Because many ISPs do not support it, the majority of routers have it disabled by default, routers have barebone IPv6 features and often times some aspects are exclusive to IPv4.
I use IPv6 in my local network but since my ISP does not support it, I cannot have publicly available IPv6 addresses.
>I use IPv6 in my local network
What are the benefits over IPv4 on LAN?
Isn't it just the same but with longer IP addresses that take more time to type?
Also why isn't local DNS a thing?
I want to type "printer.local" instead of its IP address.
>What are the benefits over IPv4 on LAN?
Optimized protocol: unicast, multicast and anycast are way more optimised communication transmissions compared to IPv4's broadcast, they of course are extremely more efficient on the internet, but it's great on a local network too.
>Isn't it just the same but with longer IP addresses that take more time to type?
192.168.1.2 vs fe80::1
Doesn't seem longer to me.
>Also why isn't local DNS a thing? >I want to type "printer.local" instead of its IP address.
It is, it's called multicast DNS.
I had to bump before thread archived
picrel is the ip my phone has currently
you will note something; this is a cgnat private ip. Other times my phone will have a 10.something address this is within the 10.0.0.0/8 private ip address space as well. The ip you see when you go to whatsmyip is the public ip of the mobile gateway not your phone. There is an exceedingly rare chance you have a public mobile ipv4 address.
you as well
[...]
I had to bump before thread archived
picrel is the ip my phone has currently
you will note something; this is a cgnat private ip. Other times my phone will have a 10.something address this is within the 10.0.0.0/8 private ip address space as well. The ip you see when you go to whatsmyip is the public ip of the mobile gateway not your phone. There is an exceedingly rare chance you have a public mobile ipv4 address.
>this is a cgnat private
it's within 100.64.0.0/10, so, yes
3 weeks ago
Anonymous
>see
meant to quote
Everyone is using ipv6 retard. You think your cell phone is getting a public v4 address?
The tech sector is currently overrun by a bunch of nepo hires and retards so nobody is updating their network infrastructure because the people they hired genuinely have no idea how to do so
Boomers who think every device having a public IP address is a security issue, when they could just tweak the firewall on their routers. They also think it's a privacy issue because they still believe IPv6 addresses include portions of their MAC addresses when there is the privacy extension and many devices spoof their MAC already without doing anything. Then, they may think that ISP's DHCP server will keep giving them the same IPv6 when that's the same for IPv4 for a long time now, and just requires a MAC spoof on the interface connected to your ISP and restart of your DHCP client. Some may think it's hard to remember addresses due to them being so long but you're supposed to be using DNS.
To anyone denying this, do you honestly think we wouldn't have had fully deployed IPv6 by now if it wasn't for NAT? Of course implementations would have been a lot better. (IPv6 only is a shitty experience in most modern OSes today. Try.)
NAT is just a cope for smooth brained boomers who don't know how to configure their firewall. IPv6 only on modern OSes works just fine unless you have fucked something up. The only way it'd suck is if your favorite websites are IPv4 only and you also want to use HTTPS since NAT64+DNS64 will fail here.
I'm almost thirty, so you must be the boomer I'm talking about. Also, NAT64+DNS64 is not tinkering. It's a late stage IPv6 transition technology, which is also used by many corporations for example Huawei. It just won't work when you want to access HTTPS for an IPv4 only host unless you got your CA to add IP: 64:ff9b::yourIPv4inhex in your certificate probably (haven't tried)
That's besides the point. OP was asking why "people" (normies) aren't using IPv6. What corps and nerds are doing to get it to work is another conversation. I myself dropped out when my ISP stopped supporting it. (IPv6 caused support tickets from normies due to shitty sites and equipment, and almost no-one cares about IPv6.)
But people are using it, they just don't know it because their carrier gives them IPv6+IPv4 addresses, and everything seamlessly runs on dual-stack. >but how does IPv6 get any traffic if IPv4 is already on?
