Slashdot Mirror

← Back to Stories (view on slashdot.org)

Whatever Happened To the IPv4 Address Crisis?

Posted by samzenpus on from the still-working dept.

alphadogg writes "In February 2011, the global Internet Assigned Numbers Authority (IANA) allocated the last blocks of IPv4 address space to the five regional Internet registries. At the time, experts warned that within months all available IPv4 addresses in the world would be distributed to ISPs. Soon after that, unless everyone upgraded to IPv6, the world would be facing a crisis that would hamper Internet connectivity for everyone. That crisis would be exacerbated by the skyrocketing demand for IP addresses due to a variety of factors: the Internet of Things (refrigerators needing their own IP address); wearables (watches and glasses demanding connectivity); BYOD (the explosion of mobile devices allowed to connect to the corporate network); and the increase in smartphone use in developing countries. So, here we are three years later and the American Registry for Internet Numbers is still doling out IPv4 addresses in the United States and Canada. Whatever happened to the IPv4 address crisis?"

95 of 574 comments (clear)

  1. NAT by hcs_$reboot · · Score: 5, Interesting

    While phones use Internet connectivity, they usually connect through the carrier infrastructure which may only allocate a few (or even 1) IPv4 addresses, thanks to NAT.

    --
    Slashdot, fix the reply notifications... You won't get away with it...
    1. Re:NAT by Rich0 · · Score: 5, Interesting

      Yup. NAT isn't really too troublesome on phones since they rarely run servers, are usually connecting to cloud-based services, and they move around so much that they'd probably have an IP change every 10 minutes if you handled them like a traditional routable IP.

      If I were using cellular service as my actual home ISP it would drive me nuts, though.

      IPv6 is needed more than it ever was. We just haven't reached the end of v4 yet.

    2. Re:NAT by aurizon · · Score: 5, Funny

      We need to get the ground work done so that IPv8 can be introduced smoothly - the galaxy demands to be properly served...

    3. Re:NAT by Bert64 · · Score: 4, Informative

      This is far more troublesome for people who *do* run servers...
      If you are getting abusive users from a mobile ISP, how do you ban those users?
      Block the IP and you block every customer of that isp.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    4. Re:NAT by jkrise · · Score: 4, Insightful

      There are 2 dimensions to the IPv4 problem - the user end; and the server end. Except for newly formed companies looking to provide internet access to their users through a proxy server; the individual users are largely oblivious to the crisis; as you rightly mentioned.

      But try hosting your own server (non-cloud provider) - your ISP forces you to acquire IPv6; and you have to jump through hoops to make it smoothly accessible over VPNs and the general inernet.

      --
      If you keep throwing chairs, one day you'll break windows....
    5. Re:NAT by gl4ss · · Score: 2

      well since you usually have to pay money anyways to get a static IP the users are unlikely to have static ip's anyhow even on their landline connections...

      --
      world was created 5 seconds before this post as it is.
    6. Re:NAT by mjr167 · · Score: 4, Informative

      Practically speaking, the IP address doesn't change unless you reboot the modem or manually do a release/renew.

    7. Re:NAT by kasperd · · Score: 5, Informative

      NAT isn't really too troublesome on phones since they rarely run servers, are usually connecting to cloud-based services

      Any sort of peer-to-peer communication is problematic, if NAT is involved. Lots of the communication you want to do on phones is peer-to-peer in its nature, but actually implementations have often chosen inferior cloud based implementations, simply to work around NAT. Why else would you involve a cloud service, when what you really want to do is to move some data from one phone to another?

      Additionally, even communication with cloud based services is problematic when NAT is involved.

      Connecting to a cloud service in order to get a notification, once there is a new email or a new chat message is something you often want to do on a phone. But you cannot do that through a NAT, unless you a prepared to send a constant stream of packets to keep a connection tracking entry alive. Now your phone has to wake up every so often just to send another keepalive packet through the NAT. This consumes battery power, it also consumes bandwidth and if everybody does it, it consumes entries on the NAT.

      If the NAT does run out of entries for connections, it will have to lower the lifetime of connections. That will lead to applications sending keepalives more frequently, and we are back in the same situation as before, only wasting more battery power and bandwidth.

      and they move around so much that they'd probably have an IP change every 10 minutes if you handled them like a traditional routable IP.

      NAT does not solve that problem, it actually makes it worse. You still have to keep track of the local IP you assigned to the phone if it is behind a NAT. The tracking of the IP address is not any harder just because it is a public address. But by introducing a CGN you introduce the requirement that all the traffic from the phone gets routed through that CGN even as the phone is moving. If you did not have the NAT layer, you only have the challenge of routing packets to the phone as it is moving, there is no need to get it through one particular NAT as well.

      --

      Do you care about the security of your wireless mouse?
    8. Re:NAT by SuricouRaven · · Score: 5, Informative

      Most ISPs assign staticish addresses. They are technically dynamic, but change very infrequently - in my case, no more than once or twice a year, baring a change of modem or network card.

    9. Re:NAT by peragrin · · Score: 4, Informative

      depending on the provider you don't get a new ip address when do those things either. from my limited experiments with Comcast and Time Warner they give the same IP address to the same Mac address every time.

      I replaced a router on both and got new ip addresses. however when i cloned the mac address from the old routers to the new I got the old ip addresses.

      Now this is really limited. 4 routers on two service providers. so take it with a grain of salt and a shot of tequila .

      --
      i thought once I was found, but it was only a dream.
    10. Re:NAT by ttucker · · Score: 2

      DHCP servers typically try to give clients their old IP addresses based on MAC address. This usually works until there is a huge demand for reservations, and the pool of free addresses runs out. This is uncommon in broadband networks where the number of clients is relatively static, and clients are rarely restarted.

    11. Re:NAT by gothzilla · · Score: 2

      One of our remote offices was connected via cellular. It was actually very usable and far more stable than you might guess. It's in a small town in Arkansas that didn't have access to anything but dialup. We couldn't even get a T1 without a huge build cost. Fortunately there were only 4 people there that needed access too. We just plugged a USB hotspot into a Cradlepoint router and it worked very well. We couldn't get a static IP but DynDNS + LogMeIn was good enough for what we needed there.

      The mom & pop cable provider there finally got internet access a few years back so we switched to that and it's so unreliable that I wish we could go back to cellular.

    12. Re:NAT by ericloewe · · Score: 2

      128 bits are enough for the whole solar system.

      Let's wait for intelligent life to show up before we scare them away with our inability to quadruple the size of our addresses.

    13. Re:NAT by SQLGuru · · Score: 4, Funny

      Until each electron, proton, and neutron needs an address......and each quark....etc.

    14. Re:NAT by Rich0 · · Score: 2

      Additionally, the DHCP server wouldn't know that a client had left, so the addresses would be occupied long after no longer being in use.

      I agree with most of what you said, but there is no reason that the DHCP server couldn't know that the client had left. The cell tower knows what phones are and are not in the area. Plus, lease times could be really short - maybe a minute or two - even if that were not handled.

    15. Re:NAT by wosmo · · Score: 2

      $100k hardware firewall? what he's talking about is an easy task for reverse proxy / load balancing. Has been for years. It's a very common setup where you have multiple worker nodes answering. It's the typical setup for 'elastic' style amazon stuff. one load balancer, as many nodes as the current load requires behind it.

