[Standards] Proposed XMPP Extension: STUN Server Discovery forJingle

Scott Ludwig scottlu at google.com
Wed Mar 28 05:03:38 UTC 2007


The updates are mostly for receiving a refreshed relay server
authentication token, since the token times out after a bit. The STUN
servers rarely change.

On 3/27/07, Peter Saint-Andre <stpeter at jabber.org> wrote:
> Gosh another top poster. :)
>
> Matt Tucker wrote:
> >
> > -Matt
> >
> >> -----Original Message-----
> >> From: standards-bounces at xmpp.org
> >> [mailto:standards-bounces at xmpp.org] On Behalf Of Peter Saint-Andre
> >> Sent: Tuesday, March 27, 2007 3:11 PM
> >> To: XMPP Extension Discussion List
> >> Subject: Re: [Standards] Proposed XMPP Extension: STUN Server
> >> Discovery forJingle
> >>
> >> Mircea Bardac wrote:
> >>> On Saturday 24 March 2007 00:17:56 XMPP Extensions Editor wrote:
> >>>> The XMPP Extensions Editor has received a proposal for a new XEP.
> >>>>
> >>>> Title: STUN Server Discovery for Jingle
> >>>>
> >>>> Abstract: This document specifies methods for discovering STUN
> >>>> servers for use in negotiation of certain Jingle transport methods.
> >>>>
> >>>> URL: http://www.xmpp.org/extensions/inbox/stun-discovery.html
> >>> I might have been influenced by the whole PEP discussion... but...
> >>> after reading this short XEP, I couldn't help imagining how this
> >>> information can be published via PEP. A server-managed
> >> PEP-node, where
> >>> clients willing to discover such could subscribe be notified.
> >> Heh, true. No need to have a specialized protocol where the
> >> server needs to keep track of who has "registered" for STUN
> >> server "events". Make it an open-access node so that anyone
> >> can request the last event. Etc. But you wouldn't get any
> >> kind of token from the XMPP server that way, only the STUN
> >> server host+port.
>
>  > I actually think getting server updates for new STUN lists is total
>  > overkill. Would anyone other than Google ever use this? :)
>
> Heh, probably not. :)
>
>  > I suppose
>  > we could have the simple IQ plus optional PEP for those
>  > services/clients that actually need updates?
>
> I think the simple IQ is probably enough.
>
> Peter
>
> --
> Peter Saint-Andre
> XMPP Standards Foundation
> http://www.xmpp.org/xsf/people/stpeter.shtml
>
>
>



More information about the Standards mailing list