Little Interest In Next-Gen Internet 351
Ant wrote in to mention a Computerworld article that is reporting on the slow acceptance of the IPv6 version of the internet. From the article: "Information Technology (IT) decision-makers, in U.S. businesses and government agencies, want better Internet security and easier network management. However, few see the next-generation Internet Protocol called IPv6 as helping them achieve their goals, according to a survey released Tuesday by Juniper Networks Inc."
just wait... (Score:3, Insightful)
Re:just wait... (Score:2, Interesting)
Re:just wait... (Score:3, Insightful)
They already have (Score:2)
They already have.
--
Don't Fight Firefox! [bobpaul.org] Let Firefox fight you!
There's no place like 0:0:0:0:0:0:0:1 (Score:5, Funny)
Re:There's no place like ::1 (Score:5, Funny)
Did you hear about the guy who went to get a
A sound point (Score:3, Insightful)
Oh Dear (Score:5, Insightful)
Translation: "There's a marketing job to be done"
I thought education was for important things which you need, and marketing was to convince you to use products and services?
Need more software and support (Score:5, Interesting)
Wait a while until there's the software backing then you'll see companies using it.
Re:Need more software and support (Score:4, Informative)
Ahem [openbsd.org].
Re:Need more software and support (Score:2, Informative)
Easy and cheap solution (Score:2)
Duh (Score:5, Insightful)
Re:Duh (Score:2)
When numbers run out???!!!??
There are over 4 billion ipv4 addresses. How many of those ip addresses are actually used? How many of those ip addresses could be easily NATted?
I couldn't imagine even 20% (800 million) are being used at any one time.
how many internet users + how many internet servers + gateways = ???
Re:Duh (Score:3, Insightful)
Re:Duh (Score:2)
Those are all easily done behind a NAT gateway. Putting devices directly onto the internet has it's purposes, but I cannot think of a good reason why a mobile phone would need a public ip address.
Can you provide a good example of the need to put a mobile phone directly onto the internet?
(and no, to get to Paris's phone easier is not a good reason)
Re:Duh (Score:2)
then again, thats all end times prediction scary kinda stuff, so maybe we should stay with ipv4.
Cheers.
Re:Duh (Score:3, Funny)
Last I heard it wasn't so much that every person would have an IP address so much as everything you wear, carry or own would have one. Supposedly the idea is that your socks will be able to email your washing machine to get the microwave to remind you to wash them whilst the fridge will IM your PVR with a list of what's in it so it can identify and record cookery shows that use those ingredients, prioritised by how close to the use by date each ingredient is. Meanwhile the coffee maker will be contacting
Re:Duh (Score:5, Informative)
IPv6 = everybody gets their own ip address, walk up to a computer, swipe your worldID card, it switches to your personal ip.
Except it doesn't work like that. IPv6 uses a hierarchical routing model, much stricter even than IPv4 classful routing.
The IPv6 address you get assigned (each customer is supposed to get their own
Now, unlike the current internet, ONLY TLA's exchange routing information with each other. Every single address withing a TLA's block MUST be routable from one of its peering points. Routing between the TLA blocks may only happen at those top-level points. Small netblocks are no longer portable, so when you change ISPs, you get new addresses. No exceptions -- doesn't matter how many you have. That also means if you want to have a redundant connection for your server (multiple ISPs), it has to have multiple IP addresses too. No more BGP tricks.
So you can't assign an IPv6 address to a person, as every terminal they use has to have a different address by definition of IPv6.
The other common misconception is that IPv6 has more addresses (2^128) than particules in the known universe. This isn't really true as the lower 64 bits are not routable. They're usually automatically derived from the 48-bit MAC address, but can be statically assigned if so desired. Even if you did statically assign them, all (2^64)-2 of them would have to be on the same (flat) subnet, which would be one huge honkin LAN.
So that really only leaves 2^64 routable networks, each of which MAY have a lot of machines but in practice probably won't have more than 100-200 max, and probably averaging much lower than that. If you take into account that the specification calls for each customer to be given their own 48-bit subnet (giving them 16 bits worth to route internally if they so desire), there isn't just a whole lot more room then IPv4 because so much is forced to sit unused. It is considerably more to work with yes, but not astronomically like many people seem to think.
