Why No IPv6?
whynoipv6.com
whynoipv6.com
If a technology is useful, you don't have to shame people into using it. At some point we will have to admit we made a big mistake.
I'm all IPv6 and i got my first network segments that are IPv6 only. I got nothing to fear. And no NAT.
The internet as consumers know it becoming only a tool to browse websites and not an actual internet connection that can do say, peer-to-peer gaming, sucks.
Everything the "alarmists" told us has already happened. IPv4 address space has already run out which now means that you often have to pay an ever increasing fee to use a public IPv4 address, if you get one at all (more and more people are sitting behing a CG-NAT). Public IPv6 blocks are completely free.
What's happened is that everybody has been forced behind NAT, completely obliterating pretty much the whole point of Internet to begin with, and that addresses that could be essentially free are super expensive. But la la la nothing has happeeeeeeeened!
Overexaggerating only makes your point come across as weak, and in my case, made me think it was all bullshit for money.
Just be honest.
And exactly who do you think is profiting?
In either case, it has little to do with IPv6, so I'll leave it there.
The functionality of the Internet is severely degraded by NAT, and has been for all of those "at least 15 years". The network can't deliver its most basic function, which is to let any node talk to any other node at will.
As a result, the designs of huge numbers of higher-layer protocols have been warped and complicated, and the whole thing has contributed to the whole maelstrom of extreme commercial centralization pressures, too.
Everybody who wasn't willing to hard cut over to IPv6 should have been forced off of the Internet 20 years ago.
There are very valid reasons to want to send traffic to a node so called "out of the blue", and suggesting otherwise suggests you have never looked deeper than HTTP.
And unless laws force them to, which will be hard to implement because individual countries do not want to hamper their own economic competitivity, the industries won't do anything more than the actual greenwashing.
Chabges usually have to be implemented BEFORE the disasters happen. In a seismic zone, we should build earthquake resistant buildings now, not after the first major earthquake.
Tell that to millions of people stuck behind CG-NAT. Tell that to ISPs trying to serve communities on shoe-string budgets:
> I work for a Native American tribe in the PNW. We scrambled to get the reservation reliable internet in the later part of 2019. We managed to cover most of the reservation with wi-max and wifi with a fiber back haul configuration. We are now slowly getting more stable and reliable fiber to the home(FttH) service installed to as many homes as we can, but it is slow process covering the mostly rural landscape doing all the work in house.
> Our tribal network started out IPv6, but soon learned we had to somehow support IPv4 only traffic. It took almost 11 months in order to get a small amount of IPv4 addresses allocated for this use. In fact there were only enough addresses to cover maybe 1% of population. So we were forced to create a very expensive proxy/translation server in order to support this traffic.
> We learned a very expensive lesson. 71% of the IPv4 traffic we were supporting was from ROKU devices. 9% coming from DishNetwork & DirectTV satellite tuners, 11% from HomeSecurity cameras and systems, and remaining 9% we replaced extremely outdated Point of Sale(POS) equipment. So we cut ROKU some slack three years ago by spending a little over $300k just to support their devices.
* https://community.roku.com/t5/Features-settings-updates/It-s...
* https://news.ycombinator.com/item?id=35047624
IPv4-only products and services is a problem right now.
reasons range from cdn prevalence to cgnat and v6 "islands" but the old internet seems to be chugging along
The lack of adoption by major players (GitHub, twitch, IMDb, etc) is a financial one, not a technical one.
This is why it's much more likely that you'll continue seeing things like CGNAT instead of IPv6 - the IPv6 standard is just an utter engineering failure no matter how you look at it.
We've been using NAT and expanding it to decades which makes it an easy proof that it brings value.
IPv6 is still barely adopted and barely available in two decades after having "ipv6 parties". What does that tell you about its value?
https://rex.apnic.net/statistics?allocationType=asn,ipv4,ipv...
Do the same with IPv6, omit the NAT, and you do not need to keep that state in sync. P2P applications stop being a routing/port forwarding problem and are only a matter of firewall settings. Cut away all that SIP and IPsec debugging.
I think it is to do with country wise allocation. I think India is seeing wide adaption of IPv6.
Guess why people are being forced to move to CGNAT? Precisely because we have run out of addresses.
The dismissal attitude of HN towards IPv6 ("we ran out of IPv4 and everything still works!") will always baffle me. Because things are _failing_ to work.
