From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from wout1-smtp.messagingengine.com (wout1-smtp.messagingengine.com [64.147.123.24]) (using TLSv1.2 with cipher ADH-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id C2B693B29D for ; Fri, 8 Dec 2023 06:54:43 -0500 (EST) Received: from compute3.internal (compute3.nyi.internal [10.202.2.43]) by mailout.west.internal (Postfix) with ESMTP id 8FD5C3200AA3; Fri, 8 Dec 2023 06:54:42 -0500 (EST) Received: from imap43 ([10.202.2.93]) by compute3.internal (MEProxy); Fri, 08 Dec 2023 06:54:42 -0500 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=honson.au; h=cc :cc:content-transfer-encoding:content-type:content-type:date :date:from:from:in-reply-to:in-reply-to:message-id:mime-version :references:reply-to:sender:subject:subject:to:to; s=fm1; t= 1702036482; x=1702122882; bh=cu4OouSBl75ohwiCBZM5xmyXnckU2ZS42eA MgwRffnU=; b=JU7ggTa85clJJKpYmtH+rWt1nEN8mzfk9Q6YUsSZAqt7gKpi5ZQ h6/xAh117sMqwb3Eyxk9bzdyXghxqR/krohjE2A995bbteIIJ6B4eVxW+p+Z9Z+j BspItgAbJ56ntp64kVRPHg8JKofG23rsNGeoFsi23HmbEqB9RBkoSDCtbTHQtM6U 0uLsAeeA8LWipDyZqb0YijFkTGcPans6Yi3yi22VI+nKVACfPSKrLW/E57bIKJ+Q /z+IFQkXHbSqAMXsAJodUEaxt5N6TZp9fL6qMhvIp/zjMkhD4ji/0TxXBp/rdBrq DebVPvAB+fEvtVeojC7/L9W9CaBk7maxlgQ== DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d= messagingengine.com; h=cc:cc:content-transfer-encoding :content-type:content-type:date:date:feedback-id:feedback-id :from:from:in-reply-to:in-reply-to:message-id:mime-version :references:reply-to:sender:subject:subject:to:to:x-me-proxy :x-me-proxy:x-me-sender:x-me-sender:x-sasl-enc; s=fm1; t= 1702036482; x=1702122882; bh=cu4OouSBl75ohwiCBZM5xmyXnckU2ZS42eA MgwRffnU=; b=ZaiIAkgauLAVg0vwNVxRXHZpoGWK77BqhsRKB/m5T92ZUQRK4tl 5RciFbYMRYieeR4LR+T73SxruWjIUXf63QjWSdBEp382KlV+z4axDtWutakxLCNn L2l5z0XfBRc0cHWHQEFmiiFvGs8ViHcisJV6JrYYuXnDkLm6cTAn6G36vuaq+fTG 5YQTH+XwHWyUNY60W62OV60SfzrRPMlqfKbFgd2pWDR3zbSNPSWGFu25VdyeysgH cfHbDhlrjA8049MGMraaW7FhmgaxXT/hR+RkKep3T9mfAwjhiyTLCwytqDTioY5R 6fxpRMvjCAvl8WHQNaIR3yTnzDMu3cLmBnQ== X-ME-Sender: X-ME-Proxy-Cause: gggruggvucftvghtrhhoucdtuddrgedvkedrudekiedgfeehucetufdoteggodetrfdotf fvucfrrhhofhhilhgvmecuhfgrshhtofgrihhlpdfqfgfvpdfurfetoffkrfgpnffqhgen uceurghilhhouhhtmecufedttdenucesvcftvggtihhpihgvnhhtshculddquddttddmne cujfgurhepofgfggfkjghffffhvfevufgtgfesthhqredtreerjeenucfhrhhomhepufht vghvvghnuceosghufhhfvghrsghlohgrthdqlhhishhtshesshhtvghvvghnrdhhohhnsh honhdrrghuqeenucggtffrrghtthgvrhhnpedvieffiefhteegieduveduteehleekgefh veduheffjeekgfefvedtkedujefhudenucffohhmrghinhepghhithhhuhgsrdgtohhmpd hsthgrrhhlihhnkhhishhprdhnvghtpdhrihhpvgdrnhgvthdpshhpvggvughtvghsthdr nhgvthdpuhhvihgtrdgtrgdpuggvrdgssgdpnhhtthdrnhgvthdpsghufhhfvghrsghloh grthdrnhgvthenucevlhhushhtvghrufhiiigvpedtnecurfgrrhgrmhepmhgrihhlfhhr ohhmpegsuhhffhgvrhgslhhorghtqdhlihhsthhssehsthgvvhgvnhdrhhhonhhsohhnrd gruh X-ME-Proxy: Feedback-ID: iaed446c2:Fastmail Received: by mailuser.nyi.internal (Postfix, from userid 501) id B5CEC2D4007D; Fri, 8 Dec 2023 06:54:41 -0500 (EST) X-Mailer: MessagingEngine.com Webmail Interface User-Agent: Cyrus-JMAP/3.9.0-alpha0-1178-geeaf0069a7-fm-20231114.001-geeaf0069 MIME-Version: 1.0 Message-Id: In-Reply-To: References: <361ee9d0d01da491a6e2b132317d170b@ausics.net> <75790520d6c8d35beb7248f5f5dff952@ausics.net> <924492fc-6ce4-4e62-82d9-c8f0b50ee96d@gmail.com> <8a807397-edef-4dcc-bfa6-f9c1598cbc77@gmail.com> <69da68cf-8746-45f8-9537-8fd8252de115@gmail.com> <7a20ea4f-4f16-4584-9f53-1ee3c72afd6f@app.fastmail.com> Date: Fri, 08 Dec 2023 22:54:21 +1100 From: Steven To: "Alexandre Petrescu" , "Freddie Cash" Cc: starlink@lists.bufferbloat.net Content-Type: text/plain;charset=utf-8 Content-Transfer-Encoding: quoted-printable Subject: Re: [Starlink] Info on IP country ranges X-BeenThere: starlink@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: "Starlink has bufferbloat. Bad." