[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: 6to4 tunnel trouble



Bill Walker wrote:
> Brian,
>
> You only need IPv4 Transit between you and HE. HE have a good config sample on their website once you create the tunnel:
>
> modprobe ipv6
> ip tunnel add he-ipv6 mode sit remote 64.71.128.82 local x.x.x.x ttl 255
> ip link set he-ipv6 up
> ip addr add 2001:xxxx:xxxx:xxxx::2/64 dev he-ipv6
> ip route add ::/0 dev he-ipv6
> ip -f inet6 addr
>
> The above uses the route2 package and if you use the show config button on your tunnel info page it has all the details filled in. They also have a code sample using net-tools
>
> ifconfig sit0 up
> ifconfig sit0 inet6 tunnel ::64.71.128.82
> ifconfig sit1 up
> ifconfig sit1 inet6 add 2001:xxxx:xxxx:xxxx::2/64
> route -A inet6 add ::/0 dev sit1
>
>   
Debian style in /etc/network/interfaces:

iface hurricane inet6 v4tunnel
        address  xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx
        netmask  64
        local    x.x.x.x
        endpoint y.y.y.y
        ttl      64
        up       ip link set mtu 1280 dev hurricane
        gateway yyyy:yyyy:yyyy:yyyy:yyyy:yyyy:yyyy:yyyy


xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx:xxxx <- your end of the tunnel v6
yyyy:yyyy:yyyy:yyyy:yyyy:yyyy:yyyy:yyyy <- their end of the tunnel v6
x.x.x.x <- your end of the tunnel v4
y.y.y.y <- their end of the tunnel v4

and maybe add "ipv6" in /etc/modules

No messing around :) After all, this is a Debian specific mailing-list.

Kind regards,
Martin List-Petersen

> Cheers,
>
> Bill
>
>
> -----Original Message-----
> From: Brian McCullough [mailto:bdmc@bdmcc-us.com] 
> Sent: Monday, 22 December 2008 11:49 a.m.
> To: Martin List-Petersen
> Cc: debian-ipv6@lists.debian.org
> Subject: Re: 6to4 tunnel trouble
>
> On Sun, Dec 07, 2008 at 09:46:38PM +0000, Martin List-Petersen wrote:
>   
>> But not every provider has a 6to4 gateway and not every provider is 
>> ready for IPv6. On top of that, the different providers apply filtering 
>> on what they accept on their BGP session and what not. Some do this 
>> based on the whois databases, which means 2002::/16 and 192.88.99.0/24 
>> will be filtered out crossing from provider to another unless otherwise 
>> agreed.
>>
>> This is both positive and negative. It means that not anybody actually 
>> has a route to a 6to4 gateway (or the opposite path).
>>
>> A fault with kame is not unlikely. Sometimes things break in the 
>> internet. A lot more often in IPv6 than in IPv4, because a lot of the 
>> carriers and provider don't take is serious, yet. I have for example one 
>> transit provider, that doesn't give us an SLA on IPv6, because they 
>> themselfes only have one transit provider with IPv6. That has in the 
>> past caused things to break for longer than you'd expect (as in days).
>>
>> The successrate is higher with a tunnel, because there is a clear 
>> defined end-point to you.
>>
>> With 6to4 the traffic flows encapsulated to the first known 6to4 gateway 
>> in your path. The return-traffic flows to the first known 6to4 gateway 
>> in your destinations path. And they are nearly always different. So if 
>> one is broken, traffic doesn't flow.
>>
>> With a tunnel the traffic always originates and returns to your 
>> tunnelprovider over the same tunnel-server.
>>     
>
>
> Martin ( and group ),
>
> I may have quoted more than necessary, but I wanted to get the correct
> context.
>
> I have been interested in IPv6 for several years, but always seemed to
> run into issues.
>
> You recommended SixXS or HE.  I have ( a while ago ) been in touch with
> HE ( and had an account with xs26, but they seem to have disappeared ).
>
> However, to use that service ( the HE tunnel ), do I need to have an
> IPv6-compatible ISP, or can I get there through an IPv4 tunnel?  I
> suspect that my questions will be somewhat elementary, but it has been
> quite a while since I last spent time working on this, and several of
> the information and documentation sites seem to have gone away.
>
>
> Thanks,
> Brian
>
>
>
>   


-- 
Communication is the beginning of understanding
   -- AT&T


Reply to: