Hacker News Reader

Before I start I must say, I built IPv6 only datacenters and enterprise networks. I am saying from experience.

> Can’t just simply put one IPv6 router/firewall behind another. Not all IPv6 routers support DHCP-PD, and even if they did, you could have 2-3-4 levels of routers/firewalls at a business.

You cannot do this with IPv4 either. Routing has to work. It works for both IPv4 and In But if you are saying, you add 2-3-4 layers of NAT, this is just crazy.

> Can’t just simply open a firewall rule on the main site router to forward say HTTPS to an internal service. Why? Not all ISPs give static IPv6 prefixes, not all PCs/servers/devices support DHCP6 for static leases, and then there’s IPv6 privacy addresses.

ISPs not providing static prefixes is a particular ISP problem, not an IPv6 problem. If your ISP cannot hand out static IPv4, you have a similar problem and tge workaround would be what? Dynamic DNS? Admittedly we had /32 and own ASN, but in remote offices where you don’t buy transit, you live on PA addresses (and of course they are static). From what I know, some ISPs don’t issue static prefixes to residential customers, but if you want to serve services from your prefix, use a stable prefix and a decent ISP.

Privacy addresses are intended for client-initiated traffic, so no matter how many privacy addresses your server has, the stable one is still there.

> WAN failover / multiple ISPs is hard. You have a fibre primary feed, and a secondary cellular/5G feed. Each has different IPv6 addresses. How do you ensure the right ISP is used at any given point?

By sending RA and expiring the old prefix on the failover. I admit, this is not an obvious solution and it may not be suitable for some edge cases.

>You cannot do this with IPv4 either. Routing has to work. It works for both IPv4 and In But if you are saying, you add 2-3-4 layers of NAT, this is just crazy.

Is it? I'm behind two layers of nat and everything works fine.

That NAT "does not work" is an old myth that certain people love to use. If you show it does work, their final reply is always: but inbound connections are hard.

I do not know who needs inbound connections to all of their devices in a corporate network, nor at home, since we solved that shit long time ago.

It only counts as working if you are ok with a giant pile of hacks and pain. Ever had to debug NAT-related SIP issues? Not fun. A lot of protocols fell out of favor, because they poorly interact with NAT (and stupid corporate firewall rules on the other hand so everything gets tunneled via https now..).

And wanting inbound connections is really not that strange a request. Maybe I want to host some website at home? How about a weather station/other IoT.. How about multiplayer gaming?

There are so many things that would benefit from being able to poke some holes in the firewall. With carrier grade NAT becoming more and more common you can forget about that, since you have no control over their infrastructure to forward any ports.

Datacenter networks are an entirely different beast to SME and SOHO business networks. You don’t get the same budget for nice equipment, nor dedicated IT persons to run it, nor any remotely technical people on the ISP side to help you. SME and SOHO are very price conscious too. Think low margin retail and petroleum and fastfood.

> You cannot do this with IPv4 either. Routing has to work. It works for both IPv4 and In But if you are saying, you add 2-3-4 layers of NAT, this is just crazy.

Yes, you literally can stack a dozen commercial IPv4 firewall/routers behind each others networks and a PC plugged into the very last one will have no problems access Google on the browser. Because of NAT, which all IPv4 capable personal/SOHO/SME routers do by default.

Sure, VoIP may not work, but actually chances are it will because the VoIP service will be NAT-free via VPN to the provider.

And in this retail/petro/fastfood environment things like p2p file sharing are irrelevant and not required. All traffic is either to websites or to corporate systems across VPNs where NAT is not present or easily managed.

> By sending RA and expiring the old prefix on the failover. I admit, this is not an obvious solution and it may not be suitable for some edge cases.

It’s a slow failover by comparison to IPv4 methods. Too slow for some common use cases in industries I deal with. It also doesn’t allow for load balancing across links (eg route all important traffic out link 1 by default and all backup uploads out link 2 by default).

> Datacenter networks are an entirely different beast to SME and SOHO business networks.

SOHO networks just use what providers gives them. I am not sure, why are you jumping from large “oil-and-gas corps” to SOHO, though.

For SOHO providers will be building IPv6-only networks with NAT64 to access the legacy IP. I am seeing it happening: I saw multiple situations like this with guest wifi already. Obviously, this trend is going to continue.

> Yes, you literally can stack a dozen commercial IPv4 firewall/routers behind each others networks and a PC plugged into the very last one will have no problems access Google on the browser. Because of NAT, which all IPv4 capable personal/SOHO/SME routers do by default.

