Performance hit on Query logging

[groups]

Eivind Olsen wrote, On 10/07/2010 04:36 PM:
> --On 7. oktober 2010 14.15.37 -0500 CT <groups at obsd.us> wrote:
>> 1) How do I deternine the number of threads Bind is currently using ?
>> per the man page
>
> You could check the syslog, or use rndc:
>
> vimes# /usr/local/sbin/rndc status
> version: 9.7.1-P2
> CPUs found: 1
> worker threads: 1
> ...
>
>
>> 2) What is the preferred way to determine "named" utilization ?
>> Are there measurable impacts to Query response not reflected in CPU load,
>> Memory or IO?
>
> Not sure what you're after. Parameters to measure? Latency / response time?
>
> Regards
> Eivind Olsen
>

Eivind
Thx for the response..
I had forgot about the simple rndc solution..

We are having a discussion to determine if disabling query logging 
improves DNS response time, when you have a machine that is under 10% 
CPU utilization, I was looking to measure query times with and without
query logging..
I know Disk I/O would have a an impact but I feel it would be very little.

One party thinks that disabling query logging will give enormous 
performance gains, while 30% is a lot.. IMHO it is very negligible in 
CPU cycles when the named process only is taking up > 10% CPU..
and less than 10% in RAM...

Just looking for any suggested tests..

Thx
Charles