[Standards-JIG] Re: What happened to the ACK proposal?
justin-keyword-jabber.093179 at affinix.com
Tue Aug 16 17:15:02 UTC 2005
On Tuesday 16 August 2005 08:15 am, Ian Paterson wrote:
> One thing I would like to see added to RFC 3920bis is something that no
> single-hop transport layer can solve. If a proxy finds it can't deliver
> a stanza to a client then it needs to be able to bounce the stanza back
> to the user's server. The server can then deliver the stanza the next
> time the user logs in, even if that is via a different proxy or a direct
Probably the best answer is for the proxy to not count as a real XMPP hop,
even if it is "smart" enough that it could be (such as an HTTP bind proxy).
For example, a SOCKS proxy won't know anything about XMPP anyway, and so even
if the path is Client->Proxy->Server, it is still just a single XMPP hop. If
the Server doesn't receive an Ack from the Client about a stanza, then the
Server could choose to redeliver it at a later time.
More information about the Standards