"happy eyeballs"
lol it's a child that thinks it's an adult. amazing watching the newer gen grow up so uneducated on basic computing. just on adoption alone you'll never make it in this business. go flip burgers this will be far more useful than the maintenance routine jobs designed for mid-level minds you are currently destined for.
>NAT64+DNS64 is not tinkering.
That's besides the point. OP was asking why "people" (normies) aren't using IPv6. What corps and nerds are doing to get it to work is another conversation. I myself dropped out when my ISP stopped supporting it. (IPv6 caused support tickets from normies due to shitty sites and equipment, and almost no-one cares about IPv6.)
>IPv6 caused support tickets from normies due to shitty sites and equipment, and almost no-one cares about IPv6
xe can't overcome this because they see this issue as their fault and correspond that to "their problem". that is why I recommended flipping burgers vs. opening a burger stand.
We should strike your pension and see how you deal with that, boomer.
3 weeks ago
Anonymous
Not a boomer. I'm also smart enough to know boomers often know their shit. Frustrated drones couldn't understand that. LOL ...
I know exactly what I'm talking about.
Sure, but you don't know what other people are talking about. That's a (You) problem and you don't seem capable of taking up the challenge.
All you've done is make excuses and seethed about your dad.
3 weeks ago
Anonymous
Then you shouldn't have any issue when we collectively strike all pensions, right?
>cope
3 weeks ago
Anonymous
Other people are being refuted and educated by me. You're just in denial. Go take your meds grandpa. If you were around before NAT then you are a boomer. You haven't even really said anything about anything.
3 weeks ago
Anonymous
Then you shouldn't have any issue when we collectively strike all pensions, right?
[...]
>And there's nothing you can do about it. >Absolutely nothing. >ipv6lets would rather learn base 16 than just extend ipv4 to be like 1.2.3.4.5.6.7.8 with existing ipv4 addresses being 0.0.0.0.1.2.3.4
it has no NAT, meaning all your insecure IP cams are wide open for scanning skiddies to attack. I've disabled IPv6 on my entire network, until it's more mature
With a typical consumer broadband connection you can scan the whole IPv4 space in a couple of days, but scanning the whole IPv6 space would take billions of years. Also, even if you could scan it in a day it would be useless for anything connected to a router newer than 5 years because of changing addresses.
NAT isn't a security feature, retard. Learn how to write firewall rules. >implying skiddies can't 0 day your router
I am retarded and replied to the wrong post, see
Good look scanning IPv6.
With a typical consumer broadband connection you can scan the whole IPv4 space in a couple of days, but scanning the whole IPv6 space would take billions of years. Also, even if you could scan it in a day it would be useless for anything connected to a router newer than 5 years because of changing addresses.
::ffff: is shorter than 0.0.0.0. and also easier to type.
3 weeks ago
Anonymous
What are the ffff for?
>0.0.0.0.1.2.3.4 >::1.2.3.4
3 weeks ago
Anonymous
https://en.wikipedia.org/wiki/Reserved_IP_addresses
That is how IPv6 displays IPv4-mapped addresses.
Or I can simply continue using IPv4.
Enjoy your higher ping and higher number of internet routing hops then.
3 weeks ago
Anonymous
Getting higher than zero numbers isn't difficult. Being used in the first place, now THAT is difficult.
3 weeks ago
Anonymous
You got used to 192.168.x.x, 255.255.255.x or whatever other standard IPv4 address/subnet mask, you will eventually get used to the IPv6 ones once ISPs will start charging for IPv4 addresses, AWS has started doing just that.
3 weeks ago
Anonymous
No, I don't think I will. Because I'd rather quit.
3 weeks ago
Anonymous
>so ffff >no, wait, a colon >no, wait, a double-colon >then ffff >then a double-colon >no, wait, no colon >no, wait, a single colon
How does one manage to convey so much brain damage in a single post such as yours?
3 weeks ago
Anonymous
It is very easy if you know the 3 rules of IPv6 compression.
