Re: [MMUSIC] SCTP-SDP: Virtual Connection impact

[Paul Kyzivat <[e-mail]>]

On 4/7/15 7:13 AM, Christer Holmberg wrote:
> Hi,
>
> Any comments on this from others? We need to solve it.
>
> Anyone has issues with allowing connection:new for SRTP-DTLS and
> SCTP-DTLS, and use that as a trigger to re-negotiate DTLS?

ISTM that this is being considered tactically. Maybe that is the best 
for now, but perhaps there should be consideration of a strategic 
approach to this as well.

What I mean is that this seems to be a back handed way to introduce 
multipath support into protocols (specifically DTLS) that don't support 
it. Perhaps multipath ought to be officially added to all the protocols 
that potentially need ICE, and then ICE would simply be one way to 
populate the paths.

	Thanks,
	Paul

> Regards,
>
> Christer
>
> *From:*mmusic [mailto:[e-mail]] *On Behalf Of *Christer
> Holmberg
> *Sent:* 3. huhtikuuta 2015 21:00
> *To:* Roman Shpount
> *Cc:* [e-mail]; Paul Kyzivat
> *Subject:* Re: [MMUSIC] SCTP-SDP: Virtual Connection impact
>
> Hi,
>
> There are obviously "combinations" that are ambiguous, but you'll find
> that elsewhere too.
>
> Or, we simply define an order in which things (5-tuple changes,
> attribute values etc) are checked.
>
> Or, we do NOT use the attribute for non-ICE.
>
> Or, we do something else - but we DO need to solve this.
>
> Regards,
>
> Christer
>
> Sent from my Windows Phone
>
> ------------------------------------------------------------------------
>
> *From: *Roman Shpount <mailto:[e-mail]>
> *Sent: *‎03/‎04/‎2015 20:00
> *To: *Christer Holmberg <mailto:[e-mail]>
> *Cc: *Paul Kyzivat <mailto:[e-mail]>; [e-mail]
> <mailto:[e-mail]>
> *Subject: *Re: [MMUSIC] SCTP-SDP: Virtual Connection impact
>
> On Fri, Apr 3, 2015 at 10:40 AM, Christer Holmberg
> <[e-mail] <mailto:[e-mail]>>
> wrote:
>
>
> In any case, I don't think it would be too late to allow usage of the
> attribute for SRTP-DTLS. It would be a straight forward mechanism, which
> could be used both in the ICE and non-ICE case.
>
> I agree this is not too late to allow usage of the attribute for
> SRTP-DTLS or any other protocol running on top of DTLS, but there are
> several backward compatibility questions:
>
> First of all, should the DTLS connection be reset on the transport
> (5-tuple) change in the non-ICE case if fingerprint stayed the same and
> the connection attribute is not present? Current specification says that
> it should, so this needs to be preserved for backwards compatibility.
>
> Also, should sending connection attribute set to "existing" and changing
> the underlying transport be allowed? If backwards compatibility to be
> maintained, it should not.
>
> _____________
>
> Roman Shpount
>