Re: [rtcweb] Stephen Farrell's Discuss on draft-ietf-rtcweb-stun-consent-freshness-15: (with DISCUSS and COMMENT)

[Xavier Marjou <[e-mail]>]

>(2) WebRTC does not require STUN or TURN servers for some calls, even if
it does for many.

In a typical WebRTC communication with A and B as WebRTC endpoints:
1 - Optionally A and/or B can use a STUN server to discover a public IP
address ("server reflexive" address as per rfc5245)
2 - Optionally A and/or B can use a TURN server to tunnel/relay WebRTC
messages with the help of a TURN server (c.f. draft-ietf-tram-turnbis-05)
3 - A and B have to implement ICE (rfc5245) and thus have to exchange STUN
connectivity check messages between their candidates (to try "open" flows",
... and maybe even to transport "consent" information).

Therefore, there can be STUN messages transported in three different
"sectors" :
1 - between the WebRTC endpoint and its STUN server, if used
2 - between the WebRTC endpoint and its TURN server, if used
(draft-ietf-tram-turnbis-05 states that "Most, though not all, TURN
messages are STUN-formatted messages" so there are precisely
"STUN-formatted", not "STUN", but I mention them to be exhaustive with
different sort of messages appearing as STUN messages in tools like
Wireshark).
3 - between the WebRTC endpoints (in this case, the STUN connectivity check
messages are in the same flow that transports SCTP/DTLS and SRTP messages).

Some newcomers think that STUN messages only happen in cases 1 and/or 2,
they forget case 3. Maybe this explains the misunderstanding.



On Wed, Aug 5, 2015 at 3:22 PM, Christer Holmberg <
[e-mail]> wrote:

> Hi Stephen,
>
> >(2) WebRTC does not require STUN or TURN servers for some calls, even if
> it does for many. Why is it ok to require such a server be present in all
> calls (which I think this means) espcially
> >when that means exposing additional meta-data (calling parties in a case
> where the servers weren't needed and call duration in all
> >cases) to those servers when that is not always necessary?
>
> Could you please refer to the text which you think mandates STUN or TURN
> servers?
>
> If there are no NATs, the STUN requests can be sent between the endpoints,
> without STUN or TURN servers.
>
> Regards,
>
> Christer
>
> _______________________________________________
> rtcweb mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/rtcweb
>