[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / s / t / u / v / vg / vm / vmg / vr / vrpg / vst / w / wg] [i / ic] [r9k / s4s / vip] [cm / hm / lgbt / y] [3 / aco / adv / an / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / pw / qst / sci / soc / sp / tg / toy / trv / tv / vp / vt / wsg / wsr / x / xs] [Settings] [Search] [Mobile] [Home]
Board
Settings Mobile Home
/g/ - Technology

Name
Options
Comment
Verification
4chan Pass users can bypass this verification. [Learn More] [Login]
File
  • Please read the Rules and FAQ before posting.
  • You may highlight syntax and preserve whitespace by using [code] tags.

08/21/20New boards added: /vrpg/, /vmg/, /vst/ and /vm/
05/04/17New trial board added: /bant/ - International/Random
10/04/16New board for 4chan Pass users: /vip/ - Very Important Posts
[Hide] [Show All]


[Advertise on 4chan]


File: shutterstock_85570423.jpg (103 KB, 1000x667)
103 KB JPG
Why not ipv6?
>>
>every device on your network has its own public IP address and can be pinged from the internet so it needs its own software firewall
No thanks, I'll stick with my comfy ipv4 address, have fun being trivially tracked per device by glowies and websites
>>
>>109648287
>so it needs its own software firewall
This is already handled by your router, retard. Unless you have an ISP for each device you're using.
>>
>>109648336
Wrong, why did you make a thread about ipv6 if you don't even know how ipv6 works?
>>
>>109648287
>have fun being trivially tracked per device by glowies and websites
Source?
>I'll stick with my comfy ipv4 address
You are doubting that you can be tracked by glowies because of ipv4? Ipv4 is old tech, the older it is the more deprecated it gets, the more vulnerable it becomes.
>>
File: covergn.jpg (67 KB, 800x450)
67 KB JPG
>>109648411
>the older it is the more deprecated it gets, the more vulnerable it becomes.
>>
it's called a router because it routes, it should selectively not route and drop unsolicited incoming packets unless a port forwarding is setup.
>>
>>109648424
Alright retard you are not gonna provide a source about tracking per device or you just gonna be retarded?
>>
File: OIP-2785285264.jpg (14 KB, 474x155)
14 KB JPG
You rang?
>>
>>109648479
>port forwarding
NAT concept
ipv6 doesn't use NAT
>>
>>109648343
He's right. You can easily test this yourself from a VPS:

On your computer run:
python3 -m http.server -b ::


On a laptop or phone run:
curl -I http://[XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX]:8000


You should see a request come in from it:
Serving HTTP on :: port 8000 (http://[::]:8000/) ...
XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX - - [26/Aug/2026 05:09:40] "HEAD / HTTP/1.1" 200 -


Now ssh into your VPS and run the same command:
curl: (7) Failed to connect to XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX:XXXX port 8000: Connection refused
>>
>>109648840
Great way to access my home network securely without crap like Tailscale or a full blown VPN.

I use NAT64 (Tayga) and DNS64 and a HTTP proxy over Yggdrasil. Self-hosted sites I run go direct via Yggdrasil instead of the proxy.
>>
File: lMobmqk.png (102 KB, 462x306)
102 KB PNG
>>109649263
I went out of my way to disable ipv6 on my router and on all of my devices, so I can't do all that, sorry.
>>
>>109649283
>Goes out of his way to disable shit over a problem that isn't real
Also, regarding ping you can block / filter ICMPv6 in the firewall too as long as you allow the control messages necessary for Path MTU Discovery:
https://en.wikipedia.org/wiki/Path_MTU_Discovery

You can block those control messages too but you'll potentially cause connectivity problems if there's MTU issues caused by a larger / smaller MTU.

The only ICMPv6 messages that are absolutely required are those for neighbour discovery so the router advertisement works.
>>
>>109648274
>>109648842
Do isp’s give out unlimited ipv6 addresses for free?
>>
>>109649315
My ISP won't give me any ipv6 address.
>>
>>109649302
I mean, I COULD read that wikipedia page and do all that. Or I could just continue using ipv4+NAT which works on my machine.
>inb4 you start making up desperate lies like "you're gonna get hacked" like this anon did >>109648411
>>
>>109649315
They give you a /56 prefix in most cases. That gives you 256 individual /64 subnets to use for different VLANs, etc. That gives you a total of 4.72 sextillion addresses according to Gemini.

