[Freeswitch-users] Getting FreeSWITCH to use RTP port + 1 for RTCP
James Mortensen
james.mortensen at synclio.com
Tue Dec 3 20:58:36 MSK 2013
Hello,
We followed the steps in this document to set up TLS:
http://wiki.freeswitch.org/wiki/SIP_TLS#If_your_SIP_profile_fails_to_start_upand
followed these troubleshooting instructions. We have the certs in
${prefix}/conf/ssl
oot at ip-10-188-136-13:/opt/freeswitch-dev/conf/ssl# l
total 12
-rw-r----- 1 root root 579 Dec 3 10:27 cafile.pem
drwxr-x--- 2 root root 4096 Dec 3 10:28 CA
-rw-r----- 1 root root 999 Dec 3 10:28 agent.pem
Also, I've verfied that libssl-dev is indeed installed, version 1.0.1.
When we try to load (or reload) the internal profile with the wss-binding
parameter enabled for port 7066, we see the following output:
nta.c:2389 agent_init_via() nta: agent_init_via: SIP/2.0/udp 50.18.X.X:5080
(sip)
nta.c:2389 agent_init_via() nta: agent_init_via: SIP/2.0/tcp 50.18.X.X:5080
(sip)
nta.c:2389 agent_init_via() nta: agent_init_via: SIP/2.0/ws
10.188.136.13:5066 (sip)
nta.c:2255 nta_agent_add_tport() nta: Via fields initialized
nta.c:2263 nta_agent_add_tport() nta: Contact header created
tport.c:1620 tport_bind_server() tport_bind_server(0x7fbf9c003ca0) to wss/
10.188.136.13:7443/sips
tport.c:1690 tport_bind_server() tport_bind_server(0x7fbf9c003ca0): calling
tport_listen for wss
nua.c:575 nua_set_params() nua: nua_set_params: entering
tport.c:628 tport_alloc_primary() tport_alloc_primary(0x7fbf9c003ca0): new
primary tport 0x7fbf9c005e80
nua_stack.c:529 nua_signal() nua((nil)): sent signal r_set_params
nua_params.c:480 nua_stack_set_params() nua: nua_stack_set_params: entering
soa.c:405 soa_set_params() soa_set_params(static::0x7fbfa8001ca0, ...)
called
tport.c:734 tport_listen() tport_listen(0x7fbf9c003ca0): unknown(pf=2
wss/[10.188.136.13]:7443): No such file or directory
nta.c:2237 nta_agent_add_tport() nta: bind(10.188.136.13:7443;transport=wss):
No such file or directory
nua_stack.c:195 nua_stack_init() nua: initializing SIP stack failed
nua_stack.c:212 nua_stack_deinit() nua: nua_stack_deinit: entering
sresolv.c:190 sres_sofia_update() sres_sofia_update(0x7fbf9c0038a0, -1, -1)
sresolv.c:190 sres_sofia_update() sres_sofia_update((nil), -1, -1)
tport.c:562 tport_destroy() tport_destroy(0x7fbf9c003ca0)
su_epoll_port.c:166 su_epoll_port_deinit()
su_epoll_port_deinit(0x7fbf9c0008c0) called
2013-12-03 17:47:36.815063 [ERR] sofia.c:2545 Error Creating SIP UA for
profile: internal (sip:mod_sofia at 50.18.X.X
:5080;maddr=10.188.136.13;transport=udp,tcp)
The likely causes for this are:
1) Another application is already listening on the specified address.
2) The IP the profile is attempting to bind to is not local to this system.
And when trying to reload the profile, this is what we're seeing:
freeswitch at ip-10-188-136-13> sofia profile internal restart
Invalid Profile [internal]
freeswitch at ip-10-188-136-13>
I'm a little unclear as to whether or not any certs actually need to be
installed on Android. The webrtc.freeswitch.org demo seems to work without
needing to do anything on the client side. What's more, that shouldn't have
any effect on the server's ability to start the profile. What are we
missing here? Thank you!
James
On Wed, Nov 27, 2013 at 10:14 PM, Anthony Minessale <
anthony.minessale at gmail.com> wrote:
> *HINT* you need to use a valid cert for the dtls and the ssl websocket
> Android is 100 times more paranoid about certs and doesn't bother to
> explain.
> A common theme in google code for webrtc to not take the time and effort
> to explain errors.
>
> go to /usr/local/freeswitch/certs
> put both the key and the cert in wss.pem
> put the same key in dtls-srtp.key and the same cert in dtls-srtp.crt
> put a valid chain cert in ca-bundle.crt
>
>
>
>
>
>
>
> On Wed, Nov 27, 2013 at 7:09 PM, James Mortensen <
> james.mortensen at synclio.com> wrote:
>
>> I have limited information at this time, but we're basically seeing RTP
>> packets flowing from Chrome to FreeSWITCH and from FreeSWITCH to Chrome,
>> but very minimal activity between Android and FreeSWITCH.
