Re: [ippm] 答复: WGLC for STAMP Extensions

[Greg Mirsky <[e-mail]>]

Hi Yuezhong,
great questions, thank you! Please find my answers below in-line tagged
GIM3>>.

Regards,
Greg

On Wed, Jun 3, 2020 at 8:51 PM Songyuezhong (songyuezhong, IP technology
Research Dept) <[e-mail]> wrote:

> Hi Greg,
>
>
>
> ok, for the new STAMP application document we can discuss off-list.
>
> And for this draft, I still have some problem. such as the Follow-up
> Telemetry TLV, Follow-up Timestamp is record by egress when the packet
> Return to sender, And this timestamp will be carried next time?
>
GIM3>> You are absolutely correct. Obtaining a timestamp on the
transmission of a packet is challenging and various techniques have been
used to improve the consistency of this process, i.e., minimize the
variable delay between when the timestamp value is obtained and the
transmission of the packet actually begins. Separating the part of reading
the time value from the transporting it is what the Follow-up extension
provides to STAMP-based systems. As you've noted, the time value is read
and saved at the Session-Reflector very close to the physical transmission
of the reflected packet. And then that value may be included in the next
reflected STAMP packet in the Follow-up TLV.

> And the last problem is why HMAC TLV is needed in authenticated mode, I
> think this mode has this function inherently.
>
GIM3>> I'll note that only the base STAMP packet is protected by HMAC per
RFC 8762. The HMAC TLV is to protect STAMP extensions. I hope that makes it
clearer.

