Re: [rtcweb] Transports: RFC 6062 support

["Hutton, Andrew" <[e-mail]>]

Personally I think leaving in the "MAY" statement is a good idea unless there is a really reason why we don’t want an implementation to do this.  Including such a candidate does not appear to do anybody any harm so I think we should say it "MAY" be supported.

So I suggest modifying the first line the text Magnus suggested as follows:

Using TURN to provide TCP relay candidates [REF] MAY be supported however such candidates are not
seen as providing any significant benefit. First, use of TURN TCP would
only be relevant in cases which both peers requires to use TCP to
establish a PeerConnection. Secondly, that use case is anyway supported
by both side establishing UDP relay candidates using TURN over TCP
[REF]. Thirdly, using TCP only between the endpoint and its relay may
result in less issues with TCP in regards to real-time constraints, e.g.
due to head of line blocking.


Andy

> -----Original Message-----
> From: rtcweb [mailto:[e-mail]] On Behalf Of Magnus
> Westerlund
> Sent: 12 March 2014 07:52
> To: Justin Uberti
> Cc: [e-mail]
> Subject: Re: [rtcweb] Transports: RFC 6062 support
> 
> On 2014-03-12 01:58, Justin Uberti wrote:
> > The reason I suggested SHOULD NOT is because I think this technique
> is
> > unnecessary for WebRTC, and typically net harmful for the reasons
> that
> > Matthew mentions.
> >
> > That said, I would be fine with saying MAY and including some
> > explanatory text; my main concerns was to avoid the ambiguity that
> would
> > arise from leaving this out altogether.
> 
> Okay, as we have considerations, I would leave the RFC 2119 terms out.
> I
> think a statement along the lines of the below might be most suitable.
> 
> Using TURN to provide TCP relay candidates [REF] was considered but not
> seen as providing any significant benefit. First, use of TURN TCP would
> only be relevant in cases which both peers requires to use TCP to
> establish a PeerConnection. Secondly, that use case is anyway supported
> by both side establishing UDP relay candidates using TURN over TCP
> [REF]. Thirdly, using TCP only between the endpoint and its relay may
> result in less issues with TCP in regards to real-time constraints,
> e.g.
> due to head of line blocking.
> 
> Please provide feedback or word smith.
> 
> Cheers
> 
> Magnus
> (As individual)
> 
> >
> >
> > On Mon, Mar 10, 2014 at 8:33 AM, Magnus Westerlund
> > <[e-mail]
> <mailto:[e-mail]>>
> > wrote:
> >
> >     On 2014-03-06 14:34, Simon Perreault wrote:
> >     > Le 2014-03-06 13:23, Justin Uberti a écrit :
> >     >> can we agree that TURN TCP candidates are a SHOULD NOT?
> >     >
> >     > Not a SHOULD, sure. SHOULD NOT, no.
> >     >
> >
> >     I would guess that the simplest is to remove discussion of TURN
> TCP
> >     altogether from Transports. That would not recommend it nor make
> it
> >     disallowed. If we want to be explicit, then simply motivate why
> we don't
> >     believe it necessary.
> >
> >     Justin, what is the reason you wanted SHOULD NOT?
> >
> >     cheers
> >
> >     Magnus Westerlund
> >
> >     -----------------------------------------------------------------
> -----
> >     Services, Media and Network features, Ericsson Research EAB/TXM
> >     -----------------------------------------------------------------
> -----
> >     Ericsson AB                 | Phone  [phone]
> >     <tel:%2B46%2010%207148287>
> >     Färögatan 6                 | Mobile [phone]
> >     <tel:%2B46%2073%200949079>
> >     SE-164 80 Stockholm, Sweden | mailto:
> [e-mail]
> >     <mailto:[e-mail]>
> >     -----------------------------------------------------------------
> -----
> >
> >
> 
> 
> --
> 
> Magnus Westerlund
> 
> ----------------------------------------------------------------------
> Services, Media and Network features, Ericsson Research EAB/TXM
> ----------------------------------------------------------------------
> Ericsson AB                 | Phone  [phone]
> Färögatan 6                 | Mobile [phone]
> SE-164 80 Stockholm, Sweden | mailto: [e-mail]
> ----------------------------------------------------------------------
> 
> _______________________________________________
> rtcweb mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/rtcweb