> > What I am saying is IPv6 has to work without LLMNR today. > > > > So lets assume the application obtains address to send to from > > foo.here.now.com by getaddrinfo not from user interface? > > > > You said LLs should not be in DNS. > > > > Hence. This case will not work for LLs. > > you can resolve the name "foo.here.now.com" into > link-local address > with scope id by LLMNR.
Uh........I said today? No one has deployed LLMNR as product and I applaud our colleague at Apple for doing what I recall was mdns. But it don't exist and its unclear it will be widely currently. Plus this now implies the node has this yet other software running called LLMNR? Not sure I agree that is good deployment model at this point. Ergo "today" it don't work. > > > Lets assume app uses command line and app name is mickeyfinn > > > > csh> mickeyfinn fe80::1 > > > > May or may not work correctly. > > correct. it won't work as the specified address is ambiguous. > error mode might be OS dependent. In addition to what Brian stated using sin6_scope_ID in this case we will try round robin if it fails but the fear is now what if it goes to the wrong node? > > > If > > > > csh> mickeyfinn fe80::1%eth0 > > > > How did the user know to use "eth0" (That is the issue.........) > > > > Lets just discuss the above for now for clarity? > > LLMNR (or icmp6 node info query). Neither have consenus or support in the market for products. At this point. But again I am unclear this is a good deployment mode for general applications. thanks /jim > > itojun > -------------------------------------------------------------------- IETF IPng Working Group Mailing List IPng Home Page: http://playground.sun.com/ipng FTP archive: ftp://playground.sun.com/pub/ipng Direct all administrative requests to [EMAIL PROTECTED] --------------------------------------------------------------------
