As with the earlier articles, there are a few basic parts that rules can be filtered for. These four parts are as follows:

- IP Addresses: a single address (94.76.240.128), or multiple addresses in a range (94.76.240.128-150)

- Interface: eth1,ppp0,wlan0 (these are usually names of physical network ports on your computer)

- Ports: a single port (http 80) or a range of ports

- Protocols: icmp, tcp, udp, all

(for a detailed description, see my first article)

In the previous article I used the loopback/localhost address (127.0.0.1) for many examples (iptables –A INPUT –i lo –j ACCEPT). To expand on that idea, I will use two. Imagine you have two physical interfaces on your computer. eth1 connects to the local network and ppp0 to the ISP for internet access.

We could set a rule: iptables –A INPUT –i lo –j ACCEPT iptables –A INPUT –i eth1 –j ACCEPT . You wouldn’t want to allow packets on the ISP interface (ppp0). Doing so would cause internet traffic through effectively making the firewall pointless.

In the real world, opening up an interface fully may not be secure enough. You will likely want a bit more control over what users can access but deny others. For example lets say on the local LAN, it uses the private network of: 192.168.1.x. You have someone in the IT department that you want to allow to with an IP address of 192.168.1.10. You would append a rule: iptables –A INPUT –s 192.168.1.10 –j ACCEPT

To explain this rule: As previously mentioned –A means append to the Input table. The –s flag means source (the IP address allowed through). –j (jump to accept).

Next while you could type in single lines for each individual IP address you want to allow. If you have a lot of computers on the network, manually doing so is very tedious. A much quicker method is to type an address range instead. To do that you can use a subnet mask or use slash notation. For a working example say we wanted to allow the entire 192.168.1.0 network range, which goes from 1-254

Network mask: iptables –A INPUT –s 192.168.1.0/255.255.255.0 –j ACCEPT

Slash notation: iptables –A INPUT –s 192.168.1.0/24 –j ACCEPT

Apart from uses specific IP addresses, another method is to filter by the MAC (media access control) address. Each network interface in the world has it’s own address that is burned into the card. No two cards in the world has the same address. To do filtering via MAC address, you load the MAC module. In an earlier article if you read through them in order. You might recall that we used module “state” to filter packets based on Established and Related connections. Using Mac filtering is pretty much the same.

Example of MAC filter rule: iptables –A INPUT –s 192.168.1.10 –m mac –mac-source 00:50:56:c0:00:08 –j ACCEPT

Flag breakdown: -m mac means to load the MAC module. The –mac-source specifics the MAC addresses of the interface. Which in these examples has the IP address of: 192.168.1.10

To find out the MAC address of an interface varies depending on the OS you use. But in general you go to a command line a type in: Windows (ipconfig /all), Linux/OSX (ifconfig) for wireless interfaces you may have to use (iwconfig)

You might be wondering what purpose using MAC filtering is over IP addresses. If you filter with MAC as well this can help prevent people from trying to gain access by pretending to be 192.168.1.10 . I will say that there are ways for hackers to spoof MAC addresses as well. That topic is beyond the scoop of this article.

I will end the discussion here and cover the remaining content in the next and last article on the basics of IPtables. I would encourage experimenting with these concepts on your local home network. As well as doing further research on anything you might be unfamiliar with. With practicing you can clear out all existing rules and then type everything in by hand. Which is very useful for memorizing the various flags and options well. If you have been reading in order. You can add these new lines to the bash script & experiment that way as well.

Until next time, I hope you have found this article interesting.

I highly recommend that these be practiced on a home machine where you have admin rights and aren’t using IPtables as your active firewall. Since deleting the configuration will leave the computer vulnerable to attack. Under no circumstances should this be practiced on a live production network (i.e at work). Myself nor GBServers will be held responsible for any possible damages by ignoring these suggestions. I’d go as far as suggesting: unplugging your computer from the internet. You can practice with computers on your home network. You don’t necessarily need to be connected to the internet.

Also be aware that while Linux distros are more or less the same. There can be some differences in how services are started such as IPtables. But if you have read the first article, you hopefully have gotten IPtables installed as well as learned how to turn it on and off. With that said, lets get started..

