Hi Greg, 

> On Jul 21, 2025, at 12:34 PM, Greg Mirsky <[email protected]> wrote:
> 
> Hi Acee,
> thank you for sharing your comments and suggestions. Please find my notes 
> below tagged GIM>>.
> 
> Regards,
> Greg
> 
> On Mon, Jul 21, 2025 at 6:07 PM Acee Lindem <[email protected]> wrote:
> As requested by the chairs, I'm reposting my RTGWG comments on this draft. 
> 
> Like the other draft, I'd like to hear about your use cases for deployment of 
> more than 3 VRRP routers for the same Virtual Router (same LAN segment and 
> VRID).
> GIM>> Is there any limitation or a recomendation in Operational Consideration 
> in RFC 9568 for the number of VRRP routers in the given VRID? 

There is no limitation. But instead of avoiding my question with a question, 
can you actually respond to my question? 

> 
> This draft requires implementation of P2MP BFD (RFC 8562) as a pre-requisite. 
> Who has implemented this draft?
> GIM>> In the course of publishing RFC 8562, at least one released 
> implementation was noted. As for the draft, to the best of my knowledge, PoC 
> was completed. 

Can you be more explicit in your response? 


> 
> 
> The draft shouldn't need to modify the VRRP packets. Given the P2MP BFD and 
> itsanw applicability to a Virtual Router needs to be enabled on the 
> participating VRRP routers, you shouldn't need to modify the VRRP encodings. 
> The BFD encapsulation in the draft can be used and the Virtual Router can be 
> determined by the IPv4/IPv6 source address. If RFC 8562 requires the 
> descriminator, it could configured or your document could modify RFC 8562 to 
> avoid bastardizing BFD. 
> GIM>> That is an interesting suggestion; thank you. Will look into 
> constructing My Discriminator using VRID. 
> Note that other protocols using BFD for fast failure detection do not require 
> BFD information to be advertised in the protocol packets (e.g., OSPF and 
> IS-IS).

That would work as well, though it would seem more flexibility just to remove 
the unnecessary P2MP BFD requirement to know the descriminator in advance. 


> GIM>> AFAIK, IGPs use respective extensions to advertise My Discriminator for 
> use in Seamless BFD (RFC 7880). Does that count as BFD-related information 
> advertised by the IGP? 

No - this is basic usage of BFD for protocol fast failure detect. 

> 
> 
> The draft in under specified as it doesn't include of the additions need to 
> the RFC 9568 pseudo code specifications. However, I REALLY don't want to see 
> any additions to section 7.
> GIM>> The intention of the proposal is to have no impact on Section 7 of RFC 
> 9568. Am I missing something here? 

You're not missing anything if you remove the gratuitous extension of the VRRP 
advertisement encoding just to use BFD. 

Acee


> 
> 
> Thanks,
> Acee
> 
> 
> 
> 

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

Reply via email to