Hacker News

points by andix 15 hours ago | 0 comments

IPv6 got some real traction in the last decade. From around 10% adoption up to 50%. It's not dead, like many are claiming. It works fine, just keep implementing it.

[–] throw0101c 14 hours ago

> From around 10% adoption up to 50%.

Google sees 70-80% from some countries:

* https://www.google.com/intl/en/ipv6/statistics.html#tab=per-...

[–] andix 14 hours ago

Sure, around 50% is a worldwide estimate.

[–] _rwo 11 hours ago

ah yeah, golden times of crawling google when they actually didn't monitor ipv6 traffic properly against scraping (seo monitoring tools were first to drive 'adoption', lol)

[–] Sesse__ 5 hours ago

I don't know what you count as “golden times”, but at least when we started the measurements, there was no indication bot traffic would drive up these stats at all. (You wouldn't trigger it just by fetching a search result page.)

[–] akho 8 hours ago

I wonder how they measure.

[–] Sesse__ 5 hours ago

I don't know exactly how they _measure_ since I'm not working in IPv6 in Google anymore, but I know how we _measured_ when we started these stats:

https://research.google/pubs/evaluating-ipv6-adoption-in-the...

[–] usui 13 hours ago

I keep IPv4 off on a media computer all the time out of curiosity. IPv6-only is still unusable because only about 5% to 10% of top sites that I find during a Google Search work. It's almost guaranteed that a website will not load on all the top results.

So even if it's true 50% of clients support IPv6, only about 5% to 10% of website hosts do.

Sure a person can do a dual-stack setup or do some IPv4-IPv6 translation, but seriously IPv6-only should be plug-and-play by now.

[–] simoncion 8 hours ago

> ...IPv6-only should be plug-and-play by now.

The obsession about "IPv6 only" has never made sense to me. Any decent firewall software will let you write rules that apply regardless of IP version. The "software complexity" argument is bunk... you already have v4 and v6 IP implementations in your OS, and every major OS has provided IP-version-agnostic socket interface code for much more than ten years. If you are using software that is IPv6-capable, it's very likely that it's using IP version agnostic library functions.

Unless stewardship of the Internet is wrested from the hands of the responsible, level-headed type of folks that run it now and transferred to the hands of folks who think nothing of making backwards-incompatible changes every six months, IPv4 won't stop running across the Internet within the lifetime of anyone who's currently reading HN. This delay is fine. The only thing about IPv4 that's a problem is that its address space is 32 bits wide. Had it been defined with a 128+ bit address space, we wouldn't have seen another version of IP on the Internet. Now that IPv6 is around to reduce address pressure, there's no big need to do anything about IPv4.

[–] ebiederm 6 hours ago

I believe the folks worrying about IPv6 only are just looking ahead to a possible future. Especially when they are looking at the question of do all the pieces work locally, vs do popular domains support it.

There are domains like T-mobile US that are IPv6 only already. T-mobile runs NAT in the network so that all of IPv4 is NAT'd to IPv6.

I do think it is reasonable to expect there will become a time when IPv6 is pervasive enough, that new deployments will target users with IPv6 rather than IPv4.

Once IPv6 becomes the choice to support to get new customers, it is reasonable to expect that IPv4 support will start deteriorating and will eventually fade away.

I will be very surprised if we don't hit that tipping point in my lifetime. Say the next 30 years.

[–] simoncion 5 hours ago

> I believe the folks worrying about IPv6 only are just looking ahead to a possible future.

I'm certain that that is not what the folks who scream very loudly about how IPv6 has been and continues to be a colossal failure because it's not the only version of IP on the Internet are doing.

> I do think it is reasonable to expect there will become a time when IPv6 is pervasive enough, that new deployments will target users with IPv6 rather than IPv4.

Sure. That's the future I find most plausible: For "residential" connections, provide IPv6 GUAs and IPv4 via CGN... [0] for "commercial" connections, provide IPV6 GUAs and give the customer the option of one or more IPv4 GUAs.

> I will be very surprised if we don't hit that tipping point in my lifetime. Say the next 30 years.

I do hope we get to that point within the next five years, if not ten! But notice that my claim is that IPv4 will still be crossing the Internet after everyone reading HN today is dead. [1]

[0] I'm calling the various mechanisms that only provide an IPv6 address to an end user and do fancy gyrations to provide IPv4 access "CGN", even if might not strictly be CGN.