  2. Probably the home router... by neilo_1701D · · Score: 4, Insightful

    When that particular comment was made, the ubiquity of the home router dolling out DHCP addresses probably wasn't considered. Nowadays, you only need one IP address for your home and let the router sort it out.

    There's still a problem, but people seem to prefer to adapt and come up with (very) clever workarounds rather than get some new solution shoved down their throat that renders existing equipment obsolete for no good reason.

    1. Re:Probably the home router... by wisnoskij · · Score: 3, Insightful

      How does it even work any other way?

      Are you/the article saying that it is possible to have a single connection to your ISP, but for every computer, fridge, toaster, TV, etc. to have its own global IP address?

      Your ISP can give you a block of dynamic/static IP addresses, which your router assigns instead of 192.168.1.X?

      --
      Troll is not a replacement for I disagree.
    2. Re:Probably the home router... by sribe · · Score: 4, Informative

      Good luck with web apps that use IP based sessions.

      Are you kidding me??? That stopped being even remote practical about 20 years ago.

    3. Re:Probably the home router... by exabrial · · Score: 2

      If you're using IP based sessions you're a moron. Only the RIAA/MPAA makes that argument.

    4. Re:Probably the home router... by Chris+Mattern · · Score: 4, Interesting

      Are you/the article saying that it is possible to have a single connection to your ISP, but for every computer, fridge, toaster, TV, etc. to have its own global IP address?

      Yes, that is exactly how IPv6 is supposed to work.

      Your ISP can give you a block of dynamic/static IP addresses, which your router assigns instead of 192.168.1.X?

      Possibly, but not necessarily even that. You could be set up to simply automatically generate IPv6 addresses from your MACs, and the ISP doesn't even explicitly grant you an address block.

    5. Re:Probably the home router... by Chris+Mattern · · Score: 3, Insightful

      Being horrified by NAT is all well and good, but the fact is, ISPs look at the horrible bandaids that work 80% of the time and say, "Good enough. Now I don't have to rebuild my entire infrastructure for IPv6." You may want something that works 100% of the time, but the people who own the equipment don't want to *pay* for something that works 100% of the time.

    6. Re:Probably the home router... by Antique+Geekmeister · · Score: 4, Informative

      As it stands, your carier does NAT themselves and gives your router one IP address, typically in the 10.0.0.0/8 address space. Your home router then does another layouer of NAT, and gives internal devices their own IP address range in the 1902.168.1.0/16 address space. The advantagie is that one can support a _tremendous_ backend infrastructure without public IP addresses. This is also a tremendous security advantage: it reduces the exposed attack surface for script kiddies and casual network scanners to attack your home devices, they have to successfully gain control of the router or another device inside your network to pass along their attack.

      The disadvantage, which dismays some people, is that NAT channels _publication_ of services through those NAT enabled routers or through externally hosted web space. It effectively makes the allocation of IP addresses and ports for exposed services require more thought, and allows easier throttling or monitoring of traffic at those NAT routers. I've found it to be a tremendous security and network management improvement: it makes firewall and routing design _much_ more stable and helps prevent people from running dangerous, unauthorized services from office networks, such as running public NFS servers without telling anyone aware of the security implications.

    7. Re:Probably the home router... by smash · · Score: 4, Informative

      This is exactly how IP (irrespective of version) is supposed to work... NAT is an ugly hack that breaks shit.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    8. Re:Probably the home router... by tlhIngan · · Score: 2, Insightful

      Are you/the article saying that it is possible to have a single connection to your ISP, but for every computer, fridge, toaster, TV, etc. to have its own global IP address?

      Yes, that is exactly how IPv6 is supposed to work.

      And this is where fundamental assumption #1 of IPv6 falls flat. Even with IPv6, every endpoint will not be reachable.

      This is the age of firewalls and all that (and even NAT provides a very basic level of firewalling). There's no guarantee that despite an endpoint having a publicly available address that it'll be reachable. Even today if a company has dozens of public IPv4 addresses for hosts, there's no guarantee that it'll be accessible.

      Which means everything still breaks just as if NAT was present.

      Even if IPv6 took the world over by storm, firewalls will still be around breaking connectivity. Even worse than NAT, you can't easily detect this condition. You might have a publicly visible address, but the firewall prevents you from establishing a connection. Or you may bind a port to serve something and the firewall blocks access.

      In fact, the early days of NAT had those problems, but these days it's largely mitigated because of many techniques.

      Possibly, but not necessarily even that. You could be set up to simply automatically generate IPv6 addresses from your MACs, and the ISP doesn't even explicitly grant you an address block.

      But it may decide that you get a static IP and firewalls everything else off. E.g., even though you're advertised a /64, your ISP may filter out everything but <prefix>::1. If you ask for another "IP", because ISPs love to sell you more, they'll just hand you another prefix.

      And finally, the biggest hurdle for IPv6 is NAT. Because NAT has a very nice side effect if you're maintaining a network of any size - it isolates internal network numbering for external network numbering. It doesn't matter what IP your ISP hands you for IPv4 - because NAT automatically hides it from internal clients. All they need to know is if they can see the router and magic happens.

      With IPv6, you lose this handy feature - your ISP decides to change your prefix? Well, damn, they haven't done that in 5 years and now everything has been hardcoded with the old prefix in it - all your internal services used it.

    9. Re:Probably the home router... by hjf · · Score: 2

      (proper) CGNAT uses 100.64.0.0/10, so it doesn't collide with RFC1918 reserved addresses. See: RFC6598.

    10. Re:Probably the home router... by kasperd · · Score: 4, Informative

      Which means everything still breaks just as if NAT was present.

      That is not correct. A properly configured firewall does not cause nearly the same level of breakage as a NAT does.

      And finally, the biggest hurdle for IPv6 is NAT.

      That is true. NAT is hurting IPv6 deployment in many ways. Had NAT never been invented, we could all have been running IPv6 years ago, and the transition would have gone smoother. For example a large part of the difficulties in using IPv6 through tunnels is entirely due to the IPv4 connections being infested with NATs.

      With IPv6, you lose this handy feature - your ISP decides to change your prefix?

      With IPv6 there are enough addresses, that this should happen very rarely.

      Well, damn, they haven't done that in 5 years and now everything has been hardcoded with the old prefix in it - all your internal services used it.

      Then use DNS and/or RFC 4193.

      --

      Do you care about the security of your wireless mouse?
    11. Re:Probably the home router... by ttucker · · Score: 2

      Your carrier does not give you these addresses. Anyone can use them on a local network. http://en.wikipedia.org/wiki/P...

    12. Re:Probably the home router... by znark · · Score: 3, Informative

      As it stands, your carier does NAT themselves and gives your router one IP address, typically in the 10.0.0.0/8 address space. Your home router then does another layouer of NAT, and gives internal devices their own IP address range in the 1902.168.1.0/16 address space.

      Not where I live, and that sounds quite limiting! Thank ${DEITY}, ISPs here in Finland assign their customers genuine public IPv4 addresses, usually via DHCP. Typically, you can even get several of them – the maximum on a consumer connection could be something like 5. (I’m using 2 right now.) Only something like the port 25 (SMTP) is blocked for inbound connections so you’re free to run a personal web server, SSH box, VPN to your home network, etc.