Ok, sorry for the rant, but just trying to make sure reasonably accurate information gets posted somewhere
Re:Duh (Score:3, Informative)
The certs can and frequently do live on each server. The cert needs to match the URL domain name you are hitting, otherwise browsers pop-up a warning.
(There are other factors which trigger pop-ups as well.)
In otherwords, one IP can serve an entire farm of hosts. In fact, one IP can serve more than one domain, by using different webservers running on dif
Re:Duh (Score:2)
Well, not in the US at least. NAT boxes and the fact that US sites have already claimed a good deal of the IP space means fewer addresses for the rest of the world. Asian has seen the most widespread adoption. The fear was that countries like China and India are getting online slowly but may require 1 billion addresses between them in the next few decades. Since IPv4 can only handle 4 billion, this would have been a problem.
artificially scarce resources (Score:3, Insightful)
One problem is that the united states has a lot more IPs per population than most of the rest of the world (does anyone have exact numbers for this?), so we'll be one of the last to run out, and therefore one of the last to adopt ipv6, which puts us in a very bad position.
A similar problem on a smaller scale is that those who own a lot of IPv4 addresses now have a competetive advantage over those who don't, and these are exactly the
NAT works... (Score:5, Insightful)
Once NAT+Firewalls became popular enough, the requirement for large IP chunks for offices and stuff disappeared.
No backward compatibility, ugly naming scheme (tell me , who like
Re:NAT works... (Score:5, Insightful)
Too many things have to work around NAT problems.
I run a small network and all the users running filesharing programs have problems. I have to give them each a port.
What happens when more than one of them wants to run server for a protocol which needs a specific port? SMTP?
Why shouldn't people be able to have full IP connectivity? NAT does not provide that, and UPNP is not enough to fix that.
90/10 (Score:3, Insightful)
That's what makes IPv6 acceptance so slow: your ISP isn't going to rebuild its infrastructure so that you can run a SMTP server. Certainly not for the measly (from their POV) $50 a month you and your friends are paying for that line. If you want a static IP, or a few, you can have it, but you'll start paying $150 a month or more for the service.
Some day, those necessary static IPs will be
Re:NAT works... (Score:2)
Re:NAT works... (Score:5, Insightful)
NAT itself is okay, but using private IP ranges behind it doesn't really work for large organizations, especially large organizations that can (and do) need to merge with other such large organizations.
I've been on the receiving end of a couple of these situations; it can cause a LOT of pain.
Actually... (Score:2)
Never invest in today's technology, always wait until it's tomorrow.
Re:Actually... (Score:3, Funny)
Re:NAT works... (Score:3, Informative)
No backward compatibility
ugly naming scheme (tell me , who like
And 127.0.0.1 is better? Easier to type? Shorter? This is a moot point, this is why ISC gave us bind, from which flows an endless bounty of hostnames.
the requirement for large IP chunks for offices and stuff disappeared.
What rock have you been hiding under? Game playing through NAT is nearly impossible. Scratch that... if there is more than one person try
Information Technology (Score:5, Funny)
ivp6 is so not cool man (Score:4, Funny)
all the cyber people need to support teh AOL in their awesome efforts to make the internets better for everybody.
can't believe you peeps havent seen the cool AOL comercials!!!!111
"want a better internet?"
"you belong to america online!!!"
More than one internet (Score:2)
By which, you are referring to both the Internet and Internet 2, I assume...
Doesn't achieve their goals? (Score:4, Insightful)
How about providing static IP addresses to DSL and cable modem users, so we can actually use simple DNS (or even just memorized IP addresses) to host things with servers in our living rooms? Seems to me that would be a huge value proposition for any ISP to its customers.
Your Goals != ISP's Goals (Score:4, Insightful)
Not to mention that by making dynamic IPs the industry standard, they can treat "static IP" as an extra feature and charge through the nose for it. (Much like text-messaging & ring-tones on cell phones.)
