[Pdns-users] Big mean response

[Piotr Zamojski]

Ok, I start from show my configuration:

/etc/powerdns/pdns.conf:
allow-recursion=127.0.0.1
cache-ttl=60
config-dir=/etc/powerdns
daemon=yes
disable-axfr=yes
distributor-threads=10
guardian=yes
lazy-recursion=yes
local-address=0.0.0.0
local-port=53
log-dns-details=no
log-failed-updates=no
max-tcp-connections=100
module-dir=/usr/lib/powerdns
negquery-cache-ttl=120
query-cache-ttl=60
receiver-threads=4
recursive-cache-ttl=60
setgid=pdns
setuid=pdns
socket-dir=/var/run
version-string=powerdns
include=/etc/powerdns/pdns.d

/etc/powerdns/recursor.conf:
allow-from=127.0.0.0/8, ::1/128, fe80::/10
client-tcp-timeout=4
local-address=127.0.0.1
local-port=5300
max-cache-entries=1000000
max-negative-ttl=3600
max-tcp-clients=10240
quiet=yes
setgid=pdns
setuid=pdns
stack-size=400000

/etc/powerdns/pdns.d/pdns.local
launch=gmysql
gmysql-dbname=...
gmysql-host=...
gmysql-password=...
gmysql-user=...
master=yes
slave=yes
recursor=127.0.0.1:5300
allow-recursion=127.0.0.0/8, ...
allow-axfr-ips=...
disable-axfr=no
logfile=/var/log/dns.log
webserver=yes
webserver-address=...
webserver-password=...
webserver-port=8012

Variables from webserver when "Uptime: 30 minutes":
corrupt-packets 9 Number of corrupt packets received
deferred-cache-inserts 130 Amount of cache inserts that were deferred
because of maintenance
deferred-cache-lookup 76 Amount of cache lookups that were deferred because
of maintenance
latency 0 Average number of microseconds needed to answer a question
packetcache-hit 42862
packetcache-miss 9157
packetcache-size 10495
qsize-q 0 Number of questions waiting for database attention
query-cache-hit 23712 Number of hits on the query cache
query-cache-miss 17267 Number of misses on the query cache
recursing-answers 7737 Number of recursive answers sent out
recursing-questions 7737 Number of questions sent to recursor
servfail-packets 0 Number of times a server-failed packet was sent out
tcp-answers 0 Number of answers sent out over TCP
tcp-queries 7 Number of TCP queries received
timedout-packets 118 Number of packets which weren't answered within
timeout set
udp-answers 51921 Number of answers sent out over UDP
udp-queries 13004 Number of UDP queries received
udp4-answers 44184 Number of IPv4 answers sent out over UDP
udp4-queries 52039 Number of IPv4UDP queries received
udp6-answers 0 Number of IPv6 answers sent out over UDP
udp6-queries 0 Number of IPv6 UDP queries received

My server is:
Linux dns1 3.0.0-vs2.3.1-pre8-beng #1 SMP Wed Aug 3 10:13:46 BST 2011
x86_64 GNU/Linux
2 x Dual-Core AMD Opteron(tm) Processor 2220
8GB RAM
Debian Squeeze

And now when i run ./namebench.py, my DNS have very slow response time:

Fastest individual response (in milliseconds):
----------------------------------------------
*MyDNS DNS2      # 0.69880
MyDNS DNS1      # 0.73814*
...
Google Public DN ################################### 26.13997
OpenDNS-2        ####################################### 29.60110
OpenDNS          ####################################### 29.62804
UltraDNS-2       ############################################ 33.03885
DynGuide-2       ################################################## 37.79197
UltraDNS         ###################################################
39.05392
Google Public DN #####################################################
40.62200

Mean response (in milliseconds):
--------------------------------
Google Public DN ## 52.20
Google Public DN ### 62.24
OpenDNS-2        ###### 163.05
OpenDNS          ######## 214.32
UltraDNS-2       ######## 216.79
DynGuide         ########## 269.92
DynGuide-2       ########## 276.10
Neostrade/TPnet  ############## 379.00
*MyDNS DNS1      ############### 419.96
MyDNS DNS2      ################ 425.80*
UltraDNS         #####################################################
1499.95

Can I change something else in order to speed up response time?


© 2012
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.powerdns.com/pipermail/pdns-users/attachments/20120518/28a9e99d/attachment.html>