Re: [rtcweb] [BEHAVE] FW: I-D Action: draft-muthu-behave-consent-freshness-04.txt

["Avasarala, Ranjit (NSN - IN/Bangalore)" <[e-mail]>]

Hi Muthu
So if I understand correctly, this functionality is on top of ICE functionality and is optional for the gateways to support it. Right?

Regards
Ranjit




-----Original Message-----
From: ext Muthu Arul Mozhi Perumal (mperumal) [mailto:[e-mail]] 
Sent: Wednesday, July 17, 2013 7:48 PM
To: Avasarala, Ranjit (NSN - IN/Bangalore)
Cc: [e-mail]
Subject: RE: [rtcweb] [BEHAVE] FW: I-D Action: draft-muthu-behave-consent-freshness-04.txt

|-----Original Message-----
|From: Avasarala, Ranjit (NSN - IN/Bangalore) [mailto:[e-mail]]
|Sent: Wednesday, July 17, 2013 2:36 PM
|To: Muthu Arul Mozhi Perumal (mperumal)
|Cc: [e-mail]
|Subject: RE: [rtcweb] [BEHAVE] FW: I-D Action: draft-muthu-behave-consent-freshness-04.txt
|
|My comments inline <Ranjit>
|
|Regards
|Ranjit
|
|-----Original Message-----
|From: ext Muthu Arul Mozhi Perumal (mperumal) [mailto:[e-mail]]
|Sent: Wednesday, July 17, 2013 2:16 PM
|To: Avasarala, Ranjit (NSN - IN/Bangalore)
|Cc: [e-mail]
|Subject: RE: [rtcweb] [BEHAVE] FW: I-D Action: draft-muthu-behave-consent-freshness-04.txt
|
||If many public STUN servers and call servers that implement STUN or ICElite
||functionalities, are not expected to support
|
|I should have said they have no role to play. It is similar to peer-to-peer ICE connectivity checks --
|those checks don't go through public STUN or call servers at all.
|[Ranjit] But there can be scenarios where RTCWebBrowsers do talk to WebRTC Gateways which could be co-
|located with call handling servers. Then it won't make sense to have this feature.

An RTCWeb gateway wouldn't have to do anything fancier than what it already does for ICE connectivity checks. 

|
|The difference between ICE connectivity checks and consent freshness checks is that the former is
|performed before session establishment and the later is performed after session establishment.
|[Ranjit] Why do you need to check consent of the peer after session is established? Ideally I should
|check consent before session establishment. If the peer does not want to receive traffic, then session
|itself should not be established.

See draft-ietf-rtcweb-security-arch:

<snip>
4.4.  Communications and Consent Freshness

   From a security perspective, everything from here on in is a little
   anticlimactic:  Alice and Bob exchange data protected by the keys
   negotiated by DTLS.  Because of the security guarantees discussed in
   the previous sections, they know that the communications are
   encrypted and authenticated.

   The one remaining security property we need to establish is "consent
   freshness", i.e., allowing Alice to verify that Bob is still prepared
   to receive her communications so that Alice does not continue to send
   large traffic volumes to entities which went abruptly offline.  ICE
   specifies periodic STUN keepalizes but only if media is not flowing.
   Because the consent issue is more difficult here, we require RTCWeb
   implementations to periodically send keepalives.  As described in
   Section 5.3, these keepalives MUST be based on the consent freshness
   mechanism specified in [I-D.muthu-behave-consent-freshness].  If a
   keepalive fails and no new ICE channels can be established, then the
   session is terminated.
</snip>

|
||then who would have interest to support this functionality?
|
|RTCWEB browsers and browser-like platforms.
|[Ranjit] browser may implement, but it would be very customized one. I do not think the intermediate
|entities would want to support.

You seem to be completely missing the very purpose of consent.

|
||Ideally getting consent for receiving traffic or call should be part of signaling.
||I do not think using STUN is appropriate.
|
|ICE already does that. You think it is inappropriate?
|[Ranjit] I am saying that using STUN for consent check is not appropriate as STUN has other dedicated
|functionality to do - like connectivity checks, keep alives, etc. 

See draft-ietf-rtcweb-security and draft-ietf-rtcweb-security.

|Also in cases of Symmetric NATs where STUN fails, TURN is needed. Then the proposed feature would anyway fail.

The solution is built on top of ICE, so should work across these.