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 08 Dec 2023 11:54:43 -0000 Alexandre, I'm confident it is a Starlink provided /56 for a number of reasons. 1. I am obtaining the /56 using DHCPv6-PD. A random DHCPv6-PD server out= there on the Internet outside of Starlinks network would not be able to= receive nor reply to my DHCP requests unless Starlink were actively for= warding this to them. 2. I am not obtaining it using tunneling. I understand these concepts we= ll and am confident there isn't 6to4, a VPN, or any other mechanism at p= lay here. 3. The /56 is from a Starlink allocation, as verified by bgp.tools. 4. Numerous others have also obtained a /56 using DHCPv6-PD from Starlin= k. A quick internet search will yield many others. Regards, Steven On Fri, 8 Dec 2023, at 10:48 PM, Alexandre Petrescu wrote: > Steven, > > Thanks for the reply. > > Le 08/12/2023 =C3=A0 11:31, Steven a =C3=A9crit=C2=A0: >> Alex, >> >> You get that /56 from Starlink. >> >> I've only seen it done using a non-Starlink router (MiktoTik - see ht= tps://github.com/tyd/mikrotik-starlink-ipv6 for example config) in place= of the supplied one, with an appropriate PoE injector to power the dish. >> >> It is definitely a Starlink provided /56, and not obtained via a tunn= el or a third party. > > How do you know the /56 is provided by Starlink and not by somebody el= se? > > If it were to me: > > Quickly, I'd type 'what is my ipv6 address' in google and check whethe= r=20 > the first 40 bits of the reported address are equal with the prefix=20 > listed for my location at https://geoip.starlinkisp.net/feed.csv > > For example, if I were in Berlin I'd compare the first 40bits and 42bi= ts=20 > of the=C2=A0 address reported by google ('what is my...') with each of= thsese=20 > 40 and 42bits '2a0d:3344:500::/40,DE,DE-BE,Berlin,' ,=20 > '2a0d:3344:2400::/42,DE,DE-BE,Berlin,' and=20 > '2a0d:3344:1500::/40,DE,DE-BE,Berlin,' reported by that .csv file. > > The result of that comparison would be a hint (not a decisive answer)=20 > easy to obtain of whether the /56 is provided by starlink or not.=C2=A0=20 > Because the comparison is only on 40bits (or 42bits) and there could b= e=20 > a /41, or a /43, or larger, routed elsewhere than to starlink, and sti= ll=20 > match that /56. > > For more assurance, I would put wireshark on the Ethernet interface of=20 > that MikroTik router connected to Dishy; then check whether wireshark=20 > reports encapsulated packets, check the dst address of the outer packe= t,=20 > and find the owner of that address in the RIR database (RIPE.net in Eu= rope). > > For more hints, I'd check the user manual of MikroTik and IPv6. Some=20 > URLs about MikroTik and IPv6 tell 'tunnelling', tell 6to4, tell=20 > 'tunnelbroker' (HE - Hurricane Electric).=C2=A0 The 6to4 is outdated, = should=20 > not be used.=C2=A0 The HE is outside starlink. > > Of course, maybe I'd try first=C2=A0 whether the latency over IPv6 is = lower=20 > than with IPv4, or equal. (consulting www.speedtest.net can be done vi= a=20 > IPv4 or via IPv6).