[mpls] Re: Poll: IOAM and PSD

[Loa Andersson <[e-mail]>]

Joel,


Den 04/08/2024 kl. 06:02, skrev Joel Halpern:
> I wondeer if we can find some better terminology. 
I have no opposition to "better terminology", only that I think what we 
have will serve unless there are proposals for improvements.

As always when I try to write definitions in English, please remember 
that English is not my mother tongue, try to understand what I mean and 
help me to improve.

What we currently have I think was proposed Stewart and says that 
information is considered "mutable" if that information in a flow is 
changes per packet or is changed by one nodes forwarding packets 
belonging to the flow.

I still think that is good enough, but would welcome improvements.

> You are correct that the sequence number on packets is not the same in 
> successive packets in a flow.  So, depending upon how you use the term 
> "mutable" it is mutable between packets in a flow.  It is not changed 
> (mutatated) by the devices that process it.  We have other use cases 
> where we use "mutable" to mean that.   Can we disentangle them?

Maybe something like this (though I guess we should give some context):

- there are two types of mutable information in MPLS IOAM packets

   -- packets where the ingress LER put in for example a sequence number 
that is
      increased per packet.
   -- packets carrying information that one or more node in the 
(non-deterministically)
      changes prior to forwarding the packet

Caveat: modulo grammar

And, yes I think this is what Stewart said from the start.

/Loa
> Thanks,
>
> Joel
>
> On 8/3/2024 11:44 PM, Loa Andersson wrote:
>> Greg,
>>
>> I think Michael is right.
>>
>> DEX is useful and we need it, but it does not solve all and any problem.
>>
>> I thought DEX was about transporting data/info generated by a node, 
>> and outgoing packet. This data is by defiintion immutable and is for 
>> example never hashed for load sharing purposes
>>
>> The issue with mutable data, e.g. the sequence number, is that it is 
>> in an incoming packet and mutable. If the network is doing ECMP by 
>> scanning the label stack, then the sequence number can't in the first 
>> 20 bits of an LSE.
>>
>> Right?
>>
>> /Loa
>>
>> Den 04/08/2024 kl. 22:58, skrev Greg Mirsky:
>>> Hi Michael,
>>> I don't think that "IOAM requires mutable data" is an accurate 
>>> statement. RFC 9326 IOAM Direct Export 
>>> <https://datatracker.ietf.org/doc/rfc9326/> indeed defines optional 
>>> parameters, Flow ID and Sequence Number, to ease the correlation 
>>> process of collected telemetry data. For some applications, e.g., 
>>> DetNet over MPLS <https://datatracker.ietf.org/doc/rfc8964/>, both 
>>> these parameters already are present as part of the DetNet MPLS 
>>> encapsulation as S-label (Service) and Sequence Number in d-CW or 
>>> d-ACH, respectively. Furthermore, the IPPM WG draft on collecting 
>>> telemetry data 
>>> <https://datatracker.ietf.org/doc/draft-ietf-ippm-hybrid-two-step/> can 
>>> be used without the use of the optional IOAM-DEX fields, i.e., 
>>> Flow-ID and Sequence Number.
>>>
>>> Regards,
>>> Greg
>>>
>>> On Sat, Aug 3, 2024 at 6:43 AM Michael Menth 
>>> <[e-mail]> wrote:
>>>
>>>     Hi all,
>>>
>>>     we're not operators but researchers who implemented some IOAM
>>>     flavors in combination with SR using P4 on Tofino. We've presented
>>>     some results at an interim meeting before:
>>> https://datatracker.ietf.org/meeting/interim-2024-mpls-01/session/mpls
>>>
>>>     IOAM requires mutable data. In combination with multiple labels,
>>>     e.g., for SR, these mutable data need to be near the bottom of the
>>>     stack, otherwise they get popped along the way. Moving them to PSD
>>>     makes encoding more efficient without increasing the effective
>>>     header length that needs to be parsed. Therefore, we think PSD is
>>>     the better option for applications like IOAM.
>>>
>>>     Regards
>>>
>>>     Michael
>>>
>>>     Am 30.07.2024 um 17:25 schrieb Tony Li:
>>>>
>>>>     [WG chair hat: on]
>>>>
>>>>     Hi all,
>>>>
>>>>     We’ve had many discussions about IOAM and PSD over the last few
>>>>     years. We need to reach consensus on the problems that need to be
>>>>     addressed in these areas. Therefore, we would like to hear from
>>>>     everyone, especially independent operators:
>>>>
>>>>     1.
>>>>
>>>>         There are many flavors of IOAM.  Which ones would you like to
>>>>         deploy/implement with MNA?
>>>>
>>>>     2.
>>>>
>>>>         Do you have other applications of MNA that have not been
>>>>         proposed yet?
>>>>
>>>>      This poll will close in two weeks, at 9am PDT, Aug 13.
>>>>
>>>>     Regards,
>>>>
>>>>     MPLS chairs
>>>>
>>>>
>>>>     _______________________________________________
>>>>     mpls mailing list [e-mail]
>>>>     To unsubscribe send an email [e-mail]
>>>
>>>     --     Prof. Dr. habil. Michael Menth
>>>     University of Tuebingen
>>>     Faculty of Science
>>>     Department of Computer Science
>>>     Chair of Communication Networks
>>>     Sand 13, 72076 Tuebingen, Germany
>>>     phone: [phone]
>>>     mailto:[e-mail] <mailto:[e-mail]>
>>>     http://kn.cs.uni-tuebingen.de
>>>
>>>     _______________________________________________
>>>     mpls mailing list -- [e-mail]
>>>     To unsubscribe send an email to [e-mail]
>>>
>>>
>>> _______________________________________________
>>> mpls mailing list -- [e-mail]
>>> To unsubscribe send an email to [e-mail]
>>

-- 
Loa Andersson
Senior MPLS Expert
Bronze Dragon Consulting
[e-mail]
[e-mail]