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

The world in which IPv6 was a good design (2017)

apenwarr.ca

Post reply on HN
Print view View original

221–230 of 318 posts

Re: The world in which IPv6 was a good design (2017)

#221

Post by jandrese »

JohnFen wrote: post #201
jandrese wrote: post #177

Earlier quoted context omitted.

Neighbor Discovery is basically ARP wearing a trenchcoat. The only people "hurt" by the change are the ones who were parsing the output of the arp command for whatever reason. IMHO the biggest problem is that IPv6 address autoconfiguration was half-baked. There is no mechanism to inform anybody about which address you have configured for yourself, unlike IPv4's DHCP where a central server knows everyone's address and…

> that IPv6 address autoconfiguration was half-baked. Indeed. I thought I was being an idiot and just not understanding how this was supposed to work, until I learned that it just doesn't do a lot of important things. So when the day comes that I have to move my network to IPv6, I plan on continuing to use DHCP because I want the omitted functionality. Of course, I still might be being an idiot and not understanding.…

Be aware that DHCP6 doesn't work the same as DHCP. It's really intended for configuring routers, not hosts.

Re: The world in which IPv6 was a good design (2017)

#222

Post by Dagger2 »

pknomad wrote: post #185
api wrote: post #91

Earlier quoted context omitted.

Residential broadband often has it too. The laggards are corporate networks and cloud where IT is ultra conservative and “if it’s not broke don’t fix it.” The only thing that will make corporate environments change is if something they need starts requiring it, and not a second before. Cloud is slowly getting it. Slowly. GitHub still doesn’t have it though, which makes pure v6 nodes annoying for a lot of use cases.

I work for a federal space where IPv6 native is a mandate. As someone who's working on k8s, we often have to build and patch everything ourselves to support that mandate. Want to pull a helm chart from a github repo? Gotta either dual stack the node or run a reverse proxy to make that happen. I'd love to be at a point where everyone just dual stacked everything so that one day we can flip off the IPv4 switch.

Try NAT64. It'll let v6-only clients reach v4-only websites.

Re: The world in which IPv6 was a good design (2017)

#223

Post by systems_glitch »

otabdeveloper4 wrote: post #152
systems_glitch wrote: post #147

Earlier quoted context omitted.

I really really don't get ISPs' difficulties in deploying IPv6. The only conclusion I can reach that actually makes sense is that they don't have the in-house talent to deploy it and refuse to hire someone who does. I guess there's just not enough pain in staying IPv4-only or halfway implementing IPv6 to make them get up and do something about it?

IPv6 doesn't actually solve any problem people wanted to solve. IPv4 (as used in practice) has 48 bits of addressing, we don't need more. What we do need is a standard way to do address translation for routing decisions, to replace the 1001 half-baked solutions for VPN and overlay networks that are used today. (Linux has something like five or six "standard" ways to tunnel IP over IP. WTF?)

That's assuming end-to-end connectivity is, for some reason, no longer desirable. There's a bunch of stuff where all the NAT-associated problems would just go away when switching to IPv6, like SIP.

Re: The world in which IPv6 was a good design (2017)

#224

Post by jandrese »

patmorgan23 wrote: post #206
jandrese wrote: post #170

Earlier quoted context omitted.

That's not quite true. Even after corporate users have switched the government will still be on V4. The US military is drowning in IPv4 addresses and feels little pressure to switch.

Azure AD only got IPv6 support this year. Most corporate networks have not switched, whole most federal agencies have implemented IPv6(due to mandates), mobile carriers are heavily utilizing IPv6 and so are residential ISPs (Comcast and Time Warner have been deploying IPv6 since 2011)

Federal agencies turned on the Cloudflare flag that auto-translates incoming IPv6 packets to the IPv4 that they support for front facing websites. That checked the box and little progress has been made since then.

Re: The world in which IPv6 was a good design (2017)

#225

Post by vetinari »

api wrote: post #91
mike_d wrote: post #78

Earlier quoted context omitted.

IPv6 adoption is just the traffic shift from desktop to mobile. IPv6 kinda makes sense in mobile because it solves a problem of needing multiple addresses per person (phone, tablet, gaming device, etc) and the whole stack is maintained by two entities (the phone OS manufacturer and the carrier). It probably would have worked even better if it was far less complex and only solved the problem that was needed. https://w…

Residential broadband often has it too. The laggards are corporate networks and cloud where IT is ultra conservative and “if it’s not broke don’t fix it.” The only thing that will make corporate environments change is if something they need starts requiring it, and not a second before. Cloud is slowly getting it. Slowly. GitHub still doesn’t have it though, which makes pure v6 nodes annoying for a lot of use cases.

Corporate networks are not lagging for being ultra conservative; they lag, because the moment they start thinking about any change, their vendors start pushing all manners of products and licenses on them, that they insist are absolutely necessary for that change and it won't work without it. Sharks feel the blood in the water, and they want to make whatever is possible on it.

