Proper way to run bind9

[Doug Barton]

On Tue, 28 Sep 2004, Juha Saarinen wrote:

> On Mon, 27 Sep 2004 18:54:01 -0700 (PDT), Doug Barton <dougb at freebsd.org> wrote:
>> A couple of them actually. We do not want to edit the files as they come
>> from the vendor without a really good reason, and this isn't one.
>>
>> I have a long term plan to write some patches to turn the pid file path
>> into a --configure defineable variable and send it to the ISC folks, but
>> it's frankly not that high a priority.
>
> Humm, that does seem like the right way to do it, instead of working
> around the issue by changing the PID file location in two different
> places.

Thanks.

>> If you use the system as installed, and/or start from the default files,
>> it's all there for you. If you choose to vary from that path, it's
>> pretty much up to you to know what you're doing and why. There are only
>> so many bullets you can take out of the foot-shooting gun.
>
> True -- however, this is likely to bite people who migrate from other
> platforms where you don't have to specify the PID file location in
> named.conf, unless you want it in a non-default location. But, people
> have plenty of toes I suppose... :-)

*nod*  Now that I've committed the chroot defaults, I may consider 
changing this back to /var/run/named.pid .... I'll wait to see how the 
chroot stuff falls out for people.

>> What would your goal be? With the current behavior, '/etc/rc.d/named
>> stop' can recover from situations where 'rndc stop' fails. Why would you
>> want to take that functionality away?
>
> Well, rndc is the vendor-supplied tool for controlling the operation of named.

I think you missed the part of my previous message where I talked about 
how the current system offers the maximum in terms of features and 
flexibility.

> The man page for named(8) says:
>
> "In  routine  operation, signals should not be used to control the name-
> server; rndc should be used instead."

That same man page then defines the behavior for SIGINT and SIGTERM. 
Killing named with a signal in this case is harmless, and should be 
functionally equivalent to 'rndc stop', except in those cases where rndc 
is buggered for some reason.

> Incidentally, shouldn't the 'rcvar" command print out all the options
> used in rc.conf for running named?

You might want to follow up with this question on 
freebsd-rc at freebsd.org.

Hope this helps,

Doug

-- 

     This .signature sanitized for your protection