[1] ...as long as the the responsible, level-headed type of folks that run it now aren't replaced with the "making backwards-incompatible changes to everything every six months is A-OK!" type of folks.

[–] spauldo 3 hours ago

I can totally understand why. IP addresses are hard to get for a large chunk of the global population. There's twice as many people as there are IP addresses.

Plus dual-stack means more work, and more possibilities for misconfigurations or vulnerabilities in your network. It'd be way easier if you only had to worry about one or the other.

I live in the US and my ISP gives me one address I have to share across all the machines in my house. That's a tragedy - by all rights I should have a subnet or two. You can't directly address any of my servers from the outside on IPv4 which means I have to do hacky crap like port forwarding and can't use a lot of the older protocols that aren't NAT friendly. And I've got it good compared to most of the world!

IPv6 gives us back a large chunk of the original premise of the Internet - that all nodes are peers. IPv4 hasn't been able to do that for decades, and it's warped the way we think about networking by separating everyone into consumers and providers.

[–] simoncion 1 hour ago

> Plus dual-stack means more work, and more possibilities for misconfigurations or vulnerabilities in your network.

Nope. Your OS vendor already did the work to make both versions of IP work. And as I mentioned, if you're using a decent firewall you can write IP-version-agnostic rules just as easily as you can ones that depend on the version.

Plus, most SOHO firewalls default to dropping unsolicited non-ICMP traffic. That's the exact same security stance that IPv4 + NAT gives you. You have to go out of your way (or be using fancy gear that is aimed at people who know what they're doing) to expose your LAN hosts.

[–] spauldo 1 hour ago

My OS vendor didn't set up my firewall rules, lay out my subnets, or configure my routing. I did that. Twice. Because I've got a slightly more advanced network layout than your typical home user. And that's all much less complicated that what your typical corporate network would be, which is much less complicated than what an ISP or COLO provider has to deal with. And then there's the cloud...

Yeah, sure, if your requirements can be met by a cable modem, then you're basically good to go. Until DNS breaks for some machines and not others, and it's because your ISP screwed something in their IPv6 routing but it still works on IPv4. Or Microsoft's authentication servers decide they don't like your IPv6 subnet anymore and Minecraft refuses to start unless you turn IPv6 off. Or... You get the idea.

[–] cowlevel 10 hours ago

That's very surprising. Cloudflare supports ipv6 and it would be surprising if only 10% of websites you visit use Cloudflare.

[–] andix 9 hours ago

I guess a lot of sites just turn it off. Even if Cloudflare proxies it over IPv4 the IPv6 source address still shows up in server logs. Features like geolocating users or fraud prevention might need additional work for IPv6.

[–] andix 12 hours ago

Cloud providers and hosters seem to have the least interest in implementing IPv6. Sitting on a huge supply of IPv4 addresses gives them a competitive advantage. So they try to disable IPv6 as a default.

[–] idatum 12 hours ago

And Windows 11 now supports CLAT.

[–] commandersaki 13 hours ago

But it's not really useful until it has critical mass at say 80 or 90%.

[–] cyberax 10 hours ago

IPv6 is dead. Nobody cares about it.

I will believe otherwise once even one US hotel chain supports it. I have _never_ seen a hotel in the US with IPv6 WiFi. Not once.

And I once stayed in a hotel (in Mountain View, CA) that gave out public IPv4 addresses.

[–] Sesse__ 5 hours ago

Last time I was in a California hotel, it had IPv6 just fine. That was 2012. Perhaps you choose the wrong hotels :-)

[–] throw0101a 3 hours ago

> IPv6 is dead. Nobody cares about it.

Except for everyone on T-Mobile US:

* https://www.youtube.com/watch?v=d6oBCYHzrTA

Or all of FaceMeta's internal infrastructure:

* https://engineering.fb.com/2017/01/17/production-engineering...

Or 86% of France:

* https://www.google.com/intl/en/ipv6/statistics.html#tab=per-...

[–] andix 9 hours ago

Okay, if you say so. Hotel Wifis are known for being on the bleeding edge of technology, and as the main innovator pushing new internet technology. Everything starts with hotel Wifis. /s

[–] cyberax 8 hours ago

Exactly my point. IPv6 failed to reach the "boring, just works" level after 30 years of deployment. So much so that we are now deploying solutions with 10+ years of expected service lifetime that do not support IPv6.

Meanwhile, CGNAT really solved most issues with IPv4. Yes, even the voice/video chat issues.