      Finnish cellular carriers – as opposed to the actual fiber/copper/cable ISPs – have a different practice, though: they will usually NAT the 3G/4G customers by default, which is quite understandable, as you generally do not want inbound connections to a cellphone. Still, at least my carrier (Saunalahti) lets advanced customers choose a different APN which will give a public IPv4 address even for a 3G modem or a cellphone, which is quite nice and handy as well for some situations.

    13. Re:Probably the home router... by Macrat · · Score: 3, Informative

      Being horrified by NAT is all well and good, but the fact is, ISPs look at the horrible bandaids that work 80% of the time and say, "Good enough. Now I don't have to rebuild my entire infrastructure for IPv6."

      And yet Comcast is rolling out IPv6. I'm on IPv6 at home today.

    14. Re:Probably the home router... by AdamHaun · · Score: 2

      Your ISP can give you a block of dynamic/static IP addresses, which your router assigns instead of 192.168.1.X?

      That's how the internet works to begin with, and it used to be the norm for IPv4 networks. A lot of large networks still do it that way -- the computer I'm on at work has a globally unique IP address. You can still get a block of static IPs if you buy a business-class connection. That used to be almost the definition of a business connection, back when more people ran their own servers instead of using hosting services. IP addresses cost money, so ISPs try to have as few as possible. NAT came about when people started getting multiple computers per household but didn't want to pay for a business connection. It was never meant to prop up the internet as a whole.

      --
      Visit the
    15. Re:Probably the home router... by InvalidError · · Score: 3, Insightful

      Most carrier-grade equipment has a useful service life of 7-8 years and practically all carrier-grade equipment that got on the market in the last 10 years does support IPv6.

      At the customer edge of the network, those upgrades are necessary to enable VDSL2 and DOC3. In the network core and backbones, router upgrades are necessary every ~7 years because new router generations have 3-4X the routing capacity per RU and bandwidth per watt as older equipment which is a major saving in floor space, power and cooling bills. Trying to cope with the 40-70%/year traffic growth using hardware from 6+ years ago would be practically impossible.

      Until traffic growth collapses, carriers and everyone else involved in large-scale transit does not have a choice to refresh large chunks of their network periodically to accommodate demand.

    16. Re:Probably the home router... by Dagger2 · · Score: 2

      Grownups that work at proper telcos replace their entire infrastructure on a regular basis anyway, so that part has already happened.

      v4 to v6 migrations are pretty well thought out, people just need to actually do them.

    17. Re:Probably the home router... by kasperd · · Score: 2

      It is most likely to happen if you change ISP, or your ISP is taken over by another company that already serves your area.

      Agreed. And for those reasons you should avoid being dependent on the assignment from the ISP being static forever.

      There was an attempt at building some renumbering logic into the DNS records (with the A6 records). But A6 records were eventually dropped with the reasoning that AAAA records were simpler and using proven methodology (due to their similarity to A records), and that the renumbering logic could be done in tools for generating zone files without needing complicated standards to be applied at lookup time, and finally that looking up an AAAA record required fewer roundtrips than looking up A6 records.

      The tools that should have emerged to help in renumbering AAAA records haven't shown up at any of the DNS providers I have been working with, but the lack of such tools have certainly not given me any reason to prefer IPv4 over IPv6.

      So all in all, it is a solved problem. One just need to pick the preferred solution among those that exist.

      --

      Do you care about the security of your wireless mouse?
    18. Re:Probably the home router... by Bacon+Bits · · Score: 2

      Please stop arguing that NAT gives you a security advantage. NAT in and of itself does not provide any additional security. The advantage is simply that of a stateful firewall, which is typically what is used to provide NAT -- except you can't really configure it. If you want security, run a stateful firewall and manage your services correctly at that firewall. NAT is lazy, NAT is sloppy, and NAT doesn't allow you to prevent users from connecting to remote services you don't want them to.

      --
      The road to tyranny has always been paved with claims of necessity.
  3. the skynet is falling the skynet is falling by Anonymous Coward · · Score: 3, Funny

    had to say that

  4. CGN, perhaps? by Zocalo · · Score: 5, Interesting

    Just a guess, but maybe widespread adoption of Carrier Grade NAT might have given IPv4 a bit of a longer shelf life. It's either that or the kind of fun and games that I once read that Hutchison (Orange) was doing on their mobile network, with no less than seven separate instances of the 10/8 network being juggled around at once.

    Still, even ARIN is now starting to tighten the screws on the size of netblocks they are assigning out, so I suspect providers are being a lot more careful about how they subnet and assign out IP addresses than they used to be. I suspect that just moving stuff like DB servers and other backend infrastructure onto private IP space instead of just dumping them in the DMZ for convenience has helped a bit too, not too mention being a better security practice.

    --
    UNIX? They're not even circumcised! Savages!
    1. Re:CGN, perhaps? by aztracker1 · · Score: 2

      I think it comes down to understanding and ability... not to mention capabilities of certain hardware. It's relatively easy for an engineer to keep a series of IPv4 address blocks, and routes in memory... not so much with IPv6. IPv6 offers a lot of ability, but the complexity exceeds what most people can keep in their heads at a given time, and this scares a lot of people, and is for many more complicated than NAT band-aids.

      I've had getting more into IPv6 and DNS (Bind9) on my todo list for about two years now... I don't have to work with it, and it's not my highest priority as a programmer. But I do know that it has its' own complications over IPv4.

      --
      Michael J. Ryan - tracker1.info
    2. Re:CGN, perhaps? by smash · · Score: 2

      carrier grade NAT double NAT, etc. is a lot more complex than an IPv6 network. the only real complexity in and ipv6 environment (excluding bugs in firmware, but that isn't TOO bad these days) is having to maintain dual stack until the laggards wake the fuck up and upgrade.

      --
      I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
    3. Re:CGN, perhaps? by DamnOregonian · · Score: 3, Interesting

      It's not a superior solution. I'm a senior network engineer at a local ISP. Our infrastructure is IPv4 and IPv6, with a chunk of fiber customers running on CGNAT. We're not even that big, but equipment that can route IPv6 with line-speed forwarding throughout the core and distribution side of the network (as well as supporting the dynamic routing protocols necessary to manage the network) is fantastically more expensive than either purchasing a CGNAT setup, or building one out of Linux (our solution). I can't even imagine the cost for someone with a large network.

      That doesn't even get to the myriad of major problems with customer-facing IPv6. The specification with regard to deployment is frankly garbage (the people who wrote the spec[s] clearly had little background in actual customer distribution networks). We couldn't be more eager to get every single one of our customers running on it, especially given how quickly our ARIN allocations are drying up, and the unlikeliness of people our sized being able to acquire more, short of acquiring the blocks of ISPs that we purchase.

      I think it's really easy for a lot of arm-chair network engineers to scoff at the speed of the ISP-side IPv6 roll-out, but the costs and technical limitations of the spec, which have required many bandaids and workarounds just to make function in a way that could even remotely be called reliable for residential customers, scales with the size and diversity of our customer base. It's a bitch.

    4. Re:CGN, perhaps? by Anrego · · Score: 2

      I think the key word is customer transition.

      The appealing thing about carrier grade NAT is most consumers won't even notice.