>And there's nothing you can do about it. >Absolutely nothing. >ipv6lets would rather learn base 16 than just extend ipv4 to be like 1.2.3.4.5.6.7.8 with existing ipv4 addresses being 0.0.0.0.1.2.3.4
Cause for the home there's no point. A typical class C IP can handle 253 devices + 1 reserved for the router. I daresay that's a lot of IP cams and other shit Then lets say off chance you do go beyond that. They make a private Class A and B range you know. You can run whatever IP range you wish at home. Long as it's valid and private. Your ISP don't give a fuck. All the world sees is whatever your IP is the ISP gives you.
it's 253 devices total so 252 if you reserve one for the gateway.
a class c /24 has 2 unusable host addresses, the top most broadcast bit and the bottom most network address bit.
There's a lot of really retarded opinions in here functionally ipv6 can work the exact same as ipv4 it just has more overhead the fields are basically all the same and some of the trivial ones removed like total length, slaac is a bit dumb and Router advertisement only working on /64 networks is also dumb but ipv6 was made to be literally as handholding as possible because the IETF knew two things
1 - the state of building computer networks should not continue to be stuck at having techs fuck with dumb shit like dhcp and basic routing it should work out of the box which is a good idea
2 IT people are retarded and lazy and were already filtered by IP(v4) so v6 should handle itself except at the carrier or campus level, which it does
surprise surprise everyone is still filtered and can hardly understand IPv4 as is anyways.
The real reason ipv6 is dead on arrival is that IT is a brainlet industry and is far too important and far too easy to get into the level that people need you at generating far too many mundane workers to move the industry en masse. Very very few of IT types are even remotely close to understanding how computer networks actually work much less protocol design your dumb little kubernetes containers that you struggle to get talking to a single gateway much less 2 isn't hard and you got filtered by TCP flow control in technical reading of course you wouldn't understand the concequences of Router Alert or Next Header.
>Cause for the home there's no point
parsing ipv6 headers is less computatonally intense than ipv4 thanks to better byte alignment and a lot less bit twiddling one needs to do in order to parse the header
another benefit of ipv6 is that you eliminate a whole class of cache coherency bugs by being able to randomly pick an internal address and have virtually zero probability of collision
almost every competent sysadmin will use ipv6 internally (unless there is a huge amount of legacy cruft forcing you on IPv4) and set up an ipv4 ingress
IPv6 is being adopted at a steady rate of ~5% per year according to google.com/ipv6
As more boomer IT admins retire and are replaced with freshly graduated minds who aren't terrified of new technology.
Anyone still advocating for IPv4 is an ignorant retard. Case in point:
homosexual doesn't even know the IP class system for IPv4 has been obsoleted many years ago. How can one expect retards like him to know anything about IPv6 when they don't even know IPv4?
It is being used and it is much more commonplace then you think. IPv4 is slowly but surely being regress to exist only in legacy networks and equipment
I once heard it put like this: >the only benefit of IPv6 is the larger address space >a larger address space only benefits people who don't have addresses, that is, people who aren't on the internet >all the work of deploying it has to be done by people who already have addresses, that is, people who are already on the internet >so you're asking tons of people to do a bunch of work that doesn't benefit them for the benefit of other people >oh and the people who aren't on the internet but want to get on want to reach the people who are already on the internet. that is, they want to reach IPv4 addresses, so a v4 address is valuable to them and a v6 address mostly isn't
I read that in a sysadmin's blog post like a decade ago when people had already been banging the drums for IPv6 for over a decade and were dismayed that it wasn't ubiquitous. And here I am without working IPv6 in 2023 and I've never noticed a problem with it. This'll keep going for several more decades I bet.
>the only benifit of IPv6 is the larger address space
already wrong and this is why I said the IT crowd aren't nearly as smart as they think they are it's just the new generation of skilled labor and not actual intelligence. ipv6 has a ton of benefits over ipv4 people are just dumb and only focus on global addressing which is still a point of contention but sysadmins are lazy so they pretend its not an issue.