Yup. Spending money CG-NAT gear:
> I work for a Native American tribe in the PNW. We scrambled to get the reservation reliable internet in the later part of 2019. We managed to cover most of the reservation with wi-max and wifi with a fiber back haul configuration. We are now slowly getting more stable and reliable fiber to the home(FttH) service installed to as many homes as we can, but it is slow process covering the mostly rural landscape doing all the work in house.
> Our tribal network started out IPv6, but soon learned we had to somehow support IPv4 only traffic. It took almost 11 months in order to get a small amount of IPv4 addresses allocated for this use. In fact there were only enough addresses to cover maybe 1% of population. So we were forced to create a very expensive proxy/translation server in order to support this traffic.
> We learned a very expensive lesson. 71% of the IPv4 traffic we were supporting was from ROKU devices. 9% coming from DishNetwork & DirectTV satellite tuners, 11% from HomeSecurity cameras and systems, and remaining 9% we replaced extremely outdated Point of Sale(POS) equipment. So we cut ROKU some slack three years ago by spending a little over $300k just to support their devices.
* https://community.roku.com/t5/Features-settings-updates/It-s...
* https://news.ycombinator.com/item?id=35047624
IPv4-only products and services is a problem right now.
Good for you! I hope my thorough ignoring the issue will not hurt your feelings.
But IPv4 people won't be able to connect to IPv6 services no matter how, so don't go surprised_pikachu.jpg whenever someone decides they don't want to pay for the IPv4 tax: https://www.bottlecaps.de/index/ https://loopsofzen.uk/
> I work for a Native American tribe in the PNW. We scrambled to get the reservation reliable internet in the later part of 2019. We managed to cover most of the reservation with wi-max and wifi with a fiber back haul configuration. We are now slowly getting more stable and reliable fiber to the home(FttH) service installed to as many homes as we can, but it is slow process covering the mostly rural landscape doing all the work in house.
> Our tribal network started out IPv6, but soon learned we had to somehow support IPv4 only traffic. It took almost 11 months in order to get a small amount of IPv4 addresses allocated for this use. In fact there were only enough addresses to cover maybe 1% of population. So we were forced to create a very expensive proxy/translation server in order to support this traffic.
> We learned a very expensive lesson. 71% of the IPv4 traffic we were supporting was from ROKU devices. 9% coming from DishNetwork & DirectTV satellite tuners, 11% from HomeSecurity cameras and systems, and remaining 9% we replaced extremely outdated Point of Sale(POS) equipment. So we cut ROKU some slack three years ago by spending a little over $300k just to support their devices.
[…]
* https://community.roku.com/t5/Features-settings-updates/It-s...
Which is a policy issue (implemented through rules on a firewall), and not an inherent limitation of the technology (like NAT is).
Non-authoritative answer:
Name: google.com
Addresses: 2607:f8b0:4002:c00::8b
2607:f8b0:4002:c00::66
2607:f8b0:4002:c00::65
2607:f8b0:4002:c00::8a
74.125.136.100
74.125.136.138
74.125.136.101
74.125.136.113
74.125.136.102
74.125.136.139This was known since the incipit, still we moved on <facepalm/>.
> And so the kinda-working status quo makes moving to the properly-working system much harder.
It's called evolution: there is no "properly-working system" unless it works at every stage. If it is at disadvantage in an intermediate stage, it will never be selected.
I don't think there's any evidence that things have to be evolved in the tech world.
paradigm shifts happen all the time.
Look, all the good of OOP has not obliterated C. All the excellence of functional programming has not even made a dent in OOP adoption. Hell, COBOL is still running fine!
Paradigm adoption is a matter of evolution: the fittest survives. When IPv6 will offer an immediate advantage to its users, people will switch to it. Until then, it is confined to those companies who are betting on it, other companies who have some spare money and prefer to spend it rather than have it taxed, and the stereotypical nerds who haven't anything more interesting to ponder about.
My point is paradigm shifts happen quite often. Evolution happens as well but is by no means the ONLY way as you're claiming.
If IPv6 is truly useless you wouldn't be able to visit my site at all, since Cloudflare is unable to make contact with the origin server.
I think you're looking at this the wrong way.
"If it ain't broke, don't fix it!" is probably the correct way. Amazon is surely aware of IPv6 but if everybody who is on IPv6 can be tunneled through IPv4 to your storefront why bother changing the site?
