Alvaro: Yes – this is the format I was indicating – In some BGP-LS document I’ve seen tables that map the BGP-LS TLVs to the corresponding ISIS/OSPF TLVs — with the understanding that the format is different. One such example is in §2.4/2.5 of draft-ietf-idr-bgp-ls-segment-routing-ext [1]. Is that the alignment you’re asking about? If so, then I think that such a table would be useful. This makes the implementations easier to create. More importantly helps those who deploy the protocol additions to understand how to debug issues. Having started out my Internet career as a OPS person, I like to see things that make life easier for the OPS/NM people. We can comments in each draft or put pointers in a single draft (living or RFC). … just wondering what would help.. I agree with your ADs comment on the text – that said “The semantic of the TLV is described in [RFC7810] and [RFC7471]” (in -14) to "The semantics of the value field in the TLV are described in [I-D.ietf-lsr-isis-rfc7810bis] and [RFC7471]” (in -15)…and the text about "TLV formats follow the rules defined in [RFC7752]” was added, as Les pointed out. Your comment makes the text correct. Thanks for catching that point. I thought Les already agreed to that change – so I did “ack” it or expand on it. Sue From: Alvaro Retana [mailto:aretana.ietf@xxxxxxxxx] On December 14, 2018 at 9:49:28 AM, Susan Hares (shares@xxxxxxxx) wrote: Sue: Hi! I think we might be talking about two different things…or I got lost, which is completely possible. :-) I think Yoshi is asking about the format of the TLVs, which happens to be the same as the OSPF ones (rfc7471), but slightly different than ISIS (rfc7810bis): the type and length fields are different. I had already pointed this out in my AD review, so the text changed from "The semantic of the TLV is described in [RFC7810] and [RFC7471]” (in -14) to "The semantics of the value field in the TLV are described in [I-D.ietf-lsr-isis-rfc7810bis] and [RFC7471]” (in -15)…and the text about "TLV formats follow the rules defined in [RFC7752]” was added, as Les pointed out. I think this is enough. When you say "align with IGP TLVs”, are you talking about the format of the TLV, or are you referring to something else? In some BGP-LS document I’ve seen tables that map the BGP-LS TLVs to the corresponding ISIS/OSPF TLVs — with the understanding that the format is different. One such example is in §2.4/2.5 of draft-ietf-idr-bgp-ls-segment-routing-ext [1]. Is that the alignment you’re asking about? If so, then I think that such a table would be useful. Thanks! Alvaro.
|