If that's not enough for you then you can get /48 prefix with most business Internet connections which allows you to make 256 different /56 prefixes.
>>
>>109649330
This is stupid. The glowies can track you regardless of your Internet Protocol version.
Your ISP knows who owns what address and will gladly share that information with them. If you're behind CGNAT there is even a record of every IP address : Port mapping probably with extensive auditing kept.
>>
File: yz57WRm.png (237 KB, 447x393)
237 KB PNG
>>109649350
How I feel continuing to use ipv4 forever while seething glowies throw temper tantrums and kick and scream to try to convince me to switch to ipv6
>>
>>109649375
Seriously, look up how your beloved NAT actually works and pay special attention to that IP : Port mapping. Your router is probably already pozzed and if you want to avoid it the solution isn't to disable the Internet its to layer additional protection (like Tor or I2P or a VPN, etc) on top.
>>
>>109648274
It's already overtaking IPV4 in usage [in India]
>>
>>109648825
It is you who is stupid.
It’s not deprecated in the way software code might be. IPv4 is simply a numbering scheme.
In many ways it’s likely stronger than ipv6 for anonymity due to the layers of NAT (assuming ip schema is the only thing involved, which it isn’t).
>>
enjoy your free public ipv4 address while it lasts.
pay extra for static IP or free CGNAT is inevitable.
>>
>>109649563
NAT weakens your anonymity because of all the connection tracking involved. Just recently there was a massive vulnerability in this so called "security" mechanism that so many luddites insist is airtight.
https://natjack.io/
>>
>>109648274
I can type dotted decimal quickly on a num pad but hex addresses suck to write
>>
>>109649684
You shouldn't be typing any addresses period. Copy paste them or use DNS.
>>
For me, it's
https://stats.labs.apnic.net/ipv6/AS5650

>But that says 20%
I'm in the 80%, apparently.

>>109648287
>have fun being trivially tracked per device by glowies and websites
Either would find browser fingerprints more effective.
https://amiunique.org/fingerprint
https://coveryourtracks.eff.org/

>>109649315
In four words: Literally, no. Practically, yes.
>>
>>109649703
>You shouldn't be typing any addresses period
You do understand someone has to configure these networks from scatch at some point right?
And that it's entirely possible to remember IPv4 addresses from memory so you don't need reference material to copy from?

I don't need to bring up my home documentation just to configure a basic network, but for IPv6 I would need to. I can tell you the static IPv4 addr. of every device in my network but couldn't tell you a single link-local IPv6.
>>
>>109649976
IPv6 auto-configures. You've done something very wrong if you're needing to manually type in individual link-local addresses.
>>
>>109648274
Because it's worse for privacy compared to NATted v4. Not much worse, but still worse.

And IPv4 just werks. No reason to change. Nothing will work better if I start using v6.
>>
>>109650005
You can NAT6 your IPv6 too if you really want to, but when NAT is vulnerable to flaws like >>109649651 I'm not sure I'd call it an improvement.
>>
>>109650011
>You can NAT6 your IPv6 too
Yeah, but why?
IPv6 doesn't have any benefits for me.
>>
>>109648274
Because I actively prevent its implementation everywhere I can
t. sysadmin
>>
>>109650030
It gives you Internet and more addresses than the single /32 prefix your ISP gives you which means you can't even self-host multiple servers behind the same port number.

It's basically the same thing IPv4 gives you but better.
>>
>>109650037
Are you too old to learn a new 30 year old protocol?
>>
>>109649651
>attack requires attacker having privileged access to your LAN
oh wow, a nothingburger
>>
>>109650071
That's not a nothingburger. Some retard with a laptop could easily bring down your Starbucks Wi-Fi with this.

There's actually a lot of scenarios where untrusted devices sit behind your router, public Wi-Fi being the most obvious but even someone being able to plug into an Ethernet port somewhere could wreak havoc.
>>
>>109650089
Many ISPs these days also turn on "public" Wifi on your gateway without your permission to provide hotspot services to other customers too. They could be affected by this.
>>
>>109650089
Having privileged access to someone's LAN is the vulnerability, not some specific vector with a logo, a cool name and an .io domain. At that point even IPv6 won't save you.
>Some retard with a laptop could easily bring down your Starbucks Wi-Fi with this.
Go ahead and try it yourself.
>>
>>109650134
Providing public services for people to use the Internet is a thing that exists though. You can call that a vulnerability but that's the entire point of the service.
Hell, every consumer slop router out there comes with a Guest network feature and your "Guests" could exploit this.
>>
>>109650147
>Providing public services for people to use the Internet is the same as giving them privileged network access
I'd call you a retard and tell you to GTFO, but it wouldn't be an IPv6 thread without people like you.
>>
>>109650178
You'd literally not be affected by this shit if you deployed IPv6 mostly networks with NAT64 to provide IPv4 compatibility.