The value provided by the flow label alone for developers or large network operators like campus size networks is huge, having to classify traffic at ingress on network devices is always a point of pain instead imagine just being able to set a flow label, also next header completely solves packet length processing overhead as well as aggregating more protocols into IP, example: segment routing, more will follow.
the only IPv6 success stories I've heard involved places with datacenters big enough that they ran out of space in 10/8
datacenters big enough to use up a /8 don't use IP routes they use mac routes and multiple overlays over an IP fabric for that reason as well as others. You will not run out of private addresses this way and there is no feasible datacenter design that would fill a /8 while also keeping every address in the same routing domain it literally doesn't happen no one is that retarded.
ipv6 is used because of the routing tunneling and qos features like stated above. It's a tool for actual networks not whatever you think the internet actually is or some small business sysadmin who got filtered by IP and for him it's 1 day job to transition from ipv4 to ipv6 since you don't even need to strip ipv4 in the first place and ipv6 just at the gateways works just fine there's multiple translation mechanisms for this but IT people are dumb and do not read.
ipv6 is so so much more than an address space if you are really a brainlet and think that the current ipv4 block alloc is fine.
I distinctly remember that that was one of the first major "yeah we switched to v6" stories, Facebook in I think it was 2014 or so. They had a slide deck about it. And they specifically said the reason they did it was because they ran out of RFC1918 v4 space in their datacenters. That was the use case, that was the benefit they went to the trouble of a v6 deployment to get, more address space.
(a few years after that I remember some news stories about various places squatting on 11/8, which was assigned to the US DoD but hadn't been announced for decades, so some people just used it as extra private address space. Then one day the glowies announced some routes for it and they got caught with their pants down)
facebook used vpnv4 route vrf's and now uses a BGP EVPN control plane for MAC routes wherein you cannot exhaust addresses there wouldn't be enough devices you could buy to exhaust it.
large datacenters with even a million transit nodes do not share the same routing domain or address space with hosts and generally as a hyper-scalar all hosts do not need to be able to talk to all other hosts so it now makes sense you can use overlapping (not really because they will be tagged with a community, or have a route distinguisher, or something) addresses. I simply cannot imagine what the fuck they were doing to run out of a /8 this seems like incredibly retarded design especially since no one else has this problem not Amazon not Apple not Microsoft not Cloudflare
ISPs only hire diverse tranny engineers now days. As such nobody has figured out how to turn v6 name space into a product mass market product yet.
> 7777:1488:69:JOHN::PC01
Not to mention all of the enduser friendly megaport-esk middleware that still needs to be conceived and developed to implement such things at scale.... Imagine the multicast risks associated with every device having its own v6 address. Boot loaders go brrrp
Because it's shit.
Because many ISPs do not support it, the majority of routers have it disabled by default, routers have barebone IPv6 features and often times some aspects are exclusive to IPv4.
I use IPv6 in my local network but since my ISP does not support it, I cannot have publicly available IPv6 addresses.
>I use IPv6 in my local network
What are the benefits over IPv4 on LAN?
Isn't it just the same but with longer IP addresses that take more time to type?
Also why isn't local DNS a thing?
I want to type "printer.local" instead of its IP address.
>he doesn't know about System32driversetchosts
I use Linux, but the real problem is I have many devices including mobile devices.
>It's called multicast DNS.
I'll look into that, thanks.
>I use Linux
>doesn't know about /etc/hosts
Stop talking, it's only getting worse for you.
>I use Linux
But like a typical Windows user, you know nothing about the OS you use lmao
>why isn't local DNS a thing?
It's called multicast DNS. Your printer just announces its name as printer.local, then everything else knows about it.
>What are the benefits over IPv4 on LAN?
Optimized protocol: unicast, multicast and anycast are way more optimised communication transmissions compared to IPv4's broadcast, they of course are extremely more efficient on the internet, but it's great on a local network too.