      Some people might claim that ipv6 could be done transparent to the end user, but personally I think that's a load of BS, and I suspect so do ISPs.

      How to transition from ipv4 to ipv6 would to me seem the most important consideration when designing ipv6, but form appearances it seems like it was an afterthought, which is probably why we'll have ipv4 for quite some time.

    5. Re:CGN, perhaps? by Dcnjoe60 · · Score: 2

      carrier grade NAT double NAT, etc. is a lot more complex than an IPv6 network. the only real complexity in and ipv6 environment (excluding bugs in firmware, but that isn't TOO bad these days) is having to maintain dual stack until the laggards wake the fuck up and upgrade.

      That may be true, but carrier grade NAT, double NAT, etc. run on today's infrastructure.

    6. Re:CGN, perhaps? by SuricouRaven · · Score: 3, Informative

      Switches shouldn't need to do IPv6. They operate at the next level down, so it makes no difference, except for the management interface. That's the idea anyway - there are sometimes issues with programers assuming things they should not assume.

    7. Re:CGN, perhaps? by kasperd · · Score: 2

      The appealing thing about carrier grade NAT is most consumers won't even notice.

      Some people might claim that ipv6 could be done transparent to the end user, but personally I think that's a load of BS, and I suspect so do ISPs.

      Either approach will cause problems for the end users. In both cases the users will have no clue what is the root cause of the problem, and they will believe whatever bullshit their ISP tells them. My ISP ran out of IPv4 addresses, thus some of my devices got no reply from the DHCP server. The ISP tried to convince me that it was due to a defective network interface on my end.

      How to transition from ipv4 to ipv6 would to me seem the most important consideration when designing ipv6, but form appearances it seems like it was an afterthought, which is probably why we'll have ipv4 for quite some time.

      That is true. I saw this coming more than a decade ago, unfortunately I was not in a position to do anything about it. Why didn't any of the people who were in a position to do something about it see it coming?

      Last year I did some calculations on how fast the transition should have been going to complete on time (i.e. before IPv4 addresses ran out). For the past year the transitioning has been going at the speed, which I calculated. Meaning at the start of 2013 we were 13 years behind schedule and at the start of 2014 we were still 13 years behind schedule. I'd say my calculations would have been realistic, if only people had started soon enough rather than waiting.

      The core problem was the lack of incentive to get started. A good incentive would have been rationing of IPv4 addresses much earlier. Instead of handing out 98% of the addresses before rationing IPv4 addresses, it might have been a good idea to only hand out 50% and then start rationing. The rule could then have been that you could get more IPv4 addresses if you were rolling out native dual stack. With such a rule the amount of IPv4 only deployments could have been kept static while dual stack was growing. By the time IPv4 ran out, we would have had 50% dual stack already.

      But looking back and pointing out what should have been done 15 years ago isn't going to solve the problem we have today. And IPv6 deployment is still not large enough for market forces alone to give incentive to upgrade. There are ISPs rolling out CGN solutions with no plan about deploying native IPv6. How do we give those ISPs a strong incentive to start upgrading?

      I have been trying to produce some incentive to move in the right direction. I am open to more ideas on what I personally can do to give ISPs the right incentive.

      --

      Do you care about the security of your wireless mouse?
  5. Only if you can't get addresses by Anonymous Coward · · Score: 5, Interesting

    It's only a crisis if it affects you. (sic)

    That's basically what is happening, a giant stand off between the access networks and the hosting providers looking who will blink first.

    From then end user perspective, you should see what happens to Skype and games when both end-users are behind a double NAT, it's hilarious. But most people seem to cope just fine.

    For the hosting providers then fun really starts when you can't get a public IPv4 for your new webserver, that'll be fun. There's no NAT workaround for that, some european hosting providers are already feeling the crunch in their IPv4 blocks, you can only host so many servers. So what can you do? Jack up the prices ofcourse, isn't the free market wonderful!

    If you are a business in the EMEA and you still want or need your own PI space for BGP, tough cookies, you can't get it anymore.

    1. Re:Only if you can't get addresses by C3ntaur · · Score: 4, Interesting

      For the hosting providers then fun really starts when you can't get a public IPv4 for your new webserver, that'll be fun. There's no NAT workaround for that, some european hosting providers are already feeling the crunch in their IPv4 blocks, you can only host so many servers. So what can you do? Jack up the prices ofcourse, isn't the free market wonderful!

      This. This is why IPv4 will stick around for decades to come. There is too much profit potential in it, and IPv6 costs too much money to implement.

      --
      Loading...
    2. Re:Only if you can't get addresses by vanyel · · Score: 2

      Having implemented ipv6...bs. It does cost some time and effort, but it's not huge, particularly if you do it incrementally and dual stack. It's fear of change that's holding it back, not cost and effort, and as a result people are missing out on getting out from under that shackles that ipv4 puts around everything you do. But "the devil you know" rules in all too many cases.

    3. Re:Only if you can't get addresses by Dynedain · · Score: 2

      For the hosting providers then fun really starts when you can't get a public IPv4 for your new webserver, that'll be fun. There's no NAT workaround for that, some european hosting providers are already feeling the crunch in their IPv4 blocks, you can only host so many servers. So what can you do? Jack up the prices ofcourse, isn't the free market wonderful!

      There is certainly a NAT-like workaround for lack of IPv4 for webservers. It's called a load-balancer. Since the domain name requested is in the HTTP header, it's easy to route the request to different hardware behind the front machine based on domain name. In fact, typical Apache configuration relies heavily on domain name being in the HTTP header.

      --
      I'm out of my mind right now, but feel free to leave a message.....
    4. Re:Only if you can't get addresses by Lanboy · · Score: 2

      Large corporate entities are also selling address space. Bought a class B for a Million last year. Not personally, but the corporate entity I represent.

      If the ipv6 standards group had made an incremental change to address address space and left the rest of the protocols, then things might be different. As it was they threw in a bunch of features that no one wanted, and no one needs. IPv6 is a rehash of the failed and unused OSI transport and intranetwork protocols, which were soundly rejected by the market. The problem was that no one really took a good look at the protocols outside of the OSI because the working engineers were too busy keeping the networks on.

      The good thing is that the longer adaption period is enabling older OSs, routers and incompatible switches to drop out of use and the market. Most carriers have had it in the lab for 10 years, vendor bugs have been discovered and fixed without melting down the internet. So what we are waiting for now is the cost benefits for ipv6 to show up. Considering that fixed space ipv4 is now something you can SELL, widespread use of ipv6 is actually of negative value for the carriers.

      Meanwhile, the large planned ip6 conversions haven't happened. The US Army passed their 2008 deadline and said F it, we will switch when we need to. Comcast uses it for cable box and element addressing, but not for internet access. APAC should be driving the adaptation, but OSs and router versions are primitive in most areas of the APAC, so the people with the most to gain have the riskiest road forward.

  6. I'm waiting for by rossdee · · Score: 4, Funny

    IPv8.1

  7. RFC 1918 by toupsie · · Score: 2

    I guess enough people finally got around to reading it.

    --
    Strange women lying in ponds distributing swords is no basis for a system of government.
  8. They probably looked at the last mile problems by Marrow · · Score: 3, Interesting

    and figured out they better find a better solution than ipv6. There is too much ipv4 only hardware out there to abandon it all. It would just be insane.

