BIND pauses on loading large zones

[Vinny Abello]

Hello all,

I've heard of this issue before and remember a discussion but no resolution.

I have a zone that is approximately 35MB in size. When I add it to the 
named.conf file and do an rndc reconfig, the server stops answering queries 
for probably around 10 seconds or more while it's loading that zone. Also, 
if I modify it and do an rndc reload zonename the same thing happens.

I heard that this is because there is only one worker thread by default if 
BIND can't detect more than one CPU. I have 4 CPU's in a system 
(technically 2 with hyperthreading) and it doesn't appear to detect them. I 
haven't tried running it in the foreground to see how many CPU's... but I 
did try changing the startup option to include -n 4 and also tried -n4 as 
well but it didn't make a difference, and the OS didn't indicate any more 
threads were created than without that setting.

This is on Windows Server 2003 with BIND 9.2.4rc4.

Any takers? :)

Vinny Abello
Network Engineer
Server Management
vinny at tellurian.com
(973)300-9211 x 125
(973)940-6125 (Direct)
PGP Key Fingerprint: 3BC5 9A48 FC78 03D3 82E0  E935 5325 FBCB 0100 977A

Tellurian Networks - The Ultimate Internet Connection
http://www.tellurian.com (888)TELLURIAN

There are 10 kinds of people in the world. Those who understand binary and 
those that don't.