Dear All <br><br> This is in reference to the performance tuning , i had already gone through the mailing list archives , but could not find answer to my <br>specific query mentioned here. <br><br> I had installed bind as a caching name server for test purposes and planning to test performance that could give me around 10000 qps. <br>
<br>The os running Centos 5.4 64 bit , with minimal packages installed. The server is dual quad core Intel Xeon 2.53 GHz ,16 GB RAM with 300 GB Hdd( Raid 1) . <br><br><u>Bind version installed is bind 9.6.1-P3 . Source extracted and compiled with the following options. </u><br>
<br>./configure --enable-epoll --enable-atomic --enable-ipv6 --enable-chroot --with-openssl --with-randomdev=/chroot/named --disable-openssl-version-check --with-libtool --enable-threads<br><br>Chrooted bind installation is done. <br>
<br>Only named, ssh and ntp services are running on the servers. <br><br>Right now i am using queryperf to test the performance with sample query file of thousand entries. Right now <br>i am getting only 2000 to 2300 qps . I am writing querylogs to a separate partition with noatime enabled for the <br>
partition. <br><br> OS hardening is done by removing unwanted services, closing all unneccesary ports and securing the running services.<br><br> My system is now using only 3 GB of RAM of total 16 GB. <br><br><u>Following is the output of "uptime;free -m" during performance testing</u><br>
<br>[root@localhost ~]# uptime;free -m<br> 22:19:52 up 1 day, 6:06, 3 users, load average: 2.03, 2.06, 1.34<br> total used free shared buffers cached<br>Mem: 16047 3183 12864 0 238 2037<br>
-/+ buffers/cache: 907 15140<br>Swap: 8189 0 8189<br><br><br><br><br><u>Following is my named.conf</u><br><br>acl testsetup_net {<br> <a href="http://10.201.31.0/26">10.201.31.0/26</a>; };<br>
<br>acl blacklistnets {<br> <a href="http://192.0.2.0/24">192.0.2.0/24</a>; <a href="http://224.0.0.0/3">224.0.0.0/3</a>; <a href="http://10.0.0.0/8">10.0.0.0/8</a>; <a href="http://192.168.0.0/16">192.168.0.0/16</a>;<br>
};<br><br>// Main options defined here<br>options {<br> directory "/conf";<br> dump-file "named_dump.db";<br> statistics-file "named.stats";<br> pid-file "/var/run/named.pid";<br>
allow-recursion { localhost; ; testsetup_net; };<br> allow-query { localhost; testsetup_net; };<br> allow-query-cache { localhost; testsetup_net; };<br> allow-transfer { none; };<br> blackhole { blacklistnets; };<br>
recursive-clients 20000;<br> version "Not old!";<br> datasize default;<br> notify yes;<br>};<br><br>// Logging options are defined here.<br>logging { // logging option for named process<br>
channel "default_debug" {<br> file "/logs/named.log" versions 10 size 50m;<br> print-time yes;<br> print-category yes;<br> severity dynamic;<br> };<br>
<br> channel "queries" { // logging option for queries to named<br> file "/logs/query.log" versions 20 size 100m;<br> print-time yes;<br> print-category yes;<br>
severity dynamic;<br> };<br><br> category default { "default_debug"; };<br> category queries { "queries"; }; // uncomment this to log queries<br> category config { "default_debug"; };<br>
category security { "default_debug"; };<br> category network { "default_debug"; };<br> category lame-servers { null; };<br> category edns-disabled { null; };<br>};<br><br><br>zone "." in {<br>
type hint;<br> file "db.rootcache";<br>};<br><br>zone "localhost" in {<br> type master;<br> file "db.local";<br> notify no;<br>};<br><br>zone "0.0.127.in-addr.arpa" in {<br>
type master;<br> file "db.127.0.0";<br> notify no;<br>};<br><br><br> Kindly guide me for improving the bind performance from 2000 qps to nearly 10000 qps. Which are the parameters i should change for improving<br>
the performance? Any os level parameters to be changed for improving the performance? <br><br>thanks in advance<br><br>Regards<br><br>Shiva Raman <br><br><br><br><br>