Bind 8 -> 9+ errors ?
sixx
xyz at neoscene.com
Fri Nov 10 05:28:22 UTC 2000
Well some things i found out through testing ...
At 06:09 PM 11/9/00 -0800, you wrote:
>I had the exact same problem, except I rebuilt my server (for several other
>reasons). I don't have a copy of my error log, but it looked exactly like
>yours. I'll be happy to see what your results are!!.
>
>Sabrina
>
>
> >From: sixx <xyz at neoscene.com>
> >To: comp-protocols-dns-bind at moderators.isc.org
> >Subject: Bind 8 -> 9+ errors ?
> >Date: Thu, 09 Nov 2000 10:41:30 +0800
> >
> >
> >I've been trying to convert my bind8 servers and new installations to bind9
> >but i'm encountering the following errors on intel linux 2.4.0test10
> >kernel.
> >
> >Has anybody come up with any solutions or meet similar problems ?
> >
> >Nov 9 09:35:18 rose named[9231]: starting BIND 9.0.1rc2
> >Nov 9 09:35:18 rose named[9233]: loading configuration from
> >'/etc/named.conf'
> >Nov 9 09:35:18 rose named[9233]: the default for the 'auth-nxdomain'
> >option is now 'no'
> >Nov 9 09:35:18 rose modprobe: modprobe: Can't locate module net-pf-10
> >Nov 9 09:35:18 rose named[9233]: no IPv6 interfaces found
set "alias net-pf-10 off" in /etc/conf.modules
> >Nov 9 09:35:18 rose named[9233]: listening on IPv4 interface lo,
> >127.0.0.1#53
> >Nov 9 09:35:18 rose named[9233]: listening on IPv4 interface eth0:0,
> >192.168.0.254#53
> >Nov 9 09:35:18 rose named[9233]: dns_master_load: primary/rev.192.168.0:1:
> >no TTL specified
> >Nov 9 09:35:18 rose named[9233]: dns_zone_load: zone
> >0.168.192.in-addr.arpa/IN: database primary/rev.192.168.0: dns_db_load
> >failed: no ttl
> >Nov 9 09:35:18 rose named[9233]: loading zones: no ttl
> >Nov 9 09:35:18 rose named[9233]: exiting (due to fatal error)
> >
> >rev.192.168.0
> >
> >@ 28800 IN SOA ns.example.com. postmaster.example.com. (
> > 2000110101 ; Serial
> > 28800 ; Refresh 8 hours
> > 7200 ; Retry 2 hours
> > 604800 ; Expire 7 days
> > 28800 ) ; Minimum TTL 8 hours
> > IN NS ns.example.com.
> > IN NS ns2.example.com.
> >
> >I've tried setting the TTL for SOA to either 0 or more but i still get the
> >same message.
> >
seems like for bind 9 is very particular about a TTL value before the SOA
records.
so something like "@ 3600 IN SOA ..." would work fine. Or another way is a
"$TTL 3600"
in the first line of your files. Note this has to be done for *ALL* zones
that you have
referenced in your named.conf =( sort of a hassle.
Another point is that a lot of functions are not implemented in bind 9 yet
(eg. logging
of queries, ndc and such)
I wonder if the recent DOS from xfer'ing slaves to masters would be
correctly rectified soon ...
More information about the bind-users
mailing list