No need for IPv6?

IPv6 is ten years old, it’s been a promising technology for 6 years now, and nobody uses it. I wrote an Introduction to IPv6 some time ago, and since then one question has intrigued me: Will IPv6 ever be used widely? One reason I can think of (and that I explained earlier), is the need for more IP-addresses. But there has been created work-arounds, like DHCP (dynamic assigning of IP-adresses). So there is not an urgent need for IPv6 as I see it. Is it possible that the market will wait for another version of IP, like ver. 7 or 8?

Opinions?

It is likely the first uses of some form of ipv6 with be auto regestistered - like dhcp in otherwords. It will probably find its way into embedded items first, such as auto location devices.

Quote:

---

IPv6 is ten years old, it’s been a promising technology for 6 years now, and nobody uses it. I wrote an Introduction to IPv6 some time ago, and since then one question has intrigued me: Will IPv6 ever be used widely?

---

I don't think IPv7 or 8 is comming out any time soon, Because IPv6 provides new capabilities for high quality streaming, multimedia communications such as real time audio and video. Huge opportunities for research and experimentation exist in this area of communication, espically as the internet continues to grow.

It also improved support for mobile computing..

Nobody uses it hhmmm? You are right you won't find many places using IPv6? Becauses lots of places are testing it. There is an IPv6 backbone called 6bone.. The 6 bone is a worldwide test network consisting of over 400 sites spread across 40+ countries... and still growing.
The purpose of the 6bone is to provide an IPv6 test site fro interoperability and connectivity..

Quote:

---

Originally posted by I am a cracker
lots of places are testing it. There is an IPv6 backbone called 6bone.. The 6 bone is a worldwide test network consisting of over 400 sites spread across 40+ countries... and still growing.
The purpose of the 6bone is to provide an IPv6 test site fro interoperability and connectivity..

---

Perhaps the market is waiting for the result of these tests? Maybe they want more information on how IPv6's services will improve the internet? The costs of upgrading the net are high and the ISPs wont spend lots of money on something that's not gonna pay off. Of course there's a great potential for making money in the video-streaming and the mobile terminal business. But as markets are pointing down, I doubt that the companies will take the risks that's involved in releasing new technology and make big investments. If the markets went up, then it's a whole other story, though.

IPv6 prommises a lot better security then the old IPv5 protocol

and I think we all agree... the more secure it is.. the better

It's IPv4. :) But there are workarounds for the security issues also, like Public Key Infrastructure(PKI). Of course if IPv6 made HUGE differences, but PKI works fine already.

Since when is DHCP a work around for lack of addresses? I think you mean NAT. And as for IP5, it isn't used, the big I is using IPv4.
Also, I hope that IPv6 is not widely used, that will leave us institutions that are using it a nice big fat pipe without a bunch of
cracker/script kids mucking up the works.

Quote:

---

From www.dhcp.org
The Dynamic Host Configuration Protocol (DHCP) is an Internet protocol for automating the configuration of computers that use TCP/IP. DHCP can be used to automatically assign IP addresses.

---

DHCP is the protocol that assigns dynamic IP-addresses, which is sort of a work-around for lack of IP addresses.

*beep* wrong answer. DHCP give out address yes, but only the amount that you have. Iif you only have 128 address (1/2 a class C),
DHCP will not invent more out of thin air. It will only give out 127 MAX (your dhcp server would need one of these)

Yes I see. But you can have more than 127 users as long as they don't access the internet at the same time, am I right? So instead of only 127 people with static addresses, there could be 1270 who shares the 127. Not a perfect workaround, but still a workaround. But you are right, there is still the limit of addresses. Just a more creative way of assigning them.

Yes, that could be an awkward solution, but a solution non-the-less. Its a navy term called "hot bunking". Meaning that if you had 256 machines
you would have to share the 127 open addresses. Only 1/2 of your availiable workstations could be on the network at any given time.

DHCP (Dynamic Host Configuration Protocol, also see www.isc.org) is based on a per-scope basis. You have a dhcpd.conf file that's read at start time that has both configurations per scope, pools of IP addresses to hand out for each scope, and what restrictions, if any, on said scopes (example: drop unknown clients instead of allowing them to get an IP).

There's a handshake that goes on when a client set to obtain automatically (DHCP) sends out the UDP broadcast basically asking "Is anyone willing to give me an IP?" to which a server set to answer will say "Yeah...lemme go see who you are (checks already-read-table-of-accepted-MAC-addresses). (Found) Here's an IP (if it has it) or (Not found, option to not allow unknown clients) Get lost!."

There's some more in there but I can't remember off the top of my head right now.

This about sums up my feelings on the matter.

Quote:

---

And while IPv6 yields a greater number of available IP Addresses, approximately
3.4 x 10^38, it remains slightly less than IPv8's 128 Bit Address
availability of '3.40282 x 10^38'. Furthermore, when noting the benefits
offered by IPv6, which are taunted as being advantageous. No presentation
emphasizing its high lights, can suppress the severity of the drawbacks it
maintains. In fact, IPv6 is not only cumbersome and difficult to use,
implement, and employ. But, it lacks a Mathematically Derived Logical
Structure, which results in a 'Default Addressing Structure' being
superfluously defined.

---

You can find the entire white paper at http://www.ietf.org/internet-drafts/...v7-ipv8-09.txt

IP6 ... the Novell-ization of hostnames. :) You know ... I'd doubt we'd have such an imminent need for 3.4 x 10^38 distinct addresses if we could clear out SPAM-based black-hole addresses. Mmmmm ... what a great world that'd be. :)

~N~