The first thing you’ll want to do is to verify whether the firewall is installed or not. Typically the easiest way to check this is to open up a console/terminal window and at the prompt type:  iptables –help (or iptables –h). If it is indeed installed you should see output similar to the following:

Note: The # symbol represents the command prompt. It shouldn’t be typed in, unless otherwise stated.

# iptables –help

iptables v1.2.9  Usage: iptables -[AD] chain rule-specification [options]        iptables -[RI] chain rulenum rule-specification [options]        iptables -D chain rulenum [options]        iptables -[LFZ] [chain] [options]        iptables -[NX] chain        iptables -E old-chain-name new-chain-name        iptables -P chain target [options]    <…>

After it’s verified installed, a useful option is the –L switch. Which will list the current entries (if any):

# iptables -L

Chain INPUT (policy DROP) target     prot opt source               destination ACCEPT     all  –  anywhere             anywhere ACCEPT     all  –  anywhere             anywhere            state RELATED,ESTABLISHED ACCEPT     all  –  neo               anywhere            MAC 00:50:56:c0:00:08 DROP       all  –  smith                anywhere ACCEPT     all  –  trinity                 anywhere            MAC 00:50:56:c0:00:01 ACCEPT     tcp  –  anywhere             anywhere            tcp dpt:ftp ACCEPT     tcp  –  ict                 anywhere            state NEW tcp dpt:ssh  Chain FORWARD (policy DROP) target     prot opt source               destination  Chain OUTPUT (policy ACCEPT) target     prot opt source               destination

If for some reason iptables doesn’t appear to be running. You can check to see if the module is enabled. By issuing the following command: # lsmod |grep ip ipt_mac . You should get some form of output such as:

2691  1 ip_tables      28644  3 ipt_mac,ipt_state,iptable_filter

Hopefully you have IPtables installed and verified it’s active. For this initial example we will start by writing a basic rule set, which will act as a useful stateful packet inspection firewall.

At a prompt type in:

# iptables -F iptables -P INPUT DROP iptables -P FORWARD DROP iptables -P OUTPUT ACCEPT iptables -A INPUT -i lo -j ACCEPT iptables -A INPUT -m state –state ESTABLISHED,RELATED -j ACCEPT

You can view the newly created rules as mentioned with: iptables –L . Which should output:

Chain INPUT (policy DROP 0 packets, 0 bytes)  pkts bytes target     prot opt in     out     source               destination     0     0 ACCEPT     all  –  lo     any     anywhere             anywhere     0     0 ACCEPT     all  –  any    any     anywhere             anywhere            state RELATED,ESTABLISHED  Chain FORWARD (policy DROP 0 packets, 0 bytes)  pkts bytes target     prot opt in     out     source               destination  Chain OUTPUT (policy ACCEPT 0 packets, 0 bytes)  pkts bytes target     prot opt in     out     source               destination

That wasn’t too difficult was it . We now have a very simple SPI firewall. The rules you typed out will permit all outgoing traffic but also block (or deny) all incoming traffic, along with forwarded traffic.

To help understand the commands entered. I will briefly explain them:

- [iptables –F]: This command with the –F switch. Means to erase all existing rules. So that you start from scratch & ensures no current rules will cause any conflicts.

- [iptables –P Input Drop]: Using the –P sets a default rule. So we have set the Input table to Drop packets. In English this means that if a packet does not match up with any other rule in the list. Then the packet will get dropped or ignored.

- [iptables –P Forward Drop]: As above we’ve set a default rule. That any network packets being forwarded should be automatically ignored. Since we haven’t setup the system to act as a router. There’s no need to forward any traffic on.

- [iptables –P Output Accept]: Again a default rule. This simply means that any network traffic leaving our computer should automatically be accepted or allowed to pass. Since normally we know the software we run to be safe.

Now that we have set our defaults. The last few lines define slightly more specific filtering.

- [iptables –A Input –i lo –j Accept]: The –A is append (add) the rule to a certain chain. In this example it means to append to Input. The –I refers to interface (i.e network card). Or any traffic going to the network card specified. Which here is lo, or commonly known as the loopback/127.0.0.1. The –j means to jump to the action (i.e Accept). In plain English this line will allow any traffic going to the loopback (or 127.0.0.1)

- (iptables –A Input –m state –state Established,Related –j Accept): This line is really the meat and potatoes of the firewall. As before we are –A (appending) to Input. The –m means to load a module (i.e state). In this example we are filtering against network connections that are already existing (Established or Related). If the connection didn’t originate from the system itself (New) they will be blocked. If you were to add in “NEW” to the line as well. That would allow incoming traffic not from your computer to pass through. Which is a security risk hence it’s not included in the rule.