You can keep burying your head in the sand if you want to though.
>>
>>109648274
probably useful for people on IPv4 CG-NAT. Otherwise IPv4 has enough space to last another 100 years at least.
>>
IPv6 means you get the same pattern address even if you move around to other places because it uses the MAC address to generate it. No thanks.
>>
>>109651194
That's not in use anymore. Besides, if for some reason you're paranoid about leaking your MAC address to "other places", you'd spoof it, regardless of whether the network is IPv6 enabled or not.
>>
File: holding-a-duck.jpg (343 KB, 896x1152)
343 KB JPG
>>109649263

this shows that IPv6 is perfect for tracking devices.. profit? for whom?

from a common sense IPv6 is an additional layer of complexity. its mission premise was to expand IP range to allow freemen to establish own IP and be accessible from any part of the network. basically this example shows it is not true - it cannot become accessible from any part of the network, but it can be related to a device for user tracking.

so IPv6 mission is failed. few big actors naturally accumulate IP ranges and thats it, the need of expanding is fake, mission failed.

why IPv6 is being pushed? beside tracking? probably some retards want to create an illusion of activity and get funds, budget etc.
>>
>>109650095
I did a quick google search, it seems it's only for fucking comcast, how can a company like comcast be so fucking corrupt and disgusting and still exist.Same company that has a contract to install flock cameras.
>>
https://www.youtube.com/watch?v=2k8Hr5Aw73o

IPv6 is watching you pee.
>>
>>109648842
Totally incorrect. Outbound NAT and ULA addresses do exactly this.
If people used IPv6 outside of a poorly configured home lab, they'd know and understand this (and would stop repeating rubbish like it's the truth).
>>
>be me
>set up a wireguard server in my lan for remote access
>use ipv6 because globally routable means I dont have to open up a port on my stinky ISP router webpage
>try to test with my phone
>doesn't work
>spent two hours troubleshooting
>claude tells me sim network commonly cant route ipv6
>go to phone settings, see ip, try some ipv6 test websites, some pings
>it's true
>mfw
>>109648287
You get a temporary ipv6 address, at least on loonix. The kernel will use that one if it's available
>>
>>109651606
what the fuck are you talking about? who are you stupid fucking niggers shitting up /g/?

IPv6 is as much attached to your identity as any IPv4 addr. A court order will compel and ISP to take the source port : ip <- CGNAT -> (You), just as much as an IPv6 source port:ip tuple.
>>
>>109649263
even if you don't filter inbound, don't run software that is unauth or a security issue on your fucking computer. this default behavior in upstream firewalls is unironically a cancer in it of itself.
>>
>>109651606
the entire asia-pacific region ran out of new ipv4 addresses to support chinese and indian internet growth. only way to get ipv4 addresses now is paying money to trade for them and also taking them from the north american region.
>>
>>109652174
you sure your router or computer isn't breaking ICMPv6 traffic? You cannot filter pings and what-not for a functional IPv6 device. A lot of retards default to blocking stuff like pings for IPv4, which is "ok" but completely wrong in v6. most people with issues in reachability have a broke shit.
>>
>>109652234
or slapping more layers of NATs in front of everything.
>>
>>109648274
Because I like to samefag on here.
>>
>>109652235
I tried to ping google ipv6. My wireguard works fine on another landline, it's really just the cellular network
Thankfully my ISP doesn't use cgnat so I get a public ipv4. I still opened up a port and use ddns with njalla to write A and AAAA so my clients just end using whichever they want (so ipv6 if it's routable)
>>
>>109648287
>>109650005
>>109651194
>>109651965
https://en.wikipedia.org/wiki/IPv6_address#Temporary_addresses
>>
>>109648287
Give your router an ipv6 address, keep ipv4 for your internal network and just NAT everything like you did before. But honestly if you rely on NAT to not be tracked, I have some bad news for you.
>>
>>109652248
that's also a given.
>>
>>109652174
Yes, this is annoying. It's the soul reason I have a VPS running Amnezia Wireguard that acts as a bouncer that routes to my home over IPv6.

It would be really fucking nice if I could skip that but it's not practical when many shit network operators are 30+ years stuck in the past still.
>>
>>109652297
>so my clients just end using whichever they want (so ipv6 if it's routable)
If you're using the Wireguard Android app it always uses IPv4 in a dual-stack network by the way. It has no logic to switch between them which is annoying.

