[SR-Users] Simultaneous usage of t_on_reply() and t_on_failure()

[Henning Westerholt]

Am Mittwoch, 5. September 2018, 14:03:10 CEST schrieb Володимир Іванець:
> I wanted to ask those who know if Kamailio's behavior I'm facing is
> expected or I should make some improvements to the configuration. Kamailio
> version is 5.1.0.
> 
> I have a route where RTPEngine parameters are being collected and
> *rtpengine_offer()* is called. After that *t_on_reply("REPLY_SIP_TO_SIP");*
> followed by the *t_on_failure("FAILURE_SIP_TO_SIP");* are used. The idea is
> to process all responces except 415 or 488 from UAC as usual in
> *onreply_route[REPLY_SIP_TO_SIP]* and use
> *failure_route[FAILURE_SIP_TO_SIP]* to update SDP with *rtpengine_offer()*
> if necessary.
> 
> *onreply_route[REPLY_SIP_TO_SIP]* just goes to *exit;* if *$rs* equals  415
> or 488. This works fine with Htek phone which sends 100, 180 and then 488.
> But I can not see  *failure_route[FAILURE_SIP_TO_SIP]* execution for calls
> to Zoiper which replies with 100 and immediately 415.
> 
> t_on_failure(failure_route) documentation says: "Sets failure routing
> block, to which control is passed after a transaction completed with a
> negative result but before sending a final reply." and to be honest I don't
> really understand how lacking of responce prevents  failure_route from
> executing.

Hello,

I did not understand you completely here. Did you receive a 415 and the 
failure route is not entered, or you did not receive a response as stated in 
the last sentence? 

If the failure_route is entered for the 488, then it should be also entered 
for the 415. I am not aware of a special handling of this code in tm. (Only 
that 415 is one of the replies that get a priority comparing to normal 4xx.)

I would suggest to try with debugging enabled and compare the logs for a 
working, and for a non-working call. You could also try to concentrate on one 
problem first, like the failure_route topic, and then combining this with the 
reply_route.

Best regards,

Henning

-- 
Henning Westerholt
https://skalatan.de/blog/