Probably not. Having the proper code is a really nice first step.
I agree.
To respond to some of the other discussion, the purpose of the new codes was to provide more feedback to address selection. As I think Pekka said in an earlier discussion, relying on transport layer timeouts to get this feedback is not a good idea.
I think for all of the new destination unreachable codes, if the source node has only a few addresses to choose from this feedback will very useful as there won't be many alternatives to try. If the source node has many addresses to try, then it is less beneficial. However, I don't think the latter is likely to be the case in common practice.
Overall I think that providing this additional feedback is useful and I don't think it does any harm. The ICMP destination unreachable messages are a SHOULD so the router has considerable choice if it wants to send them or not.
Bob
-------------------------------------------------------------------- IETF IPv6 working group mailing list [EMAIL PROTECTED] Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6 --------------------------------------------------------------------