When you do your DDNS I recommend making different subdomains like:

>v4.<YOUR_TLD>
Only A records
>v6.<YOUR_TLD>
Only AAAA records
>ds.<YOUR_TLD>
Dual-stack (both A and AAAA records)

That way you can force a protocol just by updating the Endpoint address.
>>
>>109652349
NAT babies don't know how to configure a network. This is why they talk about things like "default configurations" or "misconfigured networks" because they lack the ability to configure a network.

Probably, their router shouldn't even be allowed to connect to the IPv4 Internet because they're a danger to themselves, never mind IPv6.

This is why their solution is always:
>I don't understand this
>I'm going to turn it off
>Wait, I don't understand IPv4 either
>Well, shit. I'll keep that turned on though.
>>
File: 1781515443939682.jpg (36 KB, 800x450)
36 KB JPG
ugly addresses
simple as
>>
>>109654474
::1 is prettier than 127.0.0.1 if you ask me.
:: is also better than 0.0.0.0
I'm not same gay tranny that cares about the looks of an address though
>>
>>109654474
harder to memorize too
>>
>>109654530
Meh, as long as you remember your ULA range its no different.

I have my DNS resolvers on:

<ULA RANGE>:53::1
<ULA RANGE>:5353::

And yes, I do remember that.
If you dual stack you can even make the last nibble of the address the same as what you get assigned in IPv4 so:

192.168.0.2 -> <ULA_RANGE>::2
>>
There are now more people on the internet now than IPv4 address. The actual number of usable IPv4 is around 3 billion due to reserved ips and ones hoarded by the DoD and Amazon. You're either (CG)NATed to hell or using IPv6.

If you look at ip address exchanges a lot of legacy ip holders are either cashing in or renting their ips to botnets. Wikipedia and 4chan have so many rangeblocks on IPv4 due to overcrowding it's not funny.
>>
>>109654572
This also maps cleanly if you have web services that need dedicated addresses too, for example I host some different web applications on my OpenWRT router that each has a dedicated address per-app.

I use 192.168.43.0/24 for IPv4 and <ULA_RANGE>:443::/64 for IPv6 so the first web app has:

192.168.43.1 -> <ULA_RANGE>:443::1
And so on and so fourth.

Its not so difficult as long as you have a convention you follow.
>>
>>109652248
Eventually you run out of NAT ports. Imagine CGNAT is a bus, you have 1 licence plate number for 50 passengers compared to 4 for a car. But eventually you run out of road and need to upgrade to an IPv6 train.
>>
>>109654695
You run out almost immediately when your ISP assigns you a single /32 that you NAT and you then want to host multiple web servers behind that on port 80 and 443.

Oops, can't do that, now you need an inefficient load-balancer / reverse-proxy doing SNI inspection to route to the right place and pray to God it never breaks.

With IPv6 you can just route directly to the server on port 80 or 443 because you have so many addresses to use.
>>
>>109654527
>::1 is prettier than 127.0.0.1 if you ask me.
Use 127.1 then
>>
>>109655089
how do we feel about using 2130706433
>>
>>109655089
I think you're joking, but IPv4 addresses require all four components so you can't shorten them like that, even though 127.0.0.0/8 literally wastes 16,777,216 addresses when everyone just uses 127.0.0.1.
>>
>>109655128
>IPv4 addresses require all four components
They do not. See >>109655108 for Chaotic Evil IPv4: pure decimal form. 127.1 = 127.0.0.1.
>>
>>109655146
Oh, neat, you're right. I didn't you know you could collapse the zeroes in IPv4 too in the same way IPv6 does.

::1 is still shorter though.
>>
>>109648274
https://canvas.openbased.org/
>>
>>109655166
Also, it's still a huge waste of addresses. They should never have assigned that many addresses to the loopback that can NEVER be used for anything else.

IPv6 at least got that bit right by assigning a single /128 address and no more.
>>
>>109655166
It's among the more pointless features of the textual v4 address representation.

>>109655179
If you think that's bad, check out 240.0/4 which is assigned to be nothing, forever.