The final thing to do will be to save the rules we just entered. So that when the computer is rebooted, you don’t have to retype them in. Which on my system is the following:

# /sbin/service iptables save

Note: Linux versions do vary slightly. You may have to look up the correct path for Iptables on your particular Linux distro.

Finally what I typically do is create a simple bash script. That allows me to run the firewall by typing in a simple command. To do that you’ll want to open an editor. On Linux usually vi or nano is typically installed (I personally like nano).

Then you can paste the script:

#!/bin/bash

#

#

# iptables SPI firewall

#

# Flush all current rules from iptables

#

iptables -F

#

# Set defaults (INPUT, FORWARD, OUTPUT)

iptables -P INPUT DROP

iptables -P FORWARD DROP

iptables -P OUTPUT ACCEPT

#

# Permit access for loopback (127.0.0.1)

#

iptables -A INPUT -i lo -j ACCEPT

#

# Accept packets for established, related connections

#

iptables -A INPUT -m state –state ESTABLISHED,RELATED -j ACCEPT

The lines that start with # refers to a comment. So you can make your own personal notes. Apart from the first line: ” #!/bin/bash” which tells the computer it’s a bash file.

Next you’ll want to save the file for example: myfirstfirewall.sh . Once out of the editor make sure the script is executable by typing: chmod +x myfirstfirewall.sh (or whatever name you chose)

Finally run the script. By going to a command prompt and typing: ./myfirstfirewall.sh

I hope you have enjoyed this entry and will come back to read the next one.

Many people may not be aware but Linux and many unix like operating systems have a firewall built into it, which is most commonly known as iptables. Unlike most consumer-based firewalls, iptables is CLI or command line interface based. Although there are a variety of point and click graphical interfaces available. Most of the time they hide most of the functionality of the CLI. So in my opinion it’s best to learn how to operate iptables via the command line rather than rely on a GUI. At first it may seem difficult but with practice, it will become much easier.

Before I start; I’d like to note that while writing this article. It ended up quite long. I will break it down into multiple blog entries over the course of the month. So the majority of content can be kept. As well as you, the reader hopefully won’t get overwhelmed.

This initial article will cover basic background information. Which will be important for understanding the future material.

Basics:

Iptables use three main concepts: ip addresses, different protocols [tcp/udp] as well as port numbers.  For those unfamiliar with the terms I will give a general description:

IP Address: this is a code made up of numbers separated by three dots that identifies a particular computer on the Internet. Which consist of four sets of numbers from 0 to 255, separated by three dots. For example “66.72.98.236″

Protocol: When computers communicate with each other, there needs to be a common set of rules and instructions that each computer follows. A specific set of communication rules is called a protocol. In plain English: you can think of them as different languages. Two computers must be able to speak the same “language” to be able to communicate properly.

Port Numbers: This is a number that indicates what kind of protocol a server on the Internet is using. For example, Web servers typically are listed on port 80. Web browsers use this port by default when accessing Web pages, but you can also specify what port you would like to use in the URL like this: http://www.gbservers.co.uk:80

(Definition excerpts from: techterms.com)

IPtables consists of three main tables or functions. Which all the traffic flowing in/out of the network card is checked against and the appropriate response is taken. Depending on what rules are listed in the tables. The three primary tables are as follows along with a basic description.

Forward: This is used when you want the packets to simply pass through. Without any direct action being taken i.e the host is acting as a router.

Input: This means any network traffic destined for the particular host.

Output: The opposite of input. This is used for any traffic that originated from the host.

The entry that gets the majority of use is typically Input. Since creating a rule using Input can help secure your system and keep prying eyes out. Firewall rules are created using the above three tables. As well as two main actions, which are: Accept or Drop. Accept meaning the packet is allowed in, Drop means it gets discarded or thrown away. Each packet is screened against the list of rules. If a packet gets to the end, without a match. Then the default rule(s) will be applied. The most common is usually: deny all incoming traffic.

