Re: [v6ops] new draft: draft-taylor-v6ops-fragdrop

[Tom Taylor <[e-mail]>]

On 19/10/2012 2:11 PM, Templin, Fred L wrote:
>> other than causing bandwidth / pps DoS attacks, or alternatively tickling
>> obscure ipv6 stack bugs, no.  At least in theory.
>
> There's no excuse for broken IPv6 stacks that can't handle
> fragmentation and reassembly. About DoS concerns, if the
> network is going to intentionally break frag/reass out of
> concern for edge devices that might not be able to handle
> the load, then that is very bad for applications that cannot
> reduce the size of packets they send down to 1280. Again,
> tunnels fall into that category.
>
> So, if tunnels cannot rely on ICMP PTB and also cannot rely
> on IPv6 frag/reass then there is only one option remaining:
>
> https://datatracker.ietf.org/doc/draft-templin-intarea-seal/
>
...
which, as a matter of fact, came up briefly in the Interim Meeting 
discussion.