Want another fun example? Try to find a SIP telephony provider in the US that supports IPv6. Go on. I do this exercise about once a year.

[–] simoncion 7 hours ago

> IPv6 failed to reach the "boring, just works" level after 30 years of deployment.

/me glances over at my phone, which only has an IPv6 address when connected via cellular modem.

/me thinks back to the decade+ of incredibly boring IPv6 service from the residential ISPs I've used.

Sure.

If an ISP wants to provide only IPv6 addresses to CPE, they can set things up so that happens. But if you won't be satisfied until IPv4 no longer passes over the Internet, then you're going to die unhappy about that. The goal for our lifetime is one where every residential site has a global /56 and is behind IPv4 CGN, and servers hosted by commercial entities get to choose whether they assign IPv4 addresses, IPv6 addresses, or both to their servers.

[–] TobTobXX 7 hours ago

/me searching through all the mobile network providers in Switzerland (a technologically advanced country) and not finding any.

Seriously, this drives me mad. There is NO WAY to get IPv6 on mobile networks in Switzerland. Neither Swisscom, nor Sunrise, nor Salt, and of course no reseller.

And Android handles this particularly poorly. I have some servers set up in my tailscale network on private IPv6 addresses, but android doesn't query AAAA addresses if it has no ::/0 route, so I HAVE to route all my traffic through my VPN which often creates other complications (with corporate wifi etc.)

[–] andix 7 hours ago

I get your frustration. But not having access to IPv6 personally does not mean IPv6 is dead. It's very much alive and handles considerable amounts of traffic.

[–] TobTobXX 6 hours ago

Well ok, not dead. Dead sounds final and I still think that we will get more IPv6 deployed in the future. And I set up all my servers as IPv6-first. So no, I don't think IPv6 is dead.

But I'd claim it failed completely. Many very public services (eg. github.com, sentry.io, office365.com, archive.org) outright don't support it. And the fact that infrastructure providers (like mobile network carriers) don't support it either, proves that it is far from "alive". In fact, I'd say it is deader than MMS, Fax and landline. For these I can still get service. For IPv6 (on the phone) I can't.

[–] simoncion 6 hours ago

To be clear, 'TobTobXX' isn't 'cyberax'. 'cyberax' is claiming that IPv6 is dead, whereas 'TobTobXX' is -quite justifiably- complaining that both his local cellular ISPs and Android suck.

[–] cyberax 7 hours ago

You're welcome to try to solve the challenges from my sibling post. Both of them are solved in IPv4 by a $100 router from a nearby store and with zero configuration.

[–] simoncion 6 hours ago

> You're welcome to try to solve the challenges from my sibling post.

Check my reply to your post. [0] I wrote it up before I noticed your message here. ;)

[0] <https://news.ycombinator.com/item?id=50295293>

[–] Dagger2 1 hour ago

https://doc.didww.com/voice/outbound-trunks/outbound-sip-inf...?

[–] andix 8 hours ago

I don't get your point. IPv6 is at "boring, just works". The majority of the worlds devices run on it dual-stack and it handles a considerable amount of traffic. A lot of 5G networks run on IPv6 single stack with IPv4 compatibility on top.

IPv6 is very much alive. And it's so boring and working so well, that you don't even notice it.

[–] cyberax 8 hours ago

> I don't get your point. IPv6 is at "boring, just works". The majority of the worlds devices run on it dual-stack and it handles a considerable amount of traffic.

Most of the world runs on IPv4. If we shut down IPv6 on the core routers tomorrow, pretty much nobody is going to blink an eye. And now I believe that we should do it and start again.

The 50% stat is really mostly mobile phones on cellular networks, which remain the _only_ major success story for IPv6. Mostly because there's not enough private or even squattable IPv4 space available for many large mobile phone operators.

> IPv6 is very much alive. And it's so boring and working so well, that you don't even notice it.

Sigh. No, it's not. In a thread a couple of weeks ago, I gave several examples:

1. You're a SOHO-type business. How can you do multiple WAN uplinks with IPv6 without NAT?

2. How do you prevent your printer on IPv6 from breaking when your Internet connection goes down?

For context: I deployed IPv6 in my networks in 2006. I have war stories.

[–] simoncion 6 hours ago

> How can you do multiple WAN uplinks with IPv6 without NAT?

