[asterisk-dev] [Code Review] 4411: testsuite: fix a number of tests where Asterisk does not shutdown gracefully

[Corey Farrell]

> On Feb. 10, 2015, 12:29 p.m., Mark Michelson wrote:
> > The only caveat here is that you may want to watch automated runs of the SIP info_dtmf test to be sure that on awful hardware the 5 second Wait() isn't too short for the test to complete. I suspect it will be okay though.
> > 
> > On a side note, I have another test to add to my list of tests that could be rewritten to not rely on timing, though :)

If this turns out to be a problem then we can change the Wait() back to 5 seconds.  If that is the case then I think it will be appropriate to increase the timeout for 'core stop gracefully' from 5 seconds to 10.


- Corey


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviewboard.asterisk.org/r/4411/#review14433
-----------------------------------------------------------


On Feb. 9, 2015, 12:50 p.m., Corey Farrell wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviewboard.asterisk.org/r/4411/
> -----------------------------------------------------------
> 
> (Updated Feb. 9, 2015, 12:50 p.m.)
> 
> 
> Review request for Asterisk Developers.
> 
> 
> Repository: testsuite
> 
> 
> Description
> -------
> 
> * Add Hangup() to priority after Dial() where needed.  This prevents auto-fallthrough from playing 10 seconds of BUSY or CONGESTION tone.
> * Decrease Wait(10) to Wait(5) in tests/channels/SIP/info_dtmf.
> * Maintain list of AGI connections where needed so they can all be agi.finish().
> * Replace calls to reactor.stop() with self.stop_reactor(), remove test.start_asterisk()/test.stop_asterisk() from main().
> * Delay self.stop_reactor() in tests/channels/SIP/sip_tls_call by 2 seconds.  This gives the calls enough time to end and avoid shutdown timeout.
> 
> 
> Diffs
> -----
> 
>   /asterisk/trunk/tests/funcs/func_srv/run-test 6377 
>   /asterisk/trunk/tests/funcs/func_presencestate/run-test 6377 
>   /asterisk/trunk/tests/fastagi/stream-file/run-test 6377 
>   /asterisk/trunk/tests/fastagi/database/run-test 6377 
>   /asterisk/trunk/tests/fastagi/control-stream-file/run-test 6377 
>   /asterisk/trunk/tests/channels/SIP/sip_tls_call/run-test 6377 
>   /asterisk/trunk/tests/channels/SIP/sip_srtp/run-test 6377 
>   /asterisk/trunk/tests/channels/SIP/sip_cause/configs/ast1/extensions.conf 6377 
>   /asterisk/trunk/tests/channels/SIP/secure_bridge_media/run-test 6377 
>   /asterisk/trunk/tests/channels/SIP/noload_res_srtp/run-test 6377 
>   /asterisk/trunk/tests/channels/SIP/info_dtmf/configs/ast1/extensions.conf 6377 
>   /asterisk/trunk/tests/channels/SIP/hangupcause/configs/ast1/extensions.conf 6377 
>   /asterisk/trunk/tests/channels/SIP/generic_ccss/configs/ast1/extensions.conf 6377 
> 
> Diff: https://reviewboard.asterisk.org/r/4411/diff/
> 
> 
> Testing
> -------
> 
> Ran all effected tests against Asterisk 11 with REF_DEBUG.  Prior to these fixes graceful shutdown of Asterisk timed out, causing reference leaks to be reported.  These tests now shutdown gracefully and have no reference leaks.
> 
> 
> Thanks,
> 
> Corey Farrell
> 
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-dev/attachments/20150211/c151ee6c/attachment.html>