The smallest IPv6 prefix for advertising on the Internet via BGP is a /48, isn't it?
On Fri, Jun 3, 2016 at 10:11 PM, Cryptographrix <cryptograph...@gmail.com> wrote: > Nope - You'd have the /56 and only people within your /56 (or /64 if you > sliced it up nicely) would be able to do things with it routed by your ISP. > > Of course this means we'll have to get our ISPs to listen for our BGP > advertisement... > > > On Fri, Jun 3, 2016 at 10:09 PM Mansoor Nathani <mnathani.li...@gmail.com> > wrote: > >> Wouldn't the /56 get blocked as soon as Netflix detects multiple accounts >> logging in from the same IPv6 range? >> >> On Fri, Jun 3, 2016 at 9:49 PM, Cryptographrix <cryptograph...@gmail.com> >> wrote: >> >>> This is a good idea. We should do this. >>> >>> >>> >>> On Fri, Jun 3, 2016 at 9:48 PM Raymond Beaudoin < >>> raymond.beaud...@icarustech.com> wrote: >>> >>> > Make it a /56 each and you've got a deal. Hell, I'll throw in a round >>> of >>> > drinks. >>> > >>> > On Fri, Jun 3, 2016 at 8:40 PM, Cryptographrix < >>> cryptograph...@gmail.com> >>> > wrote: >>> > >>> >> We should crowdsource a /40 and split it up into /64's for each of us. >>> >> >>> >> >>> >> On Fri, Jun 3, 2016 at 9:38 PM Matthew Kaufman <matt...@matthew.at> >>> >> wrote: >>> >> >>> >> > If early adopter PI IPv6 was the same price as early adopter PI v4 >>> >> space, >>> >> > my wife would be totally on board with this solution. >>> >> > >>> >> > Matthew Kaufman >>> >> > >>> >> > (Sent from my iPhone) >>> >> > >>> >> > > On Jun 3, 2016, at 6:27 PM, Spencer Ryan <sr...@arbor.net> wrote: >>> >> > > >>> >> > > Well if you have PI space just use HE's BGP tunnel offerings. >>> >> > > >>> >> > > >>> >> > > *Spencer Ryan* | Senior Systems Administrator | sr...@arbor.net >>> >> > > *Arbor Networks* >>> >> > > +1.734.794.5033 (d) | +1.734.846.2053 (m) >>> >> > > www.arbornetworks.com >>> >> > > >>> >> > > On Fri, Jun 3, 2016 at 9:24 PM, Raymond Beaudoin < >>> >> > > raymond.beaud...@icarustech.com> wrote: >>> >> > > >>> >> > >> As an alternative, there are multiple cloud service offerings >>> that >>> >> will >>> >> > >> advertise your IPv6 allocations on your behalf direct to a >>> server in >>> >> > their >>> >> > >> data centers. It seems pretty tongue-in-cheek, and satisfying, to >>> >> turn >>> >> > >> up a *<insert >>> >> > >> favorite virtual router instance> *and then route through it. The >>> >> > Internet >>> >> > >> is such an amazing place. >>> >> > >> >>> >> > >> On Fri, Jun 3, 2016 at 8:15 PM, Cryptographrix < >>> >> > cryptograph...@gmail.com> >>> >> > >> wrote: >>> >> > >> >>> >> > >>> Yeah I RAWRed to them pretty hard whilst being as understanding >>> to >>> >> the >>> >> > CS >>> >> > >>> rep that it wasn't their fault. >>> >> > >>> >>> >> > >>> They thought I was weird as anything. >>> >> > >>> >>> >> > >>> If there are any Verizon FiOS network engineers on the thread, a >>> >> fellow >>> >> > >>> Verizon employee would thank you kindly for an off-thread email >>> >> > regarding >>> >> > >>> BGP advertisement (I'll buy the IPv6 block and the >>> drink-of-choice, >>> >> you >>> >> > >>> configure my account to listen for route advertisement). >>> >> > >>> >>> >> > >>> Strange that it has to come to this to get "legit" IPv6 service. >>> >> > >>> >>> >> > >>> >>> >> > >>> >>> >> > >>> >>> >> > >>> On Fri, Jun 3, 2016 at 9:08 PM Raymond Beaudoin < >>> >> > >>> raymond.beaud...