The new technology doesn't have to _just_ be better than the current one to cause adoption. It needs to be so much better that its worth adopting. Although judging from the fact that tiktok is on the list; it seems like new-er websites do not consider IPv6 to be better than IPv4.
If there wasn't a flag day for IP (1983-01-01) I'm sure there would have been people still insisting on using NCP.
It took several decades' worth of experience to go from NCP to IP, and now that we've had experience with IPv4, it's worth moving towards IPv6.
> If a technology is useful, you don't have to shame people into using it.
It is useful for some people. It is not useful for others: like newly-started ISPs who can't get / afford IPv4 allocations for their millions (tens of, or hundreds of, in many APAC countries) of users and are left with paying for CG-NAT to reach many parts of the Internet. (And then their users have to double-NAT IPv4 to run their local networks, which breaks all sorts of things).
The fact that the early adopters / 'legacy' folks in some parts of the world lucked out on allocations does not mean IPv4, with its limited 32-bit address space, is "useful".
Best use of IPv6 I have seen is in cjdns.
Where are all the other examples of peer-to-peer connectivity using IPv6.
Personally I do not want peer-to-peer connection with websites doing surveillance, data collection, tracking and advertising, or other commercial entities. I do want connections with other internet users. But that cuts out so-called "tech" companies, intermediaries that have nothing to offer except middleman services. Hence there is a conflict of interest in that the benefits I could see for IPv6 (e.g., cut out the middleman) and the benefits of IPv6 to the advocates trying to shame people into adopting it (e.g., people who work for middlemen) are mutually exclusive.
So is ipv4, but most people have the benefit of learning it piecewise over a decade+.
When I do networking root cause investigations involving v4, I still get glazed over looks from professionals that don’t specialize in net, but in that case they blame their own knowledge. If it’s v6 it’s the protocol’s fault they don’t know it.
CGNAT breaks the end-to-end principle of the Internet and 100% forces you to depend on centralized services, because you will absolutely not be able to receive unsolicited traffic on your IP without coordination/permission from your ISP or other various third parties. Yes, there are NAT bypass methods-but those methods all rely at least in some situations on an external server that relays connection requests.
Many people really don't care if the Internet becomes the next cable TV along with business tunneling services and 2 or 3 social networks and would really be fine with that.
If you don't want this to be the only choice in the future, you should start learning, using, and demanding IPv6 and really stop putting NAT on a pedestal and bitching about the long addresses.
When I set up IPv6 on my home network, I took some time and learned how to get it working with DNS, and now I don't have to type long addresses any more. I have an actual separate IPv6 subnet for guest network. It's pretty cool.
As a result it is easier to scrape a public website because the website would need to block legitimate users at the residential ISP or on the same mobile network, and this consequence is likely worse for the website than having to accept scraping. It's much harder for a website to determine that 100 requests/second from a single IPv4 address is a single user scraping away, or 100 customers each on their own phone/computer just wanting to use the website.
IPv6 is like the same code with a major refactor. One of the big unwieldy databases (NAT) was dropped because it turns out you don't need it if the system is smarter overall. Also, autoconfiguration works reasonably well now. For most usecases the subnetting also got much easier as you can just slap a /64 on any realistic network without hesitation. You just have to do a relatively small step to use it.
IPv6 can work side by side with IPv4 on the servers, but it cannot replace IPv4. Thus, don't act surprised when companies wonder "Why bother?"!
Facebook is famously IPv6-only internally (and as of 2016, 60% of user traffic was IPv6-only):
* https://www.internetsociety.org/resources/deploy360/2014/cas...
* https://engineering.fb.com/2017/01/17/production-engineering...
* https://www.youtube.com/watch?v=lU_24f6Wr7E
Microsoft and Google are IPv6-only single-stack (or moving in that direction) on their internal corporate networks:
* https://www.arin.net/blog/2019/04/03/microsoft-works-toward-...
* https://www.youtube.com/watch?v=nd0vgU6WbPo
* https://www.youtube.com/watch?v=hb98hAb5_W8
Also Wells Fargo:
* https://www.youtube.com/watch?v=EzTWjNUb4H4
The main 'advantage' IPv4 has is inertia.
That's correct, but it was an idiocy to underestimate it. Reality always wins.