Advertise the addresses offered by the primary uplink, then retire those addresses [0] and advertise the secondary when the primary fails. Or are you talking about load balancing, which is its own thing? Either way, this situation is inherently not boring, so the way you solve it is going to be quite a bit more complex -regardless of IP protocol version- than the very boring single-upstream situation.

> How do you prevent your printer on IPv6 from breaking when your Internet connection goes down?

I'll flip this around on you:

How do you prevent your printer from "breaking" when its ISP-provided DHCP lease on its globally-routable IPv4 address expires? How about when the ISP refuses to renew the lease on the printer's current globally-routable IPv4 address and forces it to get a new one?

Obviously, you use the ISP-assigned globally-routable IPv4 address and also assign an RFC 1918 address to the printer.

How to solve this with IPv6? Use ULAs! They're the equivalent of 1918 space in the IPv6 world. Just like in IPv4, IPv6 globally-routable addresses are for global communication and can be used for LAN-local communication. And -just like 1918 addresses- ULAs are for LAN-local communication. Just like with IPv4, if you can't control the addresses that you get from your upstream provider, and/or if your router is set up to retire your upstream provider's delegated address space when your upstream link goes down, then you use ULAs. [1]

I know that OpenWRT has a setting in LuCi to enable advertisement of a ULA prefix, and fifteen seconds of searching revealed that at least some Linksys combination router/APs seem to just generate and advertise a ULA prefix by default. [2]

So, if you're not configuring your LAN this way and upstream provider renumbering is getting in your way, then you really should bother to.

[0] ...by advertising their valid lifetime as zero seconds...

[1] ...or if you happen to like to be able to assign fixed IP addresses to certain pieces of hardware...

[2] <https://support.linksys.com/kb/article/3744-en/>

[–] andix 6 hours ago

> Use ULAs

Isn't that the default for modern routers? They just generate a random ULA during setup and assign it.

Random ULAs are also great for LAN-to-LAN VPNs, because there aren't any private IP overlaps. Everyone setting up their LAN on 192.168.0.x makes it quite a mess on IPv4.

[–] cyberax 6 hours ago

OK, do you _think_ my questions were that easy?

> Advertise the addresses offered by the primary uplink, then retire those addresses

Prefix withdrawal doesn't work reliably on Android. And neither is it required to work, btw. Neither does this solution allow for load balancing, as all traffic will just go through one connection.

Bonus points: triangular routing to your local NAS through the router.

> How do you prevent your printer from "breaking" when its ISP-provided DHCP lease on its globally-routable IPv4 address expires? How about when the ISP refuses to renew the lease on the printer's current globally-routable IPv4 address and forces it to get a new one?

Printers get an RFC1918 address and don't care about global connectivity.

> How to solve this with IPv6? Use ULAs! They're the equivalent of 1918 space in the IPv6 world.

They're not. IPv6 is nothing like IPv4 operationally. Back to my challenge, ULA will NOT work.

The scenario is very simple: the printer gets a ULA prefix and a GUA prefix. Your computer also gets a GUA and a ULA from the same router.

So far so good.

Now the uplink goes down, and the router withdraws the prefix. The printer sees the withdrawal and removes its GUA address. Your computer misses it, and still has the GUA and the ULA addresses.

Whoops. The printing is now broken somehow. Why?

[–] kstrauser 4 hours ago

> The 50% stat is really mostly mobile phones on cellular networks, which remain the _only_ major success story for IPv6.

The largest ISPs in the US are Comcast, AT&T, Charter, Verizon, T-Mobile, Starlink, and Google Fiber. They all support IPv6. The only top-10 US provide that didn't support it is Frontier Communications, who is rolling it out.

In other words, the biggest companies who actually do this for a living all demonstrate that you're wrong.

As to your questions, read up on IPv6 ULA which solves all of them. None of the hosts on my LAN depend on having an outbound connection.

[–] andix 6 hours ago

> For context: I deployed IPv6 in my networks in 2006. I have war stories.

Early IPv6 was a mess. It took a surprising amount of time for IPv6 to become mature and usable.

[–] cyberax 4 hours ago

I'd argue that it's still not mature. ULA/GUA mess is one example, I recommend reading this delightful thread with beer and popcorn: https://mailarchive.ietf.org/arch/msg/v6ops/7-SuigPyP0R-C6ZC...

Having to waste 64 bits for the local part of the address is another one. There's an RFC in progress to allow neighbor discovery with shorter prefixes, so we might be able to use it by 2045.

And Android _still_ doesn't support DHCPv6.