[Starlink] Debugging connection through Starlink (POINTR)

Sauli Kiviranta smksauli at gmail.com
Mon Mar 6 16:00:28 EST 2023


Thank you Ulrich for sharing the material!

I was finally able to get rough estimate of the test site, pictures
attached. Comparison with your satellite map from the presentation,
with my limited experience, looks like the situation is not so bad?

Really strange, maybe the only option is the local obstructions.
Alternatively it was just unfortunate routing for the call on our
side, very hard to judge. Now the pivot point is determined by the
call source.

Best regards,
Sauli


On 01/03/2023, Ulrich Speidel via Starlink
<starlink at lists.bufferbloat.net> wrote:
> On 1/03/2023 3:41 am, Sauli Kiviranta via Starlink wrote:
>>
>> Based on your location in NZ you should be among the best tests in
>> terms of worst possible location in relation to our infra for delays.
>
> Actually, probably we're anything but ideal here. Our Starlink kit is an
> RV subscription one (our campus is too built-up for a fixed one and H&S
> gets in the way of going onto any roof with students). RV subscription
> capacity is managed differently from fixed subscriptions, which is what
> most people have (we see much more variation in goodput and RTT than
> what I've seen fixed users report).
>
> Also, our location here gives us a number of "unfair advantages" in
> terms of Starlink: proximity to multiple gateways (lost of choice in
> terms of sats visible from both UT and GW to route through),
> neighbourhood on a cheap high fibre diet (not much competition for
> capacity from other users in the cell), small extended neighbourhood
> (some of which is already saturated). See my slides of a panel talk I
> gave the other day at APNIC55 under
> https://conference.apnic.net/55/assets/files/APPS314/starlink-first-impre_1677543480.pdf.
>
>
>
> So I'm not sure how comparable our situation would be with that of your
> customer.
>
>> Do you mind running POINTR network analysis and see what kind of
>> results you get? We did some tests with mining industry in Australia
>> with sat but it is already few years ago not Starlink.
>
> Australia is topologically further away from Finland than NZ - if you do
> a traceroute, you should see your traffic go via the US, and then
> depending on the cable it takes down to Oceania, it may actually transit
> NZ (Southern Cross cable) before hitting Australia.
>
> I'll see what I can do in terms of POINTR once our industrial action
> here ends (I'm currently nominally on strike, but then again I'd
> probably be on here even if I didn't get paid ;-)).
>
>> Even better if
>> you have two devices, register them then call to yourself so both
>> endpoints go through the Starlink to see if everything or anything
>> works and what type of delays you experience (double trouble!).
>>
>> Best regards,
>> Sauli
>>
>>
>> On 28/02/2023, Ulrich Speidel via Starlink
>> <starlink at lists.bufferbloat.net> wrote:
>> > I think you may find this difficult to reproduce unless you actually go
>> > and see your customer.
>> >
>> > Starlink isn't the same as Starlink. How it works depends on:
>> >
>> > * Where your dishy is and where the closest gateway(s) is/are. This
>> > determines the number of options that the system has to route
>> > between you and the gateway. The closer, the more options,
>> > generally, as the number of sats that both you and the gateway can
>> > see decreases, as a rule of thumb, with distance between you and
>> > gateway.
>> > * Where you are also matters. Satellite density is highest at 53 deg
>> > north and 53 deg south. Starlink units between these parallels will
>> > look in the direction of their hemisphere when looking for a
>> > satellite. Units at higher latitudes ... well someone tell me where
>> > they look!
>> > * How many other Starlink users there are (a) in your cell and (b) in
>> > the neighbourhood of your cell. This determines the share of
>> > capacity you'll get off the system. The fewer competing users, the
>> > better.
>> > * What sort of backhaul the gateway is on. I only see IPv4 here in NZ
>> > but some folk are on IPv6, and I guess the backhaul network beyond
>> > Starlink's control also matters in terms of bufferbloat etc.
>> > * Weather
>> > * Obstructions (Starlink can works pretty well with a moderate degree
>> > of obstruction, but obviously, as stated above, the more options the
>> > system has...)
>> > * Available bandwidth (did Starlink get all the frequencies that it
>> > would have wanted at your customer's site or just some?) - that's a
>> > regulatory issue.
>> >
>> > Basically, you could be comparing apples and pears here.
>> >
>> > On 1/03/2023 12:53 am, Sauli Kiviranta via Starlink wrote:
>> >> Thank you for the testing so far!
>> >>