>
>
> Regards,
>
> Yuezhong
>
> *发件人:* Greg Mirsky [mailto:[e-mail]]
> *发送时间:* 2020年6月3日 3:49
> *收件人:* Songyuezhong (songyuezhong, IP technology Research Dept) <
> [e-mail]>
> *抄送:* Ian Swett <ianswett=[e-mail]>; IETF IPPM WG (
> [e-mail]) <[e-mail]>; Tianran Zhou <[e-mail]>
> *主题:* Re: [ippm] 答复: WGLC for STAMP Extensions
>
>
>
> Hi Yuezhong,
>
> great, let us connect off-list to share ideas about a new STAMP
> application document.
>
> On your other questions (I brought it to the front) I've added my notes
> under GIM2>> tag below:
>
>
>
> And another question is how to use Class of Service TLV to find the
> misconfigure problem, is it enough?
> GIM>> One of the possible scenarios could be as follows:
>
> STAMP Sender sets DSCP1 to value A
> STAMP packet is transmitted with DSCP set to A
> STAMP Reflector copies DSCP value into DSCP2 field
> reflected STAMP packet is transmitted with DSCP set to A (as requested by
> the STAMP Sender)
> STAMP Sender receives the STAMP packet with DSCP A but DSCP2 value is B
> not as expected.
> I hope this little example helps. Obviously, there are many ways to use
> the CoS TLV to test CoS mappings.
>
> song>> the CoS mappings happened in Sender or other places, if DSCP value
> is not same with DSCP2 value, it means a error in which place?
>
> GIM2>> Let us assume that no CoS re-mapping expected along a path between
> the Sender and the Reflector. If the value in the DSCP2 field is different
> from the value set in the DSCP field by the Sender at the transmission,
> then the error is on the downstream leg of the path. If the value in the
> DSCP1 field is different from the value in the DSCP field of the reflected
> packet received by the Sender, then the error is on the upstream leg of the
> path. I'll note that CoS re-mapping may be used and then the determination
> of the error condition should be based on the expected behavior. I hope
> that helps.
>
> song>> and for Access Report TLV, can you explain more, for example the
> location of sender and reflector both in user side, and how to find the
> reflector status changed, very thanks!
>
> GIM2>> As noted in the last paragraph in Section 4.6:
>
>    The Access Report TLV is used by the Performance Measurement Function
>    (PMF) components of the Access Steering, Switching and Splitting
>    feature for 5G networks [TS23501].  The PMF component in the User
>    Equipment acts as the STAMP Session-Sender, and the PMF component in
>    the User Plane Function acts as the STAMP Session-Reflector.
>
> UE acts as Session-Sender and UPF - Session-Reflector.
>
>
>
> Regards,
>
> Greg
>
>
>
> On Mon, Jun 1, 2020 at 8:59 PM Songyuezhong (songyuezhong, IP technology
> Research Dept) <[e-mail]> wrote:
>
> Hi Greg,
>
> thanks for the reply from you and Ian, some of my questions have been
> answered, and there are still a few problems I don't understand,
>
> I will use the way you use with song>> tag for my reply
>
>
>
> Regards,
>
> Yuezhong
>
>
>
> *发件人:* Greg Mirsky [mailto:[e-mail]]
> *发送时间:* 2020年6月2日 0:03
> *收件人:* Songyuezhong (songyuezhong, IP technology Research Dept) <
> [e-mail]>
> *抄送:* Ian Swett <ianswett=[e-mail]>; IETF IPPM WG (
> [e-mail]) <[e-mail]>
> *主题:* Re: [ippm] 答复: WGLC for STAMP Extensions
>
>
>
> Hi Yuezhong,
>
> thank you for your comments and suggestions. Please find my notes and
> answers in-line under the GIM>> tag.
>
>
>
> Regards,
>
> Greg
>
>
>
> On Sat, May 30, 2020 at 4:56 AM Songyuezhong (songyuezhong, IP technology
> Research Dept) <[e-mail]> wrote:
>
> Hi Ian，
>
>
>
> I have read the latest version of this draft，and have some small
> suggestions, hope it is helpful for you.
>
>
>
> For part 4,there list 8 new TLVs, but it seems not detailed enough for
> each TLV about the application scenario and some terms in it, we need guess
> to understand the whole plan.
>
> GIM>> We have tried to provide a clear technical description of extensions
> to help implementers produce interoperable implementations. Describing
> various scenarios an extension may be used in was not our main objective.
> There are other SDOs that reference STAMP and STAMP TLVs in their
> documents. I can mention BBF's WT-390.2 IP Performance Measurement from IP
> Edge to Customer Equipment using STAMP, and MEF's MEF-w66 Service OAM for
> IP Services. Both documents are in advanced phase and will be published
> later this year.
>
>
>
> Especially for the people who have no background knowledge of each
> application scenario, maybe it is more hard for them to understand.
>
> GIM>> Yes, you are correct. Standard documents require a certain level of
> knowledge in the particular area of the technology.
>
>
>
> So I suggest for each TLV, there should have some pictures and background
> content to help people understand the TLV’s meaning and using method，it
> will be better.
>
> GIM>> That is very helpful suggestion and I think that it can be a basis
> for the Applicability of STAMP document. Would you be interested in working
> on the new document together?
>
>
>
> song>>We would like to work on the new document you mentioned，if there
> have some plan，we can discuss together.
>
>
>
> By the way, I have some doubt about the Location TLV, which is the last-hop router, the reflector or the router before it? And how to indicate if the STAMP packets are send to the wrong Session-Reflector from this TLV?
>
> GIM>> I hope that Henrik's response clarified one of the use case
> scenarios.
>
>
>
> And another question is how to use Class of Service TLV to find the misconfigure problem, is it enough?
>
> GIM>> One of the possible scenarios could be as follows:
>
>    - STAMP Sender sets DSCP1 to value A
>    - STAMP packet is transmitted with DSCP set to A
>    - STAMP Reflector copies DSCP value into DSCP2 field
>    - reflected STAMP packet is transmitted with DSCP set to A (as
>    requested by the STAMP Sender)
>    - STAMP Sender receives the STAMP packet with DSCP A but DSCP2 value
>    is B not as expected.
>
> I hope this little example helps. Obviously, there are many ways to use
> the CoS TLV to test CoS mappings.
>
> song>> the CoS mappings happened in Sender or other places, if DSCP value
> is not same with DSCP2 value, it means a error in which place?
>
> song>> and for Access Report TLV, can you explain more, for example the
> location of sender and reflector both in user side, and how to find the
> reflector status changed, very thanks!
>
>
>
> Thanks,
> Yuezhong
>
>
>
> *发件人:* ippm [mailto:[e-mail]] *代表 *Ian Swett
> *发送时间:* 2020年5月23日 5:26
> *收件人:* IETF IPPM WG ([e-mail]) <[e-mail]>
> *主题:* [ippm] WGLC for STAMP Extensions
>
>
>
> Hi IPPM,
>
> At our virtual interim meeting, we decided
> draft-ietf-ippm-stamp-option-tlv was ready for last call. This email starts
> a two-week WGLC for this draft.
>
> The latest version can be found here:
> https://tools.ietf.org/html/draft-ietf-ippm-stamp-option-tlv-04
>
> This last call will end on *Monday, June 8th*. Please reply to
> [e-mail] with your reviews and comments.
>
> Thanks,
> Ian & Tommy
>
> _______________________________________________
> ippm mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/ippm
>
>