Hi Acee,

Em sex., 30 de set. de 2022 às 09:32, Acee Lindem (acee) <[email protected]>
escreveu:

> Hi Renato,
>
>
>
> *From: *Renato Westphal <[email protected]>
> *Date: *Thursday, September 29, 2022 at 7:56 PM
> *To: *Acee Lindem <[email protected]>
> *Cc: *"[email protected]" <[email protected]>, "
> [email protected]" <[email protected]>
> *Subject: *Re: [email protected]: questions
>
>
>
> Hi Acee,
>
> Thanks for your response. Please see my inline comments below.
>
>
>
> Em qua., 7 de set. de 2022 às 18:05, Acee Lindem (acee) <[email protected]>
> escreveu:
>
> Thanks for you comments. However, at this point in the cycle, we’re not
> going to make any additions to the model since it is has already been
> through the complete review cycle. We will however fix things that are
> broken.
>
>
>
> Understood. I only have a couple of additional comments for things that
> might be worth fixing before the RFC is released:
>
>
> 1.
> /ietf-routing:routing/control-plane-protocols/control-plane-protocol/ietf-ospf:ospf/areas/area/interfaces/interface/interface-id
>
> This leaf's type should be u32 and not u16.
>
>
>
> Ack.
>
>
>
> 2.
> /ietf-routing:routing/control-plane-protocols/control-plane-protocol/ietf-ospf:ospf/database/as-scope-lsa-type/as-scope-lsas/as-scope-lsa/ospfv3/body/intra-area-prefix/prefixes/prefix/metric
>
> This leaf's type should be u16 and not "ospf-metric" (u32).
>
>
>
> No – ospf-metric type is correct with the range limiting the metric to 24
> bits.
>

While uint32 isn't a problem per-se, using uint16 would be more precise
since the Metric field of Intra-Area-Prefix LSAs is 16-bits wide. The
"metric" leaf of Router-LSAs uses uint16 for instance. Only Summary and
External LSAs need 24-bit wide metrics. But I think it's ok to leave it as
it is.


>
>
> 3. While the "lls" feature is present in the model, the
> "ospfv2-lsa-option" base identity is missing an identity for the L-Bit
> specified by RFC 5613. While that could be added via augmentation, I think
> it would make sense to have an "l-bit" identity as part of the base model.
>
>
>
> The L-bit is never set in an LSA so it is not needed.
>
>
>
> https://www.rfc-editor.org/rfc/rfc5613.html#section-2.1
>
>
>
> Should we ever add operational state with the options received, it would
> be needed and could be added in that augmentation.
>

Understood.


> 4.
> /ietf-routing:routing/control-plane-protocols/control-plane-protocol/ietf-ospf:ospf/areas/area/interfaces/interface/neighbors/neighbor/cost
>
> Shouldn't the description be updated to include NBMA networks in addition
> to p2mp and hybrid networks?
>
>
>
> No. You only have a single cost for an NBMA network.
>

Ok. Makes sense considering that an NBMA network is essentially a
pseudo-node, so all neighbors need to have the same cost.


>
>
> 5.
> /ietf-routing:routing/control-plane-protocols/control-plane-protocol/ietf-ospf:ospf/local-rib/route/next-hops/next-hop
>
> This list is keyed by the `next-hop` leaf, which is an IP address. Given
> that directly connected routes don't have an IP address, shouldn't this be
> a keyless list instead?
>
>
>
> Since the local RIB only contains OSPF routes, they always have a next-hop.
>
>
>
> I was referring to the local RIB, where OSPF connected routes are
> calculated as a "byproduct" of the SPF algorithm. Example:
> https://pastebin.com/raw/zyDASfm2 (the connected routes are the ones
> without nexthops).
>
> In any case, section 16.1.1 of RFC 2328 says that nexthop addresses aren't
> required for point-to-point interfaces. So I'm still wondering whether a
> keyless list wouldn't be a better option.
>
>
>
> A next-hop of zero is used in those cases.
>

Ok.

Thanks for your time!

Best Regards,
-- 
Renato Westphal
_______________________________________________
Lsr mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/lsr

Reply via email to