Slashdot Mirror
IPv6 Success Stories?
DonGar asks: "We've been hearing how IPv6 will save the world, and we've been hearing about how it will never happen. But can anyone give us real world results about what heppens after they convert? In particular, I'm wondering about small networks (home and/or small business). What ISP support commonly exists, and how much does it really matter? How many people are using ONLY IPv6, instead of both IPv4 and IPv6. What devices/applications/OS's cause the most problems with this? What things work, what breaks, and how much work is it to do the conversion? How hard is it to run things like web and email servers that need to reachable from anywhere? From a real world perspective, what do we need to know that isn't mentioned here?"
Sure there's a benefit: it's cool! After I found out how /insanely/ easy it is to port things to IPv6, I added IPv6 addresses on all the machines in my home network. It's really not that much work, you just add the IP with ifconfig like anything else. Now, home networks are easy because there's no complicated switches needed(we have a hub, which doesn't really know about IP as far as I can tell given that it works fine with IPv6 even though the hub dates from when 10baseT hubs were expensive and cool).
Well, of course it is needed for those of us with more than 4 billion tech gadgets!
Truthfully, IPv6 really needs to be at the ISP level before it is of any use to the end user. When the ISPs start to use it, then they can give each customer a block of 1000 static IPs...and then you will have to figure out what you are going to use 1000 static addresses for...
Oh, and I'll also note that the IPv6 addresses I use are /shorter/ than IPv4 addresses - fec0::1, fec0::2, etc.
I use IPv6 for my home network, so I can ssh in from around the world directly into machines rather than having to SSH into the NAT gateway, then ssh into the machine behind the NAT that I want.
:'s in them.
In general IPv6 was pretty painless to setup, my biggest problems were caused by the fact I was using 6to4 which means my IPv6 addresses are based on my IPv4 address, which isn't static, so it took a bit of scripting to get everything to happen correctly when my v4 address changed (changing routes etc).
Almost all application support v6 one way or another, however notably missing is Apache 1, you need extra patches to get IPv6 support, and most apache log analysers get confused with IP addresses with
I'm surprised that Distro's don't enable v6 by default. (If you have a non-RFC1918 address, use 6to4, if you only have a RFC1918 addresses, use teredo).
I've IPv6 enabled our local LUG server (http://www.wlug.org.nz/), you get a dancing penguin for the logo if you use v6.
When the ISPs start to use it, then they can give each customer a block of 1000 static IPs...and then you will have to figure out what you are going to use 1000 static addresses for...
For each of which they will charge you $5 per month. Don't think that they won't because they are "plentiful". Look how much you are paying for cable/satellite; a luxury item with no scarcity and basically zero-cost to add a customer.
Sorry, NAT's not going away until ISP's stop treating us like consumers. (At least not from MY house. I'd be paying $50/month for IP's alone!)
no.sig.is.good.sig.
IPv6 has working link-local addresses. Right now this probably doesn't mean much, but I think it will as time goes on and people get tired of IPv4 link local addresses.
:']
IPv6 allows you to have more than one public IP address on your home network. You can go through an IPv6 tunnel broker to make your IPv6 network visible to the world.
IPv6 isn't necessarily that far in the future in the U.S. For example, Speakeasy is claiming that they'll have IPv6 rolled out sometime in the March timeframe. I don't know how realistic that is, but it's certainly one of the reasons I'm sticking with Speakeasy - they seem to really have a clue.
1) That all works just fine on Linux, too. Has for years. If you take a look back in shared IP (aka NAT) history, you'll find Linux has understood alot of ways to do that longer than Windows. Though some complicated things may be easier to set up under Windows now than on Linux, I'd be surprised if, stacking the latest Windows against the latest Red Hat / Gentoo / Mandrake / SuSE, there was anything related to networking that Windows can do that Linux can't. Problem solved :)
:)
... this system could be a much simpler one than what exists now for IPv4 while retaining it's value. I know of a couple of times I've had to look up a block of addresses to ban a user and this would also be valuable for things like spam-holes, etc.
2) The average consumer is stuck on IPv4 accessories and applications right now. Of course IPv6 rolling out tomorrow won't help them today. The average consumer never makes use of a new technology immediately. The idea is to make it -possible- so that the average end user will get benefit from it in the future.
3) IPv6 doesn't have to make it easier for trolls to evade IP bans as long as there is a standardized block size made available to individual users -or- as long as some form of CIDR-like registration is used.
Perhaps this is a new tool someone should create. A system that lets you look up what size an IPv6 block is. It should probably point back to the ISP / company that owns the "master" block. In other words, it doesn't need to have any personal information about the person who is using that block (assuming a smaller block is assigned to it), just the range that is owned by that user so that allow / deny rules can be written properly.
An ISP who sets up individual user accounts could have a standard size, so that they simply go in and set up in advance the information rather than having to add it each time they get a new user. If a sub-block does not show up, then people write bans against the entire ISP block, which encourages participation
Coming from someone who used to have to assign and re-assign subnet ownership for an ISP, they already do work like this
That is, assuming that routing doesn't go back to the ways before of users aquiring netblocks and then having their ISP route them (instead of ISPs aquiring netblocks and subnetting them to customers). However even if that is eventually the method (and I believe that was part of the idea behind having so many addresses in IPv6) that just makes it easier to look up block owners and ban entire blocks.
As long as you don't mind a little bit of heavy handedness.
It is more productive to voice thoughtful opinions (reply) than to judge (moderate) others.