restarting bind

Will Yardley william-nospam-newdream-net at no.spam.veggiechinese.net
Sun Mar 24 04:43:03 UTC 2002


In article <a7jin1$1gk at pub3.rc.vix.com>, Doug Barton wrote:
> Warwick Brown wrote:
>> 
>> is it better to use ndc restart or kill -HUP `cat /var/run/named.pid
>> ` (which i currently use)?
>> 
>> my personal belief is that the 2nd option is the best coz it doesnt
>> reload every single zone (only the ones that have changed)
 
> Personal beliefs rarely enter into technical matters. :) The two
> things you described above are entirely different. Restarting the
> server means to kill the one that is running, and start a new
> instance. named responds to -HUP as if ndc had sent it a 'reload'
> command. Reload does actually examine each zone file, but only reloads
> the ones that have changed since the last reload. 

either way, bind has to stat each file to see if it *has* changed.

bind v9 allows you to reload / refresh individual zones using rndc,
which is really nice if you have a lot of zones on your machines.

as doug said, i think a SIGHUP is pretty much the exact same thing as
ndc reload.

with bind 9, it's generally suggested to use rndc for controlling named
(except for starting it)

-- 
No copies, please.
To reply privately, simply reply; don't remove anything.


More information about the bind-users mailing list