There are two ways you can set up the filtering. One way is to set up a default rule to allow all traffic. Then set specific rules to deny certain traffic. The second method (more common) is to set specific rules to allow only certain traffic through. All other traffic not meeting the list will be thrown away. Typically the first method would use Output to filter traffic, we normally trust traffic leaving our computers. Input would be used for traffic coming into the machine, since we wouldn’t trust it quite as readily to be safe.

I will end this first article here. If you wish you can research on how to install IPTables on your computer/server. Also you can practice starting and stopping the firewall. For the second article, we will start to get some hands-on practice with the basic commands and rule sets that were discussed in this article. The next article should be more enjoyable rather than just theory. But understanding the basics is very important. Until then I hope this has been informative.

Netstat

The main purpose of netstat is to give a quick general overview of the connections the system is using. Plus information such as, the protocol typically TCP/UDP or the current status state.

Open a DOS prompt {start-run-cmd}. Then type “netstat” you’ll see output like the following

You can further specify the output by using different flags. Filtering by either TCP or UDP are common. The syntax may differ slightly depending on the operating system. For example on my OS X system it’s: netstat –p tcp

The various connection states are:

ESTABLISHED – Both hosts are connected.

CLOSING – The remote host has agreed to close its connection.

LISTENING – Your computer is waiting to handle an incoming connection.

SYN_RCVD – A remote host has asked for you to start a connection.

SYN_SENT – Your computer has accepted to start a connection.

LAST_ACK – Your computer needs to obliterate the packets before closing the connection.

TIMED_WAIT – See above.

CLOSE_WAIT – The remote host is closing its connection with your computer.

FIN_WAIT 1 – A client is closing its connection.

FIN_WAIT 2 – Both hosts have agreed to close the connection.

To view the complete list of flags supported by netstat. On Windows type: “netstat /?”. Under Linux/OSX you can use “man netstat” which gives a detailed description of all the options.

It is also very useful for checking for any possible undesired programs or viruses that you might suspect. Typically they will try to establish out going connections.

TOP command

The TOP command is native to unix type operating systems such as Linux and OS X. It shows all the processes currently running

To sort by memory usage: press M

You can filter processes for a specific user by using the –u flag. i.e: $ top –u root

One of my most used flags is the –k (or kill). This is very useful for terminating a program or process that has locked up or otherwise won’t close by normal means.

Once you’ve located a process that needs to be killed, press ‘k’ which will ask for the process id, and signal to send.  Note: some kill processes by not be successful, unless you have proper permission to. Such as under the root/super user account

PID to kill: 1309 Kill PID 1309 with signal [15]: PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND  1309 geek   23   0 2483m 1.7g  27m S    0 21.8  45:31.32 gagent  1882 geek   25   0 2485m 1.7g  26m S    0 21.7  22:38.97 gagent  5136 root    16   0 38040  14m 9836 S    0  0.2   0:00.39 nautilus

There are many flags that can be used under TOP. Which can be fully covered in-depth with a single article. I would encourage you to look through the manual “man top” or with a Google search.

Until next time I hope you may find this article informative.

Note: there are various versions with multi functions. For most people puTTY.exe (first one listed) is the one you want.

2)   Once the file has been download, double click the EXE file. PuTTY will automatically launch the main configuration (no installation is required)

3)   Enter the hostname or IP address of your account. Port 22 is the default, so leave it as-is. Unless your host provider specifically states a different port number. “Connection Type” defaults to SSH which is what we want for this tutorial.

4)   The basic information is now complete. You can now press “Open” button.

Optionally: You can enter a name under “saved sessions” and click save. So you don’t have to keep typing the information each time.

A dialogue will open & connect to the server. If successful you should see a prompt like the following. From here you can navigate through & make changes to your account.

In the following section I will outline and describe some basic unix commands for navigating and viewing files.

The ls command is used to view files in the directory you are currently in:

The –A flag for the ls command is particularly useful for viewing system files or folders. For example I created a file and marked it as hidden as seen below (last file in the list)

To change directories you use the cd command. The unix file system is somewhat different to that of Windows. The top most directory is known as root or represented by a /. To get to root you type:  cd / , then you could issue an ls to see all the content. To move back up one directory use: cd ..