>> >> Special thanks to Nathan for the test call yesterday, looks like
>> >> everything works. I was shocked how good the Starlink performed, I was
>> >> in disbelief how well it worked since on my side I was also on 4G and
>> >> the performance was spectacular. Not the usual
>> >> 8000km-and-sat-inbetween feeling.
>> >>
>> >> So far we have not been able to reproduce the errors, if the call is
>> >> initiated from US it will route through the local infra. So this could
>> >> be one issue, if someone wants to make another test call we can route
>> >> it through EU.
>> >>
>> >> Anyone in Canada capable of making a test? Even to just run the
>> >> network analysis.
>> >>
>> >> I will meanwhile have to point the finger at the customer if there are
>> >> no signs of failure in any of the tests, I hate that. Fixing a juicy
>> >> bug is better than having configuration when things just fail "without
>> >> reason".
>> >>
>> >> I can also provide additional test tools if someone is interested in
>> >> more bolts and nuts testing. Send me an email with platform preference
>> >> and I will share the test kit.
>> >>
>> >> Best regards,
>> >> Sauli
>> >>
>> >>
>> >> On 27/02/2023, starlink-request at lists.bufferbloat.net
>> >> <starlink-request at lists.bufferbloat.net> wrote:
>> >> > Send Starlink mailing list submissions to
>> >> > starlink at lists.bufferbloat.net
>> >> >
>> >> > To subscribe or unsubscribe via the World Wide Web, visit
>> >> > https://lists.bufferbloat.net/listinfo/starlink
>> <https://lists.bufferbloat.net/listinfo/starlink>
>> >> <https://lists.bufferbloat.net/listinfo/starlink
>> <https://lists.bufferbloat.net/listinfo/starlink>>
>> >> > or, via email, send a message with subject or body 'help' to
>> >> > starlink-request at lists.bufferbloat.net
>> >> >
>> >> > You can reach the person managing the list at
>> >> > starlink-owner at lists.bufferbloat.net
>> >> >
>> >> > When replying, please edit your Subject line so it is more specific
>> >> > than "Re: Contents of Starlink digest..."
>> >> >
>> >> >
>> >> > Today's Topics:
>> >> >
>> >> > 1. Debugging connection through Starlink (POINTR) (Sauli Kiviranta)
>> >> >
>> >> >
>> >> >
>> ----------------------------------------------------------------------
>> >> >
>> >> > Message: 1
>> >> > Date: Mon, 27 Feb 2023 10:58:53 +0200
>> >> > From: Sauli Kiviranta <smksauli at gmail.com>
>> >> > To: starlink at lists.bufferbloat.net
>> >> > Subject: [Starlink] Debugging connection through Starlink (POINTR)
>> >> > Message-ID:
>> >> > <CAJbqEYCDQOqqfpt4=gEpGV37URRh2UvGLjKV1XEpNPm1bxMu3Q at mail.gmail.com>
>> >> > Content-Type: text/plain; charset="utf-8"
>> >> >
>> >> > Dear Starlink users,
>> >> >
>> >> > I have a request for help on debugging a customer issue. We have
>> users
>> >> > in Canada who are using our industrial remote collaboration tool
>> >> > POINTR (www.pointr.com
>> <http://www.pointr.com>
>> >> <http://www.pointr.com
>> <http://www.pointr.com>>)
>> >> and have issues connecting through Starlink
>> >> > (Standard HTTPS). We cannot reproduce their case as we have no
>> >> > Starlink at our disposal.
>> >> >
>> >> > If you would be kind enough to download POINTR (either from the app
>> >> > stores or from our website: https://www.pointr.com/downloads/
>> <https://www.pointr.com/downloads>
>> >> <https://www.pointr.com/downloads
>> <https://www.pointr.com/downloads>>).
>> >> The
>> >> > app itself follows standard SMS validation loop, if this is an
>> issue I
>> >> > can provide you with virtual number with email validation.
>> >> >
>> >> > Problem 1: Can you connect to the servers?
>> >> > - Once you try to complete registration loop, you should get an
>> >> > error
>> >> > that you are offline indicative that servers cannot be reached (AWS,
>> >> > Ireland/US/Brazil/Japan).
>> >> >
>> >> > Problem 2: What is the perceived network quality within the app?
>> >> > - Considering the outlined server distribution, if you run the
>> network
>> >> > analysis, can you take a screenshot and share it? (Example
>> >> > attached).
>> >> >
>> >> > Problem 3: What is the call quality between your devices?
>> >> > - If you can run network analysis, can you register another device
>> >> > of
>> >> > yours (same mobile number is ok) and call between your own devices
>> >> > to
>> >> > test how the video streaming goes through and what type of delay you
>> >> > experience? I will be also available for test calls if you want to
>> >> > establish a call between your location and Finland: +358 40 357 3272
>> >> > (group calls up to 7 users).
>> >> >
>> >> > Additionally any technical feedback on the behavior will be
>> >> > appreciated specially if you experience issues.
>> >> >
>> >> > Best regards,
>> >> > Sauli Kiviranta
>> >> >
>> >> > www.deltacygnilabs.com
>> <http://www.deltacygnilabs.com>