@icarustech.com> wrote: >>> >> > >>> >>> >> > >>>> I wasn't originally affected on my he.net tunnel, but this >>> >> evening it >>> >> > >>>> started blocking. The recommended ACLs are a functional >>> temporary >>> >> > >>>> workaround, but I've also opened a request with Netflix. >>> >> > >>>> >>> >> > >>>> On Fri, Jun 3, 2016 at 7:54 PM, Mark T. Ganzer < >>> >> > gan...@spawar.navy.mil> >>> >> > >>>> wrote: >>> >> > >>>> >>> >> > >>>>> So far I am not seeing a Netflix block on my he.net tunnel >>> yet. I >>> >> > >>>> connect >>> >> > >>>>> to the Los Angeles node, so maybe not all of HE's address >>> space is >>> >> > >> being >>> >> > >>>>> blocked. >>> >> > >>>>> >>> >> > >>>>> Not going to be disabling IPv6 here either. + HAD native IPv6 >>> from >>> >> > >> Time >>> >> > >>>>> Warner, but they decided to in their wisdom to disable IPv6 >>> >> service >>> >> > >> for >>> >> > >>>>> anyone that has an Arris SB6183 due to an Arris firmware >>> bug. And >>> >> > >> they >>> >> > >>>> are >>> >> > >>>>> taking their sweet time pushing out the fixed firmware update >>> that >>> >> > >>>> Comcast >>> >> > >>>>> and Cox seemed to be able to push to their customers last >>> fall. >>> >> > >>>>> >>> >> > >>>>> -Mark Ganzer >>> >> > >>>>> >>> >> > >>>>> >>> >> > >>>>>> On 6/3/2016 4:49 PM, Cryptographrix wrote: >>> >> > >>>>>> >>> >> > >>>>>> Depends - how many US users have native IPv6 through their >>> ISPs? >>> >> > >>>>>> >>> >> > >>>>>> If I remember correctly (I can't find the source at the >>> moment), >>> >> > >> HE.net >>> >> > >>>>>> represents something like 70% of IPv6 traffic in the US. >>> >> > >>>>>> >>> >> > >>>>>> And yeah, not doing that - actually in the middle of an IPv6 >>> >> project >>> >> > >> at >>> >> > >>>>>> work at the moment that's a bit important to me. >>> >> > >>>>>> >>> >> > >>>>>> >>> >> > >>>>>> >>> >> > >>>>>> >>> >> > >>>>>> On Fri, Jun 3, 2016 at 7:45 PM Baldur Norddahl < >>> >> > >>>> baldur.nordd...@gmail.com >>> >> > >>>>>> wrote: >>> >> > >>>>>> >>> >> > >>>>>> Den 4. jun. 2016 01.26 skrev "Cryptographrix" < >>> >> > >>>> cryptograph...@gmail.com>: >>> >> > >>>>>>> >>> >> > >>>>>>>> The information I'm getting from Netflix support now is >>> >> explicitly >>> >> > >>>>>>> telling >>> >> > >>>>>>> >>> >> > >>>>>>>> me to turn off IPv6 - someone might want to stop them >>> before >>> >> they >>> >> > >>>>>>>> completely kill US IPv6 adoption. >>> >> > >>>>>>> Not allowing he.net tunnels is not killing ipv6. You just >>> need >>> >> > need >>> >> > >>>>>>> native >>> >> > >>>>>>> ipv6. >>> >> > >>>>>>> >>> >> > >>>>>>> On the other hand it would be nice if Netflix would try the >>> >> other >>> >> > >>>>>>> protocol >>> >> > >>>>>>> before blocking. >>> >> > >> >>> >> > >>> >> > >>> >> >>> > >>> > >>> >> >>
