Re: [Stox] Stox-media: Should XEP-176 translations have Require: ice?

[Peter Saint-Andre <[e-mail]>]

Once again, my apologies for the slow reply.

On 1/9/14, 10:31 AM, Jonathan Lennox wrote:
> I was thinking about the issue that unlike RFC 5245, XEP-176's
> definition of ICE doesn't support fallback to a non-ICE mode.

Correct.

> It occurred to me that SIP has a way of saying "do ICE, or fail the
> call": putting a "Require: ice" SIP option tag (from RFC 5768) in the
> SIP INVITE.
>
> Should we recommend this?  It clearly has the right semantics, and
> will prevent interop failure when a non-ICE SIP endpoint answers a
> XEP-176 Jingle call.

Theoretically that makes sense.

> My concern, though, is whether there are a) endpoints that implement
> RFC 5245 but not RFC 5768, or b) non-media-terminating B2BUAs that
> will pass ICE parameters through, but will reject calls with Require
> headers they don't know.  In either of these cases, adding this
> Require header would cause a call that would otherwise have worked to
> fail.

That would be bad.

> Any thoughts, especially from folks who know the state of deployed
> SIP better than I do?

Inquiring minds want to know. :-)

Peter