Re: [Idr] WG adoption call for draft-abraitis-bgp-version-capability-08, to end September 25

[Jeffrey Haas <[e-mail]>]

Adding in a comment as I clear my backlog:

On Mon, Sep 28, 2020 at 06:18:21PM +0000, John Scudder wrote:
> It seems to me that the discussion hasn’t concluded and we are still having productive discussion. Hence, I would like to extend the period for discussion by another week, now to conclude on October 2.

I have general concerns about this feature.  But let me offer related
observations why my opinion is "not a fan, but of potential use":

- This capability is advertised prior to session establishment.  
  RFC 8203 got a bit of jutification for putting in text during a
  NOTIFICATION where the session is going down anyway.  Having this at
  session start means less chance for damange than if we're exchanging stuff a
  bit more interactively in something like the operational message.
- I think requiring extended optional params is a good idea for this.  It
  mitigates the necessity for having to do do the math to squeeze stuff in
  or not.

I would qualify the feature as "mostly harmless" and "useful in some
circumstancs".  Use in BGP fabrics is one I find a compelling use case.

I'm strongly supportive of the concerns with using this feature for general
ISP eBGP use cases.  I will, however, offer the related observation that an
ISP may like knowing what version of software their customers are running.
It may ease situations where service providers may want to signal a customer
to upgrade a vulnerable stack.

Consider me weakly supportive of adoption and fast WGLC.

-- Jeff