Where I see it being useful is for IoT devices, where you're never configuring them manually, and where they probably only need to speak to one endpoint.
For anything that involves manual setup, you're going to have to force me to use IPv6, because I just don't care. Voluntary adoption is a pointless endeavor.
I can even control how I connect to the device, either by using it's hostname on the LAN, it's WAN hostname, or it's VPN hostname. This has always been preferable to me over remembering IP addresses, even in the v4 realm.
The only time v6 has ever caused me any difficulty is when dealing with services/tools that don't support it, but that's getting to be less and less of a problem even.
> FWIW I really don't need to remember the IPv6 addresses of devices on my networks, DNS and my router handle that for me fairly automatically.
You do need to remember IPv6 addresses when your network is hosed when the DNS server becomes unresponsive or returns nonsense. IPv6 adds an annoying barrier to troubleshooting. Not that you can't work around it, but it does have more friction than troubleshooting IPv4 issues.
I'm personally really torn about IPv6. On one hand, it technically is better, and for more reason than just the increased address space. On the other hand, it is more difficult to grok than IPv4. Especially because IPv6 does a bunch of magic that, if you're not on high-end networking gear, isn't logged anywhere.
Who is going to remember the IP addresses of all their switches, routers, and firewalls even when they are IPv4-only? I have less than a handful of switches, and have no idea what the IPs are.
Document the values (Netbox, MediaWiki, private company Github doc repo) and look them up if needed. Or put the values in the hosts(5) file of your management network bastion host(s).
I'll be damned if I can remember a single public IPv6 DNS server.
* https://www.iana.org/domains/root/files
* https://www.iana.org/domains/root/servers
and bootstrapping the network infrastructure from there.
Really the dollar speaks here. If ipv6 is a real requirement, than companies not supporting it will be hurt fiscally.
Taking a step back, one might consider how incredibly... unwanted and unwieldy ipv6 is. It's just a pain to deal with. Pundits will say "Oh, it's easy!". Great. Now explain why something touted as utterly simplistic and with supposedly zero issues rolling out, and is cited as having immense upside, has its adoption fought by companies which have top tier talent working for them.
Sure. It's all that top tier talent that's wrong. They're confused. It's not ipv6.
The main failure case is that it’s a painful or possibly impossible to setup your own public internet server from home, but ISPs are probably happy about that and games have figured out a work-around.
If IPv6 was a project at work, it would probably get killed as not a priority.
The cost of ipv4 is ultimately being paid for by customers in the form of higher fees for their internet subscriptions. And this is exactly a case where the government can help the market operate more efficiently by demanding standardization.
IPv4 hurts innovation by basically charging for numbers more than an administrative fee hindering newcomers to enter the market. Everybody suffers a little because of the networking effect. However, IPv4 seems to be a pain for the big companies too. Some, like e.g. Facebook are really pushing it.
With the supposedly "top tier talent" why does e.g. Azure have such a bad IPv6 implementation? Why is Cloudflare or big parts of GitHub IPv4-only? Isn't that meant for developers? What about e.g. Twitter?
The biggest problem is not the infrastructure. All the important programs like DNS servers, operating systems and hardware do support IPv6 more or less for the last 10-15 years well enough. The problem is developers that are stuck in IPv4 mentally, doing GUI windows, that only accept IPv4, only testing on IPv4, specifying database columns to hold only IPv4 etc.
Does Ziggo support IPv6?
This is one example, but there are many where we need to investigate more than just, "Does my ISP support IPv6?"
I've no idea if my website is IPv6 compatible. When Let's Encrypt became a thing, I pushed a button on my webhost's console and I had HTTPS. It worked for the majority of devices without needing any reconfiguration. Where's the one-click solution for IPv6?
As to my current knowledge, this is not possible because any router not understanding these extensions would be unable to route packets with that extension correctly.
But there's still a gradient on how much the new protocol actually differs from the old in how networks are structured and how much work router, switch and client firmware needs to support it.
The fact that I haven't seen a single non-buggy router firmware (and I've deployed quite a few IPv6 networks) tells by itself just how overbloated the standard is.
It's a fundamental change at the IP protocol layer, there wasn't scope in that layer for backwards compatibility.
There's loads of transition mechanisms that were introduced, but it's not as tractable of a problem. Assumptions around IPv4 have been baked in all over the place.