So in the end you are looking both at capex, increased opex and for what? You have ipv6? Congratulation, now you go to your C-level bosses and explain to them why it was worth it. Good luck.

Re: The world in which IPv6 was a good design (2017)

#226

Post by ianburrell »

pknomad wrote: post #185
api wrote: post #91

Earlier quoted context omitted.

Residential broadband often has it too. The laggards are corporate networks and cloud where IT is ultra conservative and “if it’s not broke don’t fix it.” The only thing that will make corporate environments change is if something they need starts requiring it, and not a second before. Cloud is slowly getting it. Slowly. GitHub still doesn’t have it though, which makes pure v6 nodes annoying for a lot of use cases.

I work for a federal space where IPv6 native is a mandate. As someone who's working on k8s, we often have to build and patch everything ourselves to support that mandate. Want to pull a helm chart from a github repo? Gotta either dual stack the node or run a reverse proxy to make that happen. I'd love to be at a point where everyone just dual stacked everything so that one day we can flip off the IPv4 switch.

I think major failure of Kubernetes was not being IPv6-only from the beginning. Its model of every pod having own address works much better with IPv6 where addresses are cheap. With IPv4, it needs complicated overlay networks. The cluster boundary would also make a good place for NAT64 proxy.

Kubernetes didn't get IPv6 support until later, and it sounds like it isn't reliable yet.

Re: The world in which IPv6 was a good design (2017)

#227

Post by oezi »

otabdeveloper4 wrote: post #152
systems_glitch wrote: post #147

Earlier quoted context omitted.

I really really don't get ISPs' difficulties in deploying IPv6. The only conclusion I can reach that actually makes sense is that they don't have the in-house talent to deploy it and refuse to hire someone who does. I guess there's just not enough pain in staying IPv4-only or halfway implementing IPv6 to make them get up and do something about it?

IPv6 doesn't actually solve any problem people wanted to solve. IPv4 (as used in practice) has 48 bits of addressing, we don't need more. What we do need is a standard way to do address translation for routing decisions, to replace the 1001 half-baked solutions for VPN and overlay networks that are used today. (Linux has something like five or six "standard" ways to tunnel IP over IP. WTF?)

Even worse IPv6 solves the NAT traversal problem in a way how people don't want it to be solved.

People wanted a DynDNS kind of solution.

Re: The world in which IPv6 was a good design (2017)

#228

Post by p1mrx »

comex wrote: post #75
p1mrx wrote: post #64

Earlier quoted context omitted.

The biggest problem with 6to4 is that the anycast gateways (192.88.99.0/24 and 2002::/16) often go to a different network than the one you're paying for transit, so you can't just turn it on for production traffic and expect it to work. The anycast gateways are only used when communicating between 6to4 and native IPv6 addresses, so if 2002::/16 had been the only IPv6 address space, then it would have been more reliab…

> but then we'd be stuck with IPv4-based IPv6 addresses forever. Yep, but would that have been so bad? Certainly with IPv6's 128-bit addresses, there would still be enough space to go around… Though admittedly, when I see "IPv4 with more bits" hypotheticals, they often involve smaller addresses than IPv6.

It wouldn't be the end of the world, but it's inelegant. Take a look at your Gmail headers for example.

Re: The world in which IPv6 was a good design (2017)

#229

Post by patmorgan23 »

backendanon wrote: post #138
jeroenhd wrote: post #66

IPv6's biggest problem remains not that it's badly designed (at least not nowadays, there were problems but they were solved ten years ago) but that millions of network engineers never bothered to look deeper into IPv6 than "I don't get it, this feels off". You can't make a backwards compatible "IPv4 with more bits" like people dream of. L2 routers and middleboxes would still need to be replaced, software would still…

> People cared about the IPv6 privacy risks but never let go of that concept even after rfc4941 fixed that oversight. In my own experience in our household, IPv6 destroys privacy.

How so?

Re: The world in which IPv6 was a good design (2017)

#230

Post by JohnFen »

jandrese wrote: post #221
JohnFen wrote: post #201

Earlier quoted context omitted.

> that IPv6 address autoconfiguration was half-baked. Indeed. I thought I was being an idiot and just not understanding how this was supposed to work, until I learned that it just doesn't do a lot of important things. So when the day comes that I have to move my network to IPv6, I plan on continuing to use DHCP because I want the omitted functionality. Of course, I still might be being an idiot and not understanding.…

Be aware that DHCP6 doesn't work the same as DHCP. It's really intended for configuring routers, not hosts.

I wasn't aware. That's a real bummer, and a good example of the numerous kinds of gotchas that make this transition much more painful than it would otherwise have to be.

The more I learn about IPv6, the more I dread it.

Post reply on HN