Ballot for draft-ietf-lisp-te
Discuss
Yes
No Objection
No Record
Summary: Has 3 DISCUSSes. Has enough positions to pass once DISCUSS positions are resolved.
DISCUSS 1: I'd like to understand why this has been requested for publication as EXP? I may have missed this, but I didn't see any new protocol definitions. I do note it was chartered as an experimental extension, yet I see no explanation of why this is an experiment, nor how that experiment may in the future be declared as a success or failure. Whatever the reason, can this reason to be noted in the abstract and to be described in the introduction? DISCUSS 2: Since this is targeting EXP, what is the nature of the EXP specification: is it for a limited duration? A limited use-case? A limited scope of deployment? i.e., what is the experiment and can this document define implicit or explicit success/failure criteria, and an outcome that can be used as the basis for a future recommendation to the IETF community? DISCUSS 3: Section 7 is about a topic that I do not see explained. If this is important enough to include, please specify this. The informational reference to the ELP-probing mechanism details in [I-D.filyurin-lisp-elp-probing] seems insufficient and since this reference has expired (in 2018) as a non-WG I-D, can the whole section be removed?
I'd strongly encourage changing the opening sentence of section 2, to insert the word “Experimental” before extension, "Informational" if the document were changed to target this. Please consider the following: - Please do expand the word LISP in the title - Please expand LISP in the abstract. - Please also define all abbreviations on first use, I expect this will hardly increase the word count much and can save reader pain.
Hi Dino, Michael, Parantap, and Padma, Thank you for the effort put into this specification. Thanks to Dhruv Dhody for the OPSDIR review. Unless I’m mistaken, there was no discussion on these issues. Given the intended status, I didn't review the detailed proposed procedure but focused on the expected features. # Is this document ready for IESG Review? I failed to find follow-ups or updates to the IETF LC comments (directorate reviews are also part of the IETF LC comments). I won’t reiterate here the comments made by Dhruv and which I think are worth to consider. # Justification/Expectations for the intended Experimental status Please add similar text as what was agreed for the LISP Geo spec where we raised this same issue. Thanks. # Deployment Incentives Although this is experimental, a discussion on deployment incentives for this flavor of LISP TE (which is distinct from the one discussed, e.g., in RFC7834) is needed. Forcing some via-points (for some specific reasons) would require that domains that do not host LISP-capable routers to be upgraded to enable LISP for the sake of the service offered in this draft. What are the incentives for such domains to do so? What they get out of offering this capability? I’m not mentioning issues related to adding an extra intermediate on the path with all the risks of interception, etc. Section 4 lists some reasons to enable the proposed features, but no justification/discussion is actually provided. Please see below some comments about that part. ## Lack of control of underlying path CURRENT: * There may be a policy reason to avoid the ASes that make up the path between B and C. xTRs do not have control of the underlying path. Assuming that an internal path is made available to endpoints, avoiding some portion of a forwarding path may not be possible unless we assume that for every AS in the forwarding path a LISP RTR is enabled. That assumption is against the promise of the initial LISP design (free lisp core). ## LISP-specific for handling internal path failures is not a viable approach CURRENT: * There may be a failure on the path between B and C which makes the path unreliable. As failures are unpredictable, this assumes that the intended features to be efficient for this intended case must be enabled not only to react to the B-C leg failure but other similar failures of a path (let alone other alternative paths that may be selected by the underlying routing). That’s not a viable approach. Also, other traffic than the LISP-encapsulated one will be impacted by such failures if they occur. A more generic protection mechanism to detect/absorb the failure is more appropriate than LISP-specific. ## Why another mechanism for service chaining is needed? CURRENT: * There may be a chain of services performed at RTRs X and Y regardless if the path from ITR to ETR is through B and C. There are few service chaining approaches already standardized by the IETF. A discussion why these mechanisms are not sufficient here is needed. Cheers, Med
I would like to discuss the procedural elements of this document’s advancement. I observe that: -- The WG chair (Padma) who called the WGLC (https://mailarchive.ietf.org/arch/msg/lisp/wXb2V_yZxFyXFV0gn0EinzNvTDE/) and declared consensus to publish (https://mailarchive.ietf.org/arch/msg/lisp/TrBH2rZuqF3nhFZyV7-cYfNVwEw/) on -20 is now a listed document author as of -21. -- There don’t appear to be material changes between the -20 and -21 drafts. The diff (https://author-tools.ietf.org/iddiff?url1=draft-ietf-lisp-te-20&url2=draft-ietf-lisp-te-21&difftype=--html) suggests only additional example text. I was looking for a large blob of text changes which typically trigger late-stage changes in the author list. -- The document shepherd is also a document author. -- The shepherd’s report says, “The document was supported by 14 votes (appropriate for the WG) and there were no objections.” My review of the WGLC thread (https://mailarchive.ietf.org/arch/msg/lisp/TrBH2rZuqF3nhFZyV7-cYfNVwEw/) shows that 3 non-authors responding. As I did not follow the entire lifecycle of this document, it isn't clear what transpired.