IPv4 was supposed to connect a handful of universities across the US. We've been driving a clown car.
>>
>>109652234
>the entire asia-pacific region ran out of new ipv4 addresses

you cannot run out of IPv4 addresses, they multiplex via NAT

>to support chinese and indian internet growth

that is funny, what else should i support? Iranian islamic regime growth? btw they can support themselves with IPv6 as much they want, they can even disappear, i dont even know which internet sevices there could be of interest. Alibaba? maybe. WeChat? unlikely. unlikely they run out of IPs
>>
>>109655503
>you cannot run out of IPv4 addresses, they multiplex via NAT
That's not how that works at all. You can't NAT yourself out of a problem forever. NAT doesn't scale. There's a hard number of how many subscribers you can put on a single NAT box before the thing falls to shit and you need to deploy another one.
>>
>>109648274
Because we have subnetting
>>
>>109648274
i know i should use it but it's ipv6 is super busu so i'm not going to use it
>>
>>109655604
Have fun subnetting your /32 prefix to all of the devices in your home with a subnet mask of 255.255.255.255
>>
>>109655571
>NAT is not Web Scale.
>>
>>109655503
NAT fucking sucks for gaming or making a private network
>>
>>109648274
usecase for IPv6 when I don't get a dedicated internet-routable address (like the faggots who pushed this retarded bullshit promised) anyways?
>>
>>109655620
It's not anything scale. It's a hack. If I have to deal with NAT then I'd rather it be NAT64 routed over IPv6.
>>
>>109655636
End-to-End connectivity so your routed prefix from a VPS over Wireguard to get you that real connectivity works better without having to traverse a NAT.
>>
>>109655654
Wait, Wireguard over IPv6 does not use its own routing table?
>>
>>109655661
You can use Wireguard to push a prefix from your VPS to your actual router or server so if your ISP doesn't route inbound properly then you can essentially use the VPS like a bridge.

There's less connection tracking and state involved over IPv6 because it's fully end-to-end (no NAT) so it's theoretically a bit more efficient because of that end-to-end connectivity.
>>
>>109648274
>Why not
All the exploits that it brought along. Hell, the PS5 has a kernel-level IPv6 exploit.
>>
>>109655702
The PS5 runs FreeBSD by the way. Sony should have used a real operating system like the one that runs the Internet.

There have been kernel exploits in IPv4 over the years too though so that's not an argument. It's just that IPv6 has had less attention to it because faggots still cling to NAT boxes thinking that will solve all of their problems (It won't. You've been sold a bridge by some vendor that just wants to extract money from you)
>>
>>109655571
IP is solved. move on.

>>109655623
gamers are temporary https://www.youtube.com/watch?v=hqzY27r3DK8 they dont live long. even private VPN networks made for them dont live long. IPv6 lives long because it once infected some gov institutions around the globe. otherwise it would die out as Hamachi.

now justify why Hamachi shouldnt die out.. it is easier to just connect directly (IPv4 is simple) instead of creating private network to satisfy the game design (IPv6).

you may percieve this example wrong, it only says that people dont stick around games, maybe few of them who could go into sophistication of making private network or pay for a dedicated game server. beside, why paying is bad? you want own game server so much? pay for it, right, buy Hamachi (or how is new thing called) at least.
>>
>>109655800
How do you connect directly when you're behind a CGNAT?
>Just buy Hamachi
More snake oil. This is a workaround not a solution.
>>109655800
>IP is solved. move on.
If it were solved it wouldn't require these hacks.
>>
>>109655827
By the way, it's very funny to me that gamers worship low-latency network connectivity whilst simultaneously being against improvements like IPv6. They would rather traverse a million different NATs and STUN/TURN servers, etc, then have simple, direct, low-latency network connectivity they can use to play with their friends.
>>
>>109655827
You're arguing with a namefag. You should know they're all retarded by now. Especially this one is known for it.
>>
>>109655800
>it is easier to just connect directly (IPv4 is simple

you. cant. do. that. under. NAT. or. CGNAT.
>>
>>109655856
Gaymers, presumably just want their gaymes to work. I don't think IPv6 is the default protocol for majority of them and I'm pretty sure Steam literally doesn't even support IPv6.
>>
>>109655953
Steam is just bad for that in general but there are games that do support IPv6. You can run a Minecraft server using IPv6, for example.
>>
IPv6 sounds incomplete. I will just wait until it is production ready and hang back on IPv4 for the next few decades.
>>
>>109656001
I'm sure if you cherrypick hard enough, you'll find games that support IPv12, too.
>>
>>109656012
It's over 30 years old and has been used extensively in production since 2012:
https://en.wikipedia.org/wiki/World_IPv6_Day_and_World_IPv6_Launch_Day

The longer you wait though the more the value of all your IPv4 addresses increases by. I think that is the main reason we have not seen bigger adoption yet (although, on google.com they do see over 50% IPv6 usage on some days now).

