Hi Tony,
On 04/03/2022 17:36, Tony Li wrote:
Hi Peter,
I would prefer to address the "else clause" in a following way:
a) any FAD sub-TLV MUST only appear once in a the FAD definition for a
given algorithm from a given source
b) in case the FAD sub-TLV appear multiple times, the values in the
sub-TLV in the first occurrence in the lowest numbered LSP from a
given source MUST be preferred.
Above does not support the "merge" of the values from multiple FAD
sub-TLVs, but as we all agree that is unlikely a requirement.
So, you’re in favor of mandating that FlexAlgo has fixed limitations on
the scale and complexity of its definitions?
not at all.
I just don't want to get into business of merging info from several
FAD's sub-TLVs of the same type unless there is a compelling reason to
do so? So far I have not seen any. Asking for 100s of excluded SRLGs in
the FAD does not seem like a realistic case to me.
thanks,
Peter
Will you be taking the phone calls from our irate customers?
Tony
_______________________________________________
Lsr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lsr
