On 13.09.2012 14:01, Mark Scholten wrote:
I am confused about the results in
>http://mailman.powerdns.com/pipermail/pdns-dev/2012-June/001179.html It
>seems that powerdns is slower without the LUA rate limiting script. What
do I
>miss here?
>
>Not sure - perhaps Mark can clarify.
If the LUA rate limiting script is used the backend (MySQL in this test)
doesn't get the requests and doesn't have to answer them. Only the first 500
in result set #2 are answered, the other requests are dropped and never go
to the backend.
This is causing the performance drop as far as I can see.
A, this make things clear. So it would be interesting to test the
performance with filtering in front of the packet cache vs. packet cache.
regards
Klaus
_______________________________________________
Pdns-users mailing list
[email protected]
http://mailman.powerdns.com/mailman/listinfo/pdns-users
