In Seattle and surrounding areas, Comcast is more or less the only game in town (with Centurylink a distant, distant second) if you're not in one of a dozen high-rise apartment buildings. For all their marketing around 'extensive IPv6 rollout', I've found their deployment to be hideously broken and their support staff deeply ignorant. Been here 3 years, moved once, and it's just terrible. My latest foray into trying to get delegation to work was this past weekend, and I had a Comcast supervisor (in Business class!) tell me that "static anything is not a priority for the vast majority of their customers. You can either have static v4 addresses or v6 connectivity, but not both" after several hours of troubleshooting and 'automatically closed/resolved tickets' with their level 2 support folks. It's one of the more brazen "we dare you to stop being our customer" moves I've certainly been privvy to.
tl; dr - Comcast v6 works great on Business-class until you push a static v4 config, then it just stops responding to DHCPv6 and SLAAC requests. Ignorance and aggravation ensued from Comcast support, at all levels I was able to get ahold of. Being a monopoly is awesome.
In Seattle proper, almost 2/3rds of the city is covered by Centurylink with Gigabit fiber[1], but due to the way their cable franchise agreement is worded with the city, Centurylink has to wire the poorest neighborhoods first, and leave the richer areas as some of the last to get fiber from them.
Outside the Seattle area, upgrades happened much earlier so its a mix of VDSL2 down in Kent and surrounding areas, and ADSL to the north of Seattle proper (they got no upgrade munnies :c). Seattle has seen a ton of Capex over the past few years, and from what Clink employees tell me they are severely cutting back in Seattle this quarter, so as to wire the South end with VDSL2 and other cities with fiber. Portland also has near 100% fiber coverage by the way.
The crux of why Centurylink doesn't have a ton of customers on fiber comes down to a godawful prequal database, that they take offline starting anywhere from 3pm to 7pm and keep offline till well into the next morning. Techs will actually get orders put in for DSL, and when they go out to install it, see fiber on the pole and go "WTF?", but getting the prequal database amendment is near impossible.
Wrt static IPs on Comcast or worse yet Wave, just don't bother. A dynamic DNS entry will work just as well, while avoiding all the crazy stand on head networking antics both cable providers have to do to give you a static IP. Try portscanning the internal netblock your static IP is provisioned to, last time I did I found a (poorly protected) headend on a 10.x.x.x address.
In Seattle there is also CondoInternet - a fiber company in many of the newer development. But, I've seen an entire building on a NAT private network. It's included in your HOA.
Condointernet got acquired by Wave (branded Wave G); then they fired most of the condointernet staff it seems. Wave recently got acquired themselves so we'll see how much worse it gets.
FYI they are owned by Wave Broadband now, and Centurylink covers quite a bit of the city with fiber. Not that you'll know, cause their fiber prequal database is shut off in the afternoons/evenings/early morning cause screw you :(
My comment above gives a bunch more detail if you wanna give it a look.
I think of IPv6 like driving a stick shift car that you skip a gear with. If you're lucky, the engine won't stall but even if it doesn't, you'll be accelerating very slowly until the RPMs catch back up to where they should be. In IPv6's case they skipped 3 gears because they knew the car couldn't possibly stall, but here we are, barely accelerating after all this time.
Hm, this analogy is awkward. Definitely agree that it's taken a lot longer to become standard than I would like.
I was discussing this with a coworker today, and we were reflecting on a similar technical rollout that really did take of. The various ascii encodings -> UTF8. This was a big shift at the time and took a lot of time to fix. In fact I know of one large DB that, after a huge amount of outreach with customers, was only finally decommed and replaced with a UTF8 last year.
We decided the big difference was that UTF8 solved a huge problem that effected everyone. Was backward compatible with all basic ascii, and was an easy upgrade in many cases (only a problem if you bastardized stored character sets faking out the system and storing them in a different charset).
IPv6 is nicely backward compatible with IPv4, but in general it is not solving a problem most people have (yet). Most sites work fine with IPv4; IPv6 for many is just work with no significant benefit in general.
That being said, I really want IPv6 to become the only option for a lot of reasons; but there is no stron forcing function.
> but in general it is not solving a problem most people have (yet).
I don't think that that's actually true. People have massive problems due to NAT and overlapping address ranges and stuff ... but it's just commonly accepted as the way IP works, people don't realize that they could just deploy IPv6 and do away with all those problems.
Also, most mobile carriers have been solving real device addressing problems with IPv6 for a few years now. In the mobile space IPv6 is the present, and the transition has been mostly seamless enough most people haven't noticed outside some early hiccups.
Some people sure. But it's nowhere near the number of people that UTF8 solved for.
