Slashdot Mirror
After IPv4, How Will the Internet Function?
An anonymous reader writes "36 countries in the world have over 100% per-capita usage of mobile phones, and this is driving a real crunch on IPv4 addresses as more and more of these devices are data-capable. The mobile network operators are acting fast to deploy IPv6, and T-Mobile USA has had an IPv6-only trial going on for over 9 months now using NAT64 to bridge to IPv4 Internet content. It is interesting to note that the original plan for IPv6 transition, dual-stack, has failed since IPv4 addresses are effectively already exhausted for many people who want them. Dual-stack also causes many other issues and has forced the IETF to generate workarounds for end users called happy eyeballs (implying that eyeballs are not happy with dual-stack), and a big stink around DNS white-listing. How will you ensure that your network, users, and services continue to work in the address-fractured world of the future where some users have only IPv4 (AT&T ), some users have only IPv6 (mobile and machine-to-machine as well as developing countries), and other Internet nodes have both?"
It's turtles all the way down.
"36 countries in the world have over 100% per-capita usage of mobile phones
neat trick, having over 100% per-capita usage.
It seems ludicrous to claim that the dual stack idea has failed when more and more devices are suddenly finding themselves with IPv6 addresses and are putting them to use. My home and work LANs are dual stack and everything Just Works. For being a failed experiment, it works amazingly well in everyday usage.
Dewey, what part of this looks like authorities should be involved?
IPv6 of course.
Client sites have nothing to worry about straight away, unless they want to access the new IPv6 server sites that will be coming online. The issue will be new sites needing IP addresses will be IPv6 only. If everyone started the move to IPv6 today, then the internet, from the average joe point of view, will look pretty much the same. The problem is that they will start seeing the breakages because we are almost out of IPv4 addresses before anyone has really started upgrading their infrastructure to IPv6.
Jumpstart the tartan drive.
No country has close to 100% of its residents connected via multiple mobile Internet connections at the same time, and many countries provide a NATed private IP anyway.
Dual stack is an absolutely fine solution for the current Internet and the "many other issues" usually means someone is about to sell an over-complicated and unnecessary transition solution. But wait, "Happy Eyeballs", ah... today's salesman comes from Cisco. And I find it very difficult to read a proposed standard for seamless transition where the author cannot spell "seamless".
That already exists, it's called "using both".
Have you heard about SoylentNews?
What happened to IPv5?
Why aren't the wireless carriers using the private IP ranges? Why does my smartphone need it's own public IP address?
"We're gonna need a bigger boat"
It sucked.
Dewey, what part of this looks like authorities should be involved?
http://www.oreillynet.com/onlamp/blog/2003/06/what_ever_happened_to_ipv5.html
It was assigned to an interesting, but ultimately not implemented, protocol.
Maybe I'm wrong, but I suspect that a large part of the IPv4 space is used by smartphones, ebook readers, home and small office equipment.
Either all that stuff needs be upgraded to IPv6 or operators will need to deploy IPv6-to-IPv4 gateways.
If you're lucky you can mod your routers with OpenWRT or its derivatives.
Maybe Computers will never be as intelligent as Humans.
For sure they won't ever become so stupid. [VR-1988]
Just fine.
Pretty good is actually pretty bad.
lots of IP4 only cable / dsl modems and routers are out there. Do any of E-mta (that the cable force you to rent (if you have cable phone) do IPV6?)
What about variations on that theme we're all hearing about the Premium Internet - can they hook that stuff up to nice new IP6 addresses, with not a titty to be found, leaving the "ghetto" kids in IP4?
My first Journal Entry ever, in 8 years! http://slashdot.org/journal/365947/aphelion-scifi-fantasy-horror-poetry-webzine
Except that's wrong. AIX 7.1 isn't an 'experimental' release. HP-UX 11.31 isn't an 'experimental' release. Even if you include Linux as unix (which it isn't), 2.6.35 wasn't experimental either. Just because the major release number of older Linux kernels were tagged as experimental for odd numbers (2.4.x was stable vs 2.5.x experimental), doesn't mean that it applies any more or that it applies to the 'unix world.'
Well, but that is true for IPv5, google it, most news sites report that, at least with the IP protocol, odd numbers indicate experimental protocols.
Have you heard about SoylentNews?