Also there was more _reason_ to move to HTTPS. Virtually everything had already supported HTTPS for closing in on two decades at the point the mass rollout with LE happened. You could switch to HTTPS-only for your site without breaking it for your users because they already needed it for online banking etc.
You can't just switch to exclusive IPv6 because half your users don't even have an IPv6 connection. And virtually everyone with IPv6 has an IPv4 connection anyway so the benefit in dual stacking is nil. Unlike supporting HTTPS where at least those that supported it got the benefit.
If you run some infrastructure, you have to dual stack only at the gateway to other networks. You can run IPv6-only or IPv4-only inside and reverse-proxy or tunnel the traffic. Running IPv6 only can save you some money in the cloud today, it is not much but for small things the charge can be in double digit percent.
The trouble with IPv6 is the hen and egg. For ISPs it's troublesome as they have to run double stack, till the last site on the Internet is converted and for site owners the trouble is they have to double stack till last ISP converted.
Doing anything backward compatible isn't an option as the trouble is I. The 32 bit address limiting number of addresses on IPv4, and any form of increase causes compatibility issues.
(Of course one could build an IP protocol closer to IPv4 with just larger address space, but transition costs would still be very similar to IPv6)
Why should companies care about anything not directly linked to their customers' satisfaction?
Who? Where?
Oversubscribed CGNATs are real. [0] Customers may not care which protocol you (as the ISP) are using, but they will definitely notice dialup-era internet speed!
[0]: https://reddit.com/r/ipv6/comments/1as8dvy/is_there_a_way_to...
* https://www.youtube.com/watch?v=nNMNglk_CvE
Wells Fargo:
* https://www.youtube.com/watch?v=EzTWjNUb4H4
Microsoft:
* https://www.arin.net/blog/2019/04/03/microsoft-works-toward-...
* https://www.youtube.com/watch?v=nd0vgU6WbPo
Google:
* https://www.youtube.com/watch?v=hb98hAb5_W8
LinkedIn (pre-Microsoft):
If a server has no AAAA record, machines only connected to IPV6 simply cannot contact that server. This is fine in theory (just buy an ipv4 lolz) but in practices this means additional costs that will only go up, as the ipv4 space becomes more and more crowded. As a non-network admin, I care that services I'm a business costumer of are available to my servers, the biggest offender being GitHub. GitHub still does not have an ipv6 despite "looking into the issue" for several years, and this means that I am unable to use their services or indeed any software that requires access to GitHub for, say, updates, pulling artifacts, container images or whatever else from their servers.
It does sound like it's a bit presumptuous of me as a random IT person to demand an IPV6 out of them, but in the year of the Linux desktop, there's no reason for them not to have the infrastructure in place to simply flip a switch and be done with it, for basically free.
Of course, the big players own swathes of IPV4 space that they are able to rent out at a premium, so they are heavily disincentivized from ever supporting ipv6, as that means obsoleting their investment with no replacement, but it's bullshit that they are able to do this in the first place.
Numerous solutions could've been thought up (CGNAT the entirety of IPv6 as a source behind a part of the IPv4 space and allocate a part of the IPv6 region for "legacy" IPv4 addresses for example; that way you have 6->4 backwards compatibility), but none were tried. The result is a great technology in a vacuum, forever hampered in adoption rate because it's not compatible with old sites and there's no incentive for old site operators to switch.
Shaming bad actors doesn't do anything because IPv4 vs IPv6 is largely nerd shit in terms of computer issues; your average internet user is behind CGNAT already for IPv4 so address exhaustion means nothing to them so they don't care about it.
The problem is the inverse direction (IPv4 -> IPv6) since IPv4 is lacking any mechanism for forward-compatibility.
At another site I manage they have Starlink with v6 and CGNAT v4 and it's great, just plugged in the dish to the pfSense router, set up v6, and it's been ticking away happily ever since. The only problems we have are with the old devices (iot and such) that don't get V6 addresses, and have trouble with the CGNAT. (Seems like it doesn't reset long-running connections correctly perhaps, they just get stalled out. Killing the state on pfSense usually gets them working again.)
So I suspect it's more a your-mileage-may-vary situation depending on ISP than an issue with v6 or pfSense.
https://www.google.com/intl/en/ipv6/statistics.html ipv6 rollout is well in progress. AWS started charging extra for IPv4 ips. 32 bits is not enough. According to that site India is 70% ipv6
Websites should have the easier time providing ipv6. In the past they could keep it simpler because users weren't on ipv6, & often that was because ISPs have dragged their feet. That's changing
IPv6 is the only viable path forward. There are no redos here. Saying IPv6 can't work is the equivalent of burying your head in the sand.
