Re: [Perc] Confirmation of Consensus on PERC Double from IETF 99

Hi Steve,
I agree with what you wrote, I just wanted to point out that if we have some text on DTMF in the document it is not enough to say that it is only end to end but maybe mention that usage of DTMF for conference join and control is not handled by the MD.
Roni

From: Stephen Botzko [mailto:stephen.botzko@gmail.com]
Sent: יום ב 07 אוגוסט 2017 17:07
To: Roni Even
Cc: Paul E. Jones; perc@ietf.org; Suhas Nandakumar
Subject: Re: [Perc] Confirmation of Consensus on PERC Double from IETF 99

>>> [roni]the bridge will ask for conference number and password using voice messages
I'm not seeing how perc's encryption (as proposed in Prague) supports either audio prompts from the server to the client or DTMF from the client to the server.

If there is an adjunct web interface for control, that might not matter - since audio prompts can be streamed over that interface if that is desired, and the client can easily use http for replies.  That approach separates the conference control plane from the media distribution plane.

Though in your use case, another option is to allow an initial media connection to the server that doesn't use double for the purpose of voice/response - and once the user selects their conference, then that media is torn down and replaced by media encrypted with the conference keys.  That works for conference setup, but not other use cases (announcing new attendees joining the conference, or warning that the conference ending time is approaching for example).

Stephen

On Mon, Aug 7, 2017 at 8:33 AM, Roni Even <roni.even@huawei.com<mailto:roni.even@huawei.com>> wrote:
Hi Paul,
I agree that DTMF is not to the end user. On second thought I think that if DTMF will be used by the conference than there may be a separate entity that will handle DTMF. An example is when you call a conference dial-in number and the bridge will ask for conference number and password using voice messages.
I think this may happen also in Webex when you want to join using your phone to dial in, you get prompted for conference ID
Roni

From: Paul E. Jones [mailto:paulej@packetizer.com<mailto:paulej@packetizer.com>]
Sent: יום ב 07 אוגוסט 2017 14:59
To: perc@ietf.org<mailto:perc@ietf.org>; Roni Even; Suhas Nandakumar; perc@ietf.org<mailto:perc@ietf.org>

Subject: Re: [Perc] Confirmation of Consensus on PERC Double from IETF 99

Roni,

That particular issue is least concerning to me, since all endpoints I know won't play DTMF into a user's ear. AFAIK, only users over the PSTN might be affected, but then that suggests there's a gateway in the path and PERC is no longer securing the conference E2E. Thus, gateways need to stay out of PERC conferences.

Paul
________________________________
From: Roni Even <roni.even@huawei.com<mailto:roni.even@huawei.com>>
Sent: August 7, 2017 2:57:30 AM EDT
To: Suhas Nandakumar <suhasietf@gmail.com<mailto:suhasietf@gmail.com>>, "perc@ietf.org<mailto:perc@ietf.org>" <perc@ietf.org<mailto:perc@ietf.org>>
Subject: Re: [Perc] Confirmation of Consensus on PERC Double from IETF 99

Hi,
I do not have a strong opinion but I am a bit puzzled about the DTMF issue.
My understating so far was that the MD is the media distribution for a multipoint conferencing application. Typically in this cases the DTMF is for the multipoint application and not to the conference participants (The DTMF may even be suppressed by the conferencing bride) .
So how will such application work without DTMF if it depended on it? I think that there should be some text about it (use DTMF in the signaling when there is a PERC MD e.g. SIP INFO?)

Roni

From: Perc [mailto:perc-bounces@ietf.org] On Behalf Of Suhas Nandakumar
Sent: יום א 06 אוגוסט 2017 23:42
To: perc@ietf.org<mailto:perc@ietf.org>
Subject: Re: [Perc] Confirmation of Consensus on PERC Double from IETF 99

Hello All

   Following up on the consensus confirmation email, the chairs have considered all the inputs received from the people in the room and the inputs from people on the email list and determined there is consensus for the following 2 items.

1.   Allow MD to modify the 'M' (marker) bit.

2. Includes all the below
    - Move the OHB information from header extension to payload
    - RTX, RED and FlexFEC ordering : use the ordering of applying repair on the double-encrypted packet. (Option 'A' in the slides)
    - DTMF : PERC will only support E2E DTMF and MD will not be able to read DTMF info sent as media

Thanks for your inputs.

Cheers
Chairs

On Thu, Jul 27, 2017 at 10:14 AM, Suhas Nandakumar <suhasietf@gmail.com<mailto:suhasietf@gmail.com>> wrote:

Hi All,

At the IETF 99 meeting, we took hum on the following proposals and there was a strong consensus in the room in their favor, but we wish to gather any additional inputs on the list.
So, if there are any additional inputs that was not expressed in the room, please send them to the list by 4th August.

First Consensus Call:
   Allow MD to modify the 'M' (marker) bit.

Second Consensus called made includes all the following 3 proposals as a singleton:
    - Move the OHB information from header extension to payload
    - RTX, RED and FlexFEC ordering : use the ordering of applying repair on the double-encrypted packet. (Option 'A' in the slides)
    - DTMF : PERC will only support E2E DTMF and MD will not be able to read DTMF info sent as media

Here are the notes from the meeting:
  https://www.ietf.org/proceedings/99/minutes/minutes-99-perc-01.txt

Here are the slides corresponding to the above proposals :
  https://www.ietf.org/proceedings/99/slides/slides-99-perc-double-01.pdf

Thanks
Perc Chairs

_______________________________________________
Perc mailing list
Perc@ietf.org<mailto:Perc@ietf.org>
https://www.ietf.org/mailman/listinfo/perc