>Isn't it just the same but with longer IP addresses that take more time to type?
192.168.1.2 vs fe80::1
Doesn't seem longer to me.
>Also why isn't local DNS a thing?
>I want to type "printer.local" instead of its IP address.
It is, it's called multicast DNS.
It's literally garbage.
Everyone is using ipv6 retard. You think your cell phone is getting a public v4 address?
Mine is.
it's probably not
I do have one
you as well
I had to bump before thread archived
picrel is the ip my phone has currently
you will note something; this is a cgnat private ip. Other times my phone will have a 10.something address this is within the 10.0.0.0/8 private ip address space as well. The ip you see when you go to whatsmyip is the public ip of the mobile gateway not your phone. There is an exceedingly rare chance you have a public mobile ipv4 address.
I forgot the pic lel
Ever heard of CGNAT?
see
>this is a cgnat private
it's within 100.64.0.0/10, so, yes
>see
meant to quote
The tech sector is currently overrun by a bunch of nepo hires and retards so nobody is updating their network infrastructure because the people they hired genuinely have no idea how to do so
Boomers who think every device having a public IP address is a security issue, when they could just tweak the firewall on their routers. They also think it's a privacy issue because they still believe IPv6 addresses include portions of their MAC addresses when there is the privacy extension and many devices spoof their MAC already without doing anything. Then, they may think that ISP's DHCP server will keep giving them the same IPv6 when that's the same for IPv4 for a long time now, and just requires a MAC spoof on the interface connected to your ISP and restart of your DHCP client. Some may think it's hard to remember addresses due to them being so long but you're supposed to be using DNS.
Every computer in the world does not need to talk to every other computer in the world. That's not actually how the internet works.
Because IPv4 NAT was invented.
To anyone denying this, do you honestly think we wouldn't have had fully deployed IPv6 by now if it wasn't for NAT? Of course implementations would have been a lot better. (IPv6 only is a shitty experience in most modern OSes today. Try.)
NAT is just a cope for smooth brained boomers who don't know how to configure their firewall. IPv6 only on modern OSes works just fine unless you have fucked something up. The only way it'd suck is if your favorite websites are IPv4 only and you also want to use HTTPS since NAT64+DNS64 will fail here.
>NAT is just a cope for smooth brained boomers who don't know how to configure their firewall.
I see you wasn't around before NAT was a thing.
>IPv6 only on modern OSes works just fine unless you have fucked something up.
I'm talking about disabling IPv4 completely, including DNS. Getting it to work by tinkering doesn't count.
I'm almost thirty, so you must be the boomer I'm talking about. Also, NAT64+DNS64 is not tinkering. It's a late stage IPv6 transition technology, which is also used by many corporations for example Huawei. It just won't work when you want to access HTTPS for an IPv4 only host unless you got your CA to add IP: 64:ff9b::yourIPv4inhex in your certificate probably (haven't tried)
>NAT64+DNS64 is not tinkering.
That's besides the point. OP was asking why "people" (normies) aren't using IPv6. What corps and nerds are doing to get it to work is another conversation. I myself dropped out when my ISP stopped supporting it. (IPv6 caused support tickets from normies due to shitty sites and equipment, and almost no-one cares about IPv6.)
But people are using it, they just don't know it because their carrier gives them IPv6+IPv4 addresses, and everything seamlessly runs on dual-stack.
>but how does IPv6 get any traffic if IPv4 is already on?
"happy eyeballs"
lol it's a child that thinks it's an adult. amazing watching the newer gen grow up so uneducated on basic computing. just on adoption alone you'll never make it in this business. go flip burgers this will be far more useful than the maintenance routine jobs designed for mid-level minds you are currently destined for.
>IPv6 caused support tickets from normies due to shitty sites and equipment, and almost no-one cares about IPv6
xe can't overcome this because they see this issue as their fault and correspond that to "their problem". that is why I recommended flipping burgers vs. opening a burger stand.
