Daniel-Constantin Mierla writes:
Maybe you would also want to tune the timeout with the modparam:
modparam("rtpengine", "rtpengine_disable_tout", 5)
So detection of unavailable rtpproxy is fast, otherwise it is 60 sec by default, so you may still experience some slow start per child process.
I understand from the param description that it tells how to behave AFTER rtp proxy has been marked disabled:
Once an RTP proxy was found unreachable and marked as disabled, the rtpengine module will not attempt to establish communication to that RTP proxy for rtpengine_disable_tout seconds.
The problem that I have experienced is that it takes long time (2 minutes or so) before rtpengine modules disables a non-responding rtp proxy.
Also, no matter what the module does, it must not make the whole sip proxy non-responsive for any number of seconds.
-- Juha