> Paul Hoffman <mailto:[email protected]>
> Wednesday, February 11, 2015 3:17 PM
> On Feb 11, 2015, at 1:30 PM, Paul Vixie <[email protected]> wrote:
>> 25/sec will not be enough for large rdns plants.
>
> That sounds specific enough that you have actual data to back this up; if so, 
> I'm quite interested in it.

a busy RDNS that isn't doing Q-M often asks more than 25 bad-TLD queries
per second. see OARC DITL data.
>
>> that's why the default policy for slip and drop is so important. f-root's 
>> team must have overridden those, probably because various people have spread 
>> some FUD about drops.
>
> You might be willing to say what the f-root team did, and why they did it, 
> even without being on the team, but I'm not.

DNS RRL does not do 996 slips and four responses in a second under any
default config.

>
>> this work came out of ddos work not dns work. after the tenth anniversary of 
>> SAC004 came and went, with more rather than fewer edges lacking SAV. 25/sec 
>> of signed nxdomain is enough to overload any DSL circuit. i'd be happy to 
>> work with you to find an upper limit.
>
> OK, now it sounds like you don't have actual data yet. N'r mind.

3000 bytes X 25/sec X 13 root name servers X 8 bits = 7.8Megabits/second.

by the way that level of snark is unusual for you. i'm sorry for peeving
you out of your comfort zone.

-- 
Paul Vixie
_______________________________________________
dns-operations mailing list
[email protected]
https://lists.dns-oarc.net/mailman/listinfo/dns-operations
dns-jobs mailing list
https://lists.dns-oarc.net/mailman/listinfo/dns-jobs

Reply via email to