Authoritative record
Dave Comcast
dgattis at comcast.net
Wed Dec 4 00:54:10 UTC 2002
Fixed! Thanks to all who help unravel the mystery!
Dave
----- Original Message -----
From: "Kevin Darcy" <kcd at daimlerchrysler.com>
To: <comp-protocols-dns-bind at isc.org>
Sent: Tuesday, December 03, 2002 6:56 PM
Subject: Re: Authoritative record
>
> Dave Comcast wrote:
>
> > Take a look at this zone file and see what I'm missing.
> >
> > $TTL 3600
> > @ SOA ns3.romehosting.com. webmaster.rubymanager.com.
> > (
> > 2002120303 ; zone serial number in ccyymmddxx format
> > 21600 ; slave polls master for SOA/serial number
> > 1800 ; slave re-polls unreachable master
> > 864000 ; slave expires zone after master unreachable
> > 86400) ; TTL for negative answers
> >
> > ; Name servers
> > @ NS ns3.romehosting.com.
> > @ NS ns1.romehosting.com.
> > ;
> > ; Host names and addresses
> > ;
> > @ A 68.60.10.202
> > localhost A 127.0.0.1
> > ftp A 68.60.10.202:21
> > mail A 68.60.10.202.8080
> > ----- Original Message -----
> > From: "Kevin Darcy" <kcd at daimlerchrysler.com>
> > To: <comp-protocols-dns-bind at isc.org>
> > Sent: Tuesday, December 03, 2002 6:31 PM
> > Subject: Re: Authoritative record
> >
> > >
> > > Dave Comcast wrote:
> > >
> > > > What determines that a dns is authoritative to a domain? Any
examples?
> > >
> > > A nameserver is authoritative for a zone if it a) is an origin of zone
> > > data and b) replicates all zone data which it does _not_ originate
from
> > > one or more other authoritative servers and c) suffers no operational
> > > problems (e.g. failed validation or zone data, failed replication)
would
> > > prevent the nameserver from claiming authority for the zone.
> > >
> > > Note that the above is a very generic definition which even tries to
> > > accommodate so-called "multi-master DNS", which is not supported by
BIND.
> > > In BIND-specific terms, a nameserver is authoritative for a zone if it
is
> > > defined as "type master" and has successfully loaded all of the zone
data
> > > (typically from a zone file), or if it is defined as "type slave" and
a
> > > successful zone transfer has occurred more recently than the
> > > EXPIRE interval for the zone (EXPIRE is specified in one of the fields
of
> > > the zone's SOA record).
>
> One of your A records has a colon on its right-hand side, and another has
5
> octets. Looks like you're trying to embed port-number gunk into your A
> records, and that simply doesn't work. Just map from the name to the IP
> address and find some other way to deal with the port numbers (port 21 is
the
> default for FTP so you shouldn't even need to do anything special for that
> one).
>
> Note: if clients would ever support the SRV record type, you could use
that to
> specify alternate port numbers. But that's probably years away.
>
>
> - Kevin
>
>
>
More information about the bind-users
mailing list