>> >> <http://www.deltacygnilabs.com
>> <http://www.deltacygnilabs.com>>
>> >> >
>> >> > Downloads:
>> >> >
>> >> > Android:
>> >> >
>> >>
>> https://play.google.com/store/apps/details?id=com.DeltaCygniLabs.Pointr
>> <https://play.google.com/store/apps/details?id=com.DeltaCygniLabs.Pointr>
>> >>
>> <https://play.google.com/store/apps/details?id=com.DeltaCygniLabs.Pointr
>> <https://play.google.com/store/apps/details?id=com.DeltaCygniLabs.Pointr>>
>> >> > APK:
>> >> > https://www.pointr.com/downloads/#pointr-android
>> <https://www.pointr.com/downloads/#pointr-android>
>> >> <https://www.pointr.com/downloads/#pointr-android
>> <https://www.pointr.com/downloads/#pointr-android>>
>> >> >
>> >> > iOS:
>> >> >
>> >>
>> https://apps.apple.com/us/app/pointr-easy-remote-support/id1073131903
>> <https://apps.apple.com/us/app/pointr-easy-remote-support/id1073131903>
>> >>
>> <https://apps.apple.com/us/app/pointr-easy-remote-support/id1073131903
>> <https://apps.apple.com/us/app/pointr-easy-remote-support/id1073131903>>
>> >> >
>> >> > UWP:
>> >> >
>> >>
>> https://apps.microsoft.com/store/detail/pointr-easy-remote-support/9NBLGGH52R41
>>
>> <https://apps.microsoft.com/store/detail/pointr-easy-remote-support/9NBLGGH52R41>
>> >>
>> >>
>> <https://apps.microsoft.com/store/detail/pointr-easy-remote-support/9NBLGGH52R41
>>
>> <https://apps.microsoft.com/store/detail/pointr-easy-remote-support/9NBLGGH52R41>>
>> >> >
>> >> > Windows:
>> >> > https://www.pointr.com/downloads/#pointr-windows
>> <https://www.pointr.com/downloads/#pointr-windows>
>> >> <https://www.pointr.com/downloads/#pointr-windows
>> <https://www.pointr.com/downloads/#pointr-windows>>
>> >> > -------------- next part --------------
>> >> > A non-text attachment was scrubbed...
>> >> > Name: Screenshot 2023-02-27 at 10.42.12.png
>> >> > Type: image/png
>> >> > Size: 146717 bytes
>> >> > Desc: not available
>> >> > URL:
>> >> >
>> >>
>> <https://lists.bufferbloat.net/pipermail/starlink/attachments/20230227/6b707fb8/attachment.png
>>
>> <https://lists.bufferbloat.net/pipermail/starlink/attachments/20230227/6b707fb8/attachment.png>
>> >>
>> >>
>> <https://lists.bufferbloat.net/pipermail/starlink/attachments/20230227/6b707fb8/attachment.png
>>
>> <https://lists.bufferbloat.net/pipermail/starlink/attachments/20230227/6b707fb8/attachment.png>>>
>> >> >
>> >> > ------------------------------
>> >> >
>> >> > Subject: Digest Footer
>> >> >
>> >> > _______________________________________________
>> >> > Starlink mailing list
>> >> > Starlink at lists.bufferbloat.net
>> >> > https://lists.bufferbloat.net/listinfo/starlink
>> <https://lists.bufferbloat.net/listinfo/starlink>
>> >> <https://lists.bufferbloat.net/listinfo/starlink
>> <https://lists.bufferbloat.net/listinfo/starlink>>
>> >> >
>> >> >
>> >> > ------------------------------
>> >> >
>> >> > End of Starlink Digest, Vol 23, Issue 26
>> >> > ****************************************
>> >> >
>> >> _______________________________________________
>> >> Starlink mailing list
>> >> Starlink at lists.bufferbloat.net
>> >> https://lists.bufferbloat.net/listinfo/starlink
>> <https://lists.bufferbloat.net/listinfo/starlink>
>> >> <https://lists.bufferbloat.net/listinfo/starlink
>> <https://lists.bufferbloat.net/listinfo/starlink>>
>> >
>> > --
>> > ****************************************************************
>> > Dr. Ulrich Speidel
>> >
>> > School of Computer Science
>> >
>> > Room 303S.594 (City Campus)
>> >
>> > The University of Auckland
>> > u.speidel at auckland.ac.nz
>> > http://www.cs.auckland.ac.nz/~ulrich/
>> <http://www.cs.auckland.ac.nz/~ulrich>
>> > ****************************************************************
>> >
>> >
>> >
>> _______________________________________________
>> Starlink mailing list
>> Starlink at lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/starlink
>> <https://lists.bufferbloat.net/listinfo/starlink>
>
> --
> ****************************************************************
> Dr. Ulrich Speidel
>
> School of Computer Science
>
> Room 303S.594 (City Campus)
>
> The University of Auckland
> u.speidel at auckland.ac.nz
> http://www.cs.auckland.ac.nz/~ulrich/
> ****************************************************************
>
>
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: starlink-test-site-approximation.jpg
Type: image/jpeg
Size: 146195 bytes
Desc: not available
URL: <https://lists.bufferbloat.net/pipermail/starlink/attachments/20230306/71f9c940/attachment-0002.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: starlink-test-site-approximation-detailed.jpg
Type: image/jpeg
Size: 305523 bytes
Desc: not available
URL: <https://lists.bufferbloat.net/pipermail/starlink/attachments/20230306/71f9c940/attachment-0003.jpg>


More information about the Starlink mailing list