Apart from the root directory. Some other good ones to know are home or /home. Which by default on linux will hold all the files for each user account you may have. The full path would be: /home/Richard. If I wanted to get to that directory I could type either: cd /home , then cd Richard. Once you know the folder structure typing: cd /home/Richard is quicker.

On OS X systems: home folders are located under: /Users as seen above

Next is /var . Which is typically where the various system logs are stored.

Passwd: This command is used to change the password of the current user. At a prompt type: passwd then the new password that you wish to use. You may be prompted to type the old password.

Cat: The cat command is very handy for viewing the content of files. You simply type: cat [filename]. Note: some files won’t view very well. But for configuration files, system logs or plain text. The cat command works extremely well.

If you wish to make configuration file changes, or write files. Most unix type OSes have a basic editor usually either vi or nano. These are much like the old editor in DOS. You simply type: nano script.sh for example. In some cases only the root user is allowed to save changes especially to system files. In which case you’d use the sudo command i.e nano script.sh. Which will prompt you for the root password. This will allow you to act as the super user (root). Even if you are logged in to a standard account.

The unix file structure may seem confusing at first. Especially coming from Windows. If you take time to browse around using the LS command. You should find after a week or so. The structure isn’t so foreign & is quite logical.

For more information, there are many good sites dedicated to unix commands and navigating around. One is my favorites is: http://linuxcommand.org/index.php

If you have ever built or upgraded a personal computer. Then building a rack mounted server is very similar with some minor differences as seen in this article. One of the more important aspects being the hardware that you ultimately decide to use, which can affect the overall performance of the server. This post will show you the basic how-to of building a server from scratch. If you use a bare bones unit, much of the hardware should be included in the kit.

Hardware Installation

Processor

The processor is easily installed. Look at the bottom of the CPU. You’ll see at least one edge (possibly two) with pins missing. It will look like a diagonal. Which will match the socket for the cpu.  If it does not seem to fit, do not force it. You will bend or break pins. Double check that it is orientated correctly, so the pins line up.

(Picture taken from TomsHardware)

Once the CPU is sat correctly, you can close the load plate to secure it, just drop the load plate down and then push the arm into it’s lock position:

(Picture taken from TomsHardware)

Apply thermal paste to the top of the processor. Before attaching the heat sink. Thermal paste should be available at most electronic or computer stores.

Finally, attach the CPU cooler, typically with a clip or screws. For this example we are using a 1U heatsink, so it requires a backplate first:

And then flipping the motherboard over you’ll be able to screw the heatsink to the backplate for a secure fit:

PSU

If your rackmount case doesn’t already have the PSU installed, you will need to do this now. For our example we are using a 1U case, so this is a 1U PSU located at the back right. Your situation may vary, but it is all very standard.

Slot the PSU into the correct space to line up with the mounting screws. For our example the only screws to mount the PSU are on the back of the case into the front of the PSU. Some cases will have a piece of metal to screw to the back of the PSU near the cables and that then screws to mounting points on the case – This secures it very well.

Installing MotherBoard

Ensure that the case has the screw mounts in the correct places, if these are in the wrong place for the motherboard you are fitting then it could result in shorting the bottom of the motherboard. Once all the mounts are in the correct places you can line the motherboard up with each screw mount and screw it on.

You are now ready to connect the power cables to the motherboard.

First we start with the 20 or 24pin main power connection which you should be able to easily locate on your motherboard. Just take the power connector from the PSU and connect it to the correct slot on the motherboard:

Secondly, depending on your motherboard, you may need to plug in the 4 or 8 pin power connector. The slot to connect this to on your motherboard is usually near the CPU:

The last of the cables to connect to the motherboard are from the case itself. They usually consist of pins for the front power switch, reset switch, HDD L.E.D, power L.E.D and sometimes the network L.E.D and USB port.

You will need to check your motherboard manual for a guide to where each of the pins connect to your motherboard, some images are below for your general reference of what these connections look like.

Front Power Switch, Reset Switch and L.E.Ds:

Front USB Port:

RAM

RAM is easy to install, just make sure the two clips to each side of the module slot are released (down), line the RAM up with the module slot to ensure the notch out of the RAM meets with that of the module slot and push down at each corner of the RAM until the two clips at either side engage and secure the RAM in place.