Thank you to Peter Yee for the GENART review. The promised changes (https://mailarchive.ietf.org/arch/msg/gen-art/hz0tj9BXoB3AHlGRV9mSY-cmmPQ/) to his review have not been merged into the draft. Please do so.
Thanks to the authors and the WG for their work on this document. I support Gorry's DISCUSS related to ELP probing. It seems to me that ELP Probing (section 7) is an integral part of the LISP TE solution. Without this feature, I wondering how an operator could monitor/verify these LISP TE paths. And if my understanding is correct then, draft-filyurin-lisp-elp-probing that is listed as informative should become a normative reference. Regarding the experimental status for this draft, I note that the shepherd's report indicates that implementations exist. That conveys implementation experience has already been achieved. I guess the reason why this is experimental is because it relies on LCAF (RFC8060) which is in experimental status. This is not an issue per se with this document, but that the WG perhaps needs to reprioritize its work in promoting experimental documents to PS before sending more experimental documents for publication to the IESG. AFAIK LCAF and Vendor specific LCAFs are widely implemented and deployed LISP features and quite foundational but remain as experimental - please correct if I've got it wrong. So, more of a feedback to the WG chairs/AD. Regarding Section 8 Service Chaining, I support Med's DISCUSS point. The text says "ITR can encapsulate packets to the RTR which will decapsulate and deliver packets to a scrubber service device." - however, there is no protocol specification that indicates which packets the RTR should send to the scrubber service? Is the RTR itself running the scrubber service and "all" LISP packets sent to it are subject to the scrubber? More details and clarifications will help. I would expect that the ELPs need some indication of the RTR also doing extra service functions? Regarding Section 10 Multicast, I believe some references to LISP multicast documents are necessary. Perhaps RFC6831 can be referenced again here. I am not sure if any other is needed.
I support Gorry's DISCUSS - I am also unsure why this is an Experimental doc, and not a BCP or Informational one. nit: MUST choose to drop the packet why not simply: MUST drop the packet. The use of MUST and "choose" seem odd to me.
# Éric Vyncke, INT AD, comments for draft-ietf-lisp-te-21 CC @evyncke Thank you for the work put into this document. I am balloting NoObjection *because* it is an experimental document, else I would ballot DISCUSS (see points on section 5.4 and 7). Please find below some non-blocking COMMENT points/nits (replies would be appreciated even if only for my own education). Special thanks to Padma Pillay-Esnault for the shepherd's detailed write-up including the WG consensus and the justification of the intended status. I hope that this review helps to improve the document, Regards, -éric ## COMMENTS (non-blocking) ### Shepherd's write-up I am a little puzzled to see the shepherd listed as authors as well; this is quite unusual. I guess that the LISP chair/shepperd was added as an author late in the process for the last mile of this document. ### Section 2 Please add a reference to LISP RFC(s). More serious, I wonder whether the IETF needs to have yet another TE mechanism (The ELP really looks like SRv6 or SR-MPLS SID list), but I guess that this can be run as an experiment. ### Section 3 What is a `PETR`? ### Use of SVG graphics To make a much nicer HTML rendering, suggest using the aasvg too to generate SVG graphics. It is worth a try especially if the I-D uses the Kramdown file format ;-) ### Section 4 Suggest adding the nodes X & Y already in figure 1. ### Section 5 Where are the S-bit and L-bit defined ? Probably in RFC 8060, but let's be explicit. ### Section 5.4 Relying on the TTL only for loop avoidance is a severe limitation that would not be acceptable in a proposed standard to be clear. A hostile participant could then create a 255 tunnels loop if I understand correctly. ### Section 7 I am afraid that [I-D.filyurin-lisp-elp-probing] is a normative reference. ### Section 8 s/One example is to implement a *honey-pot* service/One example is to implement a *packet scrubber* service/ ?