12_IP_media_restrict.xml PRE-CREATION 
>   /asterisk/trunk/tests/channels/SIP/sip_hold_direct_media/sipp/phone_A_12_type2.xml PRE-CREATION 
>   /asterisk/trunk/tests/channels/SIP/sip_hold_direct_media/sipp/phone_A_12.xml PRE-CREATION 
>   /asterisk/trunk/tests/channels/SIP/sip_hold_direct_media/run-test 4044 
> 
> Diff: https://reviewboard.asterisk.org/r/2796/diff/
> 
> 
> Testing
> -------
> 
> Ran modified tests against Asterisk 12.
> Ran modified tests against Asterisk 11.
> 
> Repeated this process many times to make sure the results were consistent.
> Tracked the invites in 12 against the code that was generating them.
> 
> 
> Thanks,
> 
> jrose
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20130827/ce9beed6/attachment.htm>