Re: [rtcweb] AD evaluation: draft-ietf-rtcweb-stun-consent-freshness-11

["Ram Mohan R (rmohanr)" <[e-mail]>]

Attached is the diffs with comment from Christer incorporated. Please let
me know if any one else has comments. If not I will publish this diff as a
new revision.

Regards,
Ram

-----Original Message-----
From: Alissa Cooper <[e-mail]>
Date: Friday, 8 May 2015 6:41 pm
To: Christer Holmberg <[e-mail]>
Cc: Cisco Employee <[e-mail]>, "Tirumaleswar Reddy (tireddy)"
<[e-mail]>, Martin Thomson <[e-mail]>,
"[e-mail]" <[e-mail]>
Subject: Re: [rtcweb] AD evaluation:
draft-ietf-rtcweb-stun-consent-freshness-11

>Ok with me.
>Alissa
>
>On May 7, 2015, at 2:49 AM, Christer Holmberg
><[e-mail]> wrote:
>
>> Hi,
>> 
>> The text looks ok, but I think we can simplify/clarify it a little.
>> 
>> For example, the usage of "flow" is a little strange, since we
>>explicitly say that no media is sent :)
>> 
>> Also, "failure" is the wrong wording in my opinion, because there is no
>>failure.
>> 
>> What about:
>> 
>>   An endpoint that is not sending any application data does not need to
>>   maintain consent. However, not sending any traffic could cause NAT or
>>   firewall mappings to expire.  Furthermore, having one peer unable to
>>send 
>>   is detrimental to many protocols.  Absent better information about
>>the 
>>   network, if an endpoint needs to ensure its NAT or firewall mappings
>>do
>>   not expire, it can be done using keepalive or  other techniques (see
>>   Section 10 of [RFC5245] and see [RFC6263]).
>> 
>> Regards,
>> 
>> Christer
>> 
>> 
>> -----Original Message-----
>> From: Ram Mohan R (rmohanr) [mailto:[e-mail]]
>> Sent: 7. toukokuuta 2015 12:19
>> To: Christer Holmberg; Tirumaleswar Reddy (tireddy); Alissa Cooper;
>>Martin Thomson
>> Cc: [e-mail]
>> Subject: Re: [rtcweb] AD evaluation:
>>draft-ietf-rtcweb-stun-consent-freshness-11
>> 
>> Hi Christer,
>> 
>> How about the below text. Does it sound better ?
>> 
>> OLD:
>> 
>>   An endpoint that is not sending any application data does not need to
>>   maintain consent.  However, failure to send could cause any NAT or
>>   firewall mappings for the flow to expire.  Furthermore, having one
>>   peer unable to send is detrimental to many protocols.  Absent better
>>   information about the network, an endpoint SHOULD maintain consent if
>>   there is any possibility that a flow might be needed again.
>> 
>> NEW:
>> 
>>   An endpoint that is not sending any application data does not need to
>>   maintain consent. However, failure to send could cause any NAT or
>>   firewall mappings for the flow to expire.  Furthermore, having one
>>   peer unable to send is detrimental to many protocols.  Absent better
>>   information about the network, if an endpoint needs to ensure its NAT
>>   or firewall mappings persist which can be done using keepalive or
>>   other techniques (see Section 10 of [RFC5245] and see [RFC6263]).
>> 
>> 
>> 
>> Ram
>> 
>> -----Original Message-----
>> From: Christer Holmberg <[e-mail]>
>> Date: Thursday, 7 May 2015 2:40 pm
>> To: "Tirumaleswar Reddy (tireddy)" <[e-mail]>, Cisco Employee
>><[e-mail]>, Alissa Cooper <[e-mail]>, Martin Thomson
>><[e-mail]>
>> Cc: "[e-mail]" <[e-mail]>
>> Subject: RE: [rtcweb] AD evaluation:
>> draft-ietf-rtcweb-stun-consent-freshness-11
>> 
>>> Hi,
>>> 
>>>>>> Martin¹s statement says SHOULD here and does not mandate. ICE
>>>>>> keepalives could also be used to keep the NAT state
>>>>> 
>>>>> There needs to be a good justification for a SHOULD, and consent was
>>>>> never intended for NAT keep-alives.
