Re: [rtcweb] IP handling: Using mDNS names for host candidates

PTAL at https://github.com/juberti/draughts/pull/103.

It would be great to have some time to discuss this in Montreal.

On Tue, Jun 12, 2018 at 4:36 PM Justin Uberti <juberti@google.com> wrote:

> Given the general support here, I'll start writing up a PR to incorporate
> the aforementioned changes, with the goal to avoid emitting private IPv4
> addresses in the default mode.
>
> On Tue, Jun 12, 2018 at 12:14 PM Justin Uberti <juberti@google.com> wrote:
>
>> One option could be that Mode 2 incorporates RFC 4941 stateless IPv6
>> addresses, if supported. Otherwise, mDNS is used. This mode would never
>> include RFC 1918 IPv4 addresses.
>>
>> This would work well in more environments and still provide significant
>> privacy improvements.
>>
>> On Tue, Jun 12, 2018 at 12:08 PM Adam Roach <adam@nostrum.com> wrote:
>>
>>> I’d like to point out that, while this works fine in most residential
>>> settings, it’s pretty broken for multi-segment enterprise deployments.
>>>
>>> /a
>>>
>>> On Jun 12, 2018, at 11:55, Ted Hardie <ted.ietf@gmail.com> wrote:
>>>
>>> On Tue, Jun 12, 2018 at 2:40 AM, Justin Uberti <
>>> juberti=40google.com@dmarc.ietf.org> wrote:
>>>
>>>>
>>>> This is a significant improvement and I think we will want to
>>>> incorporate this suggestion. Is this something we could do as part of this
>>>> WGLC, or should we look for another option?
>>>>
>>>>
>>> Without having cleared this with Cullen or Sean, my personal chair-hat
>>> opinion is that we can do this in WGLC.  If a new technical solution is
>>> found during WG last call, I see no reason not to incorporate it.
>>>
>>> That said, I see two no-hats issues that will want pretty strong text.
>>> The first is that these are really UUIDs, not traditional mDNS names.
>>> We'll need text to strongly discourage the re-use of an existing mDNS name,
>>> since those can leak other information.  Second, we'll need text on what to
>>> do if this name can't be registered or resolved in a particular environment
>>> (not every network supports mDNS, after all).  Does it go back to the
>>> previous Mode 2 behavior, or skip private addresses entirely?  I think the
>>> right idea is "go back to the previous Mode 2 behavior" personally, but
>>> text on it one way or the other is required.
>>>
>>> regards,
>>>
>>> Ted
>>>
>>>
>>>
>>>
>>>> _______________________________________________
>>>> rtcweb mailing list
>>>> rtcweb@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/rtcweb
>>>>
>>>>
>>> _______________________________________________
>>> rtcweb mailing list
>>> rtcweb@ietf.org
>>> https://www.ietf.org/mailman/listinfo/rtcweb
>>>
>>>