Create a dynamic dns service with BIND


Some of us use an dynamic DNS service to access our home server or desktop, but with cheap INTERNET access prices you may think in having your own server (these days you can have 100/10MB fiber connection for 50€), that if you don’t like the cloud concept.

But what if you don’t want do use the domain name that your dynamic DNS provider give you (something like, lets say that you want You have two options:

1 – You buy/rent a custom DNS Service (arround 40€/year/domain)

2 – You build your costom DNS server with BIND or TINYDNS.

I’ll write about the second one, but before we start let me tell you the implications this setup has.

1.º – Need root access to the server.

2.º – If you have a lot of domains you’ll have a big DNS overhead, this because you’ll have to have a small TTL (time to live) on the DNS requests you server, this to keep other DNS servers and clients updated.

3.º – Need to have cron running.

4.º – Need to have BIND installed.

5.º- Need a Dynamic DNS service provider for your initial domain, don’t forget to check use wildcard option, check, install and configure the client (following the documentation) and test it afterwards.

6.º- Need direct access to port UDP port 53, check your firewall, and hosts.deny file if you use TCP Wrappers.

7.º – This post is not about DNS security, you should read about it and hard your setup furthermore.

I’m going to set this up in a OpenSuSE 11.0 (X86_64) , but this setup should be suitable for other distros with small changes.

Imagine that you have bough the domain

Lets start by BIND configuration and to do this lets edit /etc/named.conf and add the following lines

zone “” in {
file “master/”;
type master;
allow-transfer { any; };

now lets create and initial setup file by creating the file:


and add some initial content:

$TTL 60

@               IN SOA (

1249459201      ; serial

10800           ; refresh

3600            ; retry

604800          ; expiry

86400 )         ; minimum         IN MX           10         IN NS           ns1         IN NS           ns2

mail                    IN A  

ns1                     IN A  

ns2                     IN A  

*       IN A  

After this just reload named:

/etc/init.d/named reload

and test it:

dig @your_dns_server_IP

you should get something like:

; <<>> DiG 9.6.1 <<>> @localhost
; (2 servers found)
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 55310
;; flags: qr aa rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 2, ADDITIONAL: 2

;            IN      A

;; ANSWER SECTION:     60      IN      A

starwars-xpto.         60      IN      NS
starwars-xpto.         60      IN      NS

;; ADDITIONAL SECTION:     60      IN      A     60      IN      A

;; Query time: 0 msec
;; WHEN: Mon Aug 17 18:02:31 2009
;; MSG SIZE  rcvd: 120

If you got something like that it’s great, you have your BIND configuration working, if not, check the logs. Bind is really picky with the syntax, and by the way you can’t use # as comment in the config file just the ;

Now lets go to the interesting part, creating the script that will change your ip address on bind configuration whenever it changes. Just create a script wherever you want, my custom system scripts are usually in /root/bin so I’ll keep using it.

Lets create and edit the file /root/bin/update_dns not forgetting to change

just copy/past the content bellow:


BIND_DIR=’/var/lib/named/master’ ;



if [ $# -ne 1 ] ; then

echo Usage: update_dns domain ;

echo EX: update_dns;

exit ;


function get_ip ()


echo `dig | grep | grep -v ‘;\|CNAME’ | awk ‘{print $5}’` ;


function update_dns ()


DATA_SEGUNDOS=`date +’%s’`;





if [ “$IP_” != “`grep ‘IN.*A’ /var/lib/named/master/$BIND_FILE | grep -v SOA | awk ‘{print $4}’ | uniq`” ] ; then

cat $BIND_WORK_DIR_/$BIND_WORK_FILE_ | sed s/'[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}’/$IP_/ > /tmp/$DOMAIN_ ;

cat /tmp/$DOMAIN_ | sed s/'[0-9]\{10\}’/$DATA_SEGUNDOS/g > /tmp/$DOMAIN_.bind ;

cat /tmp/$DOMAIN_.bind


mv -f /tmp/$DOMAIN_.bind $BIND_WORK_DIR_/$BIND_WORK_FILE_ ;

rm /tmp/$DOMAIN_* ;

echo `date +’%b %d %H:%m:%S’` “Domain: $DOMAIN_ updated to IP: $IP_” >> /var/log/messages ;

chmod -R 755 /var/lib/named/master


echo “No need for update” ;



function restart_dns_server ()


if [ ! -f /tmp/restarting_named ] ; then

touch /tmp/restarting_named;