>>>>> 
>>>>> Also keep in mind that, with the "virtual connection" concept, there
>>>>> might be a big number of ICE connections - some of which you may
>>>>> never use. Why send consent on those, if there is no media?
>>>> 
>>>> ICE keepalives or consent is only required for candidate pairs
>>>> selected for media,
>>> 
>>> Correct. But, you may have multiple candidate pairs "selected for
>>> media", but that doesn't mean you are sending media on all of them.
>>> Very likely you are, at any given time, only sending media on one of
>>>them.
>>> 
>>>> http://tools.ietf.org/html/rfc5245#section-10 mandates sending
>>>> keepalives if no packet is sent on the candidate pair ICE is using
>>>>for 
>>>> a media component for Tr seconds. STUN Binding Indication or consent
>>>> can be used for keepalives.
>>> 
>>> Correct. My issue is why there should be a "SHOULD send consent" on
>>> candidate pairs currently not used for sending media. In such case,
>>> only the NAT bindings need to be maintained, and the keep-alives take
>>> care of that.
>>> 
>>> Regards,
>>> 
>>> Christer
>>> 
>>> 
>>> 
>>>> -----Original Message-----
>>>> From: Christer Holmberg <[e-mail]>
>>>> Date: Wednesday, 6 May 2015 12:23 pm
>>>> To: Alissa Cooper <[e-mail]>, Martin Thomson
>>>> <[e-mail]>
>>>> Cc: "[e-mail]" <[e-mail]>
>>>> Subject: Re: [rtcweb] AD evaluation:
>>>> draft-ietf-rtcweb-stun-consent-freshness-11
>>>> 
>>>>> Hi,
>>>>> 
>>>>> I don't think you need to continue doing consent because of NAT
>>>>> issues, if you are sending normal STUN keep-alives.
>>>>> 
>>>>> Regards,
>>>>> 
>>>>> Christer
>>>>> 
>>>>> -----Original Message-----
>>>>> From: rtcweb [mailto:[e-mail]] On Behalf Of Alissa
>>>>> Cooper
>>>>> Sent: 2. toukokuuta 2015 2:20
>>>>> To: Martin Thomson
>>>>> Cc: [e-mail]
>>>>> Subject: Re: [rtcweb] AD evaluation:
>>>>> draft-ietf-rtcweb-stun-consent-freshness-11
>>>>> 
>>>>> 
>>>>> On May 1, 2015, at 9:54 AM, Martin Thomson
>>>> <[e-mail]>
>>>>> wrote:
>>>>> 
>>>>>> On 30 April 2015 at 17:32, Alissa Cooper <[e-mail]> wrote:
>>>>>>> "An endpoint that is not sending any application data does not
>>>>>>> need
>>>> to
>>>>>>>  maintain consent.  However, failure to send could cause any NAT or
>>>>>>>  firewall mappings for the flow to expire.  Furthermore, having one
>>>>>>>  peer unable to send is detrimental to many protocols."
>>>>>>> 
>>>>>>> It sounds like the unstated implication here is that if you are
>>>>>>> such an endpoint, you should keep doing consent checks anyway to
>>>>>>> maintain consent. Should that be stated explicitly, or am I
>>>> misunderstanding?
>>>>>> 
>>>>>> Can you tell that this is my text?
>>>>>> 
>>>>>> Yep, the unspoken implication is that if you stop maintaining
>>>>>> consent, a flow is highly likely to break.  I'm OK with making
>>>>>> that
>>>> explicit.
>>>>>> 
>>>>>> ... .  Absent better information about the network, an endpoint
>>>>>> SHOULD maintain consent if there is any possibility that a flow
>>>>>> might be needed again.
>>>>> 
>>>>> WFM
>>>>> 
>>>>>> 
>>>>>> (Thanks for the suggestion on Sec7.  I wasn't happy with it
>>>>>> before.)
>>>>> 
>>>>> _______________________________________________
>>>>> rtcweb mailing list
>>>>> [e-mail]
>>>>> https://www.ietf.org/mailman/listinfo/rtcweb
>>>>> 
>>>>> _______________________________________________
>>>>> rtcweb mailing list
>>>>> [e-mail]
>>>>> https://www.ietf.org/mailman/listinfo/rtcweb
>>>> 
>>>> _______________________________________________
>>>> rtcweb mailing list
>>>> [e-mail]
>>>> https://www.ietf.org/mailman/listinfo/rtcweb
>> 
>