Hi Tom,

Thank you for your review and comments. I’ve submitted version -11 to address 
your comments with detailed answers below.

One open issue is about inet:ip-address, while we wait for consensus on this 
topic we can continue progress this draft.

Thanks,
Yingzhen

> On May 4, 2022, at 3:02 AM, t petch <[email protected]> wrote:
> 
> On 03/05/2022 01:01, Jeff Tantsura wrote:
>> Dear RTGWG,
>> 
>> This mail begins a Working Group Last Call (WGLC) on
>> draft-ietf-rtgwg-yang-rib-extend-10.
> 
> I find some of the language less than clear.
> 
> RFC8349 is specific that a RIB is for a single address family (which is not 
> how I see it used in other places) so I think that that needs calling out in 
> section 1.
<Yingzhen> I updated the description in section 1 to match RFC8349.
> 
> tag is defined in RFC8349 as 16 bit and specific to RIP.  Here it is 32 bit 
> and generic.  I do not understand what is meant by the YANG leaf tag in this 
> I-D.
<Yingzhen> do you mean the tag in the example model in RFC 8349 appendix C? 
Which is an example for tag might be used in RIP. The YANG left tag in this I-D 
if for a tag in a RIB, which might be achieved through different routing 
protocols. Now you mention it, I’m thinking wether this should be changed to 
leaf-list.
> 
> tree snapshot is used in several places but these are not snapshots, extracts 
> perhaps.
<Yingzhen> I’ll leave this to RFC editor for editorial change.
> 
> the YANG module is scrappy, lots of independent scraps.  This can be made 
> clearer with a
> /* comment */
> line at the start of each scrap.  Of course the comment needs to be correct; 
> the YANG description of the first augment seems at odds with the augment 
> itself.
<Yingzhen>  This is due to the augmentations are really scattered.
> 
> active route seems a different concept from RFC8349.  Here it is per protocol 
> per destination prefix; in RFC8349 it is per destination prefix.
<Yingzhen> This I-D is only augmenting with more route attributes. If you meant 
the “active-routes” in the protocol statistics, yes, this is per-protocol, and 
there is total-active-routes for one RIB. 
> 
> I assume that the choice of inet:ip-address is deliberate, following RFC8349.
<Yingzhen> yes, this is following RFC8349. I’ve been been following the 
discussions, and waiting for a community consensus.
> 
> There is an RFC somewhere specifying the representation of IPv6 addresses 
> which I cannot find but think that the JSON example gets it right and the XML 
> example gets it wrong.
<Yingzhen> do you mean "0::/0” or "::/0” ? Thanks for catching this, I change 
the xml example to “::/0” to be consistent with JSON.



> 
> Tom Petch
> 
> 
>> 
>> The authors have addressed all the comments received from the wg, YANG 
>> doctors
>> and routing directorate have been addressed, thanks authors for timely 
>> responses.
>> 
>> IPR: No IPR has been submitted against this draft.
>> 
>> Authors -  please respond to the WG whether you are aware of any applicable 
>> IPR.
>> 
>> Thanks!
>> 
>> Cheers,
>> 
>> Jeff
>> 
>> 
>> 
>> _______________________________________________
>> rtgwg mailing list
>> [email protected] <mailto:[email protected]>
>> https://www.ietf.org/mailman/listinfo/rtgwg 
>> <https://www.ietf.org/mailman/listinfo/rtgwg>
_______________________________________________
rtgwg mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/rtgwg

Reply via email to