On Fri, Feb 20, 2004 at 12:17:51PM -0800, Tony Hain wrote: > Alain, > > Recommending against all-zero's is not only a waste of time, it specifically > directs people to the thing you are trying to avoid. There is nothing that a > document can do to prevent a consultant from configuring all of his > customers with the same prefix. He can run [RANDOM] once and use that > everywhere, yet is in full compliance with the document. Even more of a > problem, an equipment manufacturer could run [RANDOM] once and burn that > value into every piece of equipment they ship. The best a document can do is > highlight the issues raised by duplicate assignments, and provide a > mechanism which solves them if used as directed. As much as some people want > to, it is not the IETF's job to legislate operational behavior. The language > in this document is appropriate and it should be published.
I agree, but I think Alain is right here so could the document not point out at least why all zeros is bad? I recall asking the same question in Minneapolis... I would add a third paragraph in 3.2.2 after the RANDOM paragraph that says something like: "The use of random IDs in the locally assigned Global ID prefix gives an assurance that any network numbered using such a prefix is highly unlikely to have that address space clash with any other network that has another locally assigned Global ID prefix allocated to it. This is a particularly useful property when considering a number of scenarios, e.g. networks that merge, overlapping VPN address space or hosts mobile between such networks." I'm sure those words could be improved, but it at least gives the carrot or incentive to the reader (who might otherwise think "why should I have to type in 8 random characters in my prefix when zero would work?"). Tim -------------------------------------------------------------------- IETF IPv6 working group mailing list [EMAIL PROTECTED] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
