dealing with many domains/zones

Vinny Abello vinny at
Sat Jan 15 18:08:40 UTC 2005

I've heard and been told that additional worker processes will help this 
and allow queries to continue while a reload is in progress. The worker 
processes are determined automatically upon startup based on the number of 
CPUs in your server. You can also force it with a switch.

In practice, I haven't been able to make more worker processes solve this 
problem for me personally. I have this when trying to reload large zone 
files (RBL zones).

In your situation, do you just have a lot of zones or some extremely large 
zones? Make sure you're not just doing a reload of every zone on the 
server. Just do a reload of the zone you want the reload to take place on.

Having said all this and personally loving BIND, I do know that other DNS 
servers don't have this issue and look forward to the day that this is 
really fixed in BIND.

At 06:36 AM 1/15/2005, Robert Montgomery wrote:
>If I have a large number of domains (500K+), is it
>faster to just rsync the zone & conf files (creating
>multiple masters) instead of using slaves & zone
>Also, is there any way to minimize the downtime on
>reloads?  I'm guessing that multiple fast CPU's and
>plenty of RAM are the best you can do to speed up
>reloads?  (I'm also guessing that disk I/O isnt a
>bottleneck, assuming that there is enough ram ?).
>I just need to do anything I can to minimize downtime
>during reloads and synching of slaves (or multiple
>Do you Yahoo!?
>Yahoo! Mail - Find what you need with new enhanced search.

Vinny Abello
Network Engineer
Server Management
vinny at
(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 (888)TELLURIAN

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

More information about the bind-users mailing list