You clearly have not heard of our solution in the lab: use complex numbers for each octet. This expands the space of addresses to Great Big, although finite due to use of integer values of the real and imaginary part.
Yes, yes I know what you're saying: it takes more bits, right? Wrong. String theorists have applied extra dimensions to the octet encoding so as to only use 4 bits in this space, with the additional values residing comfortably in The Other Ones.
Sorry to have left you out of the loop, but we knew we could keep getting by with our current modification to IPv4 .. just add an i.
> That already exists, it's called "using both".
Seems like that would be IPv10.
Warning: this article may contain humor, sarcasm, parody, and perhaps even irony. Read at your own risk.
We aren't out of IPv4 addresses, we are out of IPv4 block allocations. This started back in 1992 when Cisco and Bay Networks decided that forcing new allocations into consolidated routes was easier than building routers that could cope with 2^24 (or even 2^32) unique routes. The original / notation wasn't about talking about /16 or 24 but /36 was a way to describe taking 4 extra bits from the source and destination port range. That system would allow most existing hardware (even from the late 80s) to work without any changes and allow things that know about the newer way to cope with more advanced addressing for things like vhosts.
Thanks to finally embracing NAT64, this becomes easy.
If you are providing 'server' access, you pretty much *have* to get an IPv4 address, and preferably an IPv6, but not absolutely required for now. Short term, don't sweat it, medium term go dual stack at first opportunity that presents itself, long term you may take down the IPv4 network one day, but don't explicitly plan when that day will come. The common strategy may continue to be ignore v6 entirely, however moving dual stack at your pace ensures that in the slim, but real possibility that your next-hop provider stops IPv4 routing or starts penalizing IPv4 use via unreasonable fees won't put you in a tight spot. The scenario of next-hop penalizing/dropping v4 is the only scenario I see as sufficient motivation to get servers to bother with v6 at all. I think even brand new servers will do what it takes to secure IPv4 space, which may free up some given the next point...
If you are setting up a network as 'clients', you can get by with either IPv6 or IPv4 for a while. Giving dual stack when available is nice, but whatever you have would be sufficient. ISPs without IPv4 addresses available for new clients should rapidly pursue IPv6 for residential customers and give them most internet via NAT64 on their end. Doing IPv4 private addresses would doom them to crappy service indefinitely, whilst IPv6 would only be semi-crappy for a more temporary interval. If you *really* want v6 to catch on, then start allowing v4 addresses to be carved up more free-market style. All technical experts agree that this would completely fubar the v4 network performance in aggregate, but you would entice adoption of v6+NAT64 with the profitable opportunity to reclaim addresses and sell them to places that *really* need them. The v6 network would be nice and cleanly routed, and getting on the v6 network just becomes that much more important.
Some would argue that any sort of NAT at the carrier plays right into the hands of those who hate P2P networks, including NAT64 as those behind NAT64 are unreachable by peers who are v4 only. However, the reality is there are two possible outcomes, residences getting 10/8, 172.16/12, or 192.168/16 which *completely* breaks P2P (and probably many wireless routers presuming those prefixes won't come from the WAN), or NAT64 where the P2P graph may not be as connected, but all v6 peers can reach each other. Since P2P designs are inherently tolerant of unreliable ability to reach peers, this should suffice for a while.
Major architects in v6 world advocated the dual-stack method as the way to theoretically move on with no thought to the practical motivations to move forward. They hated NAT in every way as it breaks the peering model they hold dear. They hated accepting the practical view that most of the internet are clients and few are servers. If they had embraced it from the beginning, then I suspect most residences would be v6 by now.
XML is like violence. If it doesn't solve the problem, use more.
Just because 'most news sites' report it, doesn't make it true. IPv5 didn't become mainstream because of what it was designed for, NOT because it had an odd number.
Maybe you should Google what IPv5 was for. Here, I'll help. Read this.
This can be used to increase storage space too. I found that the simplest way to enable the OS to do this is to clear the allocation bitmap. This allows it to use the imaginary space on the disk. It's worked so far, and I've stored about 50% more on my disk. It's been holding up pretty well and I haven't seen any cross-link *&()N#%()K&K_*)%_*()hj JFIF PNG TXT
Geeks should know better. The way it is talked about, you'd think in a couple days someone will plug in a device and there'll be no more IPs. Not hardly. We are approaching the first milestone in an eventual crunch. That is that there will be no more addresses not assigned to a registrar. The remaining class-As will be handed out to the regional registrars. While that means at the highest level we are "out" that doesn't mean we are out on a user level.
