Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

IPv6 traffic crosses the 50% mark

google.com

Post reply on HN
Print view View original

571–580 of 643 posts

Re: IPv6 traffic crosses the 50% mark

#571

Post by jcalvinowens »

traderj0e wrote: post #558
jcalvinowens wrote: post #178

I consistently get 100x as many captchas from google over V6 as over V4, on many different networks: it is obnoxious and obviously broken on their end.

IP reputation is a tough problem with a lot of establishment behind v4. Even if they were to build up v6 reputation, since v6 addresses are cheaper, wouldn't they have lower rep?

The "reputation" concept has to be applied to network blocks in IPv6, because of privacy addresses, and they very clearly aren't doing that for whatever reason. I get at least one captcha a day from a stock Android phone that's been connecting from the same damn /64 for almost a whole year.

Re: IPv6 traffic crosses the 50% mark

#572

Post by GoblinSlayer »

kstrauser wrote: post #532
saltyoldman wrote: post #517

Earlier quoted context omitted.

I don't even know why clouds offer public IP addresses. In my opinion all clouds should only have a gateway that routes via host header for millions of customers. IPv4 should be a special priv for special situations at a higher price. Then these clouds could own maybe 20 IPs total instead of millions.

> In my opinion all clouds should only have a gateway that routes via host header for millions of customers. This is incompatible with TCP/IP networking. In TCP connections, (sender_address, sender_port, receiver_address, receiver_port) is a unique combination. Those numbers together uniquely identify the sender talking to the receiver. For a public webserver: * sender_address is the client machine's IP address * sen…

If you multiply by 20 shared addresses, it would be 130 connections to 200000 websites.

Re: IPv6 traffic crosses the 50% mark

#573

Post by ectospheno »

unethical_ban wrote: post #473
ectospheno wrote: post #297

Earlier quoted context omitted.

> Also, notice how Android and iOS don't support turning off IPv4. You can trivially connect an iOS device via IPv6 only.

I think they're saying you can't force disable ipv4 entirely.

I’m going to guess no one responding to me has actually tried connecting to an IPv6 only network because it works without incident. Will your apps? Who knows - apple hasn’t always been great at enforcing that.

Re: IPv6 traffic crosses the 50% mark

#574

Post by saltyoldman »

kstrauser wrote: post #532
saltyoldman wrote: post #517

Earlier quoted context omitted.

I don't even know why clouds offer public IP addresses. In my opinion all clouds should only have a gateway that routes via host header for millions of customers. IPv4 should be a special priv for special situations at a higher price. Then these clouds could own maybe 20 IPs total instead of millions.

> In my opinion all clouds should only have a gateway that routes via host header for millions of customers. This is incompatible with TCP/IP networking. In TCP connections, (sender_address, sender_port, receiver_address, receiver_port) is a unique combination. Those numbers together uniquely identify the sender talking to the receiver. For a public webserver: * sender_address is the client machine's IP address * sen…

This is actually a good point. I guess 20 IPs per cloud infra company is probably too few. But maybe these cloud companies can have 20k IPs instead of 2 million?

Re: IPv6 traffic crosses the 50% mark

#575

Post by tyingq »

GoblinSlayer wrote: post #569
mmbleh wrote: post #474

Earlier quoted context omitted.

It is because the IPv6 rollout has not been consistent. Some assign /64 per machine, some assign /64 per data center. Some even go the other way and do a /56 per machine. We've had to build up a list of overrides to do some ranges by /64 and others by /128 because of how they allocate addresses. This creates extra burden on server operators and it's not surprising that some just choose not to deal with it.

This problem exists for ipv4 too: some machines have static address, others have dynamic, so you can implement overrides.

Ipv6 is cheap though. If I want to get past your IP or per Network limit, options abound.

Re: IPv6 traffic crosses the 50% mark

#576

Post by Dagger2 »

saltyoldman wrote: post #556
seabrookmx wrote: post #529

Earlier quoted context omitted.

> host header Not all workloads are HTTP. > gateway .. for millions of customers That's basically what an AWS ALB is. It's not provisioning bespoke infrastructure when you create it.. it's just a routing rule in their shared infra. If Amazon wanted, they could easily have shared IP's but the cost of an IPv4 isn't so great that this approach has been warranted yet, clearly.

