ietf-ccamp-wg / ietf-network-inventory

3 stars 5 forks source link

Tom Petch WG adoption poll comments #58

Closed italobusi closed 9 months ago

italobusi commented 1 year ago

Too many authors for a RFC unless the AD agrees

Authors>> Ok, fixed in the -01 revision

'To be added' features 25 times, TBD 9 times, 'to be enriched' 4 times so there is plenty to do - perhaps more authors are needed!

Authors>> Ok, fixed in the -01 revision

I think that two letter prefixes should be avoided except for modules that will appear many times in path statements as e.g. te does

Authors>> The module name changed to ietf-network-hardware-inventory and the prefix changed to nhi in -01 revision to better match the draft scope

RFC6933 is in the module needs adding to I-D references

Authors>> Ok, fixed in the -01 revision

import of inet lacks a reference

Authors>> Ok, fixed in the -01 revision

ianahw is, well an IANA web site so needs a URL as a reference.

Authors>> Ok, fixed in the -01 revision (as per mail discussion)

revision clauses need to go at some point - perhaps this is an exercise to see if what the NETMOD WG is proposing is viable

Authors>> Ok, fixed in the -01 revision

lots of string with no constraint; given the prior art in SMI perhaps constraints similar to those of SMI would be appropriate.

Authors>> Discussion on the CCAMP and Netmod WG mailing lists has indicated that some people have concerns, but only when unconstrained strings are used as keys. The model is not using any unconstrained strings as keys, so no need to change

the tmforum web site tells me that the reference is to an old version of the asset

Authors>> Based on offline check with TMF experts, the reference provided in the document is correct and there is no new version of MTOSI after 4.0. The issue seems related to the TMF site maintenance.

See: https://mailarchive.ietf.org/arch/msg/ccamp/zD6gAfEUlYJ4W3qQlz6Y_gfX5TY/

italobusi commented 1 year ago

2023-01-11 Network Inventory weekly call

'To be added' features 25 times, TBD 9 times, 'to be enriched' 4 times so there is plenty to do - perhaps more authors are needed!

We need to review the document and check that all these issues are tracked in github. The plan is to resolve them through normal WG process once the document is adopted as CCAMP WG draft.

I think that two letter prefixes should be avoided except for modules that will appear many times in path statements as e.g. te does

The module name will be renamed as ietf-network-hardware-inventory to better match its scope so the prefix can be changed to nhi

RFC6933 is in the module needs adding to I-D references

Ok, agreed

import of inet lacks a reference

Ok, agreed (the reference to be added is RFC6991)

ianahw is, well an IANA web site so needs a URL as a reference.

Not sure this is required

revision clauses need to go at some point - perhaps this is an exercise to see if what the NETMOD WG is proposing is viable

Ok, agreed

lots of string with no constraint; given the prior art in SMI perhaps constraints similar to those of SMI would be appropriate.

Using string with no constraint is quite common practice in IETF YANG models but we can check with Netmod WG for guidelines

the tmforum web site tells me that the reference is to an old version of the asset

Need to check

italobusi commented 1 year ago

ianahw is, well an IANA web site so needs a URL as a reference.

  • [x] @italobusi check RFC8407 and, if needed, ask Netmod WG for input

No guidelines found in RFC8407

Mail sent to Netmod WG: https://mailarchive.ietf.org/arch/msg/netmod/uQYQRbkkCD6jScGW3jN4fdKSp78/

lots of string with no constraint; given the prior art in SMI perhaps constraints similar to those of SMI would be appropriate.

  • [x] @italobusi check with Netmod WG for guidelines (start a new thread from previous discussion triggered by Tom)

Mail sent to Netmod WG: https://mailarchive.ietf.org/arch/msg/netmod/eP6wVNQmTBGmC3ru3Nts571YasA/

sergiobelotti commented 1 year ago

the tmforum web site tells me that the reference is to an old version of the asset

Authors>>

  • [x] @sergiobelotti check with TM Forum experts

See: https://mailarchive.ietf.org/arch/msg/ccamp/zD6gAfEUlYJ4W3qQlz6Y_gfX5TY/ for my AP: Related to Tom Petch comment on TMF document "the tmforum web site tells me that the reference is to an old version of the asset" I checked with TMF expert and they told me that there is no new version of MTOSI after 4.0, so they have raised the issue to "member service" of TMF. But the important point is that the reference is correct, the problem is related to the TMF site maintenance.

italobusi commented 1 year ago

2023-03-01 Network Inventory weekly call

the tmforum web site tells me that the reference is to an old version of the asset

Authors>>

  • [x] @sergiobelotti check with TM Forum experts

See: https://mailarchive.ietf.org/arch/msg/ccamp/zD6gAfEUlYJ4W3qQlz6Y_gfX5TY/ for my AP: Related to Tom Petch comment on TMF document "the tmforum web site tells me that the reference is to an old version of the asset" I checked with TMF expert and they told me that there is no new version of MTOSI after 4.0, so they have raised the issue to "member service" of TMF. But the important point is that the reference is correct, the problem is related to the TMF site maintenance.

Agreed reply:

Based on offline check with TMF experts, the reference provided in the document is correct and there is no new version of MTOSI after 4.0. The issue seems related to the TMF site maintenance.

italobusi commented 1 year ago

2023-03-08 Network Inventory weekly call

lots of string with no constraint; given the prior art in SMI perhaps constraints similar to those of SMI would be appropriate.

Agreed reply:

discussion on the CCAMP and Netmod WG mailing lists has indicated that some people have concerns but only when unconstrained strings are used as keys. The model is not using any unconstrained strings as keys, so no need to change.

italobusi commented 9 months ago

All the comments have been addressed in draft-ietf-ccamp-network-inventory-yang-02