It would definitely make people's network connectivity simpler and less error prone, but I just don't see it as a huge pressing need.
If NATs didn't exist at all, then this would have been such a huge issue that it would have been needed. For most though, NAT is generally good enough.
I guess for most simple home users, it's currently not a huge pressing issue, true. But anything beyond that and you constantly run into issues. And that includes home users who also have to use some VPN to their workplace ...
But that does not mean that NAT really is good enough, it just means that deployed systems nowadays just take it as a given that NAT exists, and any technology that isn't compatible with NAT simply doesn't exist. Which makes it less of a pressing issue in a way, but that does not mean that it doesn't still cause huge costs even to home users in terms of missed opportunities of a NAT-free world.
I see that. And in general I agree with you. I think the only thing being debated here is the urgency of the change, and the harm being done to networks.
Many companies get away with NAT just fine; and that's the entire point. It works well enough that for many it's just not an urgent issue.
Like you said, it would make a lot of technology easier to deploy and build. The orthogonal comment about the Cellphone industry pushing the issue is right on the money. For them this significantly simplies their network management (which has tons of devices moving around). For laptops on wifi this would probably be better as well, but there so much is just solved by the fact that the Web is capable of tracking users across IPs with cookies.
Again I'm not arguing against IPv6; I'm just trying to better understand why it's deployment isn't being done as urgently as that of other things, like UTF8. I think the answer is in the fact that we've built so many workarounds that it stretched IPv4 well beyond its end-of-life.
> I think the answer is in the fact that we've built so many workarounds that it stretched IPv4 well beyond its end-of-life.
Well, that is true in a way (I mean, without NAT (or something similar), there obviously would not be any way to keep going, so, yeah, in that sense, NAT has made things somewhat bearable instead of completely unworkable, thus making the migration to v6 less urgent).
But my point is that the reason why people (companies in particular) aren't migrating to a large extent seems to me to not be because it wouldn't be worth it for them, but rather that they lack the understanding to see that it would be. There are lots and lots of admins out there who operate IP networks and essentially have no clue of IP routing. They have grown up in a world of NAT, and just understand "the router" as "the public internet termination point" or whatever you want to call it. They don't even see NAT as a workaround, but as the obvious and natural state of affairs, because, what are you gonna do if you want to connect more than one machine to your internet connection? You need a router! And router is synonymous with NAT gateway, because that's why you need the router ... or something.
If your whole mindset doesn't even allow you to see the possibility of the natural state of the internet (i.e. end-to-end addressability of all participants), you won't ever notice all the workarounds that you are using. And if, say, port forwarding doesn't occur to you as being a workaround, but rather the obvious thing you just need to do to make some internal machine reachable from the outside, then you also never get the idea that IPv6 might be the solution. You just assume that IPv6 obviously also has to have port forwarding, because you still want to make internal machines reachable from the outside, don't you?
I'm with a regional ISP in the UK (aquiss.net) and they assign you a native /56 if you ask nicely. What surprised me was how complicated it still is to get this set up compared to the DHCP and NAT setup we're all familiar with. It took me quite a while to figure out how to distribute addresses in my network and even get my pppd configured to request the prefix.
It seems to me that there are disadvantages to widespread IPv6 adoption for some big companies. For example, storage providers like Dropbox or Google Drive would take a hit if I were able to send files via a direct connection to a friend's computer, which is possible with IPv6. Right now, the best way to send large files is to upload them somewhere and then send a link (dynamic DNS is also possible, but certainly not as easy to set up as simply being granted a globally routable IP address). Another example is for ISPs or anyone who sells static IP addresses (e.g. AWS); IPv6 means severely decreasing the scarsity of a resource that they own and profit from, which I'm sure they wouldn't like.
To generalize, I think IPv6 represents more freedom for end-users in the form of better p2p services, but I think it also represents a loss in profit for some large companies. These companies play a roll in the IPv6 rollout, so maybe it's not surprising that the rollout has been slower than expected.
There are plenty of methods for direct transfer already, I don't think a different IP (with its likely non-trivial setup) would make much of a difference for Dropbox.
Keep in mind that would you describe would require both computers to be on at the same time, and that often doesn't happen either.
Dropbox would be here to stay even in an IPv6 world.
Well they have already ran out and the big old companies have their own reserved ipv4 ranges. They even buy other tech dinosaurs for their reserved ipv4 blocks in the order of $10 per adress. [1]
Well technically they don't charge a different price if you have a public ip or only a private IP on an instance. However even if they factored in that some high % of all EC2 instances will have a public IP and that's "part of the cost" what I'm really asking here is if eventually that cost will be so significant that they actually need to raise instance prices solely because of the IP price.