  9. Re:By all Means then by arth1 · · Score: 2

    Let's shitcan IPV6 right now, who needs it, because clearly because some people were concerned it's a reason to ignore it all now and keep using ipv4

    Your analogy fails, because IPv6 brings extra functionality, including routing advantages. It's not just an attempt at dumbing down for MBAs and the unwashed masses.

  10. Arin is alone by pcjunky · · Score: 2

    While things have slowed down here the other regional IP registars have run out. APNIC and RIPE both have no IP addresses left. Arin has only about 1.4 /8's left.

    1. Re:Arin is alone by PRMan · · Score: 2

      MIT was fast to the land grab and ahead of the curve. I don't believe in taking away these addresses. If they want to sell them and make some money, fine. But the addresses are theirs.

      --
      Peter predicted that you would "deliberately forget" creation 2000 years ago...
  11. Re:10 years by wonkey_monkey · · Score: 4, Informative

    Even through all addresses have been given out

    They haven't:

    the American Registry for Internet Numbers is still doling out IPv4 addresses

    ARIN currently has “approximately 24 million IPv4 addresses in the available pool for the region,” according to President and CEO John Curran.

    --
    systemd is Roko's Basilisk.
  12. Re:By all Means then by OS24Ever · · Score: 2, Funny

    That wasn't an analogy. That was sarcasm.

    --

    As a rock-in-roll Physicist once said, No matter where you go, there you are.

  13. IPv6 has this tiny problem by cowwoc2001 · · Score: 2

    "Hey Joe, what's your IP address?"
    "Oh, let me see... it's fe80:0:0:0:200:f8ff:fe21:67cf"

    Holy crap that's long. The second IP addresses become this difficult to exchange verbally, we're going to stop referring to them altogether.

    1. Re:IPv6 has this tiny problem by Sique · · Score: 5, Insightful

      That was the point of having DNS in the first place. Four octets just weren't bad enough.

      --
      .sig: Sique *sigh*
    2. Re:IPv6 has this tiny problem by Imagix · · Score: 2

      There's this really interesting service out there that converts from a human-friendly (well, friendlier anyway) form to the IP address. Perhaps you've heard of it. It's called DNS. (and BTW, you just quoted a link-local IPv6 address... so the guy who wants to talk to Joe probably can't use it anyway...)

    3. Re:IPv6 has this tiny problem by infogulch · · Score: 4, Insightful

      Fixed:

      "Hey Joe, what's your IP address?"
      "I don't have one, I'm behind a NAT and firewall that I don't control."

      Of the two problems, I find yours the lesser of two evils.

    4. Re:IPv6 has this tiny problem by cowwoc2001 · · Score: 2

      So my parents have to learn how to configure a DNS in order for me to troubleshoot their networking problems over the phone? :)

      On a more serious note, I don't see the possibility of getting non-techies to configure DNS entries for their computer.

    5. Re:IPv6 has this tiny problem by SuricouRaven · · Score: 2

      "I was going to ask if you'd start up the starcraft server and play a round."

    6. Re:IPv6 has this tiny problem by Dagger2 · · Score: 2

      Loops like for i in {1..50}; do ssh host${i}.cluster1.domain.com stuff; done work just fine with v6, and are no harder to remember than the same thing for v4 (since all you do is use AAAA records instead of A records.)

  14. Bad summary by AdamHaun · · Score: 4, Informative

    Unsurprisingly, address exhaustion still going on. APNIC and RIPE are down to their last /8 and are now handing out addresses as slowly as they can. ARIN and LACNIC will reach their last /8 this year. AFRINIC won't run out for years, so I suspect their new infrastructure will be built on IPv6. Here's the relevant data.

    There's a finite number of addresses, guys. They're not going to magically stop running out.

    --
    Visit the
  15. ISPs taking IPs back from customers by kasperd · · Score: 4, Informative

    Less than two months after RIPE introduced rationing of IPv4 addresses, I one day found my internet connectivity to be totally broken. Turns out the ISP had turned on NAT in my modem (without telling me about it beforehand). They did have a self service page where I could turn NAT off again and get functional internet connectivity again. However some of my devices no longer received any reply from the DHCP server.

    I called their support, who said the lack of reply from their DHCP server was due to the network interface on my computer being defective (which was obviously a lie). When I pointed out that their conclusion was directly contradicting the symptoms I had already explained them about, they just hanged up.

    Calling their support one more time, I was able to get to a supporter who knew what was going on, and didn't just invent a lie. It turns out they had run out of IPv4 addresses, and were now enforcing a maximum of two devices online per customer regardless of what limit had been in effect previously.

    A few days later I called them again asking for native IPv6, which I considered only fair, given that they had taken away some of the IPv4 addresses, which I were using. They promised me native IPv6 before the end of the year. That was in 2012, they still haven't delivered.

    Other ISPs are putting all new customers behind CGN unless they pay an extra fee for a static IP address. You'd think they'd give you native IPv6 along with that. But alas, according to the majority of ISPs, there is no shortage of IPv4 addresses in this country, so nobody needs IPv6. And since nobody is buying IPv6 connectivity, the ISPs will not offer it (completely ignoring the fact, that the reason nobody is buying IPv6 connectivity is that the ISPs themselves aren't offering it in the first place).

    From what I am told, native IPv6 plus CGN for IPv4 is already fairly common in Germany, but that's not enough to make me want to move across the border. I have yet to hear about ISPs putting customers who previously had a public IPv4 address behind NAT, but I would not be surprised if it happened.

    --

    Do you care about the security of your wireless mouse?
  16. Comcast and ipv6 by weave · · Score: 2

    Comcast brags (http://comcast6.net) that they are the largest ISP that supports ipv6. Oh wow, cool. I have a new modem that supports it as well as a home router.

    So I go to figure out how to do it and find that they are only assigning /128s (single IPs) to only certain markets.

    Who has a single computer hooked up to the Internet at home and nothing else?

    No wonder it's not going anywhere. Even early-adopters can't get on easily without tunneling or other hack.

    1. Re:Comcast and ipv6 by Aqualung812 · · Score: 3, Interesting

      I'm on Comcast, and I'm getting a /60 from them.

      Your WAN interface might be on a /128, and that is fine. You need to make sure your gear is telling Comcast what size of prefix you want delegated to your router.

      Of course, this varies by market, so it might really not be there yet, but read up on prefix delegation & make sure you've got your end setup correctly: http://en.wikipedia.org/wiki/P...

      Also, don't trust the tech support with this. They are clueless. According to them, IPv6 isn't available in my market.

      --
      Grammer Nazis - I mod you "troll" unless you actually add something on-topic. Yes, I know I have mispellings in my sig.
  17. The real crisis is the routing table size problem by exabrial · · Score: 3, Interesting

    Truth is NAT works just fine for the vast majority of cases, and makes a layered (IE not-eggs-all-in-one-basket) approach to security much simpler.


    The real problem is routing table size with BGP. As we continue to divide the internet into smaller routable blocks, this is requiring an exponential amount of memory in BGP routers. Currently, the global BGP table requires around 256mb of RAM. IPv6 makes this problem 4 times worse.


