• Welcome to Hurricane Electric's IPv6 Tunnel Broker Forums.

Wow. No wonder IPv6 isn't going anywhere...

Started by falconn, September 23, 2009, 06:46:52 PM

Previous topic - Next topic

falconn

Well I just signed up and tried to setup the HE broker service on my VPS'es and my home network... No wonder hardly anyone is migrating - it's a bloody PITA!!

</rant>

I'm obviously not up to speed enough on it to be implementing it yet, even for testing and learning, so any suggestions for where I should be going/reading/doing to pick it up? I've already spent several days over the last 6 months or so on and off learning about the differences between v4 and v6, address syntax etc but apparently that's still not enough.

I know I'm going to have to learn this at some stage, and judging by this morning's encounter, I'm still a long way off :s

broquea

#1
Bunch of books from this post should help you get going:

http://www.tunnelbroker.net/forums/index.php?topic=130.0

Don't feel like you can't ask question on the forums either. You can also email ipv6@he.net any questions you have about getting up and running.

Also since this isn't really about the cert program, I'm moving it to a different topic.

falconn

Quote from: broquea on September 23, 2009, 06:52:36 PM
Also since this isn't really about the cert program, I'm moving it to a different topic.
Sorry, didn't see the header above, only the "General Discussion" part.

I'll have a look through that link and see what I can find.

Is there a common fsck-up that people make that cause 2 Linux hosts on the same layer 2 network (my home network) to not be able to ping6 each other when:
- they have sequential addresses within the routed /64 allocated by HE
- route -nA inet6 show that both know that the /64 is on eth0
- there's no ip6tables running
?

jimb

Quote from: falconn on September 23, 2009, 07:03:32 PM
Quote from: broquea on September 23, 2009, 06:52:36 PM
Also since this isn't really about the cert program, I'm moving it to a different topic.
Sorry, didn't see the header above, only the "General Discussion" part.

I'll have a look through that link and see what I can find.

Is there a common fsck-up that people make that cause 2 Linux hosts on the same layer 2 network (my home network) to not be able to ping6 each other when:
- they have sequential addresses within the routed /64 allocated by HE
- route -nA inet6 show that both know that the /64 is on eth0
- there's no ip6tables running
?
As long as you've configured the IPv6 addresses properly on both systems, and no firewall is running, there should be no problems.  Are you perhaps not using ping -n and stopping it while it's timing out waiting for a RDNS reply?  Also make sure you're typing your ipv6 addresses correctly.  They're kinda long and easy to screw up.  Both in your interface configuration and your ping.  OH.  And make sure you're actually using the routed /64, not the tunnel /64.

falconn

I'm reasonably sure I checked all those boxes... Definately was using the routed /64, I copied and pasted the addresses, just changing the last 4 nibbles - 1000 for my firewall, and 1011 for my desktop. But I can't check anymore cause I threw the toys out of the pram and wiped all the config :P

I'm starting from scratch again, so I'll see how I go this time. I'll post back when I come across problems, which I no doubt will :P

jimb

Quote from: falconn on September 23, 2009, 07:45:15 PM
I'm reasonably sure I checked all those boxes... Definately was using the routed /64, I copied and pasted the addresses, just changing the last 4 nibbles - 1000 for my firewall, and 1011 for my desktop.
OH.  Well that's the problem.  You're in the southern hemisphere (Australia), so you have to reverse the last four nybbles.   ;D :D