Re: [yang-doctors] Dealing with BFD RFC 9127 client-cfg-parms for PIM, OSPF, ISIS and other BFD clients on some platforms

[Mahesh Jethanandani <[e-mail]>]

> On Nov 8, 2021, at 1:41 PM, Jeffrey Haas <[e-mail]> wrote:
> 
> Ah, now I understand.  
> 
> I think that means the leaves that use client-cfg-parms would need to be if-feature bfd and if-feature client-cfg-parms.

That is correct.

> 
> The references would have to be RFC XXXX.

Yes. I should have changed it to RFC XXXX instead of RFC 9127, as the latter does not have these feature definitions.

> 
> And perhaps if the namespacing isn't clear, perhaps the feature shoudl be bfd-client-cfg-parms.

Sure.

> 
> Add the Appendix of "changes since RFC 9127" and I think we have a document. :-)
> 
> -- Jeff
> 
> 
>> On Nov 8, 2021, at 3:01 PM, Mahesh Jethanandani <[e-mail] <mailto:[e-mail]>> wrote:
>> 
>> Hi Jeff,
>> 
>> And here is the full set of diffs that I had in mind.
>> 
>> <rfc9127-bis-diff.html>
>> 
>> 
>>> On Nov 8, 2021, at 11:12 AM, Jeffrey Haas <[e-mail] <mailto:[e-mail]>> wrote:
>>> 
>>> Mahesh,
>>> 
>>>> On Nov 8, 2021, at 1:49 PM, Mahesh Jethanandani <[e-mail] <mailto:[e-mail]>> wrote:
>>>> 
>>>> Hi Jeff,
>>>> 
>>>>> On Nov 8, 2021, at 10:19 AM, Jeffrey Haas <[e-mail] <mailto:[e-mail]>> wrote:
>>>>> 
>>>>> 
>>>>> 
>>>>>> On Nov 7, 2021, at 1:56 PM, Acee Lindem (acee) <[e-mail] <mailto:[e-mail]>> wrote:
>>>>>> 
>>>>>> 
>>>>>> From: Mahesh Jethanandani <[e-mail] <mailto:[e-mail]>>
>>>>>> 
>>>>>>> With Jeff’s diff, you’d still need two features. The base BFD feature which is in the protocol model and the bfd:client-cfg-parameters in the diff.
>>>>>>  
>>>>>> [mj] Acee, I was merely trying to point out to Reshad that we may not need a separate grouping with just the ‘enabled’ flag called ‘client-cfg-no-parms” or ‘client-cfg-enabled’ as the ‘enabled’ flag is needed by both modes of operation. I am fine with Jeff’s diffs.
>>>>>>  
>>>>>> But since you bring it up, how about moving the base BFD feature definition from the protocol drafts into the BFD types definition file, so there is one single way to enable the feature, rather than each protocol defining their own feature definition?
>>>>>>  
>>>>>> Sure.
>>>>> 
>>>>> The original diff I had sent out vs. the RFC was lost when the yang doctors were added.
>>>>> 
>>>>> Here's the diff again.  I think aside from adding appendix test and a little bit of word smithing, it's about all we need:
>>>>> 
>>>>> <rfc9127-bis-from-.diff.html>
>>>> 
>>>> All three protocol drafts (BGP, ISIS and OSPF, I did not check PIM) define their own feature statements for the base and for the client-cfg-parms in their own models. All I am suggesting is those feature definitions be moved into ietf-bfd-types YANG module as part of this update. But if the desire is to keep the change to the min, I am ok with it too.
>>> 
>>> My diff had it in that module.
>>> 
>>> Here's the full .xml for you to peruse.
>>> 
>>> 
>>> 
>>> -- Jeff
>>> <rfc9127.xml>
>> 
>> Mahesh Jethanandani
>> [e-mail] <mailto:[e-mail]>
>> 
>> 
>> 
>> 
>> 
> 

Mahesh Jethanandani
[e-mail]