I'm not saying that we don't need to move to IPv6 but people on /. keep talking like we are going to be out of every single IP address real soon. No, rather we will be starting a process of scarcity. So far there's been no real scarcity of IP addresses. That will change. However all that means is that costs will change.
That will actually probably be a good thing for IPv6 adoption. If you are a company and want some static IPs and your ISP says "Sure, you can have IPv4 addresses at $30/month each, or as many IPv6 addresses as you want for free," well maybe you decide there's good reason to go with IPv6 and upgrade your stuff.
If IP4 space isn't being used it should be reclaimed by ARIN
many countries provide a NATed private IP anyway.
Err... you mean company, right?
Countries too.
The simplest of solution would be for your phone to have a unique DNS name and then using dynamic IP service.
Which would still cause connections to time out and the parties to have to reconnect at the new IP address. Or is there a way to hand off a TCP connection from one IP to another in the same way that a cellular voice connection is handed off from one tower to another?
I'm waiting for DD-WRT to have integrated (as in the web fronted) support for ip6tables and NAT64. I'm not going to expose everything under my roof to the jungle.
BSD licensed software can't be stolen....
The remnants of the IPv4-only internet will become increasingly smaller and increasingly easier to control - the large-scale NAT is one example that takes much power away from users. I doubt it will vanish, and it may instead become the main target of censorship in the future, to the point where IPv6 users are treated with the same inherent suspicion as users of Tor and BitTorrent.
The problem is the asshats that came up with IPV6. It should be scrapped here and now. IPV6 is just plain and simple flat out stupid.
Using a hexadecimal address was pure stupidity. All you needed to do was turn each segment of an IP address into a word sized ( 64 bit addressing ) or a long sized ( the magic 128 bit ) value instead of a byte sized value since:
2600000.35.1254.1785
Is one hell of a lot easier to remember then
2001:0db8:85a3:0000:0000:8a2e:0370:7334.
And using the colon for address separation is equally as stupid since that is how we designate port numbers. Ohh wait I know don't forget to surround the unrememberable POS with square brackets!
To make IPV6 useful it requires anything and everything to have a DNS entry since it is pretty much unrememberable and quite frankly I have devices that I never want in the DNS system yet I will be pretyy much forced to since trying to remember an IPV6 address will give me a fucking stroke.
And lets not forget you omit parts of the address eg: 2001:0db8:85a3::0000:8a2e:0370:7334 but ONLY once! I mean why did they even bother with this crap, is that supposed to make it easier?
IPV6 was written by a bunch of head up their ass academics, and even if the members of the committee were not academics their head was still firmly planted in their ass.
The guys who came up with IPV4 new they would have to work with it and made it pretty damn simple in most respects, but these clowns have turned something that should have just made the address space bigger into to something that will require massive kludges to transition since it will pretty much cause a mandatory replacement of pretty much 90% of the hardware out there.
Never ever let an academic design anything. They will fuck it up every time.
Hey KID! Yeah you, get the fuck off my lawn!
Make a 'Your Own IP' feature for the cell providers which gives you the option of your own unique IP. Everyone else can just pull from a rotating pool of ___ IPs.
I don't think most average iPhone users give a crap if they have IPv4/IPv6 support or what their IP is at the moment, as long as their phone works and they can play Angry Birds.
If the only way you can accept an assertion is by faith, then you are conceding that it can't be taken on its own merits
Any guess as to how large private IP space is? Hint - it isn't big enough for any of the major operators to supply a unique IP within their networks.
How big is a service region? Each region could get its own /8 of sixteen million IPv4 addresses in 10.* for connections back to the IPv4 net.
