(B > I have a comment on issue 246 of rfc2461bis (preferred lifetime >
(B > valid lifetime).
(B >
(B > As far as I can see, the latest version of the draft
(B > draft-soliman-ipv6-2461-bis-01.txt introduces two changes for the
(B > issue:
(B >
(B > 1. Section 4.6.2 says:
(B >
(B > Preferred Lifetime
(B > 32-bit unsigned integer. The length of time in
(B > seconds (relative to the time the
(B > packet is sent)
(B > that addresses generated from the prefix via
(B > stateless address autoconfiguration remain
(B > preferred [ADDRCONF]. A value of all one bits
(B > (0xffffffff) represents infinity. See
(B > [ADDRCONF].
(B > ---> Note that the value of this field MUST
(B > NOT exceed
(B > ---> the Valid Lifetime field to avoid preferring
(B > ---> addresses that are no longer valid.
(B >
(B > 2. Section 6.2.1 says:
(B >
(B > AdvPrefixList
(B > [...]
(B > AdvPreferredLifetime
(B > [...]
(B > Default: 604800 seconds (7 days), fixed
(B > (i.e., stays the same in consecutive
(B > ---> advertisements). This value MUST NOT be
(B > ---> larger than AdvValidLifetime.
(B >
(B >
(B > As I said in a related discussion for rfc2462bis, the change in
(B > Section 4.6.2 seems overspecification according to the original
(B > consensus that led to the issue. See
(B > http://www1.ietf.org/mail-archive/web/ipv6/current/msg03034.html
(B > for more details. As I mentioned in this message, I think the
(B > restriction in Section 4.6.2 (the part marked with "---->") should be
(B > removed.
(B
(B=> Ok, but you said that 6 months after the text was added :)
(BThe concensus back then was to add the current text. If you
(Bwant we can revisit the issue before LC.
(B
(B >
(B > Additionally, it is probably not enough to say "AdvPreferredLifetime
(B > MUST NOT be larger than AdvValidLifetime", since either one
(B > or both of
(B > the lifetimes can be configured so that it decrements in real
(B > time. For example, consider the case where
(B >
(B > AdvValidLifetime = 40000 (decrementing in real time)
(B > AdvPreferredLifetime = 30000 (fixed time).
(B >
(B > These configurations seem valid according to the specification.
(B > However, the advertised valid lifetime would be smaller than the
(B > advertised preferred lifetime about 10000 seconds later.
(B
(B=> There is one issue here: These restrictions are specified in the context
(Bof the _advertised_ information on the wire. I'm sure that there are many
(Bdifferent ways of configuring both fields. The requirement is that _whenever_
(Bsending_this_on_the_wire_ the preferred liftime must not be larger than
(Bvalid lifetime. This is independent of how people configure fields or
(Bimplement their storage.
(B
(B > A. if AdvValidLifetime decrements in real time, require
(B > AdvPreferredLifetime decrement as well
(B > B. require routers to ensure the valid lifetime be equal to or larger
(B > than the preferred lifetime at sending time, regardless of the
(B > values of AdvValidLifetime and AdvPreferredLifetime.
(B
(B=> B. is what's implied already since we're talking about protocol
(Bfields. When we define protocol fields it is clear that the definitions are
(Bin the context of communication over the wire.
(B
(B
(BHesham
(B
(B
(B===========================================================
(BThis email may contain confidential and privileged material for the sole use
(B of the intended recipient. Any review or distribution by others is strictly
(B prohibited. If you are not the intended recipient please contact the sender
(B and delete all copies.
(B===========================================================
(B
(B
(B--------------------------------------------------------------------
(BIETF IPv6 working group mailing list
(B[EMAIL PROTECTED]
(BAdministrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
(B--------------------------------------------------------------------
