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

[Jeffrey Haas <[e-mail]>]

Donatas,


> On Jan 10, 2023, at 2:26 PM, Donatas Abraitis <[e-mail]> wrote:
> 
> I'm proposing to be something like this:
> identifier = product ["/" product-version]

Roughly the right idea, but your examples probably don't quite hit the level of specificity that you're looking for.  For example, Juniper has different products that ship as a given version number and may vary slightly by platform.

There's some level of desire to have something that's easy to parse so you can more readily scrape things for comparison.  E.g., "what versions are all of my juniper model X boxes running?".

Prior messages in this thread touched on URI or URL to cover the information.  Frankly, if you pointed to a given URL, that can be the location for more interesting structured data.  Examples of this sort of thing these days exists in "software bills of materials" (SBOM).  Here's a breadcrumb to an Internet-draft working on this area of technology.  You'll note that the current ISE, Elliot Lear, is one of the authors there.  I suspect he'd be happy to discuss this problem with you. 

https://datatracker.ietf.org/doc/html/draft-ietf-opsawg-sbom-access

Focusing on your updated -10:
- IANA Considerations: you have an unchanged use of capability here.
- You have a start to unicode related gotchas in the draft.  Please take a look at RFC 9003 for text we've found useful in the past when talking about this stuff in BGP-land.
- I'd strongly suggest making this feature require RFC 9072.

-- Jeff


> 
> Inline diff: https://author-tools.ietf.org/iddiff?url1=draft-abraitis-bgp-version-capability-09&url2=draft-abraitis-bgp-version-capability-10&difftype=--hwdiff <https://author-tools.ietf.org/iddiff?url1=draft-abraitis-bgp-version-capability-09&url2=draft-abraitis-bgp-version-capability-10&difftype=--hwdiff>
> On Tue, Jan 10, 2023 at 9:21 PM Robert Raszuk <[e-mail] <mailto:[e-mail]>> wrote:
> 
> I am not worried about sender side ... much more the receiver side. 
> 
> First it is unrealistic that receiver will keep up and recognize zoo of BGP code bases. So if you choose to segment available space it still needs to be free form with some boundaries. 
> 
> Thx,
> R.
> 
> 
> 
> 
> On Tue, Jan 10, 2023 at 6:16 PM Donatas Abraitis <[e-mail] <mailto:[e-mail]>> wrote:
> >While nice for the sender may not be easy on the receiver. I would rather not hint on any formatting here if you choose not to make it as part of the spec. 
> I'm thinking about something like HTTP User-Agent, e.g.: frrouting/8.2, ios/12.0.1, junos/..., etc. What do you think?
> 
> This message and its attachments may contain privileged and confidential information. If you are not the intended recipient, do not use, copy, or disclose this information. Please notify the sender and permanently delete the message from your system.
> 
> 
> -- 
> Donatas Abraitis
> Principal Systems Engineer
> @: [e-mail] <mailto:[e-mail]>
> W: www.hostinger.com <https://www.hostinger.com/>
> 
> 
> This message and its attachments may contain privileged and confidential information. If you are not the intended recipient, do not use, copy, or disclose this information. Please notify the sender and permanently delete the message from your system.