Thanks, Jean. This is helpful. Outlook is pretty crappy when it comes to formatting replies. Your responses below weren’t exactly obvious, but I think I found them.
The client support is addressed for me in Section 7.1 saying that the client MUST request these manifests explicitly. I’m good with the new text. The string length is good. If you think 1023 is right, I’m fine. At least I know what I might expect in terms of data size for this in my TSDB. Joe From: Jean Quilbeuf <jean.quilbeuf=40huawei....@dmarc.ietf.org> Date: Monday, March 3, 2025 at 09:44 To: Joe Clarke (jclarke) <jcla...@cisco.com>, opsawg@ietf.org <opsawg@ietf.org> Subject: RE: WG LAST CALL: A Data Manifest for Contextualized Telemetry Data Hi Joe, Thank you very much for your detailed comments, they should have been addressed. The diff is here: https://author-tools.ietf.org/iddiff?url2=draft-ietf-opsawg-collected-data-manifest-06 Answer to the major comments is inline. Best, Jean From: Joe Clarke (jclarke) <jclarke=40cisco....@dmarc.ietf.org> Sent: Monday, January 6, 2025 4:53 PM To: Joe Clarke (jclarke) <jclarke=40cisco....@dmarc.ietf.org>; opsawg@ietf.org Subject: [OPSAWG]Re: WG LAST CALL: A Data Manifest for Contextualized Telemetry Data Below is a review of this draft as a contributor. I have to say that overall I find it pretty hard to read and I worry that it is one might struggle to implement it. I like what Med does with the Word “track changes” thing, so I’m including that here as a PDF for the typos. I’ll focus on the substantive comments in this email. Much appreciated, thanks First of all, the Platform Manifest uses a lot of open-ended string leafs. Given that these data will be stored in a DB, I think some thought on limiting the length would be useful. I put 1023 everywhere. My biggest issues are with how the manifests are collected and how that collection would work between clients that are aware of this spec vs. those that are not. If I subscribe to some telemetry stream, do I get the manifest by default? And if so, what if I’m not expecting it? I found this part of the document very under-specified. At one point, the document states “Somehow, the collected data must include a metadata pointing to the corresponding Data Manifest”. That doesn’t give me a clear approach to performing this linkage. Clarified in Section 7 (previously 5) notably * the collector MUST subscribe explicitly to the data manifest * explicited how to retrieve the data manifest When it comes to manifest changes, how would one know if the manifest failed to generate/stream vs. it just didn’t change? In HTTP, you have a “302” to let you know the data is there but hasn’t changed. There doesn’t seem to be a similar placeholder for this manifest. And if it does fail, it’s not clear how one would know this. I think this is out of scope for this document. There is this sentence: "The reliability of the collection of the data manifest is the same as the reliability of the data collection itself, since the data manifest is like any other data." Let us know if something more explicit is needed. I did appreciate the operational considerations, and I’d like to see more in terms of the expected impact to scale when it comes to implementing this. Joe From: Joe Clarke (jclarke) <jclarke=40cisco....@dmarc.ietf.org<mailto:jclarke=40cisco....@dmarc.ietf.org>> Date: Monday, January 6, 2025 at 11:09 To: opsawg@ietf.org<mailto:opsawg@ietf.org> <opsawg@ietf.org<mailto:opsawg@ietf.org>> Subject: [OPSAWG]Re: WG LAST CALL: A Data Manifest for Contextualized Telemetry Data Happy New Year, WG. Because of the holidays I didn’t get a chance to officially close this in time. However, we did not get any WGLC feedback during the window. Only yesterday did some substantive feedback come from Alex (thanks!). As such, we are going to conduct another WGLC now that the holiday cycle is behind us. We’ll consider Alex’s comments as part of that. This begins a second WGLC to run until January 20. Please provide all comments (including a “yay” if you feel it is ready) on the mailing list. Joe From: Joe Clarke (jclarke) <jcla...@cisco.com<mailto:jcla...@cisco.com>> Date: Wednesday, November 27, 2024 at 09:39 To: opsawg@ietf.org<mailto:opsawg@ietf.org> <opsawg@ietf.org<mailto:opsawg@ietf.org>> Subject: WG LAST CALL: A Data Manifest for Contextualized Telemetry Data Hello, WG (and a Happy Thanksgiving to those of you in the US). With the IPR poll concluded (no IPR has been reported), we’d like to kick off a two week WG LC on draft-ietf-opsawg-collected-data-manifest<https://datatracker.ietf.org/doc/draft-ietf-opsawg-collected-data-manifest/>. Please review this draft and provide comments on-list. If you feel this draft is ready for publication, please respond as such on-list. We will kick off DIR reviews with OPS and YANG docs to get a couple more eyes on it. We are also in need of a shepherd for this document. If you are interested, please let the chairs know. The WG LC will conclude on December 11. Joe
_______________________________________________ OPSAWG mailing list -- opsawg@ietf.org To unsubscribe send an email to opsawg-le...@ietf.org
