[netconf] Re: Status update of draft-ietf-netconf-adaptive-subscription-05

["maqiufang (A)" <[e-mail]>]

Hi, Thomas,

Thanks a lot for the valuable comments, please see inline.

From: [e-mail] [mailto:[e-mail]]
Sent: Thursday, July 18, 2024 10:08 PM
To: maqiufang (A) <[e-mail]>; [e-mail]
Cc: [e-mail]; [e-mail]
Subject: RE: Status update of draft-ietf-netconf-adaptive-subscription-05

Dear Qiufang,

Thanks a lot for the update. I agree that the document is in a very good shape and addresses network operators needs for on demand fine grained network telemetry data.
[Qiufang] Thanks, glad to know;-)
I reviewed the document and have two comments.

Editorial: I suggest to change "Applying adaptive subscription allows publishers to automatically adjust the period and volume of telemetry data sent to receivers." to "Applying adaptive subscription allows publishers to automatically adjust the subscription period dynamically based on pre-defined threshold for finer grained network telemetry data sent to receivers."
[Qiufang] Sure, the proposed text looks good to me.
In https://datatracker.ietf.org/doc/html/draft-ietf-netconf-adaptive-subscription-05#section-2.2 you are mentioning

"Note that the adaptive subscription may not be supported by every YANG datastore node."

I suggest to augment ietf-notification-capabilities as defined in https://datatracker.ietf.org/doc/html/rfc9196#section-5 to describe which YANG nodes this can be applied to. From a workflow perspective, that matches what we described in https://datatracker.ietf.org/doc/html/draft-ietf-nmop-yang-message-broker-integration-03#section-4.1.
[Qiufang] This makes sense to me, and good point! Will augment the "per-node-capabilities" in the system-capabilities yang module with the support of adaptive-subscription.
I see that the document is missing a implementation status section (https://datatracker.ietf.org/doc/html/rfc7942) nor in the data tracker has under "additional resources" a "related implementations" link.
[Qiufang] Actually we used to have a hackathon project on this work around more than two years ago, we were targeting a real campus network scenario where gRPC based adaptive telemetry is used to collect data from AP devices, it turns out that adaptive subscription can analyze the WIFI degradation events with the minimum amount of push data, and you might want to refer to the presentation slides for some more details: https://github.com/IETF-Hackathon/ietf113-project-presentations/blob/main/ietf-hackathon-adaptive-subscription.pdf. And demo implementation results and learned experience were also brought back to the WG to better refine the document.
Would you suggest us to add an "implementation status" section to document related information in the draft? We will consider adding this in the next revision.

I encourage to provide an open-source reference implementation and would support and contribute in its validation as part of our "Validate Configured Subscription YANG-Push Publisher Implementations" activities at the IETF hackathon (https://wiki.ietf.org/en/meeting/120/hackathon) I believe this helps to improve the quality of the described standard.
[Qiufang] Happy to know your continued hackathon project on yang-push notification validation, it looks good! We've defined a new "adaptive-period-update" state change notification in the draft, personally I think it would be good if it could be validated using your implementation. We did implement this mechanism as demos and also in our products, however, I am unsure currently if we are able to open source it and if our existing implementation can completely fit into your project. Let me check internally, and I will get back to you then.

Best wishes
Thomas
Best Regards,
Qiufang
From: maqiufang (A) <maqiufang1=[e-mail]<mailto:maqiufang1=[e-mail]>>
Sent: Wednesday, July 17, 2024 11:38 PM
To: [e-mail]<mailto:[e-mail]>
Cc: [e-mail]<mailto:[e-mail]>; [e-mail]<mailto:[e-mail]>
Subject: [netconf] Status update of draft-ietf-netconf-adaptive-subscription-05


Be aware: This is an external email.


Hi, all,

As adaptive-subscription document is not going to be presented in this IETF NETCONF session and authors of WG documents are requested to provide a status update to the WG, this email is to give a status-update of this work.

The authors have submitted -05 before the submission window, and the main updates are as follows:
*         Move use cases for adaptive subscription into a separate appendix section;
*         Define adaptive-subscription specific RFC errors for "modify-subscription" RPC, in addition to the existing ones for "establish-subscription";
*         Add "Subscription State Change Notifications" new section to align with the YANG module definition;
*         Use the security considerations template described in rfc8407bis;
*         Other editorial updates.

There are not known open issues currently and the authors feel this document is close to WGLC, but would really appreciate any feedback from the WG.  Thanks a lot!

Best Regards,
Qiufang // on behalf of authors