Linux System Administration

68

|

Chapter 3: The Domain Name System

This type of error can also happen as a result of running BIND in a chroot environ-

ment, when one or more files are missing from the isolation directory. You can check

some of the essential files to be sure they are in the proper locations:

# ls -l /var/lib/named/etc/bind/named.conf

-rw-r--r-- 1 root bind 1611 2006-09-07 12:21 /var/lib/named/etc/bind/named.conf

# ls /var/lib/named/etc/bind/

db.0 db.local named.conf.local pri.centralsoft.org

db.127 db.root named.conf.options pri.opensourcetoday.org

db.255 named.conf pri.156.18.67.in-addr.arpa rndc.key

db.empty named.conf~ pri.156.18.67.in-addr.arpa~ zones.rfc1918

#

...

If these files do not exist, the chroot environment is not set up properly or com-

pletely. Return to the section “Using a chroot Environment for Security” near the

beginning of this chapter and follow its instructions carefully to ensure every file is in

place.

Once you’ve fixed the problem, you’ll need to stop and restart named to enable rndc

to reach the running server. Use the killall command sequence described in the previ-

ous section:

server1:~# killall -TERM /usr/sbin/named

server1:~# killall -KILL /usr/sbin/named

/usr/sbin/named: no process killed

server1:~# /etc/init.d/bind9 start

Starting domain name service: named.

server1:~#

Next, check your /var/log/syslog file to see whether the zone files loaded. You should

see something like this:

starting ...