Re: [Detnet] Opsdir last call review of draft-ietf-detnet-ip-oam-10

[Greg Mirsky <[e-mail]>]

Hi Yingzhen,
thank you for your review, comments, and questions. Please note that the
new version of the draft includes all updates listed earlier. We hope that
updates in versions 11 and 12 address your concerns.

Name:     draft-ietf-detnet-ip-oam
Revision: 12
Title:    Operations, Administration and Maintenance (OAM) for
Deterministic Networks (DetNet) with IP Data Plane
Date:     2024-02-08
Group:    detnet
Pages:    9
URL:      https://www.ietf.org/archive/id/draft-ietf-detnet-ip-oam-12.txt
Status:   https://datatracker.ietf.org/doc/draft-ietf-detnet-ip-oam/
HTML:     https://www.ietf.org/archive/id/draft-ietf-detnet-ip-oam-12.html
HTMLized: https://datatracker.ietf.org/doc/html/draft-ietf-detnet-ip-oam
Diff:
https://author-tools.ietf.org/iddiff?url2=draft-ietf-detnet-ip-oam-12

Abstract:

   This document discusses the use of existing IP Operations,
   Administration, and Maintenance protocols and mechanisms in
   Deterministic Networking networks that use the IP data plane.

Best regards,
Greg

On Mon, Feb 5, 2024 at 8:45 AM Black, David <[e-mail]> wrote:

> Hi Yingzhen,
>
> Thanks for the review - it turned up a number of issues that are common
> with Bernard's review, so we decided to deal with his review first (see the
> recently posted -11) version and then look at this one.  In general, please
> take a look at the -11 version, which should be clearer.
>
> David> Comments inline on specific items.
>
> Thanks, --David
>
> -----Original Message-----
> From: Yingzhen Qu via Datatracker <[e-mail]>
> Sent: Friday, February 2, 2024 1:45 AM
> To: [e-mail]
> Cc: [e-mail]; [e-mail];
> [e-mail]
> Subject: Opsdir last call review of draft-ietf-detnet-ip-oam-10
>
>
> [EXTERNAL EMAIL]
>
> Reviewer: Yingzhen Qu
> Review result: Has Issues
>
> Hi,
>
> Thanks for the draft.
>
> I am the assigned OPSDIR reviewer to conduct a last call review of this
> draft.
>
> General comments:
>
> The draft lists how existing OAM protocols/tools can be used in DetNet
> with IP
> data plane. However, I don't think it's clearly stated in either the
> abstract
> or the introduction section, so it's not clear to me what information the
> draft
> is trying to deliver. OAM principles as stated in the abstract? or how
> existing
> solutions may work in DetNet?
>
> David> The latter is intended - this should be clarified and consistent in
> the -11 version, please check.
>
> There are some editorial nits in the draft, and some of them made the draft
> hard to read. I'd suggest an editorial pass and then I can do another
> round of
> review.
>
> Detail comments with line number from idnits:
>
> 80       usually performed on-demand. These tasks achieved by a combination
> s/usually performed/are usually performed
> s/achieved/are achieved
>
> David> Fixed in the -11 version.
>
> In Section 2.1, there should be a ":" after each term/abbreviation.
>
> David> Agree, should be done for consistency.
>
> 145      packets. For example, that can be achieved with a 3-tuple
> 146      (destination and source IP addresses in combination with DSCP
> value)
> 147      used to identify the IP DetNet flow. In such a scenario, an IP OAM
> 148      session between the same pair of IP nodes would share the network
> 149      treatment with the monitored IP DetNet flow regardless of whether
> 150      ICMP, BFD, or STAMP protocol is used.
>
> Q: Do you mean as long as two protocols, for example ICMP and BFD, have the
> same DSCP value, the network treatment will be the same?
>
> David> No, please see the new text in -11, which should be clearer.
> In 20/20 hindsight, we omitted the assumption for the example that DetNet
> is using 3-tuples (e.g., and not 5-tuples) to determine which traffic
> receives DetNet handling/QoS.  The overall concern is that the OAM traffic
> use the same network path and receive the same DetNet forwarding treatment
> as the DetNet traffic that is being monitored.
>
> 158      protocol like, for example, UDP, a DetNet node must able to
> associate
> s/must able to/must be able to
>
> David> Ok, will fix.
>
> 168      protocol is one of the assigned by IANA, then the UDP source port
> can
> s/the assigned/those assigned
>
> David> Removed as part of rewriting that paragraph in -11.
>
> 167      DetNet flow. When the UDP destination port number used by the OAM
> 168      protocol is one of the assigned by IANA, then the UDP source port
> can
> 169      be used to achieve co-routedness of OAM, and the monitored IP
> DetNet
> 170      flow in the multipath environments, e.g., Link Aggregation Group
> or
> 171      Equal Cost Multipath.
> Q: It's not clear how using the same UDP source port the OAM packets will
> be
> guaranteed to follow the same path as the monitored IP flow. Please
> clarify.
>
> David> Clarification/explanation improved in -11, please take a look at
> the new text.
>
> 266      and MPLS data planes analyzed in Section 6.2 of
>
> 268      [I-D.ietf-detnet-mpls-oam]. Also, requirements and recommendations
> nits: broken line.
>
> David> I suspect that's a tooling artifact, RFC Editor will deal with this.
>
> 268      [I-D.ietf-detnet-mpls-oam]. Also, requirements and recommendations
> 269      for OAM interworking between a DetNet domain with MPLS data plane
> and
> 270      OAM of a TSN equally apply to a DetNet domain with an IP data
> plane.
> I'd suggest rewriting this sentence. I can't understand what it's trying
> to say.
>
> David> Yes, it's a bit terse and unclear for that reason.  We'll propose
> new text in a subsequent message.
>
> Thanks,
> Yingzhen
>
>