I hope I've improved the Abstract in -14 with a concise summary of the link 
metric backward compatibility.  I moved
the more detailed discussion to the Introduction. 

Thanks,
Acee

> On Dec 10, 2025, at 2:48 PM, Yingzhen Qu <[email protected]> wrote:
> 
> Hi authors,
> 
> If approved, the document will update RFC 5443, RFC 6987 and RFC 8770. The 
> update needs to be discussed in the introduction. My suggestion is to make 
> the abstract simpler and move some of the text to the introduction section.
> 
> Thanks,
> Yingzhen
> 
> On Wed, Dec 3, 2025 at 3:08 PM Acee Lindem <[email protected]> wrote:
> Ok - so the functional capabilities TLV was missing from the LSA capability 
> encodings.  Right? 
> 
> > On Dec 3, 2025, at 6:02 PM, Yingzhen Qu <[email protected]> wrote:
> > 
> > Hi,
> > 
> > Considering we now have ietf-ospf-functional-capability as a separate 
> > general purpose module, I'm thinking we should also add the functional 
> > capabilities TLV in link scope RI LSAs. Note that there is 
> > "functional-flag" as uint32 defined RI LSAs in RFC9129, which can return 
> > raw data. Now we have both uint32 and identities, which is fine. Thoughts?
> > 
> > I attached the updated module with augmentation to link scope RI LSAs for 
> > your reference.
> > 
> > Thanks,
> > Yingzhen
> > 
> > 
> > 
> > On Wed, Dec 3, 2025 at 9:17 AM Acee Lindem <[email protected]> wrote:
> > Hi Yingzhen, 
> > 
> > Thanks for the detailed review. I have incorporated all your comments in 
> > -12. 
> > 
> > > On Dec 2, 2025, at 7:40 PM, Yingzhen Qu <[email protected]> wrote:
> > > 
> > > Hi authors,
> > > 
> > > Thanks for working on this document. I have some comments for you to 
> > > consider.
> > > 
> > > 1)
> > > Section 2.2, the description and the topology of Figure 2 don't seem to 
> > > match. It seems there are two links between node A and B, A and C, C and 
> > > D, and B and D according to figure 3 and 4. Can you please confirm?
> > 
> > 
> > Yes, the base topology is wrong as the links from A to B and C to D are to 
> > be used exclusively for flex algo.
> > 
> > 
> > > 
> > >  2)
> > > "
> > > 3. LSLinkInfinity-Based Solution
> > > "
> > > Maybe change this to "LSLinkInfinity Based Solution"?
> > 
> > I don't much care. It is a compound adjective modifying "solution" similar 
> > to "YANG-based" modifying
> > "management-protocosl" in the YANG security template. However, the title 
> > reads well without the
> > hyphenation and I changed it. 
> > 
> > > 
> > > 3)
> > > In Section 3.1, "IGP metric" is used instead of "OSPF metric". 
> > 
> > There are multiples of these - I fixed them all. 
> > 
> > > 
> > > 
> > > 4)
> > > "
> > > Prior to this specification, OSPF treated links advertised as
> > > LSLinkInfinity as reachable [RFC2328].
> > > "
> > > Maybe "Prior to this specification, OSPF treated links with an advertised 
> > > metric of LSLinkInifinity as reachable."
> > 
> > 
> > Sure. 
> > 
> > > 
> > > 
> > > 5)
> > > Section 3.3
> > > RFC6987 applies to both OSPFv2 and OSPFv3, why is MaxReachableLinkMetric 
> > > only discussed for OSPFv2?
> > 
> > Right, RFC 6987 allows the use of either MaxLinkMetric or the Router-LSA 
> > R-bit. I've updated "OSPFv2" to "OSPF". 
> > 
> > Recently, I worked on a customer POC involving OSPF and LDP and realized I 
> > needed to add RFC 5443 as well. 
> > This is included in section 3.4. 
> > 
> > Thanks,
> > Acee
> > 
> > 
> > > 
> > > Thanks,
> > > Yingzhen
> > > 
> > 
> > <ietf-ospf-functional-capability.yang>
> 

_______________________________________________
Lsr mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to