[ippm] Re: STAMP extension to enable bi-directional ECN meas urement
<[email protected]> Tue, 18 Mar 2025 14:10:20 +0000
| Newsgroups | gmane.ietf.ippm,gmane.ietf.tsvwg |
|---|---|
| Message-ID | <BEZP281MB2007FB6A0090B646E1D0DC469CDE2@BEZP281MB2007.DEUP281.PROD.OUTLOOK.COM> |
--===============0694891122920174359== Content-Language: de-DE Content-Type: multipart/alternative; boundary="_000_BEZP281MB2007FB6A0090B646E1D0DC469CDE2BEZP281MB2007DEUP_" --_000_BEZP281MB2007FB6A0090B646E1D0DC469CDE2BEZP281MB2007DEUP_ Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi Greg, thanks, the draft addresses a reasonable goal, detecting transparent transp= ort of ECN codepoints along a network path. This is useful, independently o= f the question whether you try to detect changes RFC3168 or RFC9331 ECN bi= t settings. I note, that we had to debug our lab L4S set up, and we had to = debug our L4S production chain, as ECN codepoints didn't pass transparently= end-to-end, and I think to recollect that Comcast did so too. To quote fro= m Jason's draft, https://www.ietf.org/archive/id/draft-livingood-low-latency-deployment-07.h= tml: "Nevertheless, it is important to check network policies and router co= nfigurations to confirm [transparent transport of ECN codepoints] and to va= lidate it via packet captures at an end user CPE." One could of course use a STAMP implementation to test, whether a, say, EC= N-aware scheduler changes ECT to CE too. This requires to load the link whe= re that scheduler operates - STAMP is adding features to that extent. UDP= Speed Test (UDPST) may be an option too. I think, operational and security concerns shouldn't be ignored (I take the= generalized point "source shouldn't send ECN codepoints, if ECN no ECN awa= re transport is supported" as such a concern). If STAMP offers a sufficient= set of features and implementation requirements, like authentication of a = STAMP test packet sender and limiting replies/reflection of test packets to= the authenticated STAMP sender. If I recollect correctly, STAMP offers all= that (Greg Mirsky can confirm this, he knows best). Regards, Ruediger Von: Greg White <g.white=3D40CableLabs.com-Tr9gZwTxerDR74oF6e/[email protected]> Gesendet: Dienstag, 18. M=E4rz 2025 12:46 An: [email protected] Cc: [email protected] Betreff: [tsvwg] STAMP extension to enable bi-directional ECN measurement Hi all, I'd like to make TSVWG aware of a draft in IPPM that defines an extension t= o the Simple Two-Way Active Measurement Protocol (STAMP). STAMP supports "= extensions" in the form of TLVs that can optionally be included in a STAMP = test packet by a Session Sender, and that are returned in the reflected pac= ket by the Session Reflector (often with additional information inserted). This new extension allows STAMP to be used for bi-directional measurement o= f path ECN characteristics, including ECN traversal/bleaching measurements,= probing for CE-marking on paths, and differential measurement of path char= acteristics that depend on the value of the ECN field. STAMP previously ha= d the ability to be used to test bi-directional DSCP characteristics of a p= ath, but only outbound ECN characteristics. This draft was presented during the IPPM session today, and it was suggeste= d that TSVWG be made aware since it has to do with ECN marking, and by send= ing test packets with various ECN values, it could be seen as violating the= letter of RFC3168 & RFC9331 (e.g. having a host send individual ECT1 test= packets, even though it doesn't implement the Prague requirements). Time permitting, I could give a brief overview during the TSVWG tomorrow, b= ut the draft itself is pretty straightforward. The important point from th= e IPPM perspective is to seek consensus from TSVWG that it's ok to use STAM= P to measure path ECN characteristics in this way. https://datatracker.ietf.org/meeting/122/materials/slides-122-ippm-stamp-ex= tension-for-dscp-and-ecn-traversal-measurement-00 https://gwhitecl.github.io/draft-white-ippm-stamp-ecn/draft-white-ippm-stam= p-ecn.html Best, Greg --_000_BEZP281MB2007FB6A0090B646E1D0DC469CDE2BEZP281MB2007DEUP_ Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable <html xmlns:v=3D"urn:schemas-microsoft-com:vml" xmlns:o=3D"urn:schemas-micr= osoft-com:office:office" xmlns:w=3D"urn:schemas-microsoft-com:office:word" = xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" xmlns=3D"http:= //www.w3.org/TR/REC-html40"> <head> <meta http-equiv=3D"Content-Type" content=3D"text/html; charset=3Diso-8859-= 1"> <meta name=3D"Generator" content=3D"Microsoft Word 15 (filtered medium)"> <style><!