    IPv6 is a failure, we don't actually _need_ everything to have a publicly routable address. There were only two real problems with IPv4: wasted space on legacy headers nobody uses, and NAT traversal. IETF thumbed their noses as NAT (not-invented-here syndrome) and instead of solving real problems using a pave-the-cowpaths-approach, they opted to design something that nobody has a real use for.

    Anyway, I'm hoping a set of brilliant engineers comes forward to invent IPv5, where we still use 32 bit public address to be backward compatible with today's routing equipment, but uses some brilliant hack re-using unused IPv4 headers to allow direct address through a NAT.

    Flame away.

  18. It is just costing us $$$ at this point by Anonymous Coward · · Score: 2, Informative

    At work we wanted to set up some VPNs with a cloud provider but our ISP doesn't want to give us the IPs so we had to forgo the VPN and instead lease a line for $5000 a month + we'll end up with dev and production envirnments that don't match which will probably hit us as some downtime in the future (we're just using OpenVPN in dev which doesn't require an IPv4).

    So in the case of my team of eight workers the IPv4 crisis is costing $5000/mo + countless meetings and endless paperwork. Not a showstopper, but enough that I'm not yelling "What Crisis?!" from the rooftops.

  19. What happened? by GT66 · · Score: 2, Insightful

    The human tendency for hyperbole happened. It was the same for Y2k, is the same for just about every winter season snow storm, and is ceaseless in our politics. We just love the drama of a crisis. Just recently John Kerry referred to man-made global warming as weapon of mass destruction. Talk about a drama queen. [br] [br] So, as it turned out, despite seemingly needing more than billions of IP addresses and IPv4 only supplying a few billion in totality, what the world really needed was just a few million IPv4 addresses that could provide "outside" initiated connectivity into the host. ie, servers. For all the rest, outbound connectivity could be supplied by some smaller proportion of addresses using NAT and clever work around services and many systems required even less than that needing only local area connectivity and allowing IPv4 to be reused over and over. [br] [br] So, the need for IPv6 RIGHT NOW OR THE END WILL CONSUME US! was driven largely by hyperbole and the reality that IPv4 can and will continue to serve our purpose is tempered by the other human traits of conservation and ingenuity. [br] Yes, the transition to IPv6 is inevitable and necessary however, the consumption of IPv4 will not be no more a sudden catastrophic event event any more than John Kerry's belief that climate change is a weapon of mass destruction. It just never happens that way.

    1. Re:What happened? by Dagger2 · · Score: 4, Funny

      Thankfully they keep discovering new fields of IPv4 addresses. Peak IP is never going to happen!

    2. Re:What happened? by Zero__Kelvin · · Score: 5, Insightful

      ". It was the same for Y2k,"

      I'm glad you brought this up, as it is an excellent parallel. The Y2K crisis was real just as the IPv4 shortage was real. In both cases people took pro-active steps to head off disaster. Now, because those proactive steps averted the disaster all those who had no hand in it and didn't understand it proclaim: See! It was never an issue! It didn't happen!. No shit sherlock; it didn't happen because people saw the potential for disaster and took steps to avoid it.

      --
      Guns don't kill people; Physics kills people! - John Lithgow as Dick Solomon on Third Rock From The Sun
    3. Re:What happened? by WaffleMonster · · Score: 3, Interesting

      The human tendency for hyperbole happened.

      Or more accurately "does not effect me"

      It was the same for Y2k, is the same for just about every winter season snow storm, and is ceaseless in our politics.

      In the IPv6 case the projections for run out have been right on the money. The only people screaming "the world didn't end" are media people looking to whore hits to their sites. Addressing authorities and publicized events ( IPv6 Day) all included FAQs clarifying the end of the world does not happen at exhaustion.

      Just recently John Kerry referred to man-made global warming as weapon of mass destruction.

      I have a feeling if you were head of state for some dinky island nation in the middle of nowhere and you looked at the projections for sea level rise vs land area of your country effectively consumed or endangered by conditions (tides, storms) you would not be so quick to sound the hyperbole alarm.

      The same goes for small VM/hosting provider who runs out of IPs to assign to new customers... these things are a "big fucking deal" to them but for everyone else it is hyperbole or even beneficial. Climate change has winners and so does IPv4 exhaustion. CGN vendors, competitors who "planned ahead" hoarding more addresses than they were supposed to or those blessed with massive legacy allocations have market advantage with respect to IPv4 exhaustion the rest of us don't.

  20. The US has nothing to worry about but... by trparky · · Score: 5, Informative

    The United States has enough IP addresses in our pool to carry us through to the end of say... 2018. If current growth of the Internet continues we will still have enough IP addresses in our pool, we'll just have to knock a year or two off that projection. Say, may 2017 or half way through 2016. The United States has more than enough IP addresses to keep us going for some time.

    Europe and other parts of the world is a totally different story. When the Internet was created and we started handing out the IP addresses we were quite stingy when giving them to other parts of the world. The United States is one of the biggest hoarders of IP addresses in the IPv4 world while Europe and the rest of the world got relatively few IP addresses with compared to how many the US holds. There's where we are seeing the problem.

    Europe has the issue, Europe has no choice in the matter; they have to move to IPv6 or their side of the Internet is pretty much crippled. So unless we all implement 6to4 to allow United States Internet users to connect to European web site (that's fugly) or finally get on the bandwagon in converting to IPv6 in the US, there will eventually be two Internets; a US and a European Internet with IPv4 and IPv6 being the limiting factor.

  21. Re:The real crisis is the routing table size probl by Typical+Slashdotter · · Score: 5, Informative

    IPv6 is designed with such a large address space specifically to make BGP tables smaller. One of the factors causing IPv4 tables to grow is that, since addresses are scarce, people are getting clever with how they allocate blocks, divvying things up very finely so as not to waste. Since BGP entries are by block, this creates many blocks that need routing. The IPv6 designers went with 128 bits of address not because they think they need room for 2^128 hosts, but because there will be enough room to divide blocks hierarchically and logically, "wasting" addresses all along the way. This will allow global routing tables to more accurately reflect the structure there is between ISPs, shrinking their size.

  22. Already on IPv6 by TyFoN · · Score: 2

    My fiber ISP provides 6rd connectivity with a /62 prefix address space, and will bump it to /54 when they implement dual-stack on all systems.
    There are still legacy routers on the system apparently.

    However tomato on my rt-n66u handles the 6rd just fine.

    A lot of systems are on ipv6 already, and I think I have around 50/50 ipv6 and ipv4 traffic now. There is no real difference in use for a regular user. Even all the phones, tables and the chromecast use it without me having to do anything except connecting the router.

    I still have a regular fixed ip for ipv4, but all my devices are behind nat.

  23. It's there, just wait and see by Morgor · · Score: 4, Insightful

    In short, it's just too early to tell. Just because the RIRs ran out of addresses, it doesn't mean that the LIRs have yet (the ISPs).

    Based on my experience as a network engineer at an ISP, the following is happening already:

    Small ISPs and ISPs that have not been in the business for a long time* have either run out or are on the verge of doing so. They are doing the following:

      * Purchasing legacy IPv4 addresses from enterprises with /16 networks from the old days where available.
      * Deploying CGN-like solutions for their end-customers if their end-customers are residential users.