There's no technical reason, it's political and economical. People are playing a game of chicken to see how high the cost of an IPv4 address can go up by before they dump them.

>>109656017
Yes, there are games that don't support it that should. Gamers should be able to see the benefits direct connectivity brings. There's so much FUD around this subject it's unbelievable.
>Low ping good
>High ping bad
Gamers all know and understand that, yet they will still recommend disabling IPv6 for NATs despite the fact this is clearly less efficient.

If gamers weren't all fucking retards they would be pressuring all of the companies to adopt it for its obvious benefits instead of telling everyone to turn it off.
>>
>>109655713
>there have been exploits in this mature, tested, stable technology
>therefore we need to go to this unnecessary "solution" that's 500 times as complex
>it just has more exploits because it didn't get as much attention is all!
>>
>>109655827
>How do you connect directly

you buy a game server OR cheaper, you buy a VPN connection into a dedicated private network. do you have a reading/listening comprehension? same as this retard >>109655947 ?

>This is a workaround not a solution

close eyes, its not there. IPv6 push. push. yeeek yeeek, blump! its there. in the closed waters. hehehe.

IP is solved. you do no worth in advocacy for complication. complication leads to retardation.
>>
>>109656516
If you add up all the exploits ever found in IPv4 ever it will be way more than what's been found in IPv6 stacks.

Faggots like you that immediately take the "time to burn it all down because it had one bug that time" approach are the reason nothing ever progresses.
Meanwhile, you didn't ever dare think to do the same to IPv4 every time a bug was found in the network stack somewhere.

>>109656560
>How do you connect direct
>Just use this intermediary
You're not very smart are you?
>>
I despise everyone against ipv6.
It just shows a certain lack of intelligence. None of their arguments ever amount to anything. The "I don't like the addresses / can't remember it" types are the worst.
>>
>>109656054
I'll be fine. IPv4 has at least another 150 years left. I disable IPv6 on all my servers too. I try to give it back but they never want it so it must not be that great.
>>
>>109650005
> Nothing will work better if I start using v6.
Retard alert.
>>
>>109648274
>ISP pushes ipv6
>half of my mp vidya stops working
>swap back to ipv4
>>
>>109656799
dual stack
i'm sorry that your ISP is a retard tho
>>
>>109656799
How did it just "stop working"?
>>
>>109656568
>You're not very smart

ofc youre not smart, youre a dumb IPv6 shill. everybody knows what game server is - a direct connection to a dedicated server. you want to become a server yourself? buy an IP. you cannot? hehehe, hecking poor retard.
>>
>>109656896
>Just buy an IP
You're still not very smart, there aren't enough to buy which is why we have NAT and CGNAT, and if you need more than one IP address, then what? In many cases your ISP won't be able to provide this to you.

They can provide IPv6 though.
>>
>>109656560
>directly
>you buy a game server OR cheaper, you buy a VPN connection into a dedicated private network

brain dead
>>
>>109656001
steam networking opportunistically uses ipv6 though
>>
Stop saying "though".
>>
>>109656896
this just means you will have to rely more and more on corporations.
you cannot just host a game server in your house anymore. or anything.
>>
>>109656914
>there aren't enough to buy

yes, youre poor and stupid shill. poor, because you cannot buy IP (for your imaginary self-hosting needs) and youre stupid because you dont know about a whole business of selling servers and VPNs.

go to a searchbar, type "buy VPS server in India" and dilate.
>>
>>109656955
>Just rent a server from someone else
That's not direct at this point, you keep circling back to these workaround despite "IP being a solved problem" according to you.

Where is the solution to self-host directly at home with no intermediaries behind CGNAT? It doesn't exist. So fuck off with your whole "IP is solved" bullshit. No, it's not solved. IPv6 solves it though.
>>
>>109656941
just say "and stuff" instead
>>
File: 1785302103117324.png (40 KB, 152x254)
40 KB PNG
IPv6 has 79228162514264337593543950336 times as many possible IPs as IPv4. Imagine all the cheap proxies we would have if v6 reached the same level of adoption. Shit gives me wet dreams.
>>
>>109648411
>source: my ass
>>
>>109656979
With CGNAT everyone is a cheap proxy and best of all when you get banned all of your neighbours get banned too.
>>
>>109650178
this
>>
>>109652312
when I am on cellular I am behind a CGNAT. my ip is shared across many devices which is good for anonymity. this is why resiproxies work so well on mobile. when I am on fiber my isp lets me rotate my ipv4 just by resetting my router, because ipv4 is scarce enough to need dynamic allocation.
this is assuming the os is smart enough to allocate a new address for every socket connection instead of device session. I am assuming the theoretic ideal that no kiked system will ever implement, because I got sick of myself making the same refutation to retards who post wikipedia links to ipv6 temp addresses. even the ideal can't beat nat.
>>
>>109658399
You can NAT your IPv6 if you really want to. Also there are ISPs that give you a new IPv6 prefix when your reboot your router just like you're used to with IPv4.

