I am looking at http://track.sipfoundry.org/browse/XECS-839 "Can't do
gateway redundancy properly".  The problem seems to be that our internal
serial forking timer (default 20 seconds) causes us to step through
gateways if a call is not answered in 20 seconds.  This is not desirable
behaviour.  20 seconds is too short - the far end's VM doesn't even kick
in.

A simple way of improving things is to specify a longer timer for
gateways.  This can be done by putting an Expires tag in the
fallbackrules.xml file.  Changing it to 60 seconds gives a much better
user experience.

Even better might be not to implement any timer for calls going through
a gateway - only an explicit (failure) response from the gateway would
cause us to try the next one.  Does this seem preferable? (though it
sounds easy I don't think it's quite as easy to do)

Please let me know what you think.
Carolyn
_______________________________________________
sipx-dev mailing list
[email protected]
List Archive: http://list.sipfoundry.org/archive/sipx-dev
Unsubscribe: http://list.sipfoundry.org/mailman/listinfo/sipx-dev

Reply via email to