/etc/init.d/named stop;

sleep 3 ;

pkill -9 named ;

sleep 1 ;

/etc/init.d/named restart ;

rm /tmp/restarting_named


sleep 10 ;

restart_dns_server ;



IP=`get_ip` ;

if [ “$IP” != “`grep ‘[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}\.[0-9]\{1,3\}’ $BIND_DIR/$BIND_FILE | awk ‘{print $4}’ | uniq`” ] ; then


restart_dns_server ;


Finally you just need to setup cron, just type crontab -e

and add an entrace like

*/5 * * * * root /root/bin/updatedns

Just wait the 5 minutes and check if the ip changed with the dig command as wrote above.

After all this and as the last step of configuration go to the domain provider were you bought your domain (ex: and configure it to use as name server the and

and your done.

This isn’t a easy setup to do but if you have multiple custom domains it can save a few € every year, I know in my case it does.


Pedro Oliveira

Why use a redundant swap space or partition

Today I’ve had a problem in one of the servers we support, no web access, no ssh, and no console just a bunch of sentences passing so fast I couldn’t read it on the terminal. The solution a simple hard reset and the system came online, it was a hard disk failure but the system online without trouble because we were using a raid configuration. One of the disks didn’t show up in the RAID array, a few tests later and declared the hardware fault the cause of the downtime.

But why did the system came down because of a disk failure if there was a RAID system available, simple the swap was spread among the disks but not in a raid system so no redundant swap partitions, when the need for data in the swap of that file system came there wasn’t any data available and the system came to a stop.

From now on we’ll create a redundant swap partition using a RAID volume so this doesn’t happen again as a server should never stop because of a disk problem,  living and learning.


Pedro M. S. Oliveira

BTW – to reassemble the array I used mdadm, bellow there is a simple usage if you want to reassemble a previous build array:

mdadm –manage  /dev/md0 –add /dev/sda1

this command will add the partition /dev/sda1 to the raid array /dev/md0

if you want to learn more about RAID in linux just type man mdadm or mdadm –help

QOS – Service with quality using MasterShaper

QOS – “Quality of service” – what a fancy name for somethings that will allow you to manage the available bandwidth for your servers/pcs/equipments. Usually setting up QOS is not “so fancy” to setup and manage. At work I had to setup a QOS for a custumer and didn’t want to be writing rules on my own so I searched a bit and found MasterShaper.

You can read more about it here.

Lets imagine that you want to allow p2p on your network but you don’t want to waste all your bandwidth on that kind of traffic, after all you still need bandwidth for your VOIP, http, and mail traffic.

To setup this I’ll be using CentOS and Mastershaper.

After installing a VMware virtual machine with CentOS (default install with 2 network cards), I’ve setup apache, this time I didn’t use virtualhosts or any other extras as I just wanted to create a single purpose system.

To start I’ve concufigured 2 network interfaces (internal/external)

eth0 (external)

eth1 (internal)

After downloading MasterShaper decompress it to:


Install pear and pear modules as described in MasterShapper docs.

And finally create the mysql database to support the installation.

I won’t get into install details (just follow the documentation), but in the end you be able to define schedulers, hosts and services with higher or lower priority and in the end you’ll also be able to monitor all this in a pretty web interface with some graphics.

Just don’t forget to install mysql support in php, and pear, you’ll also have to configure sudo but in the end just read the documentation in mastershaper it’s very good.