Re: [Last-Call] Opsdir last call review of draft-ietf-lsr-ospf-l2bundles-06

[Ketan Talaulikar <[e-mail]>]

Hi Dan,

Thanks for your review and please check inline below for responses.


On Sat, Sep 24, 2022 at 1:25 PM Dan Romascanu via Datatracker <
[e-mail]> wrote:

> Reviewer: Dan Romascanu
> Review result: Has Issues
>
> Ready with Issues.
>
> This document defines the protocol extensions for OSPF to advertise the
> link
> attributes of L2 interface bundle members such as instances of a IEEE
> 802.1AX
> Link Aggregation Group (LAG).
>
> This is a very useful document. It's clear and well-written. Operators that
> deploy OSPF in environments where IEEE 802.1AX or other L2 bundle
> interfaces
> may be present in routers will find this document very relevant.
>
> My main issue is related to the lack of any operational or management
> considerations. I believe that the following should be mentioned
> explicitly or
> by reference in a short Operational and Manageability Issues section:
>
> - any backwards compatibility or transition issues at deployment
> - any impact on traffic
> - any configuration parameters and default values
> - management including possible extensions of YANG or MIB modules
>

KT> These aspects are covered in the introduction section - the last
paragraph and bullet list thereafter. They are not a section in this
document since these aspects are outside the scope of this document. There
are no backward compatibility considerations since this information is
not used for OSPF protocol operations. The configuration of the per member
link attributes may not be even under an OSPF model. Some of them like
link bandwidth may be obtained from the interface model. AFAIR there has
been no requirement or discussion for their inclusion in the IGP models in
the WG even during similar work in ISIS.


>
> I agree with the issues raised by the Gen-ART reviewer concerning the
> references, with the difference that I believe that [IEEE 802.1AX] should
> remain a Normative Reference. It's a full IEEE 802.1 standard, so
> referencing
> as Normative is fine.
>

KT> Thanks for the confirmation.


>
> I would also suggest sending the document for review to the IEEE 802.1 WG
> via
> the IETF-IEEE coordination team.
>

KT> I am not sure if this is necessary, but I see that John has already
reached out and so we'll wait for feedback on this one.

Thanks,
Ketan