Sebastian, The current version tries also to address what you call "normal" ALTO server discovery and the authors said that a name change will follow. There is also draft-song-alto-server-discovery-03 that tried to address the points made in previous meeting and I thought that the direction was to merge both drafts and have just one.
Roni Even > -----Original Message----- > From: [email protected] [mailto:[email protected]] On Behalf Of > Sebastian Kiesel > Sent: Tuesday, July 27, 2010 12:03 AM > To: Thomson, Martin > Cc: [email protected] > Subject: Re: [alto] the discovery problem > > Martin, > > On Mon, Jul 26, 2010 at 07:51:06PM +0800, Thomson, Martin wrote: > > draft-ietf-geopriv-lis-discovery probably provides most of what you > > will need for your application. You could possibly even use the DHCP > > option defined there. > > This might be true for the "normal" ALTO server discovery, when the > ALTO client is located in the resource consumer (e.g., peer of a p2p > app), which will eventually access the desired resource. > > The 3pdisc use case is different: > > Host "B" (e.g., a tracker) connected to an arbitrary network receives > an > inbound TCP/UDP flow from another host "A" (e.g., peer) in another > arbitrary access network. We do not want to make any assumptions on > the > L7 protocol, so all "B" knows is the source IP in the packets that > belongs to "A" (or to the outermost NAT box in front of "A", which > should be enough for our purposes, carrier-grade NAT might need > additional discussion). Knowing only this IP address, host "B" then > needs to discover an ALTO server that can give guidance for optimizing > host A's peer selection. > > This seems to be basically the same scenario as described in sec. 3.2 > of > http://tools.ietf.org/id/draft-thomson-geopriv-res-gw-lis-discovery- > 03.txt > What happened to that draft? > > > Thanks, > Sebastian > _______________________________________________ > alto mailing list > [email protected] > https://www.ietf.org/mailman/listinfo/alto _______________________________________________ alto mailing list [email protected] https://www.ietf.org/mailman/listinfo/alto
