Zone transfer problems

[Mark Damrose]

"Relay test" <glewis at elstree.co.uk> wrote in message
news:ad5h4a$b2vn$1 at isrv4.isc.org...
>
> michael.wilkinson at twcable.com wrote:
>
> >
> > [ID xxxx daemon.error] transfer of 'fwdzone/IN' from x.x.x.x #53: failed
> > while receiving responses: unexpected end of input
> >
> > What is wierd is if I manually pull it the zone will come over but it
> > expires after the TTL hits.
> >
> > # named-xfer -z fwdzone -f db.fwdzone x.x.x.x
> >
> > Anybody run into problems with BIND actiig as a slave for a CNR master?
> > Under 8.2.2P5 all was
> > well.
> >
> > Thanks
> >
> > Mike Wilkinson
>
> named-xfer no longer exists as a standalone executable in BIND 9 so the
> named-xfer you are using is left over from your old BIND 8 installation
> which as you say was working fine anyway so doing a manual transfer
> using named-xfer is not using the BIND9 code.
>
> BIND 9 includes a named-checkzone script so if it is possible to ftp a
> copy of the zone file from the master and run the script against it it
> will probably tell you what it doesn't like in the zone file. Common
> gotchas in BIND 9 are:
> 1)$TTL must exist as first line of zone file or on the first RR.
> 2)No periods allowed in serial numbers
> 3) RRs can't stradle line breaks so the SOA must have the opening
> paranthesis on the first line of the SOA record.

None of these should affect a slave, since it does not transfer the original
file.  If the master will load the zone, it should be fine.

Can you use dig to do a zone transfer?