Yeah I get all that, but the only two connection types that are useful are http/s/ and ssh. SSH can have work-arounds like the way google does. Let's let the people that want non http workloads pay more.

Remember that, at one stage, the only two types that were useful were FTP and telnet. HTTP and SSH didn't even exist.

Let's not strangle the next big thing that doesn't exist yet before it can even be born, yeah?

Re: IPv6 traffic crosses the 50% mark

#577

Post by Dagger2 »

ahartmetz wrote: post #554
Dagger2 wrote: post #376

Earlier quoted context omitted.

Questions like this suggest that 1&1 do: https://forum.openwrt.org/t/solved-dsl-over-ds-lite-with-isp... I'm less sure about Telekom. For obvious reasons, it's hard to find info in English.

My anecdata is that I'm on 1&1 fiber and I have a public IPv4 address.

German ISPs commonly seem to have an option of their "old" platform (v4 only, no v6) or their "new" one (v6 and CGNATed v4). Maybe you're on their old one? Or maybe they only use it for certain account tiers or somesuch.

Re: IPv6 traffic crosses the 50% mark

#579

Post by kalleboo »

usui wrote: post #371
kalleboo wrote: post #364

Earlier quoted context omitted.

IPv4 is provided using DS-Lite or MAP-E depending on the provider. I'm using OpenWRT and paid for a static IP so I had to manually configure all the details for the MAP-E tunnel in OpenWRT myself, I think typically the routers sold to consumers pick up the configuration automatically somehow.

Which provider are you using? I'm curious about this since there are not many OpenWrt guides for getting connected in Japan. Is your config similar to this write-up? https://github.com/fakemanhk/openwrt-jp-ipoe I didn't need to do any configuration for DS-Lite or MAP-E, as DHCPv6 with a configured prefix got IPv6 working, although DNS is still broken when turning off IPv4 entirely.

I'm using en hikari on Flets Cross (10 gigabit) with v6plus.

I set it up like 2 years ago but this was in my bookmarks: https://zenn.dev/ebi68k/articles/6df7cfec64c281

edit: looking this all up again to refresh my memory, it looks like with my static IP the provider switches from using their default MAP-E to using RFC2473 directly instead, the configuration matches "IPIP" in this guide https://qiita.com/kouhei-ioroi/items/cf0c6228c5c1faef415a

Re: IPv6 traffic crosses the 50% mark

#580

Post by Dagger2 »

mrjoe3332 wrote: post #568
azernik wrote: post #547

Earlier quoted context omitted.

IPv4 addresses are in fact a subspace of IPv6 (that's NAT64). They were not by fiat declared invalid. The actual thing I think you're complaining about - the necessity for NAT64 at all - is unavoidable, because you need a NAT/protocol-translation layer for packets to actually move between the new address space and the old one. SLAAC-by-default is not, in my experience implementing IT automation, an actual barrier for…

NAT64 is a temp bolt on and also not the same thing. If I own 1.1.1.1 in v4, that doesn't mean I own some equivalent of 1.1.1.1 in v6. They had router nat64 and relay nat64, both with dealbreaker problems. You don't exactly need a translation layer. If they just gave me 1.1.1.1:: in v6, anyone migrating v4 to v6 would have the same route to me as before, and other changes like DNS6 could be gradual. Then after v4 is…

I mean, you own 2002:101:101::/48. You also own ::1.1.1.1, ::ffff:1.1.1.1 and 64:ff9b::1.1.1.1.

> If they just gave me 1.1.1.1:: in v6, anyone migrating v4 to v6 would have the same route to me as before, and other changes like DNS6 could be gradual

That's not really how routes work. Or DNS.

I think you're mainly arguing we should import v4's allocations into v6, but about the only benefit is that people don't have to bother requesting new allocations. It doesn't help with any other aspect of the transition, and there are good reasons to avoid doing it -- the v4 address space is highly fragmented and also very unfairly allocated.

Plus we have plenty of people saying we should take back v4 allocations from companies that own them. That's not possible, but _not_ giving owners of v4 /8s an entire 1/256th of the v6 address space certainly is.

Post reply on HN