Hi, I've solved problem removing tcp NAPTR record from DNS: /domain.it. IN NAPTR 2 0 "s" "SIP+D2T" "" _sip._tcp.domain.it.
/it seems there's some problem with Snom phones registered in tcp.../ / regards Gmb ha scritto: > Gmb ha scritto: >> Dale Worley ha scritto: >>> On Fri, 2009-03-13 at 12:11 +0100, Gmb wrote: >>> >>>> Hi, >>>> I'm using sipXecs v. 3.10.2 in HA configuration, I've some problems >>>> with SNOM 300/360 phones firmware version 6.5.18, >>>> they appears registered but calls to them go directly to VM. >>>> I've noticed that some SNOM phone are registered with >>>> "transport=tcp" and someone else without "transport=tcp" on the >>>> same server, >>>> master or slave, the error appears only on phone registered in >>>> master server and registered with "transport=tcp". >>>> >>>> "5271"<sip:[email protected]> >>>> <sip:[email protected]:2089;transport=tcp;line=0ok9mp3m> >>>> 2151 sipxslave.sipx.domain.it >>>> "5265"<sip:[email protected]> >>>> <sip:[email protected]:2084;transport=tcp;line=ixyn9621> >>>> 1731 sipxslave.sipx.domain.it >>>> "5269"<sip:[email protected]> >>>> <sip:[email protected]:2059;line=g9ircgf6> 2494 >>>> sipxslave.sipx.domain.it >>>> "5267"<sip:[email protected]> >>>> <sip:[email protected]:2088;transport=tcp;line=ve1sz3a8> >>>> 354 sipxmaster.sipx.domain.it >>>> "5239"<sip:[email protected]> >>>> <sip:[email protected]:2059;line=4u4y9gvi> 588 >>>> sipxmaster.sipx.domain.it >>>> "5273"<sip:[email protected]> >>>> <sip:[email protected]:2059;line=d91l1t79> 2872 >>>> sipxslave.sipx.domain.it >>>> "5262"<sip:[email protected]> >>>> <sip:[email protected]:2090;transport=tcp;line=kr4lp4nz> >>>> 887 sipxmaster.sipx.domain.it >>>> "5272"<sip:[email protected]> >>>> <sip:[email protected]:2085;transport=tcp;line=ol0hm06l> >>>> 2405 sipxslave.sipx.domain.it >>>> "5259"<sip:[email protected]> >>>> <sip:[email protected]:2055;line=jl2mkqgc> 1239 >>>> sipxmaster.sipx.domain.it >>>> >>>> 5267, 5262 are unreachable phones. >>>> >>> >>> What about 5272? >>> >>> If I recall correctly, there is a problem with Snom phones registering >>> TCP addresses which are not correct, in that the phone is not listening >>> for TCP on the address and port that its Contact address specifies. >>> (The phone expects incoming calls only on the TCP connect which it used >>> to register.) >>> >>> One test would be to do a packet capture, filtering on the address/port >>> specified in the registration -- Since the call goes immediately to >>> voicemail, it appears that when the proxy attempts to connect to the >>> phone, the phone is sending an immediate rejection of the connection >>> (RST packet). This should be apparent on the packet trace. >>> >>> None of this solves your problem, of course. Can you determine the >>> difference in configuration between the phones that register with >>> "transport=tcp" and those that do not? >>> >>> Dale >>> >>> >>> >>> >> Hi Dale, >> I've done new tests on my system, it seems that this problem affect >> both master and slave server but >> not at the same time and only with SNOM phones registered with >> transport=tcp parameter. In previous example >> phones registered with transport=tcp parameter on sipxmaster server >> are unreacheble, other phones registered >> on sipxslave with transport=tcp parameter on sipxslave work perfectly >> (also 5272). >> After timeout and a new registration the situation has changed, now >> the unreacheble phones are on sipxslave. >> A packet capture on registration port shows that server affected from >> this problem doesn't send any packet to phone, on the >> contrary capture on reacheble phone shows a normal message exchange. >> It seems that SNOM phones are listening >> on right port but sipx doesn't send messages to phones... >> >> > > _______________________________________________ sipx-users mailing list [email protected] List Archive: http://list.sipfoundry.org/archive/sipx-users Unsubscribe: http://list.sipfoundry.org/mailman/listinfo/sipx-users
