<p><br>
On Aug 5, 2012 8:00 PM, "Doug Barton" <<a href="mailto:dougb@dougbarton.us">dougb@dougbarton.us</a>> wrote:<br>
><br>
> On 08/03/2012 05:39, Benedikt Stockebrand wrote:<br>
> > yes, in some cases you may want to filter e.g. routing headers and<br>
> > such.<br>
><br>
> Do you have references to this issue?<br>
></p>
<p><a href="http://tools.ietf.org/html/rfc5095">http://tools.ietf.org/html/rfc5095</a></p>
<p>> > More generally speaking, with new ICMP6 types possibly coming<br>
> > up you may want to whitelist rather than blacklist individual ICMP6<br>
> > types/codes.<br>
><br>
> This is the opposite of what should be done, for 2 reasons. First, you<br>
> should only blacklist things you know you're having problems with.<br>
> Second, but taking the approach you suggest you miss out if the protocol<br>
> changes and you don't update your filters.<br>
><br>
> The whole concept of blanket ICMP restrictions in v4 was bad, doing it<br>
> for ICMPv6 is really bad.<br>
><br>
> --<br>
><br>
> I am only one, but I am one. I cannot do everything, but I can do<br>
> something. And I will not let what I cannot do interfere with what<br>
> I can do.<br>
> -- Edward Everett Hale, (1822 - 1909)<br>
</p>