We are running a caching name server on the server, bound to the loopback interface only. Public nameservers use BuddyNS, for which are using:

;; NS (nameserver) records -- we're using BuddyNS as our primary DNS provider
@           IN  NS      b.ns.buddyns.com.
@           IN  NS      c.ns.buddyns.com.
@           IN  NS      d.ns.buddyns.com.
@           IN  NS      e.ns.buddyns.com.

Zone files are maintained in:

/etc/named/zones

If a change is made, be sure to update the serial in the zone file, then "rndc" to reload named.

We decided to use BIND 9, as it is well supported now. (Note that the default in Debian 5.0 is BIND 8, if you just say "bind".) We also decided to put it into a chroot jail, as it's pretty simple to do and well-documented. This will protect us from most BIND and DNS exploits.

Note that we do not cover in this document the DNS services that maintain the SLUUG.ORG domain name. The domain name info is documented on a separate page.

Everything from here to the end probably only applies to the old bud server and doesn't apply now that bud had been replaced by by amber running a different Linux configuration.

First, install the required packages:

apt-get install -y bind9 bind9-host dnsutils bind9-doc

Debian automatically starts the daemon, but we're going to change a lot of its config, so we should stop the daemon until we're done:

/etc/init.d/bind9 stop

Next build out /var/lib/named to contain enough so that bind9 can run chrooted within it:

mkdir -p /var/lib/named
mkdir -p /var/lib/named/etc /var/lib/named/dev
mkdir -p /var/lib/named/var/run/bind/run /var/lib/named/var/cache/bind
chown -R bind:bind /var/lib/named/var/*
mknod /var/lib/named/dev/random c 1 8
mknod /var/lib/named/dev/null c 1 3
chmod 666 /var/lib/named/dev/null /var/lib/named/dev/random
ln -sf /var/lib/named/var/run/bind /var/run/bind
ln -sf /var/lib/named/var/cache/bind /var/cache/bind

Copy the configuration into the chroot directory, and link back to the original locations, so we can update the configuration from the original config-file location:

mv /etc/bind /etc/bind.dist
cp -a /etc/bind.dist /var/lib/named/etc/bind
ln -s /var/lib/named/etc/bind /etc/bind

Next edit /etc/default/bind9 to tell it to start up chrooted to /var/lib/named:

sed -i -e 's:OPTIONS="-u bind":OPTIONS="-u bind -t /var/lib/named":' /etc/default/bind9

Edit /var/lib/named/etc/bind/named.conf.options and tell it which interfaces to listen on, and who to forward requests to if we don't have the answer cached. We also include a few backup forwarders commented out, in case we decide to use them at a later date.

options {
	directory "/var/cache/bind";
	listen-on {127.0.0.1;}; # only act as a DNS cache for localhost
	forwarders {205.242.92.2; 205.242.176.103;}; # ns1.primary.net, ns2.primary.net
	#forwarders {4.2.2.1; 4.2.2.2; 4.2.2.3; 4.2.2.4; 4.2.2.5; 4.2.2.6;}; # Verizon public DNS servers
	#forwarders {208.67.220.220; 208.67.222.222;}; # OpenDNS public DNS servers
	auth-nxdomain no; # conform to RFC1035
};

To get logging out of the chroot jail, we need to set up a socket within the jail, and have the syslog daemon listen to it. We configure syslog by specifying the name of the socket in a '-a' option. This is set in the SYSLOGD parameter in the /etc/init.d/sysklogd file:

sed -i -e 's:^SYSLOGD=""$:SYSLOGD="-a /var/lib/named/dev/log":' /etc/default/syslogd

Then restart the logging daemon:

/etc/init.d/sysklogd restart

Start the named server:

/etc/init.d/bind9 start

If startup fails, tail the /var/log/syslog file to look for errors. The most likely error is forgetting a semi-colon somewhere in the config file.

Edit /etc/resolv.conf to tell clients to use localhost to resolve DNS names. Again, we include a few other servers just as documentation.

domain sluug.org
nameserver 127.0.0.1
#nameserver 205.242.92.2 # ns1.primary.net
#nameserver 205.242.176.103 # ns2.primary.net
#nameserver 208.67.220.220 # OpenDNS public DNS server
#nameserver 208.67.222 .222 # OpenDNS public DNS server

We also need to delete any dns-* lines from /etc/network/interfaces, as they cause /etc/resolv.conf to be updated when the interface comes up.

sed -i -e 's/^.*dns-.*//' /etc/network/interfaces

Run nslookup and/or dig to resolve some DNS names. Make sure you get answers back from 127.0.0.1.

Run some client programs to make sure they are resolving host names properly.

Check /var/log/daemon.log and /var/log/syslog for startup/shutdown info from the bind9 daemon.

Run rndc status to check the status of the server.

Run rndc stats and then read /var/lib/named/var/cache/bind/named.stats to get server stats, including number of successful and failed DNS lookups.

These settings are for our hosting at Primary Networks. Our forwarders will need to be changed if we change hosting/ISPs.

The OpenDNS servers are publicly available for anyone to use. It probably doesn't make sense to use them on a server though, because they send unknown addresses to their own servers. Their servers contain search pages for web access; I'm not sure what happens with other services.

The 4.2.2.x addresses are supposedly Verizon's publicly-available DNS server that anyone can use.

If we move the servers, we need to change the forwarders in /var/lib/named/etc/bind/named.conf.options to the upstream ISP's DNS servers, or use some of the public DNS servers.

Much of this is based on the Bind-Chroot-Howto for Debian.