We should strike your pension and see how you deal with that, boomer.
Not a boomer. I'm also smart enough to know boomers often know their shit. Frustrated drones couldn't understand that. LOL ...
Sure, but you don't know what other people are talking about. That's a (You) problem and you don't seem capable of taking up the challenge.
All you've done is make excuses and seethed about your dad.
Then you shouldn't have any issue when we collectively strike all pensions, right?
Other people are being refuted and educated by me. You're just in denial. Go take your meds grandpa. If you were around before NAT then you are a boomer. You haven't even really said anything about anything.
I know exactly what I'm talking about.
>IPv4 NAT
You must love solving captcha.
it has no NAT, meaning all your insecure IP cams are wide open for scanning skiddies to attack. I've disabled IPv6 on my entire network, until it's more mature
Firewall issue retard
This is what link-local addresses are for and also firewalls.
>meaning all your insecure IP cams are wide open for scanning skiddies to attac
hasnt this always been the case
Good look scanning IPv6.
With a typical consumer broadband connection you can scan the whole IPv4 space in a couple of days, but scanning the whole IPv6 space would take billions of years. Also, even if you could scan it in a day it would be useless for anything connected to a router newer than 5 years because of changing addresses.
NAT isn't a security feature, retard. Learn how to write firewall rules.
>implying skiddies can't 0 day your router
I am retarded and replied to the wrong post, see
My ISP literally doesn't support it.
don't care, still not using it because i don't like the syntax.
Less human-readable addresses
A bit overcomplicated networking system
ipv4 werks
NAT exists
If it ain't broke don't fix it.
literal brainlets.
Once Amazon starts charging $90 a year for a single IPv4 they'll have no choice.
Says you.
>cope
IPV4 cucks would rather beg DoD to share some of their 175 million IPs stash than learn base 16.
And there's nothing you can do about it.
Absolutely nothing.
ipv6lets would rather learn base 16 than just extend ipv4 to be like 1.2.3.4.5.6.7.8 with existing ipv4 addresses being 0.0.0.0.1.2.3.4
You act like any of us are in the IETF and can do anything about it. Retard.
>>>
They had their shot, but they didn't use it. Now IPv6 is destined to land on the same heap of scrap that were CDs, DVDs, and Silverlight.
Pajeet solution.
Wouldn't adding just a single extra byte have been enough?
1 trillion IP addresses instead of 4 billion - 100 for every person in the world.
The solution is to nuke Africa
No significant contribution and frees a good couple million addresses
Honestly this is a pretty good solution
Why don't they make this the ipv6 standard?
Serious question
Elaborate
I don't think that explains anything
That's a shame.
boy you are describing a result, not the reason
and you know that
Quints have spoken.
An example address would be
82.50.27.73.34.85.45.88
Isn't really better than IPv6
IPv6 is overkill at this point but at least we can stay on it forever
>What is ::ffff:1.2.3.4
Trash.
It's trash, and only trash people use trash.
::ffff: is shorter than 0.0.0.0. and also easier to type.
What are the ffff for?
>0.0.0.0.1.2.3.4
>::1.2.3.4
https://en.wikipedia.org/wiki/Reserved_IP_addresses
That is how IPv6 displays IPv4-mapped addresses.
Enjoy your higher ping and higher number of internet routing hops then.
Getting higher than zero numbers isn't difficult. Being used in the first place, now THAT is difficult.
You got used to 192.168.x.x, 255.255.255.x or whatever other standard IPv4 address/subnet mask, you will eventually get used to the IPv6 ones once ISPs will start charging for IPv4 addresses, AWS has started doing just that.
No, I don't think I will. Because I'd rather quit.
>so ffff
>no, wait, a colon
>no, wait, a double-colon
>then ffff
>then a double-colon
>no, wait, no colon
>no, wait, a single colon
How does one manage to convey so much brain damage in a single post such as yours?
