[Pdns-users] How to optimize cache hit ratio

姜伯洋 15555513217 at 163.com
Mon Aug 5 14:02:50 UTC 2019

[root at ops-dns-recursor-1 ~]# rec_control get-all |grep cache

Is this a recursive node that has been running for a while, and the probability of a cache hit and a packet cache hit seems normal? Why do I feel a little less?

At 2019-08-05 17:28:25, "Markus Ehrlicher" <Markus.Ehrlicher at komsa.de> wrote:



rec-control is a command from the recursor, so why are you talking about your authoritative nameserver?

There are a too small number of queries - you can not draw conclusions from this, in my opinion.


Here’s an example from my ns:


root at ns1:~# rec_control get-all|grep cache

cache-entries   119504

cache-hits      474284

cache-misses    10249358

max-cache-entries       1000000

max-packetcache-entries 500000

negcache-entries        11014

packetcache-entries     82704

packetcache-hits        17015067

packetcache-misses      10715597



best regards,



Von: Pdns-users <pdns-users-bounces at mailman.powerdns.com> Im Auftrag von ???
Gesendet: Montag, 5. August 2019 11:14
An: pdns-users at mailman.powerdns.com
Betreff: [Pdns-users] How to optimize cache hit ratio


Hello everyone

[root at ops-dns-recursor-3 ~]# rec_control get-all|grep cache

cache-entries  121

cache-hits        0

cache-misses  12

max-cache-entries       3000000

max-packetcache-entries        500000

negcache-entries        4

packetcache-entries    7

packetcache-hits         51

packetcache-misses    12

I don't know if the data monitored on my authoritative node is normal, but it seems that there is a high probability that the miss will be lost. How can I optimize it?




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.powerdns.com/pipermail/pdns-users/attachments/20190805/78ddead2/attachment-0001.html>

More information about the Pdns-users mailing list