IPv4 space has run out. Any solution that add more space requires full adoption from client on the planet.
The thing is, IPv6 only clients can reach IPv4 only services. It's why these services don't advertise v6, they don't technically need to even if some clients would be better off.
But an IPv6 only service _can not_ be accessed by an IPv4 only client.
These limitations are true for any possible alternative to IPv6.
We need people to stop trying to deny IPv6, it IS the future.
If the USB consortium can do it, the network folks can as well.
Implementing it fully is a technical process, but it's not an impossible barrier. The real reason for the slowness is that there's little to no profit incentive to ensuring its full compatibility, in part because IPv4 is guaranteed to be (mostly) backwards compatible for decades.
It's a slow process, but it's happening. There's no need to invent yet another standard for something that's already working and will likely be good enough to last generations once it's finished being implemented.
Nothing is going to change the fact that _any_ solution needs every router to work, needs to handle legacy v4 and new v* addresses.
There is a reason why there aren't even alternative proposals to IPv6, none of the problems that the IPv6 deployment is having can be solved by a "simpler" implementation.
But that's just not true now, is it? Extending the address space didn't need a complete rework and rethink how local network routing works, removal of standard NAT and messing around with address assignment (DHCPv6, SLAAC, and all of those messes).
IPv6 isn't having problems because it's not compatible, it's having problems because they made an utter mess of how local networks works and routers most companies (and people) have aren't able to reliably deal with the complexity of what's rather easily solved by DHCPv4 on "old" networks. The fact that IPv6 decided to rework the basics of how routing works while also demanding a dual-stack deployment is the core of the problem.
If the extent of deploying IPv6 would be "we'll enable address assignment on our existing DHCP server with same rules", you'd be on mostly IPv6 right now.
I think the issue here is that IPv6, despite being the successor to IPv4, is actually quite old in itself; it was developed starting in the early 90's and so it's just a different idea of what the internet was going to be. NAT is more like a "hack" that developed around that time as well and has turned into the normal way networks are managed.
(But that was even shown when they were dreaming of deploying IPv6 without any kind of DHCP which had to be backtracked making a utter frigging mess of address assignment on networks with multiple competing approaches.)
Who? And what constitutes "actual"? Are the netengs in FAANG actual network operators? Who aren't actual operators? Why they aren't?
>Most deployments of IPv6 are mobile networks which don't actually allow their client phone IPv6 addresses to be reachable (as was the dream around getting rid of NAT).
[Citation needed]
I just tested with 2 cellular ISPs, and I was able to connect to the SSH service on my phone with either.
(It's mostly useless though, since whenever the phone is locked the app is also suspended, causing the SSH service to be dropped)
There's no fundamental difference in routing. They both use trivially simple prefix routing.
> removal of standard NAT
NAT isn't "standard". NAT is a nasty kludge that people put in because they were too lazy to upgrade their software stacks. And IPv6 was defined before NAT was commonly used, in large part as a way to avoid NAT becoming common. Because NAT is a bad idea in every possible way.
Also, IPv4 is itself much, much simpler without NAT. And you can do NAT on IPv6 if you're stupid enough to want to.
Saying that IPv6 is complicated because it removes the need for a complicated hack like NAT is just insanity.
> (DHCPv6, SLAAC, and all of those messes)
Those are not in any way more complicated or harder to use than DHCP4. Setting up a DHCP server to get automatic address assignment on an IPv4 network is actually harder than letting a router automatically do SLAAC to do address assignment on IPv6.
Also, DHCP is not part of IPv4, and DHCP6 is not part of IPv6. I was using and managing IPv4 networks before DHCP existed. Yes, it was a royal pain in the ass... which is why that functionality was integrated into IPv6. Which was done a protocol that's actually simpler than DHCP, although I didn't and still don't approve of the way it wastes address bits.
The fact that you consider SLAAC somehow equivalent to DHCP shows that - I've never seen a router (outside maybe some ciscos) where those two could be used to achieve the same stable network assignment and configure downstream service firewalls appropriately. That is - if SLAAC actually works across all clients and they're not a buggy mess.
45% adoption.
