Re: [Ioam] Internal WG Review: In-situ OAM (ioam)

Hi Spencer,

There is some expertise overlap, but there are several distinct skills
needed especially in the context of virtual network functions since it
needs a deep understanding of server/NIC architectures.

Thanks,
Ramki

On Fri, Feb 10, 2017 at 9:19 AM, Spencer Dawkins at IETF <
spencerdawkins.ietf@gmail.com> wrote:

> Hi, Ram,
>
> On Feb 10, 2017 11:09, "Ram Krishnan" <ramkri123@gmail.com> wrote:
>
> Very good discussion.
>
> Another aspect to note is that we see value for in-situ OAM for monitoring
> network functions (especially virtual) besides network interconnects. This
> is captured in Section 3 in the draft (https://datatracker.ietf.org/
> doc/draft-krishnan-opsawg-in-band-pro-sla/?include_text=1).
>
> I believe IPPM charter addresses only network monitoring.
>
>
> The responsible AD for IPPM happens to agree :-)
>
> Is the expertise required for path monitoring the same as the expertise
> required for network element monitoring?
>
> Thanks,
>
> Spencer
>
>
> Thanks,
> Ramki
>
> On Fri, Feb 10, 2017 at 7:11 AM, Mirja Kuehlewind (IETF) <
> ietf@kuehlewind.net> wrote:
>
>> Hi Frank, hi all,
>>
>> please see in-line.
>>
>> > Am 10.02.2017 um 14:38 schrieb Frank Brockners (fbrockne) <
>> fbrockne@cisco.com>:
>> >
>> > Hi Mirja,
>> >
>> > you raise an interesting point. The IPPM charter states  " Specifying
>> network or lower layer OAM mechanisms is out of scope of the IPPM
>> charter.", whereas the WG has " Submit a draft on the IPv6 Performance and
>> Diagnostic Metrics (PDM) Destination Option as Proposed Standard
>> > draft-ietf-ippm-6man-pdm-option" as a milestone. I'd assume that we'd
>> likely qualify IPv6 as a transport protocol…
>>
>> If you also cite the sentence before this, this might become clearer:
>>
>> "The IP Performance Metrics (IPPM) Working Group develops and maintains
>> standard metrics that can be applied to the quality, performance, and
>> reliability of Internet data delivery services and applications running
>> over transport layer protocols (e.g. TCP, UDP) over IP. Specifying network
>> or lower layer OAM mechanisms is out of scope of the IPPM charter.“
>>
>> It’s focused on performance measurements of data delivery services and
>> application, not metric sthat are specific to network operation e.g.
>> up-time of a router (as an example that just came to my mind).
>>
>> So to me the scope and the goals of IPPM and IOAM are overlapping
>> currently as for me "real-time telemetry of individual data packets and
>> flows“ is exactly what IPPM is doing.
>>
>> >
>> > So far I understood the main focus of the new IOAM WG to be
>> network-layer focused, i.e. piggyback OAM-meta-data onto network-layer
>> protocols - but that does not necessarily need to be always the case as you
>> implicitly highlight by drawing the link to IPPM. One could also do so
>> using e.g. TCP options. I did not read the statement on IPPM in the draft
>> charter as "not cooperating with IPPM" - I read it in a way that methods
>> that do not piggyback information on live traffic are not considered in
>> IOAM. That said, especially when it comes to export and interpretation of
>> in-situ OAM data, there might indeed be common ground between IOAM and IPPM.
>>
>> My point is not that there needs to be cooperation. My point is that we
>> already have a working group that is mostly charter to do what you want to
>> do.
>>
>> Mirja
>>
>>
>> >
>> > How about we add another sentence to the charter that underlines the
>> fact that IOAM would actively seek cooperation with other related efforts?
>> We could add something like:
>> >
>> > "The IOAM WG seeks cooperation with other appropriate standards bodies
>> and forums to promote consistent approaches, as well as definition and
>> interpretation of in-situ OAM data."
>> >
>> > This would naturally capture IETF WGs like IPPM - but also efforts like
>> INT in P4, hence we'd even cast the net a little wider.
>> >
>> > Thoughts?
>> >
>> > Thanks, Frank
>> >
>> > -----Original Message-----
>> > From: Ioam [mailto:ioam-bounces@ietf.org] On Behalf Of Mirja
>> Kuehlewind (IETF)
>> > Sent: Freitag, 10. Februar 2017 13:10
>> > To: Carlos Pignataro (cpignata) <cpignata@cisco.com>; Alvaro Retana
>> (aretana) <aretana@cisco.com>
>> > Cc: iesg@ietf.org; The IAB <iab@iab.org>; Spencer Dawkins at IETF <
>> spencerdawkins.ietf@gmail.com>; ioam@ietf.org
>> > Subject: Re: [Ioam] Internal WG Review: In-situ OAM (ioam)
>> >
>> > Hi all,
>> >
>> > also one more comment on this point:
>> >
>> >> Am 09.02.2017 um 18:18 schrieb Carlos Pignataro (cpignata) <
>> cpignata@cisco.com>:
>> >>
>> >>>> Is there any connection with IPPM?
>> >>
>> >> Yes, there is, as already mentioned above.
>> >
>> >
>> > The charter currently says:
>> >
>> > "Other ongoing OAM-related efforts in working groups(s) such as MPLS
>> and IPPM that do not piggyback information onto the live user data traffic
>> are out of scope of the IOAM WG.“
>> >
>> > which indictates that cooperation with IPPM is not planned.
>> >
>> > To me in general the relation between this work and other ongoing work
>> in the IETF is not very clear and IPPM has several documents and milestones
>> that are in scope for this work:
>> >
>> > - Submit a draft on the IPv6 Performance and Diagnostic Metrics (PDM)
>> Destination Option as Proposed Standard: draft-ietf-ippm-6man-pdm-option
>> (this draft is mainly done and silenter the publication process soon to my
>> understanding)
>> >
>> > - Submit an Experimental draft on coloring-based hybrid measurement
>> methodologies for loss and delay to the IESG: draft-ietf-ippm-alt-mark-03
>> >
>> > I don’t think that the assessment in the charter that IPPM's scope does
>> not include piggybacked information is correct. Looking at
>> draft-brockners-inband-oam-transport-02, I think that any work on IPV6
>> and IPv6 in this scope should be done in IPPM because that’s were this work
>> is already on-going and where the measurement expertise is.
>> >
>> > Mirja
>> >
>> >
>> >
>> >
>> > _______________________________________________
>> > Ioam mailing list
>> > Ioam@ietf.org
>> > https://www.ietf.org/mailman/listinfo/ioam
>>
>> _______________________________________________
>> Ioam mailing list
>> Ioam@ietf.org
>> https://www.ietf.org/mailman/listinfo/ioam
>>
>
>
>
> --
> Thanks,
> Ramki
>
>
>

-- 
Thanks,
Ramki