[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: IKEv2 Correction (editorial)
>>>>> "Radia" == Radia Perlman <- Boston Center for Networking <Radia.Perlman@Sun.COM>> writes:
Radia> I don't understand why reserved fields shouldn't be "MUST be
Radia> sent as zero". At least for implementations of this
Radia> draft. There might be a future version that uses the bits, but
Radia> that spec would no longer say "MUST be sent as zero". Instead
Radia> it would say "MUST be set to the grobnitz filter value" or
Radia> whatever the reserved bits get used for.
Radia> If the spec for this version of IKE says SHOULD, doesn't that
Radia> allow a vendor to use the bits for some proprietary purpose?
Two reasons.
1. "SHOULD" on send rules is the standard rule.
2. If you say "MUST send as zero" that creates a strong temptation in
receivers to check for zero and reject the packet if a non-zero is
found. If that is done, the whole version thing breaks.
paul