On a system with ALL addresses pointing to ONE interface and ONLY one interface intended to be used by Sipxecs, freeswitch still (auto) picks the wrong one.
This will request will only insure that freeswitch uses the SAME address as all other components configured on this server. Port 15060 is setup in the same config file, so why not the IP address. I did read the freeswitch manual briefly and i agree that multi-interface work on freeswitch is non-trivial. Just this fixing of the address will ensure that freeswitch has the same worldview as the other components. (I intend to use OpenSBC as a separate SBC on the other interface, and currently it partly works, but it still needs some more work). kind regards, Nico On Wed, 29 Sep 2010 09:08:00 -0400, Tony Graziano <[email protected]> wrote: > It does not address how the proxy will glom onto all the interfaces in a > system unless additional coding takes place. This will always create an > issue, especially if the startup script fails to run as a result due to > errors... > > How the services glom onto an interface (no matter which services) still > needs to be addressed. > > On Wed, Sep 29, 2010 at 9:00 AM, Nico (sipxecs) > <[email protected]>wrote: > >> Oops the previous message was sent prematurely... >> Hi Devs, >> >> On a system with multiple interfaces, freeswitch fails to "do the right >> thing" w.r.t. interface choice. >> >> In SIPXECS the ip address is specified on the server page. Due to a >> $${local_ip_v4} in the sofia config file it consistently chooses the >> wrong >> address there. >> The config file is allready generated to reconfigure the sip-port to >> 15060. so this should be a fairly minor fix IMHO. >> >> The request is to insert the configured ip address in the configfile for >> the sip-ip & rtp-ip tags. This makes the media-services work miracously >> in >> a multi homed system. >> instead of missing failing auto attendant, conferences & voicemail. >> It does improve stability though by relying on explicit in stead of >> implicit configuration of core functionality. >> >> Kind regards, >> Nico >> _______________________________________________ >> sipx-dev mailing list >> [email protected] >> List Archive: http://list.sipfoundry.org/archive/sipx-dev/ >> _______________________________________________ sipx-dev mailing list [email protected] List Archive: http://list.sipfoundry.org/archive/sipx-dev/