All of which is to say, ISPs see no profit from giving all their users static IPs. IPv4 is a blessing because it makes static IPs precious. Moving to IPv6 would just cut apart that revenue stream (at least in the short-run, which is all most companies seem to be concerned with).
-tom
India and China (Score:3, Insightful)
India/China IPv4 myth (Score:4, Informative)
But don't take it from me. Take it from the guy who runs the organisation that gives out addresses to India and China [com.com].
But... (Score:2, Insightful)
I know i'm not the only one who thinks like this.. all of my colleagues are happy with the v4 system, and the (less high maintenance) users know what i'm talking about when I assign IPs or mention '127.0.0.1'. None of them have a clue about '::1', and it isn't
Vested Interest (Score:2, Insightful)
Re:Vested Interest (Score:2, Interesting)
I'm pretty sure that Juniper sell IPv6 compliant kit and would love to sell more of it.
I know for a fact that they sell VPN kit - NAT & IPSEC go together like oil and water. Yes, I know it does work but it's still a pain in the ass and I say this as someone that has to do it for a living.
Remember kids, implementing IPSEC NAT-Traversal makes the baby Jesus cry.
Re:Vested Interest (Score:5, Interesting)
And Juniper Networks is pushing the idea that IPv6 is not on anybody's agenda because sell routers, NAT boxes and associated services.
I hope you are joking. Juniper would love to sell upgrades of their router's to all of their current customers to facilitate the jump to IPv6, but as they said, customers are just not very interested. I work for a company that sells network security devices and I can tell you IPv6 has been on the agenda for a long time, but most of the IPv6 support just keeps getting pushed back further and further, because no one really wants it from us. The only reason to include it is because some of the asian market is starting to ask for it. The U.S. as a rule is uninterested.
Re:Vested Interest (Score:2, Interesting)
Might be a good idea (Score:3, Funny)
Hmm... moving AOL to Tokyo or Beijing might not be a bad idea. Would be much more expensive to send out all those CDs to people here...
I have a minimal (Score:5, Interesting)
IPv6 Bittorrent (Score:2)
Bram Cohen has talked on occasion about IPv6 having some advantages for Bittorrent although I can't remember what he said.
Let me tell you (Score:2)
Re:Let me tell you (Score:2)
Corporations started building extensive VPN network a few years ago and are finally realizing what a nightmare this is.
Re:Let me tell you (Score:2)
How about fixing SMTP (Score:2)
The other major source of the perceived "insecurity" of the net is due to the insecurity of end-user devices (and end-users themselves), but that a harder issue to
Re:How about fixing SMTP (Score:2)
As for your problem with receiving mail claiming to come from your own domain, my answer to that is reject it. All of the email servers that I'm responsible for respond with a 5xx error whenever a computer connects and uses a hostname or ip addre
IPv6 experiences since 2000 (Score:5, Funny)
I first implemented IPv6 on a Cisco 7120 with a single FreeBSD 4.0 box as a host behind it - this would have been some time in late 2000. The IPv6 link came from Viagenie and this lasted a few months before I got bored with it.
I tried again last year with a couple of cable modem attached Cisco 17xx and some tunnels from Hurricane Electric. I was at a point where I wanted to do a lot more with IPv6 to get ready for my CCIE exam. HE was relentlessly useless in getting me more than what their tunnel broker system provided so I gave up again.
I tried later last year with BTexact's tunnel broker service and some other routers. Made it run, then started moving offices and lost interest.
I'm at it again - BTexact because they've got the best tunnel broker web interface and they'll give multiple tunnels, Cisco 28xx here, Cisco 17xx at a playful customer's site, and one FreeBSD 5.4 host. My CCIE gets closer and closer so this time its gotta go - web server, DNS, going to put up six total tunnels, then press for a block larger than the default
Looking at IPv6 from the outside it would appear that someone collected a bunch of people who got kicked out of IETF for mental instability, a number of disgruntled Novell employees who believed that IPX was a gift from an advanced space alien culture, and locked them all in a junior high gymnasium with a goodly supply of blotter acid and two boxes of twinkies. Its the only explanation we have for the results we see today
use this tunnel broker (Score:3, Informative)
These guys have a good tunnel broker interface:
https://tb.ipv6.btexact.com/ [btexact.com]
I used these guys a couple of years ago and they made me very sad:
ipv6tb.he.net/
why do you need a tunnel broker anyway? (Score:2)
Few articles actually address IPv6 benefits (Score:5, Insightful)
Few articles actually address real IPv6 benefits and instead pull out strawmen about a purported shortage of IP addresses. That's got to be the least significant and least relevant change between IPv4 and IPv6. Maybe that's all the 'journalists' can get their teeny minds around, or maybe it's mandated spin because certain key advertising accounts *cough*MS*cough* aren't looking to be IPv6 compliant any time soon.
