Question, do SST help the situation? Is it widely accepted enough in the protocol to provide some backup mechanism to maintaining dialog state in the event of a lost BYE.
BTW, how is a BYE accounted for in ACC when generated locally because of an expired dialog? On Thu, Mar 5, 2009 at 10:00 AM, Bogdan-Andrei Iancu <[email protected] > wrote: > Iñaki Baz Castillo wrote: > > 2009/3/5 Bogdan-Andrei Iancu <[email protected]>: > > > >> Hi Inaki, > >> > >> This is an old, hot topic. > >> > >> There are many services that are more appropriate via a B2BUA (like > acc, > >> dialog stuff, security, etc) - last time the discussion started from the > >> question if a proxy is the best place to do accounting. > >> > >> In all the case is about compromising - how much you are willing to > lose. > >> You may loose time/resources to build and implement a platform were > there is > >> no way for bad thinks to happen (you deal with all corner cases) - and > you > >> end up with a huge platform, very complex, difficult to maintain, > expensive > >> to run, etc . > >> > >> Or you can loose some corner cases and build a simpler and more > efficient > >> platform. > >> > > > > Yes Bogdan, also the approach described in the wiki is really > > appropiate and valid when the caller or callee is a PSTN gateway which > > sends a BYE if RTP is lost. > > > > But as always, pure "Internet" calls (between users) are really > > difficult to monitorize and control by a proxy. > > > > I agree on this. But this is SIP - from signalling level only you do not > have information about the call status (during the call). > > Regards, > Bogdan > > _______________________________________________ > Users mailing list > [email protected] > http://lists.opensips.org/cgi-bin/mailman/listinfo/users >
_______________________________________________ Users mailing list [email protected] http://lists.opensips.org/cgi-bin/mailman/listinfo/users
