<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body>
<p>Still not seeing any IPv6 from a roaming unit in New Zealand,
although I notice Starlink advertising global maritime coverage
now, which requires the laser links to work at least a little bit.</p>
<p>Dave, I'm currently collecting more flent data for you - did so
the other week but then realised that the directory the data was
going to was cloud synced, so wouldn't have made for a fair
measurement given that it would have synced via Starlink, too.</p>
<p>As for the jumps in performance, that's easily explained: As
dishy switches to a different satellite, it sees a different
queue, but the satellite that this queue belongs to also sees a
different set of users with different cardinality, doing different
things. So if you're switching to a satellite that sits over an
area with low Starlink user density, you're likely to see a
shorter queue (less latency) than if you're thrown in with the
crowd. If it's the BBR where a significant jump happened, then
that could just be coincidence.<br>
</p>
<div class="moz-cite-prefix">On 28/03/2023 7:54 am, David Lang via
Starlink wrote:<br>
</div>
<blockquote type="cite" cite="mid:768s6nq6-q9q8-741s-75o6-72qp279482rr@ynat.uz">
I'm seeing some IPv6 on starlink (and had to troubleshoot another
person on <br>
starlink that had a corporate access problem that was resovled by
disabling <br>
IPv6)<br>
<br>
David Lang<br>
<br>
On Mon, 27 Mar 2023, David Fernández via Starlink wrote:<br>
<br>
> Date: Mon, 27 Mar 2023 12:44:26 +0200<br>
> From: David Fernández via Starlink
<a class="moz-txt-link-rfc2396E" href="mailto:starlink@lists.bufferbloat.net"><starlink@lists.bufferbloat.net></a><br>
> Reply-To: David Fernández <a class="moz-txt-link-rfc2396E" href="mailto:davidfdzp@gmail.com"><davidfdzp@gmail.com></a><br>
> To: <a class="moz-txt-link-abbreviated" href="mailto:starlink@lists.bufferbloat.net">starlink@lists.bufferbloat.net</a><br>
> Subject: Re: [Starlink] Starlink ISL data<br>
> <br>
> I thought Starlink was not supporting IPv6, because of this:<br>
><br>
> <a href="https://twitter.com/elonmusk/status/967712110661615616" moz-do-not-send="true">https://twitter.com/elonmusk/status/967712110661615616</a><br>
><br>
>> Date: Sun, 26 Mar 2023 19:32:31 -0700<br>
>> From: Dave Taht <a class="moz-txt-link-rfc2396E" href="mailto:dave.taht@gmail.com"><dave.taht@gmail.com></a><br>
>> To: Dave Taht via Starlink
<a class="moz-txt-link-rfc2396E" href="mailto:starlink@lists.bufferbloat.net"><starlink@lists.bufferbloat.net></a><br>
>> Subject: Re: [Starlink] Starlink ISL data<br>
>> Message-ID:<br>
>>
<a class="moz-txt-link-rfc2396E" href="mailto:CAA93jw4kcfXiwMw0zKAuQV3BjpWp6ZvQ2TQFbtEy9CFRN5aY2w@mail.gmail.com"><CAA93jw4kcfXiwMw0zKAuQV3BjpWp6ZvQ2TQFbtEy9CFRN5aY2w@mail.gmail.com></a><br>
>> Content-Type: text/plain; charset="utf-8"<br>
>><br>
>> Nate gave me the opportunity to test a bit of ipv6 access
on one of<br>
>> his dishys today. I am going to give p2p a shot, also,
and he<br>
>> conveniently has bbr2 installed. It is of course
difficult to discern<br>
>> the difference between transport behaviors and their
underlying<br>
>> connectivity - for example the BBR2 result attached has a
baseline<br>
>> (idle!) latency jump of over 40ms which is hard to
explain.<br>
>><br>
>> Despite these plots saying downloads, they were
essentially uploads<br>
>> from his box over the internet.<br>
>><br>
>> The summary of the data I have so far on this direction
is:<br>
>><br>
>> Cubic, looks like cubic, usually, but not always. BBR
rarely looks like BBR.<br>
>><br>
>> On Fri, Mar 24, 2023 at 2:46 PM Dave Taht
<a class="moz-txt-link-rfc2396E" href="mailto:dave.taht@gmail.com"><dave.taht@gmail.com></a> wrote:<br>
>>><br>
>>> Vint just asked me a difficult question over here
about the<br>
>>> performance of the ISL links for my upcoming AMA next
week<br>
>>><br>
>>> <a href="https://twitter.com/mtaht/status/1639361656106156032" moz-do-not-send="true">https://twitter.com/mtaht/status/1639361656106156032</a><br>
>>><br>
>>> And to date, we really don't know. We do know it is
up!? but...<br>
>>><br>
>>> has anyone managed to measure p2p ipv6 performance
starlink to<br>
>>> starlink over an ISL link as yet? Do we know anyone
at the poles? In<br>
>>> general I always look for flent and irtt data, but
I'd settle for a,<br>
>>> oh, call it 5-10 minute long packet capture of single
iperf flow,<br>
>>> running over tcp cubic (bbr would be great too)....