It is very easy if you know the 3 rules of IPv6 compression.
Or I can simply continue using IPv4.
Midwits hate ipv6 because they can remember 4 bytes but remembering 16 bytes is too complicated for them
>And there's nothing you can do about it.
>Absolutely nothing.
>ipv6lets would rather learn base 16 than just extend ipv4 to be like 1.2.3.4.5.6.7.8 with existing ipv4 addresses being 0.0.0.0.1.2.3.4
Yes.
I will once my ISP gets its shit together, that means never.
>no answer with regards to the pensions
I'll take that as a "yes".
Cause for the home there's no point. A typical class C IP can handle 253 devices + 1 reserved for the router. I daresay that's a lot of IP cams and other shit Then lets say off chance you do go beyond that. They make a private Class A and B range you know. You can run whatever IP range you wish at home. Long as it's valid and private. Your ISP don't give a fuck. All the world sees is whatever your IP is the ISP gives you.
it's 253 devices total so 252 if you reserve one for the gateway.
a class c /24 has 2 unusable host addresses, the top most broadcast bit and the bottom most network address bit.
There's a lot of really retarded opinions in here functionally ipv6 can work the exact same as ipv4 it just has more overhead the fields are basically all the same and some of the trivial ones removed like total length, slaac is a bit dumb and Router advertisement only working on /64 networks is also dumb but ipv6 was made to be literally as handholding as possible because the IETF knew two things
1 - the state of building computer networks should not continue to be stuck at having techs fuck with dumb shit like dhcp and basic routing it should work out of the box which is a good idea
2 IT people are retarded and lazy and were already filtered by IP(v4) so v6 should handle itself except at the carrier or campus level, which it does
surprise surprise everyone is still filtered and can hardly understand IPv4 as is anyways.
The real reason ipv6 is dead on arrival is that IT is a brainlet industry and is far too important and far too easy to get into the level that people need you at generating far too many mundane workers to move the industry en masse. Very very few of IT types are even remotely close to understanding how computer networks actually work much less protocol design your dumb little kubernetes containers that you struggle to get talking to a single gateway much less 2 isn't hard and you got filtered by TCP flow control in technical reading of course you wouldn't understand the concequences of Router Alert or Next Header.
>Cause for the home there's no point
parsing ipv6 headers is less computatonally intense than ipv4 thanks to better byte alignment and a lot less bit twiddling one needs to do in order to parse the header
another benefit of ipv6 is that you eliminate a whole class of cache coherency bugs by being able to randomly pick an internal address and have virtually zero probability of collision
almost every competent sysadmin will use ipv6 internally (unless there is a huge amount of legacy cruft forcing you on IPv4) and set up an ipv4 ingress
IPv6 is being adopted at a steady rate of ~5% per year according to google.com/ipv6
As more boomer IT admins retire and are replaced with freshly graduated minds who aren't terrified of new technology.
Anyone still advocating for IPv4 is an ignorant retard. Case in point:
homosexual doesn't even know the IP class system for IPv4 has been obsoleted many years ago. How can one expect retards like him to know anything about IPv6 when they don't even know IPv4?
It is being used and it is much more commonplace then you think. IPv4 is slowly but surely being regress to exist only in legacy networks and equipment
I HAVE NO DAMN CLUE HOW DOES IPV6 WORK, BUT PLEASE USE IT. I LOVE BYPASSING RATELIMITS WITH TUNNELBROKER IPV6 TUNNELS!!!
LULZ will never support it again so I don't care about it
It actually is used though. It's not 10 years ago.
