IPV6 options draft - AD evaluation comments

[shwethab]

Martin Duke [email protected]
Apr 29, 2022, 10:23 AM (3 days ago)
to draft-ietf-ippm-ioam-ipv6-options.all

Thanks for a concise document!

My main concern was the clarity of Section 4. I was not clear on what this was specifying due to these problems:

• "Deployments select one of these extension header types depending on how IOAM is used, as described in section 4 of [I-D.ietf-ippm-ioam-data]"

That draft certainly doesn't say anything about extension header types. Guidance about which header type to use is actually later in Section 4 of this document!

• "IOAM Type: 8-bit field as defined in section 7.2 in [I-D.ietf-ippm-ioam-data]."

Sec 7.2 does not define this at all.

• I do not understand the presentation format of the five bullets after "In-situ OAM Option-Types are inserted as Option data as follows:", The previous text defines "option types" and "IOAM types" but each of these bullets provides an "IOAM Option Type". Which does this correspond to? Please don't introduce a new and confusing term here! After skimming through ioam-data, I think the "Option Type" is the bit field listed in this draft, and the "IOAM type" field is taken from the registry in Sec 8.1 of ioam-data. Is that correct? If so, it is very confusingly described.

• After several re-readings, I think there are actually only two Option Types that map to the five IOAM types. Perhaps, instead of using 'xxxxx' for all unassigned bits, you can use 'xxxxx' and 'yyyyy' to make the overlaps among these option types more clear?

NITS:

(4) s/lenght/length

(5.1) I'm not sure what you mean by "the source that inserts and leaks the IOAM data". It both inserts and leaks it?

(5.4.1) please expand ION on first use.

[shwethab]

possible table to map data option type to v6 header - and options