Although, the RFC recommends giving customers a static / sticky address for a variety of reasons sometimes that's not possible.
>>
>>109658480
>>109658399
Also the fact that you share that CGNAT box is irrelevant because they will have extensive logs of each subscriber + port mapping for the session. If law enforcement comes knocking they will know who owns what connection. Best not do anything illegal, or at the very least get yourself a VPN.
>>
>>109656914
NAT is how the internet should have been from the start. the ideal would have been CGNAT with a more lenient allocation scheme than symmetric NAT, or a dedicated open port that the isp port forwards to the internet facing router.
either way, the key contribution of a NAT is in turning the internet into one where others can only talk to you if you've shown that you're willing to talk to them at that moment. like how you need to pick your phone up for them to successfully call you.
NAT requires a specific ritual that requires both parties to participate in the act in order to establish a peer connection. the only complicating property of NAT is how implementations are not at all standardized, because standards bodies were reluctant for the longest time and still are, itself because network niggers refuse to let go of the ideal of the internet as a mail network (ie. port scannable shithole and an uninhabitable wasteland) instead of a packet switching network that looks and smells like the telephone network it was inspired by.
>>
>>109658519
>the key contribution of a NAT is in turning the internet into one where others can only talk to you if you've shown that you're willing to talk to them at that moment.
That's a fucking firewall you moronic inbred. The whole idea behind the Internet is that it was supposed to connect people together. You can't do that without public connectivity.

NAT is a hack that was invented because we ran out of addresses to give people because of how fucking mismanaged the whole IPv4 address space is.
>>
>>109658494
who the fuck cares about the government. that's not in my threat model. I don't live in a five eyes nation where the state is strictly a colonial puppet of america. this proves by contradiction that NAT has utility because it remains useful for me.
as for using a VPN in a 13 eyes nation just to get right back into their tracking scheme again, no screw you.
>>109658480
my landline isp actually does this with ipv6. I may have lucked out but I would not count on this. my point is there is an active incentive for isps to not give out sticky ipv4 that does not exist elsewhere.
>>
>>109658544
What is your threat model?
>I care about anonymity but I don't care if my ISP logs all my communications
>>
>>109658533
no, a firewall filters connections to an address. a NAT obscures the address itself.
there would be two ways to "connect people together" without having them leave their bare exploitable ass to the whole fucking globe.
you could get a singular forwarded port from your ISP that you then keep it firewalled well, because presumably you know what you're doing, and know well what the software you're running does, eg. it's not qbittorrent opening a bare unauthenticated web server by default over upnp.
you can establish a peer connection using ICE-style handshakes (but over standardized NAT implementations) where both sides agree to open a connection through another channel. this is retard proof because it nigh guarantees no third party connection can be established in the meantime without you trying very hard to make it so (ie. malware).
>>
>>109658579
A NAT merely translates A:port -> B:port
It doesn't obscure jack shit. It just maps one to another. If you only care about not exposing your internal addressing, guess what? IPv6 has a mechanism for that too, it's called NPTv6:
https://www.rfc-editor.org/info/rfc6296/
https://en.wikipedia.org/wiki/IPv6-to-IPv6_Network_Prefix_Translation

This translates your GUA to some other ULA although WebRTC leaks your internal addressing anyway so…
>>
>>109658607
it obscures the address because it is essentially an ephemeral (stateful) firewall. yeah, you can call your new special snowflake a firewall. but this would imply that the firewall can be made into a hierarchy, instead of the traffic mixing itself being the entire point. every anonymity system is a mixnet for a reason, blending in is what gives anonymity. that's why I said CGNAT would be the ideal.
also this again obscures by device session not socket connection. that would be haram for the network boomer kikes otherwise. it's probably exactly why they hate nat so much.
webrtc leaks addresses because it tries very hard to figure out your network. this is because NAT is unstandardized because IETF refuses to make NAT good for political reasons. they are elites that stumbled across a good solution (actually they didn't, network equipment manufacturers did) and refuse to research and learn for it, but instead stuck to their old ways. ipv6 is old. it is fundamentally ipv4 with more bits for ip and less bits for cruft in the packet. it makes no new insight over ipv4 that would then justify its existence. it's just what if ipv4 but uuid bit length.

