Re: [Last-Call] Opsdir telechat review of draft-ietf-detnet-ip-oam-12

[Yingzhen Qu <[e-mail]>]

Hi Greg,

Thanks for the reply. Please see my answers below inline.

Thanks,
Yingzhen


On Tue, Feb 13, 2024 at 9:48 PM Greg Mirsky <[e-mail]> wrote:

> Hi Yingzhen,
> thank you for your kind words in support of this work; much appreciated.
> Please find my notes below tagged GIM>>.
>
> Regards,
> Greg
>
> On Tue, Feb 13, 2024 at 8:09 PM Yingzhen Qu via Datatracker <
> [e-mail]> wrote:
>
>> Reviewer: Yingzhen Qu
>> Review result: Ready
>>
>> I've reviewed version 10, and this is a follow up review.
>>
>> Thanks to the authors for working on this document and addressing my
>> comments.
>>
>> The document readability has improved a lot from my last review of
>> version -10,
>> and I believe it is ready for publication.
>>
>> There are a couple of nits for the authors to consider.
>>
>> The line numbers are generated using idnits.
>>
>> 160        is being done by using the Internet Control Message Protocol
>> (ICMP)
>>
>> nits: there should be a "." in the end.
>>
> GIM>> It seems like the long sentence may confuse a reader:
>    Most of on-demand failure detection and localization in IP networks
>    is being done by using the Internet Control Message Protocol (ICMP)
>    Echo Request, Echo Reply and the set of defined error messages, e.g.,
>    Destination Unreachable, with the more detailed information provided
>    through code points.
> The intention is to note that Echo Request/Reply that are the mechanism
> used in ICMP are the common method for on-demand failure detection and
> localization, i.e., of ping and traceroute. Would a slight re-wording make
> it clearer:
> NEW TEXT:
>    Most of on-demand failure detection and localization in IP networks
>    is being done by using the Internet Control Message Protocol's (ICMP)
>    Echo Request, Echo Reply, and the set of defined error messages, e.g.,
>    Destination Unreachable, with the more detailed information provided
>    through code points.
> WDYT?
>

[Yingzhen]: Now I see I did break the long sentence wrong. Fortunately the
message of the sentence was still delivered correctly. Here is my
suggestion:
In IP networks, the majority of on-demand failure detection and
localization is achieved through the use of the Internet Control Message
Protocol (ICMP), utilizing Echo Request and Echo Reply messages, along with
a set of defined error messages such as Destination Unreachable, which
provide detailed information through assigned code points.


>> 166        traffic between DetNet nodes with IP DetNet traffic, e.g.,
>> ensure
>> 167        that such ICMP traffic uses the DetNet IP data plane in each
>> node,
>>
>> nits: somehow this sentence doesn't read well to me. "uses the DetNet IP
>> data plane" is not very clear to me, I think you're trying to say ICMP
>> traffic
>> should go through the same path/interface as DetNet traffic, for example,
>> there is ECMP between two DetNet nodes.
>>
> GIM>> Thank you for pointing this to me. You are correct, the intention is
> to stress the importance of ensuring that ICMP packets traverse the same
> set of nodes and interfaces and receieve the same QoS treatment as the
> monitored DetNet IP flow. I propose the following update:
> OLD TEXT:
>    In order to use ICMP for these
>    purposes with DetNet, DetNet nodes must be able to associate ICMP
>    traffic between DetNet nodes with IP DetNet traffic, e.g., ensure
>    that such ICMP traffic uses the DetNet IP data plane in each node,
>    otherwise ICMP may be unable to detect and localize failures that are
>    specific to the DetNet IP data plane.
>  NEW TEXT:
>    In order to use ICMP
>    for these purposes with DetNet, DetNet nodes must be able to
>    associate ICMP traffic between DetNet nodes with IP DetNet traffic,
>    i.e., ensure that such ICMP traffic traverses the same interfaces and
>    receives the same QoS treatment as the monitored DetNet IP flow;
>    otherwise, ICMP may be unable to detect and localize failures that
>    are specific to the DetNet IP data plane.
> What are your thoughts? Is the text more clear now?
>

[Yingzhen]:  The new text is definitely more clear. Here is my suggestion,
and it's up to you to decide whether to use it or not.
To utilize ICMP effectively for these purposes within DetNet, DetNet nodes
must establish the association of ICMP traffic between DetNet nodes with IP
DetNet traffic. This entails ensuring that such ICMP traffic traverses the
same interfaces and receives identical QoS treatment as the monitored
DetNet IP flow. Failure to do so may result in ICMP being unable to detect
and localize failures specific to the DetNet IP data plane.