using $INCLUDE with views

Stefan Puiu stefan.puiu at gmail.com
Tue Oct 10 20:13:04 UTC 2006


On 10/10/06, Bill Moseley <moseley at hank.org> wrote:
> On Tue, Oct 10, 2006 at 11:27:34AM +0300, Stefan Puiu wrote:
>
> [quoting out of order]
>
> > Oct 10 11:21:34 tuxws43 named[9050]: /var/lib/named/foo.common:1: no
> > TTL specified; using SOA MINTTL instead
> > Oct 10 11:21:34 tuxws43 named[9050]: dns_master_load: foo.zone:3: no
> > current owner name
>
> According to the FAQ:
>
>     This error is produced when a line in the master file contains
>     leading white space (tab/space) but the is no current record owner
>     name to inherit the name from. Usually this is the result of
>     putting white space before a comment. Forgeting the "@" for the
>     SOA record or indenting the master file.
>
> So, perhaps the $INCLUDE is forcing white space before the SOA
> record?

Bill, as far as I can see the log points to line 4 in hank.org, not
the included file. I've already explained why - each resource record
has to have an owner, but when you use the include file, the NS
records don't have any. However, if you put the SOA before the
$INCLUDE directive, you have a default owner (default domain in ARM
terminology).

So, the problem occurs when the $INCLUDE is the first thing in the
file, and then right after it come the NS records, without an explicit
owner. If you put the SOA before the $INCLUDE, it makes a difference.

Also, I would look at the permissions for /etc/rndc.key, if you want
to be able to use rndc to control named.



More information about the bind-users mailing list