… and then Google blocks it all because the amount of requests from a single IP exceeds their expectations. I have seen half of Belarus blocked by Google because their NAT pools were overprovisioned. When I was at Cisco, a customer asked us to change the NAT behaviour so, that it uses pool not “first, all ports from the first IP in the pool, then all ports from the second IP, etc” to “first port from the first IP, then the first port from the second IP, etc”. The reason was, their starting IPs in the pool were always oversubscribed and blocked. The alternative behaviour broke FTP, SIP even with application-level fixups because the protocols don’t expect signalling and data to use different IPs.

Not to mention, SOHO routers just don’t have enough capacity to maintain translation cache so connections will be randomly dropped.

> Too slow for some common use cases in industries I deal with.

Have you tried it? Renumbering event means sending two RA packets: one to expire an old prefix, another to announce a new one. While technically it’s slower than changing routers on the NAT gateway, I wonder what are your requirements.

It doesn’t allow load balancing, which is true. Work is being done in IETF to support for client-based load balancing, see RFC7157, etc. In my experience, load balancing is rarely happening in such scenarios because it will result an overloaded link when either on two goes down. Anyway, I don’t argue, it is something that could be improved.

> Not all IPv6 routers support DHCP-PD

I my experience, they all did. I'm not even sure how you could possibly deploy IPv6 without it, manually handing out prefixes?

> you could have 2-3-4 levels of routers/firewalls at a business.

It seems just a bad design, but I still don't see how IPv6 would not allow it.

> Why? Not all ISPs give static IPv6 prefixes, not all PCs/servers/devices support DHCP6 for static leases, and then there’s IPv6 privacy addresses

1. you don't need a static prefix to write a firewall rule: you can simply remove the dynamic prefix with a mask and match the EUI-64 suffix. For example with ip6tables it's something like this `::e2ab:8fff:fe12:3b6b/-64`.

2. All IPv6 devices do support SLAAC with a stable address mechanism, either EUI-64 or stable privacy address and you can use that in the firewall rule.

3. IPv6 privacy extensions, when enabled, don't preclude listening and accepting connection on the EUI-64 address. So, an inbound traffic firewall rule will just use the stable address: you shouldn't listen on a privacy address, they are for outbound connections.

> Each has different IPv6 addresses. How do you ensure the right ISP is used at any given point? IPv6 shifts this decision to the client.

I honestly don't understand the difference with IPv4. You can have multiple addresses and do load balancing on the router with both.

> IPv6 shifts this decision to the client. This makes load balancing and policy based traffic routing

I can't comment on this because I never tried it, either on IPv4 or IPv6. On top of my head, I'd say it would be possible by updating the routes priorities with an RA, if you don't want or can't do a NAT66 with a ULA prefix.

> At the cost of not having dedicated unique public IPs but these places simply don’t need them.

Wrong! Everyone needs routable addresses, even if they don't know it because they need VoIP, video calls on webRTC, FTP, p2p file sharing, online games, etc. All of these barely manage to work in a NAT by using workarounds like ALGs, UpnP, NAT-PMP, relay servers and other atrocities that greatly complicate the network design and are a security nightmare.

Sorry I could have stated the DHCP-PD issue more clearly. Comcast doesn’t support it. [1]

> I honestly don't understand the difference with IPv4. You can have multiple addresses and do load balancing on the router with both.

A PC cannot easily have 2 different IPv4 subnets in the same NIC, which is what IPv6 permits with RA. Even if a PC did have 2 networks, how does the router tell the PC to use fibre for everything except backups, and cable for backups, and cellular only if the others fail? These decisions are best made by the router, not an arbitrary client device connected behind it. With IPv6 and no NAT, how does the client PC know what IP to use as its source at any given time? Yes NAT66 works- and for a very very long time the IPv6 community has resisted any NAT, which has resulted in poor NAT support in IPv6 stacks across many vendors.

> Wrong! Everyone needs routable addresses, even if they don't know it because they need VoIP, video calls on webRTC, FTP, p2p file sharing, online games, etc. All of these barely manage to work in a NAT by using workarounds like ALGs, UpnP, NAT-PMP, relay servers and other atrocities that greatly complicate the network design and are a security nightmare.

I didn’t say a routable IP isn’t required. I said a dedicated unique public IP is not required.

A gas station or fast food business doesn’t need online games or p2p file sharing. Their VoIP is done over VPN or uses NAT ALG in the firewall. Their web browser and payment systems are happy with a NAT’d IP on a LAN behind their router (which may or may not have a public static IPv4).

1. https://forums.businesshelp.comcast.com/conversations/ipv6/i...

> A PC cannot easily have 2 different IPv4 subnets in the same NIC

i know this to be false, but i don't want to dismiss your comment because of poor wording.

so: what do you mean?

The only "problem" here that is caused by IPv6 is the SD-WAN / multi-provider scenario. Even that is solvable. There are so many IPv6 addresses that your local router can have a unique routable range that it can hand out to the local network, "masking" the external internet provider addresses in the same way that IPv4 routing does it. That's not NAT, that's just routing. E.g. I have a customer that uses a non-RFC1918 public range internally and they have 5 ISP uplinks. They don't need to renumber their internal range just to fail over to another link.