    Larger ISPs and older ISPs with allocations from ye old pre-RIR days continue to hold addresses and are often able to free large quantities of addresses from old deployments. Mind you, a lot of public IPv4 space have been "wasted" on infrastructure addressing, and management of devices that were not even connected to the internet. Devices such as modems, DSLAMs, CPEs and similar.

    One could easily speculate that the business of ISPs will be severely affected in the future, as customers will go to the old providers that have plenty of v4-space available at the cost of newer players who followed the RIR regulations of only applying for the address space they needed based on relative short-term predictions.

    If you are a registered LIR you will see a flood of SPAM from so-called IP brokers who are trying to purchase unused IPv4 space in hope of selling this to LIRs in need. That market will probably become quite desperate in the coming years.

    Oh, and by the way, I see no evidence that IPv6 deployment is taking any noticeable speed.

    *) Long as in they were in the game when classfull allocations were made.

    1. Re:It's there, just wait and see by grunthos · · Score: 2

      If you are a registered LIR you will see a flood of SPAM from so-called IP brokers who are trying to purchase unused IPv4 space in hope of selling this to LIRs in need. That market will probably become quite desperate in the coming years.

      Yeah, one contacted me about an old /16 block assigned to a company where I was network manager 20 years ago, wanting to make a deal. (Company went bankrupt, got bought up, buyer went bankrupt, got bought up, and so on). The would-be brokers are digging up every possibly-unused block they can.

      I contacted ARIN and released it back to the pool.

      Interesting, all the details that come back to mind, even though I hadn't thought about them in two decades :-)

      --

      My son's 5th grade teacher actually assigned them "write a limerick about a planet". I'm not kidding.
  24. Re:Chicken little by Anonymous Coward · · Score: 5, Informative

    You can't get new IPv4 addresses in Europe or Asia. End users are already on DS-lite, with IPv6 for their only public address. You can not initiate a connection to millions of Europeans and Asians if you don't use IPv6. Not soon, now.

  25. Re:Chicken little by Bert64 · · Score: 3, Informative

    Actually, ipv6 adoption seems to be higher in the US than anywhere else in the world... I run a bunch of dual stack websites, and v6 accounts for about 15% of american traffic and considerably less from other countries.

    --
    http://spamdecoy.net - free throwaway anonymous email - avoid spam!
  26. Re:Chicken little by Alain+Williams · · Score: 4, Informative

    It was never a crisis to begin with? This is why you don't listen to chicken littles.

    I don't know where you live, but at a guess I would put you in a country such as the USA or in United Kingdom. If you look at how many IP addresses there are per 1,000 population you will see that the USA has about 5,000, the UK 2,000 but that India has 29. So it might not be a problem for you, but for for some it is. It is not just 1st vs 3rd world, overall the EU has 19 per 1,000.

    Many people use more than one IP address (think: office, home, mobile 'phone). Yes NAT can help, but it is not the complete answer.

  27. Re:The real crisis is the routing table size probl by smash · · Score: 2

    Your phone isn't trying to route at terabits per second.

    --
    I run: Windows, OS X, Linux, FreeBSD. Just because you have a hammer, doesn't mean everything is a nail.
  28. Re:The real crisis is the routing table size probl by Antique+Geekmeister · · Score: 2, Interesting

    I sat in on a router design meeting for IPv6. It took me 20 minutes to stop laughing when I heard them seriously say that it was acceptable for the system to crash if it encountered a router loop, because users will "just be careful and that won't happen". Then I took the copy of the presentation and my notes to my stock analyst and pointed out "these people ar bozos, do not invest in them or trust anyone who has invested in them". I didn't make money, but it helped keep me from *losing* a good chunk of money when their "Cisco-killer" failed miserably.

  29. Are you an Idiots, Trolls or Innovator? by jsklein · · Score: 2

    The wining of many of the people on this list make me laugh. I heard the same thing from the Novell, SNA and AOL users about IPv4 and the Internet. As for the business case, here are five reasons to move: 1. PERFORMANCE - Performance browsing to an IPv6 enabled server is 10-20% faster, making anyone's web properties look better to their customers! 2. LOWER COST - The maintenance of IPv6 networks cost 17% less, then IPv4 only or dual stack. 3. BETTER SECURITY - IPv6 end-to-end communications along with DNSSec with DANE, perfect forward security, BPKI, and others. Reduces the chance of man-in-the-middle attacks, SPAM, and identifying source of DOS's. 4. MAINTAIN COMMUNICATIONS - If your mail servers, web server and browsing is on IPv4 only, there are website today you will not be able to access. 5. GROWING MARKET - No matter what your business, the network effect (Metcalfe's law n^2) allow you to connect to more people, and systems. To the innovators, please visit my blog at http://www.scientifichooligan.... to learn more about IPv6 features and security. TO THE IDIOTS & TROLLS - please, return to AOL where you belong.

  30. Re:10 years by wonkey_monkey · · Score: 2

    Cool! I wonder what I can get for 127.0.0.1... I never use it.

    --
    systemd is Roko's Basilisk.
  31. Excerpt from wikipedia entry July 2018 by ConfusedVorlon · · Score: 2

    After the technological meltdowns consistently failed to appear, IPv4 was finally replaced when IPv7 was adopted globally in the year 2017 as a result of a world trade agreement.

    The incongruous IPv7 clause was widely seen as the result of an unlikely alliance between the RIAA, MPAA and various repressive regimes such as China, Zimbabwe and the United Kingdom.

    Frustrated by the inability to trace internet usage to a single user via IPv4, these organisations lobbied for IPv7 to be adopted so that individual phones and computers could be mapped permanently to a single device and user. Unlike IPv6, IPv7 includes a direct mapping to the mac address of a device and the user's global internet ID, so that (in theory at least), all downloads can be linked to a specific person.

    Although the EFF and various other organisations campaigned vigorously against IPv7, the arguments around catching terrorists and preventing pedophilia prevailed.

    --
    VLC Remote for iPhone and Android
  32. Re:10 years by fuzzywig · · Score: 2

    The blocks are all doled out to different regions, but the entities in charge of those regions (ARIN in this case) haven't finished doling them out to customers.

  33. Re:IPv6 usage IS increasing by fuzzywig · · Score: 2
    From some googling apparently these ISPs do offer IPv6:

    AAISP
    Clara.net
    Entanet
    Exa networks
    Goscomb Tech
    IDNet
    Webtapestry

    Virgin announced several years ago that they would soon be offering it, they still aren't.

  34. For anybody paying attention... by Yaztromo · · Score: 5, Informative

    For anybody paying any attention over the past few years, this shouldn't come as a surprise.

    The IANA ran out of IPv4 address space available for doling out to the Regional Internet Registries (of which there are six) three years ago. APNIC (Asia Pacific) and RIPE NCC (Europe) went below a single /8 three and two years ago respectively. The IPv4 address exhaustion has already begun.

    ARIN (North America), however, has 82 /8s. If you consider that there are only 221 /8s in total (the IANA keeps 35 for reserved use), this means that ARIN has 37% of all usable Internet addresses assigned to it, for roughly 8% of the worlds population. More than a third of all possible addresses for less than a tenth of the worlds population.

    Even still, ARIN now only has about 1.3 /8s free. Projections have them running out next year. They've always been estimated to be one of the last RIRs to run out (with AfriNIC being last, as they still have just over 3 of their nearly 13 /8s free) due in part to the huge number of /8s already in use in North America (way out of proportion to the population of the continent).