These large operators have had to choose between partitioning their subscribers which makes phone-to-phone applications a mess
Or they could just require a land-based proxy server between phones for phone-to-phone applications where neither side is on an "enterprise" service level agreement. According to acceptable use policies that I've read, "running a server" isn't something that one is supposed to do on a telephone.
game servers
Have you read the typical Acceptable Use Policy of home Internet access lately? Game servers are supposed to be coloed in datacenters the way CCP, Blizzard, Zynga, etc. do it, not using one of the clients as a server the way most Xbox Live games do it.
peer 2 peer applications
Carriers have been seeking affiliations with MPAA studios in order to use "watch movies" as a bullet point to attract paying subscribers, and most noninfringing files too big for HTTP are also too big for the 5 GB/mo cap on typical 3G plans. So why would carriers make effort to allow peer-to-peer file sharing applications?
Just a note Straterra.... The major release number for linux kernels is the second number in this case. 2.6.35 is part of the 2.6.x stable kernel release. 2.7.x would be experimental, just as 2.5.x was....
I'll be happy.
I know.
Movistar in Argentina uses 10.x.x.x network addressing on Mobile phones last time I checked.
dont worry. Internet is a series of interconnected peaches. size of the peach is its bandwidth, and the icky hairs on its skin is its traffic. so, it will keep functioning even if you put it in the fridge.
Read radical news here
Guys, look at This list of Class A.
Prudential insurance? A class A? Almost 17 million addresses?
Ford motor company? General electric?
DoD has 11 class A chunks? That's almost 200 million addresses. You could give almost everybody in the united states a mobile phone with that.
These are just the most obvious ones. Does Apple really need 17 million addresses? Does HP? Xerox PARC?
This FUD has been getting spread around since the late 1990s. I think we're fine, and I think we're going to be fine for quite a while into the future.
NewslilySocial News. No lolcats allowed.
Remember back when you were a kid, and no one wanted to be stuck at the "children's table" during holiday dinner? Well businesses are similarly scared of getting stuck in the IPv6 address space on the internet. Given that only something on the order of one half of one percent of end users are actually USING IPv6 (and that includes developing countries), no business in their right mind would stand up an IPv6-only website. And as long as there is nothing interesting on the web that requires users to use IPv6, no one but a handful of uber-geeks is going to bother switching.
The "sky-is-falling" scare tactics of the IPv6 advocates are rapidly being exposed as snake oil and hokum. We are a long way from being "out of" IPv4 addresses. NAT has expanded that IPv4 address space from 2**32 or ~4 billion addresses to more like (2**32)*(2**24) or ~64 quadzillion addresses for end users. Even if IPv4 addresses were to become more scarce, that would simply raise the price of buying them.
No matter what whopping lies the IPv6 promoters tell in an effort to scare people into adoption, no one is going to "CONVERT" to IPv6 any more than the US is going to make Esperanto the national language.
Eventually there may be a small bubble of IPv6 users that rivals the market penetration of Linux onto corporate desktops, but nothing that is going to happen with IPv4 is bad enough to force people to go through the pain of conversion.
The IETF needs to admit that IPv6 was a brain-dead mistake, and go back to the drawing board for IPv7. THIS TIME, start with backwards-compatibility with IPv4, then fix the problem with iso-chronous delivery so that voice and video actually WORK, and then maybe people will start adopting it.
Until then, IPv6 is just a pipe-dream, and a sales gimmick for network gear vendors trying to con suckers into digging up and replacing their safe and stable IPv4 networks.
In the real world ( read as 'unix world' ) odd numbers are always "experimental" .. There is a 5 .. but it was never meant for mass consumption
Which is why OSPFv3 is used with IPv6.
You clearly have not heard of our solution in the lab: use complex numbers for each octet.
Oh you also oppose the tyrrany of the powers-of-two addressing space? Excellent. Personally I've been working on an Egyptian fractions representation. Imagine the entire IP addressing space between the intervals of zero and one. And we'll never need a larger space, merely subdivide more aggressively. Regular expressions and routing tables are a bit tedious of course. But, string handling technology has been neglected for years by the tyrrany of floating point accelerators, its time for a new paradigm.
http://en.wikipedia.org/wiki/Egyptian_fraction
"OK, the static address of the printer is 1/2 + 1/6 + 1/87 + 1/67289, ok no problemo."
The reverse DNS would be pretty simple, that would be laserprinter.innitech.com AAAAAA 67289.87.6.2.in-addr.arpa. (don't forget the dot at the end)
The concept of router interface addresses needs some work, I'd go back to the DECNET thing and assign an address to the machine instead of the interface.