ALL of the other issues are caused by vendors selling garbage products that have IPv6 as a "checkbox" tick without real support.

The real problem is that network engineers around the world let them get away with it, because they avoid IPv6 because of this perception that it isn't mature.

The protocol is over two decades old! It's been fully supported since Windows 2000! My phone is IPv6. My fibre LAN is IPv6. It just works.

What doesn't work is this attitude where people just like you continue to pay for products that only pay lip service to the future.

I've gone on rants here before about enormous service providers like AWS and Azure providing IPv6 "support" that is an absolute garbage fire. Totally worthless. Useless beyond belief. Literally counterproductive in the case of Azure, where turning on IPv6 anywhere will break unrelated IPv4 functionality!

Nobody ever chimes in to say: "Yes, we agree, this is bad."

Nobody ever says "Microsoft should fix this!"

Nobody has a problem. So now we have situations like the one you just described, where people are still buying routers in 2021 that that can't handle the most basic IPv6 functionality.

IPv4 addresses are now USD $40 each and rising exponentially. I really do hope the COVID case number graphs taught some of you in a visceral sense what the world "exponential" means. Hint: it doesn't mean that it'll cost $60 in a decade. Try $40000, each.

My ISP uses carrier-grade NAT on IPv4, breaking sites all over the place. I had to call them to enable a dedicated IPv4 address. Their annual investor report says that they will run out within a couple of years. Get one now while you can is the message.

Do you have any idea how much code is required for two PCs to communicate directly now, with up to 4 NAT layers as a common scenario? It's a huge library you have to integrate in your code and then you need a cloud service to assist it.

This is madness, and network engineers around the world are perpetuating it.

> The only "problem" here that is caused by IPv6 is the SD-WAN / multi-provider scenario. Even that is solvable. There are so many IPv6 addresses that your local router can have a unique routable range that it can hand out to the local network, "masking" the external internet provider addresses in the same way that IPv4 routing does it. That's not NAT, that's just routing. E.g. I have a customer that uses a non-RFC1918 public range internally and they have 5 ISP uplinks. They don't need to renumber their internal range just to fail over to another link.

Neither does anyone using IPv4.

Does your customer use BGP for that public IPv6 range? Do they have cellular links? Can they load balance some traffic over two/three links? Can they direct certain protocols over different carriers?

y u so mad?

> Literally counterproductive in the case of Azure, where turning on IPv6 anywhere will break unrelated IPv4 functionality!

not just on azure; smartphones seem to be the only environment were v6 is doing its job (lots of smartphones need lots of addresses. ipv6 has a lot of addresses, problem solved) and i guess it's because they are most user-centric client-device imaginable, the actual users don't care at all and most importantly; expect things to occasionally fail.

I'm mad because network "engineers" use that word without realising what it entails to earn that title.

The painting has been on the wall for decades. IPv4 exhaustion was predicted to the week.

Did anyone do anything about it? No.

There are people talking about how Kubernetes is the "future" on this very forum, yet it simply does not work with IPv6. Instead, it uses NAT and RFC1918 ranges... but you've got to be careful not to accidentally overlap those subnets with the rest of your network!

I have a customer that screwed this up and now they're facing a rebuild of 3 out of 4 of their clusters. Fun times.

I have another customer undergoing mergers, where renumbering their IPv4 RFC1918 ranges is going to cost them on the order of $10M.

IPv4 is the legacy, and its costing real money to keep it going decades after its expiry.

> I have another customer undergoing mergers, where renumbering their IPv4 RFC1918 ranges is going to cost them on the order of $10M.

i suppose that is way less than what a switch to v6 would cost?

Precisely.

And why should IPv6 be expensive?

Is it because we still tolerate vendors throwing up software that merely pretends to be IPv6 capable?

Or because IPv6 is magically harder than IPv4 despite being nearly identical?

IPv6 should have supported NAT like IPv4, not dragged its heels by begrudgingly implementing it years/decades later (and still not all support IPv6 NAT so the problem isn’t yet solved).

NAT, as much as you hate it, makes deployments easier and quicker and simpler for your average business. I don’t have to worry about BGP, or if my ISP has given me a /64 only or can they do a /60 or /48.

A gradual migration would have allowed both types of approaches to be used. Some would start with IPv6 using NAT, and other more switched on orgs could go straight away with NAT-free IPv6 networks. And lets face it, this has been a very gradual 20 years migration…

router that do BGP are cheap, especially if you're only taking a default route which is all you're going to need for using multiple ISPs for redundancy.

It’s not the router that is the problem with BGP, it’s the ISPs. Will your cheap as chips cellular carrier support BGP? Will your local cable provider? Probably not.