Re: [Idr] Shepherd report on draft-ietf-idr-bgp-ls-sr-policy-00

[Ketan Talaulikar <[e-mail]>]

Hi Sue,

As indicated during the presentation during IETF 117, the authors would
like to request allocations for the pending code points for this draft as
follows:

1214 (suggested) SR Segment List Identifier

1215 (suggested) SR Bidirectional Group Constraint

1216 (suggested) SR Metric Constraint

1217 (suggested) SR Segment List Bandwidth


This will help us start preparation of implementation reports for this
draft as well.

Also, thanks for your review and please check inline below for responses.


On Fri, Sep 8, 2023 at 4:02 PM Susan Hares <[e-mail]> wrote:

> This is a shepherd report for draft-ietf-idr-bgp-ls-sr-policy-00.
>
>
>
> This draft will need to be adjusted after
> draft-ietf-idr-segment-routing-te-policy is split into two drafts.
>

KT> I am not sure that would be necessary. That draft was about BGP SR
Policy SAFI and this is BGP-LS SAFI - they are different and independent.


>
>
> Why?: draft-ietf-idr-segment-routing-te-policy is an informative reference.
>

KT> There is no normative dependence on the BGP SR Policy SAFI document.
This BGP-LS draft is about (northbound) advertisement of SR Policies from
routers (headend) to controllers. The source of the SR Policy on the
headend could be local configuration (e.g., via CLI, Netconf, etc.) or via
protocols PCEP or BGP SR Policy SAFI.


>
>
> Other-things to fix:
>
>
>
> Section 5.2 –
>
>
>
> U-flag – specify what the flag means when the bit is zero.
>

KT> When clear it indicates that this CP is using the specified BSID. This
flag is ignored when there is no specified BSID. Will include this in the
next update.


>
>
> Section 7.0
>
>
>
> Thank you for putting RFC7752bis in this document.
>
> I recommend you consider adding a bit more text prior to WG LC.
>
> Consider issues on rapid change and detection.
>

KT> Could you please elaborate a bit more on this? There is nothing of the
nature of "statistics" or "counters" that are being advertised here which
could change rapidly. This is quite similar to protocol state - albeit
related to SR Policies.


>
>
> Section 9.0
>
>
>
> This section needs upgrading to present the likely problems.
>
> At least you should consider what RFC7752bis issues have been closed.
>

KT> Agree. Will update this section for the next update.

Thanks,
Ketan


>
>
> I will add this to the early review queue as soon as the
> draft-ietf-idr-segment-routing-te-policy is completed.
>
>
>
> Cheers,  Sue
>