I am running around 25 OpenVZ containers on a node right now, and all of them are assigned only IPv6 addresses that are being routed from tunnelbroker.net
I used this guide to set the IPv6 up, as I don't have native IPv6. www.useipv6.com/fedora.php
Now, none of my nodes are able to connect to the IPv4 internet(as obviously expected), and thus unable to get updates and the like.
What I want to do is assign all of these containers, all at once, a single shared IPv4 address so that they can communicate with the internet, download files, updates, etc., but I don't want to assign ports to the VPSes just yet.
I tried adding 192.169.2.1, 169.254.1.100, 10.10.10.195 one by one to a container for the local ip addresses, but all of them failed with the following message:

vzctl restart 18548
Restarting container
Stopping container ...
Container was stopped
Container is unmounted
Starting container...
Container is mounted
Adding IP address(es): <the ipv6 address> 192.168.2.1
arpsend: 192.168.2.1 is detected on another computer : 00:04:9b:f2:b0:00
vps-net_add WARNING: arpsend -c 1 -w 1 -D -e 192.168.2.1 eth0 FAILED
Setting CPU limit: 50
Setting CPU units: 1000
Setting CPUs: 1
Container start in progress...

Doesn't matter which IP I try to add, it shows this error message and doesn't work.
I do have two containers running Public IPv4 addresses and they are working well, and I have one IP assigned to the host, and one un-utilized IPv4 address atm.
I'm ready to use that one extra IP, but preferably I'd like using the hosts IP address for setting up NAT.
Since this is my first time dealing with OpenVZ or any sysadmin tasks for that matter, I am perplexed.
I don't know how to go about doing what I want.
Can somebody please guide me with it?
Also, could someone tell me how to assign IPv6 subnets(64 addresses) to a container at once? I have a /64 routed to the host, so 18 quintillion addresses to spare.
Also, I just noticed that when I run ping6 google.com from within any container, it stays empty for some time, and then returns unknown host, any ideas on how to solve that?
Help would be extremely greatly appreciated.

By the looks of it a device with Mac 00:04:9b:f2:b0:00 has that IP or atleast it thinks it does. Are you configuring the host node with those IP address first? Because if you are, that is where the conflict is, unless you have the host node configured as a router as well and using the host as the gateway.

I don't think I assigned the host node even one of those IPs, and since those ips are private and not public, I don't think my provider did either. I have five public ipv4 addresses, but they are not assigned to the host node, not even in the eth0-range(or something similar) file, and they're working just fine. I don't think there should be this problem with internal addresses.