one test in each<br>
>>> direction.<br>
>>><br>
>>> (in fact that would be great from any starlink
terminal to any of my<br>
>>> servers around the world)<br>
>>><br>
>>> I have some data from a couple of you (thx ulrich in
particular!), but<br>
>>> I have not sat down to take it apart as I have been
far, far too busy<br>
>>> with libreqos and a bunch of nice, small, competent
ISPs deploying<br>
>>> that, to worry about fixing a billionaire's network
all that much....<br>
>>> but I've set aside next week to answer AMAs about
everything from all<br>
>>> and sundry, so if you got data, please share?<br>
>>><br>
>>> --<br>
>>> Come Heckle Mar 6-9 at: <a href="https://www.understandinglatency.com" moz-do-not-send="true">https://www.understandinglatency.com/</a><br>
>>> Dave Täht CEO, TekLibre, LLC<br>
>><br>
>><br>
>><br>
>> --<br>
>> <a href="https://www.youtube.com/watch?v=tAVwmUG21OY&t=6483s" moz-do-not-send="true">https://www.youtube.com/watch?v=tAVwmUG21OY&t=6483s</a><br>
>> Dave Täht CEO, TekLibre, LLC<br>
>> -------------- next part --------------<br>
>> A non-text attachment was scrubbed...<br>
>> Name: tcp_ndown_-_starlink-ipv6-cubic.png<br>
>> Type: image/png<br>
>> Size: 145300 bytes<br>
>> Desc: not available<br>
>> URL:<br>
>> <<a href="https://lists.bufferbloat.net/pipermail/starlink/attachments/20230326/32c7fed1/attachment.png" moz-do-not-send="true">https://lists.bufferbloat.net/pipermail/starlink/attachments/20230326/32c7fed1/attachment.png</a>><br>
>> -------------- next part --------------<br>
>> A non-text attachment was scrubbed...<br>
>> Name: tcp_ndown_-_starlink-ipv6-bbr2.png<br>
>> Type: image/png<br>
>> Size: 64805 bytes<br>
>> Desc: not available<br>
>> URL:<br>
>> <<a href="https://lists.bufferbloat.net/pipermail/starlink/attachments/20230326/32c7fed1/attachment-0001.png" moz-do-not-send="true">https://lists.bufferbloat.net/pipermail/starlink/attachments/20230326/32c7fed1/attachment-0001.png</a>><br>
> _______________________________________________<br>
> Starlink mailing list<br>
> <a class="moz-txt-link-abbreviated" href="mailto:Starlink@lists.bufferbloat.net">Starlink@lists.bufferbloat.net</a><br>
> <a href="https://lists.bufferbloat.net/listinfo/starlink" moz-do-not-send="true">https://lists.bufferbloat.net/listinfo/starlink</a>
<br>
<fieldset class="moz-mime-attachment-header"></fieldset>
<pre class="moz-quote-pre" wrap="">_______________________________________________
Starlink mailing list
<a class="moz-txt-link-abbreviated" href="mailto:Starlink@lists.bufferbloat.net">Starlink@lists.bufferbloat.net</a>
<a class="moz-txt-link-freetext" href="https://lists.bufferbloat.net/listinfo/starlink">https://lists.bufferbloat.net/listinfo/starlink</a>
</pre>
</blockquote>
<pre class="moz-signature" cols="72">--
****************************************************************
Dr. Ulrich Speidel
School of Computer Science
Room 303S.594 (City Campus)
The University of Auckland
<a class="moz-txt-link-abbreviated" href="mailto:u.speidel@auckland.ac.nz">u.speidel@auckland.ac.nz</a>
<a class="moz-txt-link-freetext" href="http://www.cs.auckland.ac.nz/~ulrich/">http://www.cs.auckland.ac.nz/~ulrich/</a>
****************************************************************
</pre>
</body>
</html>