To subnet, merely add another term, so if my ISP is 1/2 + 1/98 + 1/102 (shorthand 2.98.102) they might assign me, 2.98.102.253. Then I assign my hosts with the final digit larger than 253. 2.98.102.253.512 might be my desktop.
"Science flies us to the moon. Religion flies us into buildings." - Victor Stenger
I've only been able to play with my Verizon Droid but it never gets an internet facing IP address. It always gets a 10 net address(yes while on 3G). My guess is that they do this so when you hop from one tower to another your traffic can be forwarded, but I've never verified that. I know IPv4 address are running out but are mobile phones really the reason? If every company is like Verizon(and I would assume at least a few other are) that means only one Internet IPv4 address per tower. Meaning one address could be serving thousands.
I would assume the answer is no, but I am nervous about it. Can I rely on everyone having dual stack or only ipv4 INTERNALLY for atleast the next 5 - 10 years? I write software for a living and our software more than likely will break if attempted to use with IPv6. My main problem is that the technology we use to develop the software is old (10 years or so?) and not maintained any longer (Delphi 5 + BDE). I am sure the Firebird database we use either works now with IPv6 in its current version or it will in the near future as it will be necessary, but the BDE technology we use to connect to that database is no longer being maintained as far as I know. If I can rely the internal networks all having IPv4 internal addressses then its not much of a problem. But if I have to support IPv6 internally, I think it would require a complete rewrite with a modern programming language.
I hate to admit that Facebook has a use, but it really could do here.
Phase 1: Facebook puts up notices warning users that after date X (Say, 3-6 months in the future) if they cannot access the site it's the fault of their ISP and they should complain. Make vague statements about network upgrades to improve your user experience or somesuch nonsense.
Phase 2: On date X, take Facebook IPv6 only
Phase 3: 1 month after date X, everyone and their mother has an IPv6 address allocated by their ISP
Not anymore. Linus dropped the branches between "Experimental" and release. http://en.wikipedia.org/wiki/Linux_kernel#Version_numbering
NATurally I would think.
As with all things ... check out what the Pr0n industry latches on to.
Not that I would ever suggest that the internet is a foul bastion of depravity, but it just looks that way from the outside.
When we see Pr0n lead the IPv6 uptake, we know mainstream acceptance is minutes away.
Q:I was listening to a CD in Grip and it sounded horrible! What's up? A:Perhaps you are listening to country music
they seem to come around every 4/5 weeks now its getting old
I would mod you up if I had points. NAT64 is the way to go as a long-term strategy for clients, and dual-stack as a short-term solution.
Even most direct-connect software will work fine connecting an IPv6 host to an IPv4 host. Most software will try negotiating both ways before giving up. The only people who will be broken are clients on IPv4 hosts, using NAT, without UPnP. Otherwise IPv6 hosts should be able to connect to IPv4 hosts fine for direct file transfers.
If you want to run a server on IPv6, you would be SOL for those who don't have IPv6 capabilities, but then again, if you're running a server, you probably should be running on business plan (most ISPs have clauses saying "you can't run servers" on residential plans).
P2P for the most part should be fine with NAT64.
Yes, I stand corrected, sorry, I read a decent article that pointed it out. In fact, I read the article you pointed me to, even though I only just clicked it.
Always good to know you can still make mistakes.
Have you heard about SoylentNews?
(Look at the ID ;-) )
I think this guy summarised it well -
http://www.ietf.org/mail-archive/web/v6ops/current/msg06483.html
The Internet's nature is peer to peer - 20050301_cs_profs.pdf
Oh god, every device has an infinite number of addresses.
WHAT did you say about reverse DNS ?
I've had a dual stack for a while thanks to Teredo, and quite frankly it's pretty useless. Don't get me wrong, I like the Teredo concept, I just don't believe in the migration plan. I foresee a world of NATs, just like eastern civilizations have had it for a while now, where IPv6 is regarded as a non-production protocol used mostly for P2P. Nothing is really going to change, we'll just lose our public addresses at home and get used to it; hell I'm already finding it odd that with this crunch, Vodafone is still giving me a public IPv4 address when I connect to the Internet from my cell phone.