Muthu

|
|Muthu
|
||-----Original Message-----
||From: Avasarala, Ranjit (NSN - IN/Bangalore) [mailto:[e-mail]]
||Sent: Wednesday, July 17, 2013 12:16 PM
||To: Muthu Arul Mozhi Perumal (mperumal)
||Cc: [e-mail]
||Subject: RE: [rtcweb] [BEHAVE] FW: I-D Action: draft-muthu-behave-consent-freshness-04.txt
||
||Hi Muthu
||
||If many public STUN servers and call servers that implement STUN or ICElite functionalities, are not
||expected to support, then who would have interest to support this functionality? Ideally getting
||consent for receiving traffic or call should be part of signaling. I do not think using STUN is
||appropriate.
||
||Regards
||Ranjit
||
||
||
||
||-----Original Message-----
||From: ext Muthu Arul Mozhi Perumal (mperumal) [mailto:[e-mail]]
||Sent: Wednesday, July 17, 2013 12:07 PM
||To: Avasarala, Ranjit (NSN - IN/Bangalore)
||Cc: [e-mail]; [e-mail]
||Subject: RE: [rtcweb] [BEHAVE] FW: I-D Action: draft-muthu-behave-consent-freshness-04.txt
||
||Ranjit,
||
|||Many publicly available STUN servers or those that are part of Call servers
|||may not support this.
||
||Public STUN servers and call servers are not expected to support this.
||
|||Can't there be some other neutral mechanism to query peer's consent - through
|||signaling?
||
||No. That signaling between the JS and the web server could be anything (SIP or XMPP or proprietary or
||whatever) and the browser has no control over it.
||
||Muthu
||
|||-----Original Message-----
|||From: Avasarala, Ranjit (NSN - IN/Bangalore) [mailto:[e-mail]]
|||Sent: Wednesday, July 17, 2013 11:19 AM
|||To: Muthu Arul Mozhi Perumal (mperumal)
|||Cc: [e-mail]; [e-mail]
|||Subject: RE: [rtcweb] [BEHAVE] FW: I-D Action: draft-muthu-behave-consent-freshness-04.txt
|||
|||Hi Muthu
|||
|||Though using STUN request/response may be good for querying about consent, I feel it is overloading
|||the functionality of STUN. Many publicly available STUN servers or those that are part of Call
||servers
|||may not support this.
|||
|||Can't there be some other neutral mechanism to query peer's consent - through signaling?
|||
|||
|||Regards
|||Ranjit
|||
|||
|||
|||-----Original Message-----
|||From: [e-mail] [mailto:[e-mail]] On Behalf Of ext Simon Perreault
|||Sent: Tuesday, July 16, 2013 6:34 PM
|||To: Muthu Arul Mozhi Perumal (mperumal)
|||Cc: [e-mail]; [e-mail]
|||Subject: Re: [rtcweb] [BEHAVE] FW: I-D Action: draft-muthu-behave-consent-freshness-04.txt
|||
|||Le 2013-07-16 14:43, Muthu Arul Mozhi Perumal (mperumal) a écrit :
|||> |> |>    Liveness timer: If no packets have been received on the local port in
|||> |> |>    Tr seconds, the WebRTC browser MUST inform the JavaScript that
|||> |> |>    connectivity has been lost.  The JavaScript application will use this
|||> |> |>    notification however it desires (e.g., cease transmitting to the
|||> |> |>    remote peer, provide a notification to the user, etc.).
|||> |> |
|||> |> |This seems to me like it will not fulfill the goal set in the abstract:
|||> |> |"to ensure that a malicious JavaScript cannot use the browser as a
|||> |> |platform for launching attacks". If the JavaScript is free to ignore the
|||> |> |notification from the browser, then it has no security benefits. If you
|||> |> |want to reach that goal, the browser needs to forcefully stop transmitting.
|||> |>
|||> |> That goal is fulfilled by the consent checks -- the browser would stop transmitting everything
||on
|||> |that candidate pair, including liveness checks, if there is a consent failure.
|||> |
|||> |That's not what the draft says. It says that the browser "notifies" the
|||> |JS app. It needs to say that the browser MUST stop sending.
|||>
|||> No. That section is about liveness check and its intention is just notify the JavaScript of a
|||potential connectivity loss. It is when the consent check fails the browser actually stops sending
|||everything. Does the draft need more text on the distinction between consent and liveness tests?
|||
|||Ah! No, you're right, and the text is already perfectly clear about
|||this. No need to change. I was just confused.
|||
|||> |> |>    When not actively sending traffic on a nominated candidate pair,
|||> |> |>    performing consent freshness does not serve any purpose from a
|||> |> |>    security perspective.
|||> |> |
|||> |> |I don't understand what this means. Why is the "security perspective"
|||> |> |important here? Aren't we concerned about keepalives?
|||> |>
|||> |> You mean one could use keepalives (Binding indications) for launching attacks, so consent
|||freshness
|||> |would be required for sending them as well?
|||> |
|||> |No.
|||> |
|||> |This is a section about keepalives. I just don't understand this
|||> |sentence, and I don't understand why it talks about security.
|||>
|||> Ok, let me elaborate:
|||> - Consent freshness is not necessary when the browser is not sending any
|||>   traffic on a candidate pair.
|||> - If the browser is not performing consent freshness on a candidate pair
|||>   for the above reason, it performs ICE keepalives (or RTP keepalives) to
|||>   refresh NAT bindings.
|||>
|||> Of course, the browser could continue performing consent freshness even when it is not sending any
|||other traffic on that candidate pair and skip ICE keepalives.
|||
|||Ah, ok I understand with your explanation. It makes sense. There should
|||be a way to reformulate the text to make it clearer.
|||
|||Thanks,
|||Simon
|||_______________________________________________
|||rtcweb mailing list
|||[e-mail]
|||https://www.ietf.org/mailman/listinfo/rtcweb