>horrid bugginess of IPv6 network stack
Were the IPv4 network stack _not_ buggy? Are you sure it's just not the general incompetence of network equipment people?
>where those two could be used to achieve the same stable network assignment and configure downstream service firewalls appropriately
What?
>SLAAC actually works across all clients and they're not a buggy mess
What buggy mess? I'm on SLAAC and I don't notice anything buggy?
After 28 years? Yeah, that sounds like failing adoption.
Edit: I should caveat - I don't have good data on actual lifespan of networking equipment; if you can show that it's likely to live more than a decade then that would be reasonably compelling.
I stand by my assessment that 45% adoption after 12 years is actually pretty good.
[0]: https://transparencyreport.google.com/https/overview?hl=en
(I use Google's data for both HTTPS and IPv6 stats. If you consider 3rd party data, HTTPS adoption is way worse, even to this day it hasn't topped 75%.)
This is a classic responsibility-evasion cycle:
Users: We can't get equipment that supports this (or supports it well, or makes it easy), so we won't deploy it.
Vendors: Users aren't demanding it, so we won't support it.
The result is a suboptimal local equilibrium and an obvious market failure, but you invariably get loudmouths on message boards saying "Look, the Market (Peace Be Upon It) has spoken!". Gets a bit old.
The right answer in this case is probably coercive government regulation of the form "No IPv4 packet shall cross a property line", but I'm not holding my breath.
You can scream and cry as muck as you like, IPv4 IS the present. And people in general will not care until the world will work for them.
And I trust those companies to know how to cater for people's money. When, and if, IPv6 will be the right tool, they will use it. But trying to shame them on such a meaningless technicality is the equivalent of criticising NASA for the colour of the Apollo launch station.
I still wonder why they did not enhance IPv4 by adding more digits. Maybe that would have been easier ?
The phone company did that 50 years ago. When I was a kid we only needed to dial 5 numbers for local, 7 for long distance with a 1. As time went on, additional digits were added.
I mean, that's not too far off from what IPv6 is. The main difference, at least as far as the visual addresses go, is that IPv6 also supports letters, which drastically increases the total number of addresses beyond what extending the number of characters would do.
I think part of the idea was that, since transitioning from IPv4 to IPv6 is going to be difficult, we might as well make it worthwile be ensuring that another transition never needs to happen again for the forseeable future.
there are no "additional bits" left to use and it would also require cosiderable effort to implement
Adding more digits sounds simple, but there is no good room in IPv4 packet. And then any solution using existing one would not anyways be backwards compatible. So for each round of new byte you add to header you would have different version anyway. Requiring new hardware and software.
IPv4's address space is fundamentally too small, so there has to be some kind of alternative or replacement.
> As we edge closer to exhausting the IPv4 address space, the immense address capacity of IPv6 becomes indispensable.
No explanation why this is bad. It's harder for websites to track individual users and block them. What's not to like about this?
> Beyond the scalability, IPv6 brings along robust security protocols and superior performance, making it the linchpin for modern, efficient, and secure internet communications.
Show me one person that had performance problems with ipv4
It's a double edged sword. Other websites can't track you, simultaneously that also means other internet users can't reach you.
The most obvious consequence is that to host a server, you must purchase a VPS or rent an public IP address from your ISP, and the price for a public IPv4 address is getting higher and higher.
The less obvious consequence is that you're giving up control to the VPS providers (and other centralized services).
It's rather ironic that people on HN, a website whose name literally includes the term "hackers", would support things like CGNAT which hurt hackers/hobbyists the most.
>Show me one person that had performance problems with ipv4
Me. My ISP's CGNAT gateway was temporarily overloaded during one of the holidays earlier this year. I was thoroughly unimpressed by the dialup-like experience.
That being said, I prefer shame to regulation. But both are pretty far down the list of ideal solutions.
Oh, if only I had literally any of those to pick from. I'm American, sadly, and there is no competitive landscape.
I’m pretty sure it’s not a large factor overall but it seems that economically it doesn’t make much sense for most existing actors to switch to IPv6 anytime soon.
Companies are already doing it. IPXO, for one.
When I turn IPv6 off and everything starts working normally again on my home network, it's a safe bet I'm not going to troubleshoot what the real issue is. Sure, misconfiguration at my ISP probably, but if you think I care enough to go down that rabbit hole, you don't understand how people work.