    I feel really ashamed every time this topic comes up on /. at the complete and rampant ignorance of the issues surrounding IPv4 and IPv6. We will run out of IPv4 address space, but address space is hardly the only problem with IPv4. The bigger problem is ROUTABILITY -- the IPv4 routing tables have become seriously unweildly, they are getting progressively worse (in part due to InterRIR transfers of address blocks now that Europe and Asia have run out of addresses), and they continue to need more and more compute power thrown at the problem just to keep up. The number of BGP forwarding entries has doubled from roughly 250k to nearly 500k in just the last six years. The algorithms used for determining routes in IPv4 are complex. The computability is difficult, and it's slowing down the Internet today.

    IPv6 solves a lot of the routing problems inherent in IPv4, making routability a lot easier to compute. IPv6 packets have a simpler header, routers don't need to provide fragmentation services, and there is no header checksum. IPv6 also avoids the routing anomalies present in IPv4 due to things such as the switch to CIDR. We know a heck of a lot more about packet routing now than we did in the 60s when IPv4 was first defined, and these improvements are available in IPv6.

    This is why I cringe whenever I see a post in an IPv6 address exhaustion related /. story complaining about a lack of backwards compatibility in IPv6, or anytime anyone says that NAT is good enough for everybody. As the address space fragments even further, and historic /8s and /16s are broken up into ever smaller units which are then distributed to diverse geographies, the routing table in IPv4 is going to continue to blow up, becoming ever uglier -- it simply wasn't designed to scale in the manner in which we're using it. IPv6 brings sanity to global routing again, in a way that no backward-compatible solution could achieve.

    The IANA is out of addresses. RIPE and APNIC are virtually out of addresses (with only enough reserved to aid in IPv4 - IPv6 tunnelling and translation services). ARIN is down to less than 1.5 /8s, and survives purely on the fact that it has a disproportionate number of /8s compared to the population it serves. And worst of all, IPv4 routing is an absolute mess that requires a ton of processing power and compute time to maintain. Remember these things before you post something silly about being pro-NAT, pro-some-untested-IPv4-address-extension-proposal, complaining about backward compatibility, or how people have been predicting IPv4 exhaustion for the last 25 years (just because you see the train coming towards you way off in the distance does

  35. Re:Privacy Benefits to NAT? by rusty0101 · · Score: 3, Insightful

    To embellish smash's response, no there is no privacy benefit to using NAT. If you want some sort of a privacy benefit, you still need to add a firewall to your connection that can monitor your traffic for the very same things it would have to monitor for if you use global addressing. The only thing that NAT provides is an address translation interface, too allow you to have a larger pool of addresses to use than your provider can grant. If there is a port forward for a service set up either statically or dynamically (upnp) any flaws in the service that is being forwarded can be exploited in the same way it would be if there were no NAT involved.

    --
    You never know...
  36. Re:Chicken little by Anonymous Coward · · Score: 2, Interesting

    Another part of the answer...take back the class A allotments that were given to companies/organizations early on. If you're not in the business of using the addresses to help your customers connect (Level-3, AT&T and such), you should be using NAT like the rest of us. I'm looking at GE (3.0.0.0/8), IBM (9.0.0.0/8), Xerox (13.0.0.0/8), HP (15.0.0.0/8, 16.0.0.0/8), Apple (17.0.0.0/8), MIT (18.0.0.0/8), Ford (19.0.0.0/8), CSC (20.0.0.0/8), Halliburton (34.0.0.0/8), Merit (35.0.0.0/8), Eli Lilly (40.0.0.0/8), Amateur Radio (44.0.0.0/8), Prudential (48.0.0.0/8), duPont (52.0.0.0/8), Daimler (53.0.0.0/8), Merck (54.0.0.0/8) and USPS (56.0.0.0/8).

    Between them, these organizations have almost 7% of the IPv4 address space and all of them have similar counterparts that manage to get by without a block of ~16m addresses. Address space isn't property and should be allocated by the internet community based on the common good. These organizations should be given sufficient notice to ensure that they have enough time to prepare, but they shouldn't be allowed to hold these addresses indefinitely.

  37. Re:Chicken little by RR · · Score: 4, Interesting

    Another part of the answer...take back the class A allotments that were given to companies/organizations early on.

    Why does this myth persist? Modded Interesting, even. This proves that education is the major barrier to IPv6 adoption.

    We can't "take back" the class A allotments because there is no "back" to take it to. Those were given by Jon Postel before IANA existed, and IANA does not claim any more legal authority to those addresses than anybody else. It's an unwise investment of limited resources to challenge those companies' legal departments.

    Also, with the rate that IPv4 addresses were being allocated, and the acceleration of the rate before 2011, those addresses would have postponed IPv4 exhaustion by months at best. It's surely not worth the expense to force all those companies to release their class A networks just so we could collectively fail to do our jobs, that is, switch to IPv6.

    --
    Have a nice time.
  38. Re:Chicken little by slimjim8094 · · Score: 4, Interesting

    Google agrees. They're probably a bit less US-centric.

    As bad as the ISPs in the US are, we're actually a world leader in v6 traffic. Comcast, Time Warner (the ones I have personal experience with) and apparently Verizon are all doing v6 natively and properly. That accounts for a huge percentage of customers - as they get around to replacing their gateways, it should "just work".

    -- reply ends, general comments begin --

    Just so everybody's clear what I mean by "just work" - when I moved into my new apartment, I rented a modem/router from the cableco (I of course bought my own a few weeks later like a good nerd). Out of the box, it requested a /64 prefix and delegated it to the internal network, including the v6 DNS servers. All OSes made in the last 10 years know how to do v6 properly, so everything from my desktop to my phone to my smart TV can access v6 resources just fine.

    v6 is here. It works great, and you get real IPs! Like, you can actually paste an IP to a friend so he can download a file from your box just like the old days, without doing any NAT port mapping bullshit. Want to play a game, or video chat, or VNC or something? Just open a damn socket, no STUN or UPnP or any other crap.

    I don't get why so many Slashdotters are bitching/FUDding about v6. There's no money in it - all the ISPs are doing it happily - so it's not astroturfing. And the comments don't fit the typical troll model. What gives?

    --
    I have developed a truly marvelous proof of this comment, which this signature is too narrow to contain.
  39. Re:Chicken little by war4peace · · Score: 2

    Interestingly enough, both Germany and Romania have a higher adoption rate.
    I'm from Romania and gave IPv6 a try. I have a router that allows both IPv4 and IPv6 connections at the same time, so I enabled both and worked like that for a while. For some unexplained reason, the IPv6 connection took a huge amount of time to get its IP (literally minutes) and after both connections were enabled, many things wouldn't work right. I experienced repeated loss of connectivity in pretty much all online games, Yahoo Messenger would randomly disconnect, Skype would randomly disconnect, Steam would go offline for 30 minutes in a row, Dropbox would lose connectivity, etc.

    Maybe IPv6-based PPPoE has issues, I don't know, but I was literally forced to disable it for my computers to work properly.

    Anyway, I would definitely not consider 6-7% as being a "successful" deployment. It's a start, but still a LONG way to go.

    --
    ...gis sdrawkcab (usually not responding to ACs; don't bother posting as AC)