Some of the main advantages of IPv6 over IPv4 are:
Lies. All lies. (Score:2)
Of these, only "simplified headers" really applies to IPv6 over IPv4. (Although I confess to not knowing what "improved routing" refers to.) Yes, there is QoS for IPv4, and multicasting for IPv4, and IPSEC for IPv4, and Zeroconf, etc.
The real advantage of IPv
IPv6 Myths (Score:3, Insightful)
Dude, I've been working with IPv6 for 7 years or so.
It's got built in equivalent of IPSEC. That alone would go a long way in improving most computing environments.
"built in equivalent of IPSEC"?? Huh? Rather, you mean the IPv6 standard requires that IPv6 implementations must have IPSEC, I am guessing. IPv6 with IPSEC is no more secure than IPv4 with IPSEC.
"Improved routing" refers to, among other things, route aggregation which reduces the size of routing tables which is he
There is also the "network effect" to consider (Score:3, Interesting)
First step.... Make the ISPs switch (Score:2)
Once most ISPs are IPV6 native, there'll be a lot more reason for people to play with it -- if only because it'll then be a lot easier. (Hey, I'm lazy. I expect that others are too). I had tunnelling working for a while but it broke and I haven't gotten around to getting it working again.
What was wrong with OSI? (Score:2)
IPV6 --prepare for your toaster spamming people (Score:2, Interesting)
Take that away, have loads of IPV6 addresses and un-informed consumers, and your setting yourself up for your uC driven toaster, oven, refrigerator, entertainment center etc spamming people.
It just gives me the screaming heebie-jeebies -- does anyone else remember the feeling of walking into a PeeCee site that was 'internet connected'back in the 90's and asking what they were doing and finding out every un-patched P
Please note: (Score:2, Informative)
There's currently an IPv6 conference [coalitionsummit.com] at which they're appearing as well. The conference ends today (2005-May-26).
There's a Washington Post [washingtonpost.com] article [washingtonpost.com] on the summit.
I'm posting from the summit, where they have a IPv6 802.11 network up for visitors use.
First mover disadvantage? (Score:2)
Of course they will have missed the innovation boat (and profits) by then and will be users rather than providers of new technologies.
Could have had a V8 - IPV8 that is! (Score:2, Informative)
Try IPV8! its a hell of a lot better!
its backward compatible with IPv4 - not necessary to change all the internet hardware or BS
and it has a LOT more addresses than Ipv6 ever will.
Dont like it? then try IPV16 !!!
sheesh you guys are behind the times... really!
Simple Question: (Score:2)
If they would stop killing p2p apps! (Score:2)
Hopefully, VoIP and VCoIP will catch fire and providers will realize that its much easier to provide these services without every user using NAT.
IPv6 problems (Score:2)
IPv4 subset of IPv6 (Score:3, Interesting)
Right now we've got a catch-22 it seems. Why would I offer an IPv6 ONLY service, if that means a ton of my users will be locked out? As long as I offer an IPv4 service, why would my users switch? They can just use IPv4 up the stack.
If IPv4 address were subsets of IPv6, couldn't an IPv4 users request an IPv4 address. Once it hits their ISP, check routing and prefix if possible with IPv6 prefix. This could happen anywhere along the line, including just the last hop. My server can just run an IPv6 stack, and know that the rest of the internet, IPv4 and 6, can reach it.
Instead, we've got a "fresh start" approach, which seems like a bit of a stretch.