also if ipv6 had actual direct addressing (leveraging its uuid ass length) it would open actual new doors ipv4 could not, and I would salivate at what could now be done and would clamor for ipv6 instead. but again this would never be approved by the control freak types in standards bodies.
>>
>>109658657
>instead of the traffic mixing itself being the entire point. every anonymity system is a mixnet for a reason
That isn't what NAT is.

It's not going to give you the same sort of anonymity as something like Tor or I2P would.
There are plenty of ways to identify someone behind a NAT still.
>>
>>109658668
as of now I could do my absolute best in privacy and still leak IP. I understand that every attempt to cloak my IP without fatally sacrificing performance only shifts the privacy risk to whoever owns the proxy I'm routing over. onion networks are all susceptible to global passive adversaries if you're a server btw. especially with the guard node poz.
my main point regardless is NAT provides superior security by obscuring addresses at the socket connection level. you can't just mass gather IPs somehow and cold call my address.
>>
>>109658691
They can still scan the entire IPv4 Internet and find your router though. Ultimately, Internet connectivity requires exchanging addresses with people in order to talk to them.

What you're doing with NAT is like going through a middleman but everyone knows who that middleman is. It's not giving you any anonymity.
>>
>>109658691
I meant to say low latency onion networks
>>109658704
this only matters if (1) your router's default isp firewall is retarded, (2) your router has upnp enabled, or (3) your router runs a clandestine web server or some other form of isp backdoor. all of which could be resolved with proper standards.
it gives me anonymity vis a vis whoever is behind that nat. I blend in at home with those who I live with, and everywhere else with everyone in the building. and the most common form of nat today is mobile, even residential cgnat.
>>
>>109658730
Except you don't blend in at all. Each of devices has unique fingerprints and there's a million different ways to track each and every one of your devices.

If you care about privacy and anonymity then honestly the best thing you can do is to make an airgapped network and not connect to the Internet in the first place.
>>
>>109658735
the most common form of fingerprint, and the only form once you install ubo, is IP and cookies/localstorage. the biggest impact on your digital life is the traceability of your ip address. VPNs are actually a pragmatic solution if it weren't so overpriced compared to its alternatives.
>>
>>109648274
I would prefer to not expose the internal network structure to the entire world to be completely honest with you.
Yes, you can work around that, but NAT does that for you for free.
IPv6 by default also increases the traceability of devices across networks, if you don't take precautions.

If you need IPv6, you are simply too poor and need to suck up to the people forcing you into submission.
>>
>>109658792
How much money do you think a home user is reasonably going to spend on IP addresses?

To give you an example, I have two Docker hosts on my network with native IPv6 configured, each gets a dedicated /80 out of my /56 prefix. I don't even think my ISP would be able to provide me that many routable IP address blocks for IPv4 even if I wanted to.
>>
>>109648274
Retarded to not make it backward compatible with IP4. Literally all they had to do was add four more octets, ISPs get 18446744073709552000 addresses, I get my NAT and dynamic IP, and everyone would be happy.

Also, stop going on about NAT giving privacy, I like it because it makes managing the internal network so much easier.
>>
>>109658792
>If you need IPv6, you are simply too poor and need to suck up to the people forcing you into submission.
One day, CGNAT will come for us all. The clock is ticking.
>>
>>109659056
>I like it because it makes managing the internal network so much easier.
Actual reason to use it if you have many statically configured devices and don't want to have to change them. This is the real reason to use NAT.

When Company A buys Company B and wants to connect them to their network they can't just go and do that because there will be collisions that will fuck everything up so what do they do? They buy a magic NAT box with smoke and mirrors inside and connect both sides of the network to that.

How does it work? Nobody knows. Better hope the magic box doesn't break. But it means they can remain ignorant of IPv6 or huge network renumbering and go on about their day.
>>
>>109659216
And this is why the first adopters of IPv6 were the biggest ISPs. Companies like Comcast and the mobile operators were already an amalgamation of lots of smaller ISPs that were bought up over the years so their private networks were already an unmaintainable mess, with too many devices to even fit in private RFC space. IPv6 greatly simplified their networks.



[Advertise on 4chan]

Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.