Because ISP rather put everyone under CGNAT
I once heard it put like this:
>the only benefit of IPv6 is the larger address space
>a larger address space only benefits people who don't have addresses, that is, people who aren't on the internet
>all the work of deploying it has to be done by people who already have addresses, that is, people who are already on the internet
>so you're asking tons of people to do a bunch of work that doesn't benefit them for the benefit of other people
>oh and the people who aren't on the internet but want to get on want to reach the people who are already on the internet. that is, they want to reach IPv4 addresses, so a v4 address is valuable to them and a v6 address mostly isn't
I read that in a sysadmin's blog post like a decade ago when people had already been banging the drums for IPv6 for over a decade and were dismayed that it wasn't ubiquitous. And here I am without working IPv6 in 2023 and I've never noticed a problem with it. This'll keep going for several more decades I bet.
>the only benifit of IPv6 is the larger address space
already wrong and this is why I said the IT crowd aren't nearly as smart as they think they are it's just the new generation of skilled labor and not actual intelligence. ipv6 has a ton of benefits over ipv4 people are just dumb and only focus on global addressing which is still a point of contention but sysadmins are lazy so they pretend its not an issue.
The value provided by the flow label alone for developers or large network operators like campus size networks is huge, having to classify traffic at ingress on network devices is always a point of pain instead imagine just being able to set a flow label, also next header completely solves packet length processing overhead as well as aggregating more protocols into IP, example: segment routing, more will follow.
datacenters big enough to use up a /8 don't use IP routes they use mac routes and multiple overlays over an IP fabric for that reason as well as others. You will not run out of private addresses this way and there is no feasible datacenter design that would fill a /8 while also keeping every address in the same routing domain it literally doesn't happen no one is that retarded.
ipv6 is used because of the routing tunneling and qos features like stated above. It's a tool for actual networks not whatever you think the internet actually is or some small business sysadmin who got filtered by IP and for him it's 1 day job to transition from ipv4 to ipv6 since you don't even need to strip ipv4 in the first place and ipv6 just at the gateways works just fine there's multiple translation mechanisms for this but IT people are dumb and do not read.
ipv6 is so so much more than an address space if you are really a brainlet and think that the current ipv4 block alloc is fine.
I distinctly remember that that was one of the first major "yeah we switched to v6" stories, Facebook in I think it was 2014 or so. They had a slide deck about it. And they specifically said the reason they did it was because they ran out of RFC1918 v4 space in their datacenters. That was the use case, that was the benefit they went to the trouble of a v6 deployment to get, more address space.
(a few years after that I remember some news stories about various places squatting on 11/8, which was assigned to the US DoD but hadn't been announced for decades, so some people just used it as extra private address space. Then one day the glowies announced some routes for it and they got caught with their pants down)
I found it but I cannot imagine what the fuck they were doing to run out of a /8
https://atscaleconference.com/videos/a-history-of-ipv6-challenges-in-facebook-data-centers/
facebook used vpnv4 route vrf's and now uses a BGP EVPN control plane for MAC routes wherein you cannot exhaust addresses there wouldn't be enough devices you could buy to exhaust it.
large datacenters with even a million transit nodes do not share the same routing domain or address space with hosts and generally as a hyper-scalar all hosts do not need to be able to talk to all other hosts so it now makes sense you can use overlapping (not really because they will be tagged with a community, or have a route distinguisher, or something) addresses. I simply cannot imagine what the fuck they were doing to run out of a /8 this seems like incredibly retarded design especially since no one else has this problem not Amazon not Apple not Microsoft not Cloudflare
>I cannot imagine what the fuck they were doing
¯_(ツ)_/¯
the only IPv6 success stories I've heard involved places with datacenters big enough that they ran out of space in 10/8
fuck remembering those addresses
don't start talking about dns that shit breaks all the time
2001:4998:44:3507::8001
This is the type of address we will need to remember.
ISPs only hire diverse tranny engineers now days. As such nobody has figured out how to turn v6 name space into a product mass market product yet.
> 7777:1488:69:JOHN::PC01
Not to mention all of the enduser friendly megaport-esk middleware that still needs to be conceived and developed to implement such things at scale.... Imagine the multicast risks associated with every device having its own v6 address. Boot loaders go brrrp