=C2=A0 This could tell whether IPv6 is more advantag= eous=20 > than IPv4 from a latency perspective, but there could also be dooubts=20 > about where and how are the speedtest servers placed, IPv4 vs IPv6. > > Alex > >> >> I haven't seen this tried using their included router, but would be i= nterested in hearing if DHCP-PD on the LAN segment of the included route= r yields a prefix delegation too. >> >> Cheers, >> Steven >> >> On Fri, 8 Dec 2023, at 9:22 PM, Alexandre Petrescu wrote: >>> Le 08/12/2023 =C3=A0 09:40, Steven a =C3=A9crit=C2=A0: >>>> You do indeed get a /56, so are able to assign unique /64s to each = of your networks. >>> Do you get that /56 from starlink or from somebody else? >>> >>> If you use a non-starlink router, it might be that the IPv6 feature = of >>> that non-starlink router gets that /56=C2=A0 from the outside of sta= rlink domain. >>> >>>> Your router obtains an address using SLAAC for your WAN interface f= rom a /64 (not sure if this /64 is unique to each customer or shared). >>> Ok, good to know. >>>> You can then request a /56 using DHCP-PD (separate to the /64 used = on the WAN interface). >>> Yes, it might be indeed that the router (provided by starlink router= , or >>> not by somebody else) runs a DHCPv6-PD server. >>> >>> Alex >>> >>>> Cheers, >>>> Steven >>>> >>>> On Fri, 8 Dec 2023, at 7:30 PM, Alexandre Petrescu via Starlink wro= te: >>>>> Le 08/12/2023 =C3=A0 06:57, Freddie Cash a =C3=A9crit=C2=A0: >>>>>> Dishy gets a /64 >>>>> IF Dishy gets a /64 from the starlink operator then I am afraid on= e cant >>>>> make subnets in home, because each other subnet needs a distinct /= 64. >>>>> >>>>> >>>>>> and I've tested DHCPv6 on both my Firewalla and my USG. They do p= refix >>>>>> delegation to distribute that as a /56 locally. >>>>> I am afraid it is not possible to make a /56 out of a /64 (the inv= erse >>>>> is true). >>>>> >>>>> Alex >>>>> >>>>>> No NAT required for IPv6 (incoming or outgoing) connections. And = there >>>>>> doesn't appear to be any restrictions on IPv6 traffic. >>>>>> >>>>>> This is with the round Dishy. >>>>>> >>>>>> Cheers, >>>>>> Freddie >>>>>> >>>>>> Typos due to smartphone keyboard. >>>>>> >>>>>> On Thu, Dec 7, 2023, 3:54 a.m. Alexandre Petrescu via Starlink >>>>>> wrote: >>>>>> >>>>>> >>>>>> Le 04/12/2023 =C3=A0 19:17, J Pan via Starlink a =C3=A9crit= =C2=A0: >>>>>> > yes, starlink does respond to its customers' complaints, = although >>>>>> > sometimes slowly. its ipv4 address acquisition is scatter= ed >>>>>> around as >>>>>> > a latecomer to the isp world, and as a global local isp, = it's more >>>>>> > troublesome. ip packets have to be tunneled back to its h= ome pop >>>>>> where >>>>>> > nat and other functions happen, sometimes around the worl= d, >>>>>> causing a >>>>>> > much higher minimum rtt fluctuation in 15-second handover >>>>>> > intervals---bad for network protocols and applications. i= pv6 can do >>>>>> > better but currently follows the same route as ipv4---an >>>>>> incentive to >>>>>> > promote ipv6 ;-) >>>>>> >>>>>> Excellent incentive! >>>>>> >>>>>> It would be good to know whether the dishy router obtains a= /56 or >>>>>> a /64 >>>>>> prefix from the starlink ISP.=C2=A0 That is easy to find ou= t by just >>>>>> looking >>>>>> at the packets.=C2=A0 This would tell whether a NAT can be = avoided at >>>>>> home, >>>>>> and hence more apps made possible. >>>>>> >>>>>> IT would also be good to=C2=A0 know whether the claimed IPv= 6 access on >>>>>> dishy >>>>>> is via a tunnel (IPv6 in IPv6, or IPv6 in IPv4) or it is 'n= ative' (no >>>>>> tunnel).=C2=A0 That will tell many things about additional = latency that >>>>>> might >>>>>> be brought in by IPv6.=C2=A0 (we'd want less latency, not m= ore). >>>>>> >>>>>> After that, one can look more at promoting IPv6.