-- /* Font Definitions */ @font-face {font-family:"Cambria Math"; panose-1:2 4 5 3 5 4 6 3 2 4;} @font-face {font-family:Calibri; panose-1:2 15 5 2 2 2 4 3 2 4;} @font-face {font-family:Aptos;} /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal {margin:0cm; font-size:12.0pt; font-family:"Aptos",sans-serif; mso-ligatures:standardcontextual;} a:link, span.MsoHyperlink {mso-style-priority:99; color:#467886; text-decoration:underline;} .MsoChpDefault {mso-style-type:export-only; font-size:10.0pt; mso-ligatures:none;} @page WordSection1 {size:612.0pt 792.0pt; margin:72.0pt 72.0pt 72.0pt 72.0pt;} div.WordSection1 {page:WordSection1;} --></style><!--[if gte mso 9]><xml> <o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" /> </xml><![endif]--><!--[if gte mso 9]><xml> <o:shapelayout v:ext=3D"edit"> <o:idmap v:ext=3D"edit" data=3D"1" /> </o:shapelayout></xml><![endif]--> </head> <body lang=3D"DE" link=3D"#467886" vlink=3D"#96607D" style=3D"word-wrap:bre= ak-word"> <div class=3D"WordSection1"> <p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;mso-fareast-language= :EN-US">Hi Greg,<o:p></o:p></span></p> <p class=3D"MsoNormal"><span style=3D"font-size:11.0pt;mso-fareast-language= :EN-US"><o:p> </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt;mso-f= areast-language:EN-US">thanks, the draft addresses a reasonable goal, detec= ting transparent transport of ECN codepoints along a network path. This is = useful, independently of the question whether you try to detect changes </span><span lang=3D"EN-US" style=3D"fon= t-size:11.0pt">RFC3168 or RFC9331 ECN bit settings. I note, that we h= ad to debug our lab L4S set up, and we had to debug our L4S production chai= n, as ECN codepoints didn’t pass transparently end-to-end, and I think to recollect that Comcast did so too. To quote fro= m Jason’s draft,<o:p></o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt"><o:p= > </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt;mso-f= areast-language:EN-US"><a href=3D"https://www.ietf.org/archive/id/draft-liv= ingood-low-latency-deployment-07.html">https://www.ietf.org/archive/id/draf= t-livingood-low-latency-deployment-07.html</a><span style=3D"color:black">: “Nevertheless, it is important to check network policies and router = configurations to confirm [transparent transport of ECN codepoints] and to = validate it via packet captures at an end user CPE.”<o:p></o:p></span= ></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt;color= :black;mso-fareast-language:EN-US"><o:p> </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt;color= :black;mso-fareast-language:EN-US">One could of course use a STAMP implemen= tation to test, whether a, say, ECN-aware scheduler changes ECT to CE= too. This requires to load the link where that scheduler operates – STAMP is adding features to th= at extent. UDP Speed Test (UDPST) may be an option too.</span><span lang=3D= "EN-US" style=3D"font-size:11.0pt;mso-fareast-language:EN-US"><o:p></o:p></= span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt;mso-f= areast-language:EN-US"><o:p> </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt;mso-f= areast-language:EN-US">I think, operational and security concerns shouldn&#= 8217;t be ignored (I take the generalized point “source shouldn’= ;t send ECN codepoints, if ECN no ECN aware transport is supported” as such a concern). If STAMP offers a sufficient set of f= eatures and implementation requirements, like authentication of a STAMP tes= t packet sender and limiting replies/reflection of test packets to the auth= enticated STAMP sender. If I recollect correctly, STAMP offers all that (Greg Mirsky can confirm this, he knows b= est). <o:p> </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt;mso-f= areast-language:EN-US"><o:p> </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt;mso-f= areast-language:EN-US">Regards,<o:p></o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt;mso-f= areast-language:EN-US"><o:p> </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt;mso-f= areast-language:EN-US">Ruediger<o:p></o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt;mso-f= areast-language:EN-US"><o:p> </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"mso-fareast-language:E= N-US"><o:p> </o:p></span></p> <div> <div