The migration was terribly planned, they had over a decade to come up with a better solution (which exists), but instead of doing it they insisted in going on with the same plan ignoring the fact that nobody's gonna change their networks unless they're broken, and these people are the so called experts. Having two networks that can barely talk to each other and hoping that people simply migrate when one of them is considered highly optional is ridiculous. Does the IETF have any real engineers or is it just the name?
Parent post illustrates how IPV6 is the Vista of networking technology.
IPV4 works for me today, so I'll just wait for IPV7.
IPv24?
IPv1296?
IPv4096?
The whales have already shown the time is now to move to ipv6: ipv6.google.com, ipv6.netflix.com, ipv6.weather.yahoo.com, www.v6.facebook.com, ipv6.t-mobile.com, ipv6.comcast.net, ipv6.cnn.com, www.brocade.com, www.ipv6.cisco.com, and the list goes on. These companies are not going dual stack for fun on their servers, they are doing it because ipv6-only users are on the horizon and they all know it. Without a native IPv6 setup, they will be screwed going via a proxy on NAT64, and they dont want that.
I have been using IPv6 on the T-Mobile USA beta and it works fine, the user experience is just like ipv4. Politics and dogmas aside, it's just a number for an end node and it works. If you google around, you can find links to try it out yourself on T-Mobile USA.
... will be which one?
now we need to go OSS in diesel cars
In the real world ( read as 'unix world' ) ...
As opposed to the Windows world where every release is considered experimental? ;-)
It was an experimental version. IP is like Linux (x.y.z - the y) - the odd numbers are experimental and the even numbers are production. (IIRC)
Dual stack has not failed - it's the only sensible way to migrate to IPv6. - Today, the Internet consist of two logical networks, IPv4 and IPv6 - Currently few customers are conected to both networks, but that will change in 2011 av beyond - In the future (20-100 years?), only IPv6 will be needed - By running dual stack on the PC/Server, the DNS will resolve which logical network you will use to access services - By running dual stack, you will (over time) migrate traffic from IPv4 to IPv6, and you don't have to bother that this process is going on. - By running av 6to4 tunnel, you never get rid of IPv4 (which eventually is not needed any more), so the tunneled access to IPv6 is a bad idea The dual stack adoption rate has been slow because of - Lack of SOHO routers with dual stack support (still a problem for $100 routers), but this is expected to change in 2011 - Lack of ISP routers capable of doing IPv6 in hardware (this problem is mostly solved now, but ISPs must invest in new routers, particularly at the edge of the network) - Lack of demand from customers - they have so far got IPv4 addresses ...
- ISPs have only recently enabled dual stack in their core network, a configuration thing really
Lack of IPv4 addresses will lead to IPv6 migration
- Lack of IPv4 adresses will lead to NAT-ing of IPv4 for new customers which will have impact on services
- Only IPv6 will provide you with ENOUGH public IP adresses (NOT-NATED)
- Most ISPs are now ready with dual stack IPv4/IPv6 (I work for an ISP, so I know this)
- Dual stack is not complicated, migration is toughest for those customers that can't software upgrade their firewall,
beacause they must convert the firewall policies which can be complicated
- Those who will provide services will serve on both IPv4 and IPv6
What's missing
- Google, Youtube and others should annonce NEW services on IPv6 2-4 weeks before they do on IPv4, that would lead to IPv6 demand
- At the moment, IPv4 and IPv6 are to separate networks that lead to the same supermarkets - with the same products. This is bad
- IPv6 supermarkets should be more sexy
At the end
- Lots of public IP addresses are needed per user in the future, only IPv6 can provide us with all these addresses
- Dual stack is the ideal way to migrate to IPv6, the DNS will migrate trafiic from IPv4 to IPv6 over time.
Well, expect internal networks to migrate to IPv6-only after they have an IPv6 address to migrate to, the sysadmins did all the testing, and something (like router or DNS) at the IPv4 net breaks. Or maybe, forget about that testing step.
Now, you had 10 years to migrate that IPv4 only application, like everybody else. Maybe it already works with IPv6 (Delphi 5, maybe, Firebird almost certanly, and BDE, I have no idea) maybe you should test it sometime. If not, I'm sure somebody will come with yet another baind-aid over it (like making it run on a virtual machine, with IPv4 connectivity that is translated by the host to IPv6) that will take more work to maintain than just replacing it for once. In short, don't despair, it'll be business as usual.
Rethinking email