=C2=A0 Othe= rwise, IPv6 >>>>>> might >>>>>> still look as a hurdle, an obstacle, additional work that i= s too >>>>>> little >>>>>> necessary, or might even be worse than IPv4. >>>>>> >>>>>> Alex >>>>>> >>>>>> > -- >>>>>> > J Pan, UVic CSc, ECS566, 250-472-5796 (NO VM), Pan@UVic.C= A, >>>>>> Web.UVic.CA/~pan >>>>>> > >>>>>> > On Mon, Dec 4, 2023 at 4:04=E2=80=AFAM Noel Butler >>>>>> wrote: >>>>>> >> Thanks, it seems they are trying it on then :) >>>>>> >> >>>>>> >> On 04/12/2023 10:44, J Pan wrote: >>>>>> >> >>>>>> >> starlink advertises its customer ip address location at >>>>>> >> http://geoip.starlinkisp.net (not always updated but goo= d enough in >>>>>> >> most cases and traceroute can confirm to some extent as = well) >>>>>> >> -- >>>>>> >> J Pan, UVic CSc, ECS566, 250-472-5796 (NO VM), Pan@UVic.= CA, >>>>>> Web.UVic.CA/~pan >>>>>> >> >>>>>> >> On Sun, Dec 3, 2023 at 4:15=E2=80=AFPM Noel Butler via S= tarlink >>>>>> >> wrote: >>>>>> >> >>>>>> >> >>>>>> >> I run an open access usenet server, but only for those w= ithin >>>>>> my CC, so access is by IP based on our CC allocations from = APNIC. >>>>>> >> >>>>>> >> Because IPv4 exhaustion this changes sometimes with buyi= ng >>>>>> allocations from other regions, and if they get denied acce= ss I >>>>>> encourage them to let us know so we can keep ACL's updated,= I've >>>>>> had a request from a starlink user who claims they are here= , but >>>>>> traceroute shows them in .DE >>>>>> >> >>>>>> >> tracing some 217.foo.ad.dr >>>>>> >> >>>>>> >> ... >>>>>> >> 9 ae-6.r21.frnkge13.de.bb.gin.ntt.net >>>>>> (129.250.3.183) >>>>>> 290.223 ms 290.180 ms ae-1.r20.frnkge13.de.bb.gin.ntt.net >>>>>> (129.250.7.35)= 280.523 ms >>>>>> >> 10 ae-1.a03.frnkge13.de.bb.gin.ntt.net >>>>>> (129.250.3.152) >>>>>> 290.109 ms 289.667 ms 292.864 ms >>>>>> >> 11 ae-0.spacex.frnkge13.de.bb.gin.ntt.net >>>>>> (213.198.72= .19) >>>>>> 279.611 ms 278.840 ms 279.592 ms >>>>>> >> 12 undefined.hostname.localhost (206.224.65.189) 280.127= ms >>>>>> 278.506 ms 284.265 ms >>>>>> >> 13 undefined.hostname.localhost (206.224.65.209) 284.198= ms >>>>>> undefined.hostname.localhost (206.224.65.201) 274.663 ms 27= 3.073 ms >>>>>> >> 14 * * * >>>>>> >> >>>>>> >> >>>>>> >> As it is our policy to not collect any user info or issue >>>>>> user/pass's and=C2=A0 only allow access by IP, I'm hoping s= omeone here >>>>>> knows if they are full of it, or does starlink really assign >>>>>> addresses from anywhere? That one hardly makes sense for us= er >>>>>> experience, or maybe starlink has so few users in this coun= try >>>>>> they haven't bothered changing anything yet? >>>>>> >> >>>>>> >> -- >>>>>> >> >>>>>> >> Regards, >>>>>> >> Noel Butler >>>>>> >> >>>>>> >> >>>>>> >> _______________________________________________ >>>>>> >> Starlink mailing list >>>>>> >> Starlink@lists.bufferbloat.net >>>>>> >> https://lists.bufferbloat.net/listinfo/starlink >>>>>> >> >>>>>> >> >>>>>> >> -- >>>>>> >> >>>>>> >> Regards, >>>>>> >> Noel Butler >>>>>> >> >>>>>> >> >>>>>> > _______________________________________________ >>>>>> > Starlink mailing list >>>>>> > Starlink@lists.bufferbloat.net >>>>>> > https://lists.bufferbloat.net/listinfo/starlink >>>>>> _______________________________________________ >>>>>> Starlink mailing list >>>>>> Starlink@lists.bufferbloat.net >>>>>> https://lists.bufferbloat.net/listinfo/starlink >>>>>> >>>>> _______________________________________________ >>>>> Starlink mailing list >>>>> Starlink@lists.bufferbloat.net >>>>> https://lists.bufferbloat.net/listinfo/starlink