[asterisk-dev] [Code Review] 4006: Test to validate 503 not generated on INVITE retransmissions

Torrey Searle reviewboard at asterisk.org
Fri Sep 19 10:09:52 CDT 2014



> On Sept. 19, 2014, 2:34 p.m., Mark Michelson wrote:
> > There are a couple of problems with this test:
> > 
> > 1) It's quite a bit more complicated than it needs to be. What's actually being tested here is that Asterisk does not send a 503 in addition to a 486 on an INVITE retransmission. This only requires a UA to send and retransmit the INVITE and Asterisk. This should be doable entirely within a SIPp scenario and does not need voxcallcontrol, a Perl AGI load balancer, or grepping any logs.
> > 2) The included sip.conf file is about 95% comments. This makes reviewing the config file much more difficult than it needs to be.
> > 
> > I think this test can be accomplished using the SIPpTestCase and defining the test details entirely within test-config.yaml, with no run-test file necessary. If you're unfamiliar with how this is done, there are several examples of this in the testsuite. A simple example can be found at tests/channels/SIP/directrtpsetup/test-config.yaml. In that test, there is a test-modules section that tells the testsuite to use sipp.SIPpTestCase as the main test object for the test (it also has an unnecessary add-test-to-search-path option set. You can ignore that). The corresponding test-object-config provides details about the SIPp scenarios to run. In that test, there are two scenarios run, but I suspect that for your test, you would only need a single scenario to run. If you're curious about what options are available for configuring the SIPpTestCase, you can look in sample-yaml/sipptestcase-config.yaml.sample for some more details. Your test can pass or fail based on whether the SIPp scenario succeeds or fails.

The problem is that I have to ignore INVITES for awhile by using a pause block.  The 503 happens during that ignoring time after loosing a day and a half I couldn't find a better way then to pause then grep the log to see what happened.  If you can make a version that can work entirely in sipp let me know :-)

agiload balancer and voxcallcontrol is there by mistake, (thought I had removed all references to them already)  I'll check again and remove the references and post the new patch


- Torrey


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/4006/#review13368
-----------------------------------------------------------


On Sept. 19, 2014, 8:09 a.m., Torrey Searle wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/4006/
> -----------------------------------------------------------
> 
> (Updated Sept. 19, 2014, 8:09 a.m.)
> 
> 
> Review request for Asterisk Developers.
> 
> 
> Bugs: ASTERISK-24335
>     https://issues.asterisk.org/jira/browse/ASTERISK-24335
> 
> 
> Repository: testsuite
> 
> 
> Description
> -------
> 
> This is a test for the test suite to reproduce the issue described in ASTERISK-24335
> 
> 
> Diffs
> -----
> 
>   /asterisk/trunk/tests/channels/SIP/tests.yaml 5608 
>   /asterisk/trunk/tests/channels/SIP/invite_retransmit/test-config.yaml PRE-CREATION 
>   /asterisk/trunk/tests/channels/SIP/invite_retransmit/sipp/A_PARTY.xml PRE-CREATION 
>   /asterisk/trunk/tests/channels/SIP/invite_retransmit/run-test PRE-CREATION 
>   /asterisk/trunk/tests/channels/SIP/invite_retransmit/configs/ast1/sip.conf PRE-CREATION 
>   /asterisk/trunk/tests/channels/SIP/invite_retransmit/configs/ast1/extensions.conf PRE-CREATION 
> 
> Diff: https://reviewboard.asterisk.org/r/4006/diff/
> 
> 
> Testing
> -------
> 
> test passes when 4003 patch applied, fails when patch not applied
> 
> 
> Thanks,
> 
> Torrey Searle
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20140919/7072fa09/attachment-0001.html>


More information about the asterisk-dev mailing list