[Standards] Proposed XMPP Extension: STUN Server Discovery forJingle

Robert McQueen robert.mcqueen at collabora.co.uk
Wed Mar 28 14:31:12 UTC 2007


Peter Saint-Andre wrote:
> Matt Tucker wrote:
>>  * Shouldn't the port value be optional? STUN typically operates at a
>> well-known port of 3478. If no port is specified, it would be a good
>> idea for the client to check DNS SRV for the host first, then use the
>> default port if that fails. 
> 
> True, I'll fix that.

This strikes me as kinda inconsistent, saying in the preamble that the
intention of this XEP is to avoid the requirement for SRV lookups on the
server (in case this is not possible to implement), and then later
saying that if you don't get given a port number then you have to do a
SRV lookup. Is the rationale that you want to allow servers to get away
without SRV records, or do you want to allow clients to get away without
implementing SRV lookups?

Other than this, this XEP gets a +1 from me. Might be worth saying that
the server's disco info reply should be used to find whether this
protocol is supported.

> I'm open to suggestions. Though I think the "please tell me my external
> IP address" feature is a different beast, no?

As well as the "get me my IP" function, if you're using a raw UDP
connection to your peer, STUN requests have the side-effect of creating
a mapping in your NAT for that port, which allows direct media to flow
in the cases where the NAT doesn't consider the source address/port
(just the incoming address/port) for routing incoming UDP. This makes an
XMPP-based address discovery significantly less useful.

> /psa

Regards,
Rob



More information about the Standards mailing list