If that's the case, then it may make sense to lock the instance price down, and start charging for in-use v4 IPs, not just unused allocated ones.
Let's say the next block of 1 Million IPs that AWS wants to buy from say, GE fails to go through, and AWS is looking at a pool of 500K IPs starting to shrink at the rate of 30k per day. The only way to ensure they don't run out would be to separate the price out and jack up the price considerably so they don't run out.
What I'm saying is that at some point, WHEN the number of IPs that people want outpaces the number AWS (or Google or other cloud provider) can obtain from other companies, they will be forced to change something.
What would be interesting would be to see AWS etc, to try to get their customers to migrate off IPv4 for administrative and back-office services since there really isn't a strong reason to have VPNs and SSH or RDP sessions backed by a v4 address.
Has the price per elastic IP ever gone up though on AWS? I don't think so, but you'd expect price to rise over time as less addresses become available.
In San Francisco of all places, neither my home nor my office have IPv6. I wonder at what point the FCC or a similar regulating entity will require IPv6 rollouts (do they even have the power to do that?).
I have Webpass (now owned by Google Fiber) at home and Sonic.net at work.
Googling shows that they have some deployments but at least as far as I can tell from my network settings (and http://ipv6-test.com) I do not have an IPv6 address. Perhaps they never got around to updating my apartment building.
My ISP (Spectrum) only enables IPv6 to a subset of their "supported" cable modem list. In other words, if you don't have the right cable modem, you may get IPv4 but don't get IPv6.
Charter includes the modem with their plans though, so why would you not just ask for an IPv6 approved modem? Its not as though they charge an equipment rental fee like Comcast or Cox...
A gigabit switch is a heck of a lot cheaper than a 48 port router or a basic Linux box with proper VLANs and tagging on the switch. These were also likely configured in the early days of Webpass, so they prolly won't touch it unless it breaks.
IPv6 is not a requirement in all situations. At most enterprises need IPv6 at their edge to avoid traffic flowing through carrier grade nats. Internally private IPv4, plus nats for egress are generally fine for most use cases.
Internal IPv6 is much cleaner though, and less prone to failures (e.g. anyone VPNing in from a coffee shop that uses the same private IPv4 range will have a bad time). If you've already gone to the effort of getting IPv6 to the edge, surely at that point it's worth pushing it all the way through.
Not just that, also in the case when a big company acquires/merges with another big company. The suggestion to have loads of servers on a private range such as e.g. 10.x.x.x is nice as it should be unique.. until this assumption breaks due to combining two networks together. It'll result in conflicts, discovery that lots of places had the IP address instead of a hostname, etc. It seems much saner to use IPv6.
VPN from the same subnet simply doesn't work at all on Macs built-in VPN client, last time I looked into it. Windows built-in VPN client hasn't had a problem dealing with this for a long time.
I started to read the article but found the font too distracting. tl;dr I guess.
I see that my devices get IPV6 and IPV4 addresses. Beyond that I don't know "if it works." From outside my firewall (pfSense) I wouldn't know how to access resources within (which are walled off in the first place.)
If you want to see if things on your network are reachable from outside, you'll need something else with IPv6 connectivity. In my own case, a T-Mobile handset with LTE works nicely; if I open up a port to an IPv6 host on my LAN, I can reach it from the phone.
My home IPv6 connection is via a /60 provided by Comcast, but I'm currently only using a single /64 out of that pool.
Don't forget you can always use IPv6 as there are gates to 6th internet over 4th internet.
After all the words Inter Net mean this technology can work over and between any network, be it Ethernet, Wi-Fi, Token Ring or whatever. And of course it can work over IP of another version.
One particular mechanism is 6to4: https://en.wikipedia.org/wiki/6to4. It works through anycast address 192.88.99.1, so you don't need any provider-specific setup or subscription.
What's the best way to set up and use such an ipv6 gateway? After a bit of fiddling, I managed to set something up with sshuttle, but it was a bit of a hack.
Some feedback: we have been successfully using HE for years at our office, because of a lousy ISP that keeps on piling up ridiculous excuses. Latest one was that we couldn't get IPv6 because we had a "special" IPv4 block.
Also used SixXS but the aiccu+AYIYA/6in4 requirement isn't ideal when you have some network gear that supports 6to4 tunnelling natively.
tl; dr - Comcast v6 works great on Business-class until you push a static v4 config, then it just stops responding to DHCPv6 and SLAAC requests. Ignorance and aggravation ensued from Comcast support, at all levels I was able to get ahold of. Being a monopoly is awesome.