[ippm] FW: Comments on draft-tgraf-opsawg-ipfix-on-path-telemetry-00.txt

[[e-mail]]

Dear Al,

Thank you very much for the review and the comments. Much appreciated. I merged them in the next -01 version as following:

https://www.ietf.org/rfcdiff?url1=https://raw.githubusercontent.com/graf3net/draft-tgraf-opsawg-ipfix-on-path-telemetry/main/draft-tgraf-opsawg-ipfix-on-path-telemetry-00.txt&url2=https://raw.githubusercontent.com/graf3net/draft-tgraf-opsawg-ipfix-on-path-telemetry/main/draft-tgraf-opsawg-ipfix-on-path-telemetry-01.txt

Best wishes
Thomas

From: MORTON JR., AL <[e-mail]>
Sent: Monday, October 31, 2022 2:22 PM
To: Graf Thomas, INI-NET-TCZ-ZH1 <[e-mail]>; [e-mail]; [e-mail]
Subject: Comments on draft-tgraf-opsawg-ipfix-on-path-telemetry-00.txt

Hi Thomas and Benoit,

I reviewed your draft at
https://github.com/graf3net/draft-tgraf-opsawg-ipfix-inband-telemetry/blob/main/draft-tgraf-opsawg-ipfix-on-path-telemetry-00.txt<https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fgraf3net%2Fdraft-tgraf-opsawg-ipfix-inband-telemetry%2Fblob%2Fmain%2Fdraft-tgraf-opsawg-ipfix-on-path-telemetry-00.txt&data=05%7C01%7CThomas.Graf%40swisscom.com%7C01c6ddf4df83435d8ef608dabb42e467%7C364e5b87c1c7420d9beec35d19b557a1%7C0%7C0%7C638028193212406417%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=KqX%2BYwKyrIckXDnwh32Sh%2FIhW%2FvRz2F94050XriFGas%3D&reserved=0>


I like the fact that you have appreciated the IPPM work and IPfix work that precedes your efforts.  This draft is fully useful in ippm and the OPSarea, IMO.

You might mention that Section 2 uses the template of RFC8911 and the Performance Metrics Registry. Your draft is the first to adopt this template – now that the RFC is fully published!

I’ll add some notes as I scan through:

End of 2.3.1
   The Traffic Filter at the OP is configured to observe a single IP
   connection.
We don’t have connections at the IP-layer, only a stream of datagrams...

BTW, when I used the simplifying convention of
2.4.2.5.  Metric Units

   The <statistic> of one-way delay is expressed in seconds, where
   <statistic> is one of:

   *  Mean
...

in RFC 8912, I ultimately had to help IANA prepare complete and separate sections for each <statistic>. So, while I completely agree with re-using the <statistic> simplification in the draft, be prepared to create the 4 new versions of Section 2, one for each <statistic>, when this eventually reaches IANA for implementation. 😊

and that’s it for now!

I can easily see how much research and work you have committed to this draft. There seem to be some other comments to address, but the solutions appear within reach.

regards,
Al