>>
>> Before my colleague hit the road for the day, we did notice there were
>> two separate INVITES being sent to Android, and that's where we stopped.
>> It is possible there's a port mismatch somewhere, assuming we really did
>> see two INVITES in the FreeSWITCH SIP trace. To be sure, we registered
>> Chrome and Android on different networks so there was absolutely no
>> confusion.
>>
>> I've attached a SIP trace from several of these call attempts. Just look
>> from the top where you see a JsSIP client send an INVITE from 50.43.x.x and
>> then scroll down to see FreeSWITCH send two INVITES to two separate ports
>> on the 115.11.x.x network. Moreover, the candidates in both INVITES
>> contain different ports. Here are those two INVITES inline, where user
>> 1000 is the Android client on 115.11.x.x and 1100 is the Chrome JsSIP
>> client at 50.43.x.x, JsSIP initiates the call:
>>
>> ------------------------------------------------------------------------
>> send 2031 bytes to udp/[115.11.X.X]:9322 at 23:25:08.408616:
>>
>> ------------------------------------------------------------------------
>> INVITE sip:1000 at 115.11.X.X:9322;transport=udp SIP/2.0
>> Via: SIP/2.0/UDP 50.18.X.X:5070;rport;branch=z9hG4bKrv59pQQvDaagc
>> Route: <sip:1000 at 115.11.X.X:9322;transpo;lr;ovid=c5ab5428>
>> Max-Forwards: 68
>> From: "James" <sip:971200xxxx at 10.188.136.13>;tag=UX0748rrX2Bgm
>> To: <sip:1000 at 115.11.X.X:9322;transport=udp>
>> Call-ID: fed8b6a7-d25d-1231-2cad-22000abc880d
>> CSeq: 52461290 INVITE
>> Contact: <sip:mod_sofia at 50.18.X.X:5070>
>> User-Agent:
>> FreeSWITCH-mod_sofia/1.5.8b+git~20131127T162035Z~bbe1fe1a31~64bit
>> Allow: INVITE, ACK, BYE, CANCEL, OPTIONS, MESSAGE, INFO, UPDATE,
>> REGISTER, REFER, NOTIFY, PUBLISH, SUBSCRIBE
>> Supported: precondition, path, replaces
>> Allow-Events: talk, hold, conference, presence, as-feature-event,
>> dialog, line-seize, call-info, sla, include-session-description,
>> presence.winfo, message-summary, refer
>> Content-Type: application/sdp
>> Content-Disposition: session
>> Content-Length: 975
>> X-FS-Support: update_display,send_info
>> Remote-Party-ID: "James" <sip:971200xxxx at 10.188.136.13
>> >;party=calling;screen=yes;privacy=off
>>
>> v=0
>> o=FreeSWITCH 1385562204 1385562205 IN IP4 50.18.X.X
>> s=FreeSWITCH
>> c=IN IP4 50.18.X.X
>> t=0 0
>> a=sendrecv
>> a=msid-semantic: WMS oXSGoBbKflmyggzbJEZP120R0qAOBfsQ
>> m=audio 32504 RTP/SAVPF 0 8 101 13
>> a=rtpmap:0 PCMU/8000
>> a=rtpmap:8 PCMA/8000
>> a=rtpmap:101 telephone-event/8000
>> a=fingerprint:sha-256
>> 91:9E:50:FA:2E:A5:8E:02:41:1E:FC:A5:8A:B0:0C:1F:98:1F:1B:49:BB:D6:44:A8:E1:01:EA:82:7C:B6:59:1D
>> a=rtcp-mux
>> a=rtcp:32504 IN IP4 50.18.X.X
>> a=ssrc:177794596 cname:2bmhDoRIz36GawIJ
>> a=ssrc:177794596 msid:oXSGoBbKflmyggzbJEZP120R0qAOBfsQ a0
>> a=ssrc:177794596 mslabel:oXSGoBbKflmyggzbJEZP120R0qAOBfsQ
>> a=ssrc:177794596 label:oXSGoBbKflmyggzbJEZP120R0qAOBfsQa0
>> a=ice-ufrag:1xoTpfAgv0VX5dDX
>> a=ice-pwd:eQ5I4MguGd1GAzfr
>> a=candidate:5290007956 1 udp 659136 50.18.X.X 32504 typ host
>> generation 0
>> a=candidate:5290007956 2 udp 659136 50.18.X.X 32504 typ host
>> generation 0
>> a=crypto:1 AES_CM_128_HMAC_SHA1_80
>> inline:TTl8L9nzxRRf4R3jWWukpQaOSmkB5tncyLIuwvaA
>> a=ptime:20
>>
>> ------------------------------------------------------------------------
>> send 2034 bytes to udp/[115.11.X.X]:62163 at 23:25:08.409283:
>>
>> ------------------------------------------------------------------------
>> INVITE sip:1000 at 115.11.X.X:62163;transport=udp SIP/2.0
>> Via: SIP/2.0/UDP 50.18.X.X:5070;rport;branch=z9hG4bKS5y2rj8ZaK02Q
>> Route: <sip:1000 at 115.11.X.X:62163;transpo;lr;ovid=c5ab5428>
>> Max-Forwards: 68
>> From: "James" <sip:971200xxxx at 10.188.136.13>;tag=v6S0639UtB22F
>> To: <sip:1000 at 115.11.X.X:62163;transport=udp>
>> Call-ID: fed8d029-d25d-1231-2cad-22000abc880d
>> CSeq: 52461290 INVITE
>> Contact: <sip:mod_sofia at 50.18.X.X:5070>
>> User-Agent:
>> FreeSWITCH-mod_sofia/1.5.8b+git~20131127T162035Z~bbe1fe1a31~64bit
>> Allow: INVITE, ACK, BYE, CANCEL, OPTIONS, MESSAGE, INFO, UPDATE,
>> REGISTER, REFER, NOTIFY, PUBLISH, SUBSCRIBE
>> Supported: precondition, path, replaces
>> Allow-Events: talk, hold, conference, presence, as-feature-event,
>> dialog, line-seize, call-info, sla, include-session-description,
>> presence.winfo, message-summary, refer
>> Content-Type: application/sdp
>> Content-Disposition: session
>> Content-Length: 975
>> X-FS-Support: update_display,send_info
>> Remote-Party-ID: "James" <sip:971200xxxx at 10.188.136.13
>> >;party=calling;screen=yes;privacy=off
>>
>> v=0
>> o=FreeSWITCH 1385569930 1385569931 IN IP4 50.18.X.X
>> s=FreeSWITCH
>> c=IN IP4 50.18.X.X
>> t=0 0
>> a=sendrecv
>> a=msid-semantic: WMS YGqq3KBdA3doGvggsuWTkskoGwvg99oY
>> m=audio 24778 RTP/SAVPF 0 8 101 13
>> a=rtpmap:0 PCMU/8000
>> a=rtpmap:8 PCMA/8000
>> a=rtpmap:101 telephone-event/8000
>> a=fingerprint:sha-256
>> 91:9E:50:FA:2E:A5:8E:02:41:1E:FC:A5:8A:B0:0C:1F:98:1F:1B:49:BB:D6:44:A8:E1:01:EA:82:7C:B6:59:1D
>> a=rtcp-mux
>> a=rtcp:24778 IN IP4 50.18.X.X
>> a=ssrc:177896132 cname:FFeyfGCFzFU610ck
>> a=ssrc:177896132 msid:YGqq3KBdA3doGvggsuWTkskoGwvg99oY a0
>> a=ssrc:177896132 mslabel:YGqq3KBdA3doGvggsuWTkskoGwvg99oY
>> a=ssrc:177896132 label:YGqq3KBdA3doGvggsuWTkskoGwvg99oYa0
>> a=ice-ufrag:lZ3vhdKOA6UA59oA
>> a=ice-pwd:EtZJZsepYZmPPpZ1
>> a=candidate:6658001960 1 udp 659136 50.18.X.X 24778 typ host
>> generation 0
>> a=candidate:6658001960 2 udp 659136 50.18.X.X 24778 typ host
>> generation 0
>> a=crypto:1 AES_CM_128_HMAC_SHA1_80
>> inline:CKFCzTuq04VQ1E0QJMmo0WjXDUtp+GTuK4CLgAS7
>> a=ptime:20
>>
>>
>> It seems like something is amiss here. This does seem odd. Hope this
>> helps!
>>
>>
>> James
>>
>>
>> On Wed, Nov 27, 2013 at 4:44 PM, Michael Jerris <mike at jerris.com> wrote:
>>
>>> Try looking at a pcap and comparing to the android logs. Are they
>>> sending the stun requests to the right place and are they getting to
>>> freeswitch?
>>>
>>> On Nov 27, 2013, at 7:38 PM, James Mortensen <
>>> james.mortensen at synclio.com> wrote:
>>>
>>> The issue is that Android is showing STUN ping timeouts in the libjingle
>>> logs. This doesn't happen with Asterisk, so we just asked ourselves, "What
>>> is different in the SDP's" and we noticed it's the way RTCP is handled....
>>>
>>> Now, I'm not suggesting using the port +1 method would solve the
>>> problem; it was mostly just something we thought to try to see if we'd get
>>> different results. If it's not worth trying, we'll try something else. :)
>>>
>>> I don't have the Android debug logs handy at this time, but we'll work
>>> on putting them together.
