[asterisk-users] asterisk pjsip webrtc rtp to private IP

[Joshua C. Colp]

On Thu, Dec 12, 2019 at 7:57 AM marek <cervajs64 at gmail.com> wrote:

> with wireshark i need decrypt traffic every call which is time consuming.
> get debug from pjnat through asterisk is not possible because of technical
> reasons or nobody did it?
>
>
> in my case its strange that ice candidates are the same
>
> good call
>
> v=0
> o=- 3669976329745317845 2 IN IP4 127.0.0.1
> s=-
> t=0 0
> a=msid-semantic: WMS EoNIdKcMZvWBLULGqGPJTDe12ujjFEemeapo
> m=audio 52421 RTP/SAVPF 8 0 101
> c=IN IP4 10.2.152.36
> a=rtcp:9 IN IP4 0.0.0.0
> a=candidate:3607370648 1 udp 2122260223 10.2.152.36 52421 typ host
> generation 0 network-id 1 network-cost 10
> a=candidate:2575820648 1 tcp 1518280447 10.2.152.36 9 typ host tcptype
> active generation 0 network-id 1 network-cost 10
>
> bad call
>
> v=0
> o=- 2602173234285924157 2 IN IP4 127.0.0.1
> s=-
> t=0 0
> a=msid-semantic: WMS aDrO7zRNTqNWKodpSG62Co1IDoHReEpT8Ga3
> m=audio 63249 RTP/SAVPF 8 0 101
> c=IN IP4 10.2.152.36
> a=rtcp:9 IN IP4 0.0.0.0
> a=candidate:3607370648 1 udp 2122260223 10.2.152.36 63249 typ host
> generation 0 network-id 1 network-cost 10
> a=candidate:2575820648 1 tcp 1518280447 10.2.152.36 9 typ host tcptype
> active generation 0 network-id 1 network-cost 10
>
>
> but RTP looks like
>
> bad call (1.1.1.1 is "public" ip of PSTN SIP GW)
>
> Got  RTP packet from    1.1.1.1:13460 (type 08, seq 002433, ts 000160,
> len 000160)
> Sent RTP packet to      10.2.152.36:63249 (type 08, seq 022470, ts
> 000160, len 000160)
> Got  RTP packet from    1.1.1.1:13460 (type 08, seq 002434, ts 000320,
> len 000160)
> Sent RTP packet to      10.2.152.36:63249 (type 08, seq 022471, ts
> 000320, len 000160)
> Got  RTP packet from    1.1.1.1:13460 (type 08, seq 002435, ts 000480,
> len 000160)
>
> good call (1.1.1.1 is "public" ip of PSTN SIP GW, 2.2.2.2 is public IP of
> router)
>
> Got  RTP packet from    1.1.1.1:15026 (type 08, seq 021197, ts 000160,
> len 000160)
> Sent RTP packet to      *10.2.152.36:52421 <http://10.2.152.36:52421>
> (type 08, seq 032328, ts 000160, len 000160)*
>
> [Dec 11 16:59:53] DEBUG[44360]: res_rtp_asterisk.c:6049
> ast_rtp_remote_address_set: Setting RTCP address on RTP instance
> '0x7faa14005408'
>
> Got  RTP packet from    1.1.1.1:15026 (type 08, seq 021198, ts 000320,
> len 000160)
> Sent RTP packet to      2.2.2.2:52421 (via ICE) (type 08, seq 032329, ts
> 000320, len 000160)
> Got  RTP packet from    1.1.1.1:15026 (type 08, seq 021199, ts 000480,
> len 000160)
> Sent RTP packet to      2.2.2.2:52421 (via ICE) (type 08, seq 032330, ts
> 000480, len 000160)
> Got  RTP packet from    1.1.1.1:15026 (type 08, seq 021200, ts 000640,
> len 000160)
> Sent RTP packet to      2.2.2.2:52421 (via ICE) (type 08, seq 032331, ts
> 000640, len 000160)
> Got  RTP packet from    1.1.1.1:15026 (type 08, seq 021201, ts 000800,
> len 000160)
>
> looking for the part where RTP engine switch from *10.2.152.36 to *
> *2.2.2.2*
>
> it looks like its somewhere in the learning phase
>

You need to look at the ICE candidates given by Asterisk as well, and
ensure that if it is behind NAT it is configured in rtp.conf to do some
mapping of candidates, as well as ensuring the firewall is open. The
wireshark capture like I said will provide insight into what ICE is doing.

ICE is what is used to figure out the path and determine the IP
address/port to use. If that fails, then it won't work.

I would also urge you to learn more about the lower level details of WebRTC
if you plan on deploying it. You really need some understanding of
ICE/STUN/DTLS-SRTP if deploying, as those are fundamental aspects and stuff
doesn't just work in all cases. Digging into why it's not working takes you
down to those.

-- 
Joshua C. Colp
Senior Software Developer
Sangoma Technologies
Check us out at www.sangoma.com and www.asterisk.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-users/attachments/20191212/99a3c717/attachment.html>