Linux IPv6 routing strange behaviour

[Pim van Pelt]

On Aug 15, 2013 1:27 PM, "Phil Mayers" <p.mayers at imperial.ac.uk> wrote:
>
> On 15/08/13 12:14, Pim van Pelt wrote:
>>
>> Just ad a datapoint to Max' last remark, at sixxs we moved away from
>> kernel based routing by implementing ipv6 routing in userspace (taking
>> tap input and raw socket output) largely because of neighbor cache
>
>
> Interesting. Was this custom/proprietary software or is it available
somewhere?
It's proprietary, not open source and not necessarily even a complete
routing stack (but it will scale to 100k interfaces). Rather it is a
specific implementation of a tunnelbroker with filtering, RPF, proto41,
AYIYA and heartbeat support.

We could not find any platform (commercial or open source) that would
perform and scale beyond the ~1000s of tunnels, as well as support
TIC/AYIYA (not surprising as the tunnelbroker field is a true niche and we
made it worse by offering our own draft protocols which folks don't
necessarily share across platforms), so we went ahead and implemented our
own client (open source, called aiccu) and server (closed, called sixxsd).

Pim
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cluenet.de/pipermail/ipv6-ops/attachments/20130815/b8aa6bf3/attachment.htm>