>>>
>>> Thanks!
>>> James
>>>
>>>
>>>
>>>
>>> On Wed, Nov 27, 2013 at 4:14 PM, Michael Jerris <mike at jerris.com> wrote:
>>>
>>>> We are using rtcp mux where it uses the same port. This is the default
>>>> for webrtc and I don't think we have a way to disable it for outbound
>>>> invites for webrtc media, but there shouldn't ever be a reason to disable
>>>> it as all the browsers support it and its required by the spec. As for why
>>>> its not working with chrome for android, I'm not sure, do you have a debug
>>>> log of the call?
>>>>
>>>>
>>>> On Nov 27, 2013, at 5:56 PM, James Mortensen <
>>>> james.mortensen at synclio.com> wrote:
>>>>
>>>> > Hello,
>>>> >
>>>> > I've searched the list and documentation, and it's not yet clear if
>>>> FreeSWITCH is able to send candidates in the same manner as we were doing
>>>> with Asterisk 1.5 where it sends the RTCP to the RTP port +1. In other
>>>> words, I'd see candidates that looked like this:
>>>> >
>>>> > a=candidate:3441596188 1 udp 659136 10.3.1.1 20622 typ host
>>>> generation 0
>>>> > a=candidate:3441596188 1 udp 659136 50.18.X.X 20622 typ srflx
>>>> generation 0
>>>> > a=candidate:3441596188 2 udp 659136 10.3.1.1 20623 typ host
>>>> generation 0
>>>> > a=candidate:3441596188 2 udp 659136 50.18.X.X 20623 typ srflx
>>>> generation 0
>>>> >
>>>> > For instance, if the RTP port was 20622, the RTCP port would be
>>>> 20623. We're getting no audio when trying to make Chrome to Android and
>>>> Android to Android WebRTC calls, and we thought we'd try a different RTCP
>>>> method as we did have it working with Asterisk but want to use FreeSWITCH
>>>> instead.
>>>> >
>>>> > I've checked out the latest FreeSWITCH code from the master branch.
>>>> >
>>>> > Not sure if FreeSWITCH can do this yet, but if anyone knows that
>>>> would be helpful.
>>>> >
>>>> > Thank you!
>>>> > James
>>>> >
>>>>
>>>
>>>
>>> _________________________________________________________________________
>>> Professional FreeSWITCH Consulting Services:
>>> consulting at freeswitch.org
>>> http://www.freeswitchsolutions.com
>>>
>>>
>>>
>>>
>>> Official FreeSWITCH Sites
>>> http://www.freeswitch.org
>>> http://wiki.freeswitch.org
>>> http://www.cluecon.com
>>>
>>> FreeSWITCH-users mailing list
>>> FreeSWITCH-users at lists.freeswitch.org
>>> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users
>>> UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users
>>> http://www.freeswitch.org
>>>
>>>
>>
>> _________________________________________________________________________
>> Professional FreeSWITCH Consulting Services:
>> consulting at freeswitch.org
>> http://www.freeswitchsolutions.com
>>
>>
>>
>>
>> Official FreeSWITCH Sites
>> http://www.freeswitch.org
>> http://wiki.freeswitch.org
>> http://www.cluecon.com
>>
>> FreeSWITCH-users mailing list
>> FreeSWITCH-users at lists.freeswitch.org
>> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users
>> UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users
>> http://www.freeswitch.org
>>
>>
>
>
> --
> Anthony Minessale II
>
> FreeSWITCH http://www.freeswitch.org/
> ClueCon http://www.cluecon.com/
> Twitter: http://twitter.com/FreeSWITCH_wire
>
> AIM: anthm
> MSN:anthony_minessale at hotmail.com
> GTALK/JABBER/PAYPAL:anthony.minessale at gmail.com
> IRC: irc.freenode.net #freeswitch
>
> FreeSWITCH Developer Conference
> sip:888 at conference.freeswitch.org
> googletalk:conf+888 at conference.freeswitch.org
> pstn:+19193869900
>
> _________________________________________________________________________
> Professional FreeSWITCH Consulting Services:
> consulting at freeswitch.org
> http://www.freeswitchsolutions.com
>
>
>
>
> Official FreeSWITCH Sites
> http://www.freeswitch.org
> http://wiki.freeswitch.org
> http://www.cluecon.com
>
> FreeSWITCH-users mailing list
> FreeSWITCH-users at lists.freeswitch.org
> http://lists.freeswitch.org/mailman/listinfo/freeswitch-users
> UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-users
> http://www.freeswitch.org
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.freeswitch.org/pipermail/freeswitch-users/attachments/20131203/6f36acbe/attachment-0001.html
Join us at ClueCon 2013 Aug 6-8, 2013
More information about the FreeSWITCH-users
mailing list