style=3D"border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0cm = 0cm 0cm"> <p class=3D"MsoNormal"><b><span style=3D"font-size:11.0pt;font-family:"= ;Calibri",sans-serif;mso-ligatures:none">Von:</span></b><span style=3D= "font-size:11.0pt;font-family:"Calibri",sans-serif;mso-ligatures:= none"> Greg White <g.white=3D40CableLabs.com-Tr9gZwTxerDR74oF6e/[email protected]> <br> <b>Gesendet:</b> Dienstag, 18. M=E4rz 2025 12:46<br> <b>An:</b> [email protected]<br> <b>Cc:</b> [email protected]<br> <b>Betreff:</b> [tsvwg] STAMP extension to enable bi-directional ECN measur= ement<o:p></o:p></span></p> </div> </div> <p class=3D"MsoNormal"><o:p> </o:p></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt">Hi a= ll,<o:p></o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt"><o:p= > </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt">I= 217;d like to make TSVWG aware of a draft in IPPM that defines an extension= to the Simple Two-Way Active Measurement Protocol (STAMP). STAMP sup= ports “extensions” in the form of TLVs that can optionally be included in a STAMP test packet by a Session Sender, and that are retur= ned in the reflected packet by the Session Reflector (often with additional= information inserted).<o:p></o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt"><o:p= > </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt">This= new extension allows STAMP to be used for bi-directional measurement of pa= th ECN characteristics, including ECN traversal/bleaching measurements, pro= bing for CE-marking on paths, and differential measurement of path characteristics that depend on the value of the ECN fi= eld. STAMP previously had the ability to be used to test bi-direction= al DSCP characteristics of a path, but only outbound ECN characteristics. <o:p></o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt"><o:p= > </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt">This= draft was presented during the IPPM session today, and it was suggested th= at TSVWG be made aware since it has to do with ECN marking, and by sending = test packets with various ECN values, it could be seen as violating the letter of RFC3168 & RFC9331 (e= .g. having a host send individual ECT1 test packets, even though it doesn&#= 8217;t implement the Prague requirements).<o:p></o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt"><o:p= > </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt">Time= permitting, I could give a brief overview during the TSVWG tomorrow, but t= he draft itself is pretty straightforward. The important point from t= he IPPM perspective is to seek consensus from TSVWG that it’s ok to use STAMP to measure path ECN characteristics = in this way.<o:p></o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt"><o:p= > </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt"><a h= ref=3D"https://datatracker.ietf.org/meeting/122/materials/slides-122-ippm-s= tamp-extension-for-dscp-and-ecn-traversal-measurement-00">https://datatrack= er.ietf.org/meeting/122/materials/slides-122-ippm-stamp-extension-for-dscp-= and-ecn-traversal-measurement-00</a><o:p></o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt"><a h= ref=3D"https://gwhitecl.github.io/draft-white-ippm-stamp-ecn/draft-white-ip= pm-stamp-ecn.html">https://gwhitecl.github.io/draft-white-ippm-stamp-ecn/dr= aft-white-ippm-stamp-ecn.html</a><o:p></o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt"><o:p= > </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt">Best= ,<o:p></o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt">Greg= <o:p></o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt"><o:p= > </o:p></span></p> <p class=3D"MsoNormal"><span lang=3D"EN-US" style=3D"font-size:11.0pt"><o:p= > </o:p></span></p> </div> </body> </html> --_000_BEZP281MB2007FB6A0090B646E1D0DC469CDE2BEZP281MB2007DEUP_-- --===============0694891122920174359== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: base64 Content-Disposition: inline X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KaXBwbSBtYWls aW5nIGxpc3QgLS0gaXBwbUBpZXRmLm9yZwpUbyB1bnN1YnNjcmliZSBzZW5kIGFuIGVtYWlsIHRv IGlwcG0tbGVhdmVAaWV0Zi5vcmcK --===============0694891122920174359==--