Or am I missing something obvious here? It sure looks to me at this point that running an IPv6 only server is a bit complicated unless you set up a broker or something else manually.
It won't happen until Microsoft MAKES 'em. (Score:3, Interesting)
Guess what? Apple's already done it, (with Airport Extreme and Express, with eight octet groups right on the hardware,) but they're not making a big deal out of it because Apple's customers are not tech savvy enough to know what the fuss is about anyway.
All Apple need to do is start making a noise and Microsoft will once again play 'catch up.'
I'm running IPv6 on my friggin LAN and the WAN is only running IPv4. Go figure?
Lost in the debate (Score:3, Informative)
Everyone keeps talkign about NAT and its problems and support for apps and services. The real reason that IPv6 isn't being adopted is because core backbone providers aren't forcing it. No one has made a real commitment to IPv6, so it is not used at the enterprise level.
If you start with service providers, I don't believe that there is a lot of IPv6 even at that level. This is only really my conjecture, but as a consultant in the network management space, I don't hear customers begging for products that support IPv6. And until the backbone providers , and the IETF, decide that IPv6 must go forward, NAT is going to work for most people, and not much will change.
IPv6 is going to be a tough row to hoe, it will necessitate a lot of updates to libraries and software before it can be fully supported. A lot of companies spend a lot of money every year to monitor and manage their business systems with IPv4 based applications, and aren't going to risks the expense until IPv6 is necessary and vendors fully support it.
Re:Give me an easy upgrade path (Score:2, Interesting)
Re:Give me an easy upgrade path (Score:5, Informative)
I run ipv6 and ipv4 together on the same net all the time.
For ipv4 I have one static internet address on my router machine which provides NAT for all the other internal machines.
For ipv6, I have a
Also the applications don't need writing for one or the other. By using the ipv6 API you automatically get support for ipv4 at the same time.
There is an easy upgrade path. What will really get folks upgrading is when more and more sites become ipv6 only. For example I am setting up a nature webcam site which will be ipv6 only for exactly this reason.
Re:Give me an easy upgrade path (Score:5, Insightful)
Until the porn world goes IPv6 only there will be no major rush to upgrade. I really hope that your "nature webcam site" is really nude women running around in the forest or I just don't see it making a difference.
Re:Give me an easy upgrade path (Score:2)
Heh. not quite.
Actually it's small mammals running around somewhere in England. (or will be when I get things sorted out properly)
Re:Give me an easy upgrade path (Score:3, Funny)
Young, petite, teens?
Re:Give me an easy upgrade path (Score:2)
-Charles
Re:Give me an easy upgrade path (Score:2, Informative)
Re:Give me an easy upgrade path (Score:3, Informative)
IPv4 and IPv6 can easily coexist, and IPv6-only programs don't even have to be modified to accept IPv4 connections (keyword "IPv4-mapped addresses").
A miracle?
The best way to speed up adotion (Score:5, Funny)
Second, move all the porn sites to IP v6.
Actually, screw step 1. If you can manage step 2, and keep any new upstarts from taking over the vacated IP v4 porn market, then IP v6 will become the standard within a year. We'll all have a laugh over the contorted and convoluted arguments PHBs with little to no technology understanding will come up with to justify switching their corporate networks to IP v6.
Re:The best way to speed up adotion (Score:2)
IPv6 is not really suitable for porn because it's expected that end users receive static address assignments, which makes porn download less anonymous.
Re:The best way to speed up adotion (Score:2)
I mean NEVER.
The first industry to make money online was Porn, and I would guess it's still among the largest. You can't change the underlying architecture by ignoring one of the major uses of the existing technology. It'd be like making a new TV standard that didn't have sound.
Re:Give me an easy upgrade path (Score:3, Interesting)
I can access any ipv4 address or ipv6 address without any problems.
but I guess that doesn't count as an "easy way...to try it out"
Re:Give me an easy upgrade path (Score:2, Informative)
when i bought my powerbook 15 months ago, reverse tunnels stopped working, and it took me awhile to figure out why.
normally you'd run something like:
ssh -R 8080:localhost:8080 user@remotedevbox.com
to let a remote server access tomcat running on your laptop.
i suspect OSX routes "localhost" to the ipv6 address by default in this case. the solution was to change the tunnel:
ssh -R 8080:1
Re:Give me an easy upgrade path (Score:2)
The simple reason is that when people were working out the ideas for IPV6, the internet was basically a flat network where everybody had their corner and everybody had a class B, etc. Their fear was we'd run out of addresses. Well, we did and people simple solved the problem with cleverness instead of just expanding the address space.
