Site-local addresses are more problematic - they cross more architectural/political/layer 8 boundaries. As you suggest, there are issues with address assignment and delegation, routing, provider independence, etc. I'm not sure we have a tight/complete set of requirements that are not met by global addresses to guide the design of some other form of addressing ... so I don't know that we can effectively discuss whether one solution or another is the right solution.
- Ralph
At 12:43 PM 8/22/2003 -0700, Fred Templin wrote:
Hi Ralph,
I think I'm beginning to notice a trend in these discussions. There seem to be strong arguements against site-local addresses, link-local addresses, multi-addressing, and even "limited-range" addresses such as those proposed in the Hinden/Haberman draft. Without any of these options, it seems to me that all we would have left to consider would be globals.
But, how can a host get a global if it comes up in an ad-hoc network with no router? The only option I can see is for the host to have a "burned-in" global address that comes from, e.g., a centralized addressing authority and is assigned, e.g., by the manufacturer, by the user's manual configuration, etc. So, the global would be just like a MAC address in this sense - right? This would all be fine, but we are hit once again with the question of how this would impact routing scalability. Any ideas?
Fred [EMAIL PROTECTED]
Ralph Droms wrote:
At 10:00 PM 8/21/2003 -0700, Tony Hain wrote:
This is a clear capability & advantage that IPv6 brings over IPv4. The only thing holding it back is the obstinate views of those who don't want to make the scenarios work. After-all they don't work in IPv4, so they must not be really needed, right???
Tony - (assuming "they" == IPv6LL) can you explain why IPv6LL will work while "they don't work in IPv4"? My experience with IPv4LL has been uniformly bad; I've never intentionally used an IPv4LL address and the automatic assignment of an IPv4LL address has on several (many?) occasions silently interfered with my ability to assign a non-LL address and use greater internet connectivity. I will admit ignorance and am happy to hear success stories about IPv4LL.
I don't know that some folks "don't want to make the scenario work". I don't understand the advantage to IPv6LL from the scenario you described. We can assign IPv4LL addresses today in a one-link, no router ad-hoc network. But we have to enter the addresses manually and those addresses get in the way when full Internet connectivity becomes available. What's different with IPv6LL?
- Ralph
-------------------------------------------------------------------- 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] --------------------------------------------------------------------
-------------------------------------------------------------------- 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] --------------------------------------------------------------------
-------------------------------------------------------------------- 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] --------------------------------------------------------------------
