Turn off IPV6_USE_MIN_MTU?

[Mark Andrews]

In message <20160531164208.GA1020 at mycre.ws>, Robert Edmonds writes:
> Mark Andrews wrote:
> > UDP is reliable if you don't have idiots breaking the protocol by
> > dropping fragments.
> 
> This is like Taco Bell saying that 88% of their meat is 100% pure beef.

No it isn't.  You don't get some patties (paths) that are a 100%
pure beef and some that patties that aren't.  You also don't get
customers saying give me the adulterated patties and then complaining
that they are adulterated.

> Here is the third largest wireless carrier in the U.S. declaring that
> they rate-limit UDP traffic by both bps and pps and advocating that
> other network operators do the same:
> 
>     https://tools.ietf.org/html/draft-byrne-opsec-udp-advisory

Which was not accepted by the IETF nor was it published as a RFC
via the ISE.  There are lots of bad ideas published as internet
drafts.  Lots of operators also said it was a bad idea at the time
it was published.

> This thread is also interesting:
> 
>     https://groups.google.com/a/chromium.org/d/msg/proto-quic/09L5YD2u5xU/EsZ
> gXHJq0o4J
> 
> I wouldn't be surprised if exactly these sorts of L3/L4 policers
> deployed by network carriers accounted for a significant proportion of
> the dropped fragments Geoff detected.
>
> -- 
> Robert Edmonds
> _______________________________________________
> bind-workers mailing list
> bind-workers at lists.isc.org
> https://lists.isc.org/mailman/listinfo/bind-workers
-- 
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742                 INTERNET: marka at isc.org