Bind99 and a slave named server

[LuKreme]

On 20 Aug 2013, at 14:38 , Alan Clegg <alan at clegg.com> wrote:
> To convert master to slave:

[snip]

> Bazinga!

OK. Not Bazinga.

$ grep covisp named.conf
zone "covisp.net" { type slave; file "slave/covisp.net"; masters { 75.148.117.92; }; };
$ rndc status
version: 9.9.3-P2
CPUs found: 2
worker threads: 2
UDP listeners per interface: 2
number of zones: 117
debug level: 0
xfers running: 0
xfers deferred: 0
soa queries in progress: 5
query logging is OFF
recursive clients: 0/0/1000
tcp clients: 0/100
server is up and running
$ grep listen named.conf
        listen-on { 75.148.117.93; 75.148.117.91; 127.0.0.1; };
$ dig @localhost covisp.net | grep -A2 ";; ANS" | tail -2
$ dig @75.148.117.91 covisp.net | grep -A2 ";; ANS" | tail -2
$ dig @ns1.covisp.net covisp.net |grep -A2 ";; ANS" |tail -2
covisp.net.             86400   IN      A       75.148.117.93
covisp.net.             86400   IN      A       75.148.117.90

in /var/log/messages:
Aug 20 20:40:23 mail named[81006]: the working directory is not writable<1>
Aug 20 20:40:23 mail named[81006]: all zones loaded
Aug 20 20:40:23 mail named[81006]: running

Oh, and slave/ is empty.

$ grep covisp named.conf-master 
zone "covisp.net" { type master; file "master/covisp.net";  };
$ diff /var/named/etc/namedb/master/covisp.net /var/named/etc/namedb/slave/covisp.net
$ cp /var/named/etc/namedb/named.conf-master /var/named/etc/namedb/named.conf
$ rndc reload
$ dig @75.148.117.91 covisp.net | grep -A2 ";; ANS" | tail -2
covisp.net.             86400   IN      A       75.148.117.93
covisp.net.             86400   IN      A       75.148.117.90

<1> (the working directory is not writeable comes up every time because /var/named/etc/namedb is owned by root and changing it causes bind to first change it back, and then log the error anyway).


-- 
LOOSE TEETH DON'T NEED MY HELP Bart chalkboard Ep. AABF16