The fact is now, entire corporations can run on a coupl
How about this? (Score:2)
Re:How about this? (Score:2)
Re:Wintel? (Score:2)
Re:Wintel? (Score:2)
Re:What does ipv6 get you? (Score:2)
Not many, but forget about dialup and you still have a startage of addresses.
I know a few people who struggle now to get 2 ip addreses now for business purposes and the problem will only get worse.
ipv6 is the solution and so far the best arguments against it seem to revolve around not being able to read the addresses as easily. Well I say wake up and use DNS. That's what it's there for.
NAT is a hack and it broken for many protocols anyway. That's why all sorts of conne
Re:What does ipv6 get you? (Score:2)
Re:What does ipv6 get you? (Score:2)
Right, because that's the only thing that was available at the time. Playing QuakeWorld over dialup on a p-100 as opposed to playing a comparable game today, with today's bandwidth, with today's computers. Even playing over dialup back when people played QuakeWorld, there'd always be someone that would get on with a resnet and own everyone.
Worldwide, it's the majority
Re:What does ipv6 get you? (Score:3, Informative)
- New header format (less overhead in routers)
- A new Efficient and hierarchical addressing and routing infrastructure (again....less overhead in routers)
- Stateless and stateful address configuration (You could theoretically dump your DHCP servers)
- Built in IPsec
- Better support for QOS (Quality of service) in the protocol fields
- It's extensible (more headers can be added..it's in the protocol)
and more...
Re:What does ipv6 get you? (Score:2)
1 dialup user != 1 IP address. 1 IP address is needed per phone line, but in broadband, 1 IP address is needed per user...
Re:What does ipv6 get you? (Score:2)
How about cable, satelllite, etc. NAT setup even with customers' routers (using NAT)?
Re:What does ipv6 get you? (Score:2)
Unless that user is attempting ftp with an old server that doesn't support passive mode.
Unless that user wants to establish a P2P session with a [g]AIM user and exchange files or photos.
Unless that user is doing IPSEC with the corporate office and NAT traversal is unavailable for some reason.
More than you think: 41% of home Internet users still use dialup according to this report. [websiteoptimization.com]
Re:What does ipv6 get you? (Score:2)
Removing those dialup users would also severely cut
Re:What does ipv6 get you? (Score:2)
Don't get me wrong, I TOTALLY understand the need for public space in some cases, but there should be different level of services depending on what the customer expects from the connection. If someone were to do a study that said 99% of AOL's customers simply want to use the internet for web/e-mail, and AOL could save $50 million/year by not having to pay ARIN for ip space (and lowered s
Re:Industry study say govt must spend billions... (Score:3, Funny)
Re:Gotten used to NAT (Score:2)
the teqniques needed are in RFC 3489 [faqs.org]
the main issue is that to use it you have to do everything over udp and that requires far more programmer effort to do well than tcp.
Re:Gotten used to NAT (Score:2)
Most corporate NATs, and more and more consumer NATs remember the (protocol, source port, source IP, destination port, destination IP) tuple for each state. STUN depends on the NAT remembering no more than (protocol, source port, source IP, destination port); anything built on Linux IPTables, or OpenBSD's pf already remembers too much.
Re:will IPv6 give me an unique identifier? (Score:3, Informative)
You can run dhcp on ipv6. Your isp can choose how they set up things, just like with ipv4.
Autoconfiguration with MAC addresses might not be the answer for ISP networks, since an ISP might not have enough control about people's NICs or how they will use MAC numbers. I suspect they will offer two solutions: one DHCP-like with 1 ip for cheaper accounts, and one with
In 2000, I had an ISP that had native ipv6 support, and I was assigned one random ipv6 address throug