Re: [tram] [Tsv-art] Tsvart last call review of draft-ietf-tram-turnbis-25

[Joe Touch <[e-mail]>]

Notes below.

> On Jun 18, 2019, at 6:03 AM, Konda, Tirumaleswar Reddy <[e-mail]> wrote:
> 
> HI Magnus,
>  
> Please see inline [TR]
>  
> From: tram <[e-mail] <mailto:[e-mail]>> On Behalf Of Magnus Westerlund
> Sent: Tuesday, June 18, 2019 4:57 PM
> To: Konda, Tirumaleswar Reddy <[e-mail] <mailto:[e-mail]>>; Joe Touch <[e-mail] <mailto:[e-mail]>>
> Cc: [e-mail] <mailto:[e-mail]>; [e-mail] <mailto:[e-mail]>; [e-mail] <mailto:[e-mail]>; Brandon Williams <[e-mail] <mailto:[e-mail]>>; [e-mail] <mailto:[e-mail]>
> Subject: Re: [tram] [Tsv-art] Tsvart last call review of draft-ietf-tram-turnbis-25
>  
> CAUTION: External email. Do not click links or open attachments unless you recognize the sender and know the content is safe.
> 
> On 2019-06-17 12:44, Konda, Tirumaleswar Reddy wrote:
> You need to address these issues - e.g., by a version of the response above.
> 
> [TR1] Okay, will add the following lines:
> 
> TCP multi-path [RFC6824] is not supported by this version of TURN because TCP multi-path is not used by both SIP and WebRTC protocols [RFC7478] for media and non-media data. If the TCP connection between the TURN client and server uses TCP-AO [RFC5925], the client must secure application data (e.g. using SRTP) to provide confidentially, message authentication and replay protection to protect the application data relayed from the server to the peer using UDP.  Note that TCP-AO option obsoletes TCP MD5 option.  Unlike UDP, TCP without the TCP Fast Open extension [RFC7413] does not support 0-RTT session resumption. The TCP user timeout [RFC5482] equivalent for application data relayed by TURN is the use of RTP control protocol (RTCP). As a reminder,  RTCP is a fundamental and integral part of RTP.
> 
>  
> 
> Sorry, I find this very confusing. On the Client to TURN Server leg if that is using a TCP connection, then there are no issues with using either TCP-MP or TCP-AO options on that leg. They will have no impact on the TURN messages carried inside the TCP connection or how its outgoing IP/UDP packet. So from my perspective there are no necessity to discuss these as not supported.
> 
I’m not weighing in on whether they’re supported or not, but support needs to be discussed one way or the other.
> [TR] The comment from Joe is : if TCP-AO is used, application data is authenticated in the TCP leg but the data can be faked when relayed from the server to the peer using UDP. I tried to address this comment by saying if secure application data (SRTP) is used message authentication is available at the application layer even if UDP does not support authentication option.
> 
> What is discussed in this document are the options for the IP/UDP packet being sent or received by the turn server to the peer. Those IP/UDP Fields that can be controlled and have meaning are discussed here. Like the DSCP, where TCP will force the use of a single one, when UDP to UDP can support many as the field value is taken from the packet on the client to server leg, rather than an internal TURN message field.
> 
> I guess the issue here is that there is a lack of requirement making clear that for a number of the fields in the IP/UDP handled on the server to peer leg that needs to handling rules for the client to server transport.
> 
> I think what Section 14 and 15 talks about should be labeled relaying and not translation. And thus focus on how one avoid information or intent loss in the relay process. I understand why section 13 is called translation, but when we get down to it, it is the same here, but a focus on how the relay process maintain information of the IP layer, especially in the face of address family differences.
> 
> [TR] Agreed, replaced translation with relaying in both sections 14 and 15.
> 
That would help, but it’s more than just replacing a single word.

The way in which those sections discuss IP options is misleading. It should talk about them more clearly in terms of individual UDP messages and TCP connections, not in terms of ‘ when you get this TURN message, here’s how to set these values in IP packets” - because the latter (even if not a direct quote) is misleading in implying these settings can change throughout a TCP connection.

I agree that presenting this as relaying would be much more clear, but it affects more than a simple word change.

Joe