In my opinion with servers, RAM shouldn’t be chosen solely for maximum speed, but should be selected for reliability. Modules with ECC support (Error Correction Code) are a good choice. Since it can compensate for single-bit errors.

The use of dual channel, should be given consideration when buying memory. Because two DIMMs are not any more expensive, than a single module that is twice as large. In my experience it offers significantly better performance.

Typically with RAM modules with ECC support. Smaller amounts of memory (2x 512 MB rather than 1x 1 GB) is often more economical. The downside is if you ever wish to upgrade the memory. Chances are you’ll replace the current RAM with two modules of higher capacity. Whereas using a single module, you’ll typically have slots free. There are pros/cons to both paths.

Hard Disks

To install the hard disks for our rackmount case, you are required to unscrew the mounts from the case, and then screw them to each side of the HDD.

Once this has been done you can then screw the mounts back into the case to secure each HDD.

Once the drives are slotted in the power and data cables are connected. You will need to connect the data cable to your HDD and then terminate the cable at your motherboard or raid controller.

Above: drives are all connected and ready for operation.

Your server basics are now complete. If you have a raid controller or other PCI cards to install, you should be able to consult your hardware manual for this.

Cable management is important, spend some time ensuring cables are as tidy as possible, and that they do not restrict too much airflow from any case fans.

Wrapping Up

The last process would be to power up the unit and make sure every thing seems to be running fine & POST check.

The final step would be to install the operating system of your choice. We have an image step by step tutorial of how to install CentOS, you can find that here.

This basic tutorial will show you how to set up a CentOS5.x (CentOS5.5).

Once you have completed this tutorial. You should hopefully have a basic up and running system. I would like to point out that this is not a concrete or the only method of doing it. You may find slight differences depending on your server specs or different versions of the various services.

1) Requirements

To install CentOS you will need to download the ISO image(s). Which can be found at: http://centos.org/modules/tinycontent/index.php?id=15

Once downloaded, you will want to burn the ISO file(s) to cd/dvd. The process of which is not covered in this article. If you do a Google search, there are some good step by step articles written by others.

Note: I’m using a made up hostname with private IP addresses and gateway. These are entirely for demonstration purposes. You should adjust them to suit your real world settings as required.

2) Install The Base System

Boot from the first CentOS CD (Disc 1). Press <ENTER> at the boot prompt:

If you wish you can run a check of the installation disc. But it can take a long time, so I always skip it. I run the test, if the install fails or suspect the disc didn’t burn correctly.

Next the CentOS installer splash screen appears. Click on Next:

Select your language,  then click next:

Select the keyboard layout:

I’m installing CentOS 5.0 on a new system, so I chose Yes to the question: “Would you like to initialize this drive, erasing ALL DATA?”.

Select the partitioning for the installation. For simplicity’s sake, I use Remove linux partitions on selected drives and use default layout. Which will create a small /boot and a large / partition as well as a swap partition. You should alter these to suit your needs.

“Are you sure you want to do this?” select Yes

For the network settings. The default is to configure the network interfaces using DHCP. But when installing a server, use of static IP addresses can be helpful. To alter settings Click on the Edit button. Again use of DHCP/static should be based on your needs.

For using static: In the window that comes up: uncheck “Use dynamic IP configuration (DHCP)” and “IPv6 support” and give your network card a static IP address. Also enter the netmask (e.g. 255.255.255.0) & other entries as required.

Choose your time zone:

Enter a password for the root user:

Next select the software to install. Uncheck everything but “Servers”. Also don’t check Packages from CentOS Extras. Select the “Customize now” option at the bottom, then click Next:

Select the package groups to install. In this example I chose: Editors, Text-based Internet, Development Libraries, Development Tools, DNS Name Server, FTP Server, Mail Server, MySQL Database, Server Configuration Tools, Web Server, Administration Tools, Base, and System Tools and click on Next:

The installer checks the dependencies of the selected packages:

Click on Next to start the installation:

The installer will ask if you have all the required discs. Click Continue:

The hard drive is then formatted:

The installation begins. This could take some time depending on your system:

The installation is then complete, you can remove your CD from the computer and reboot.

After the reboot, you will see this screen. Select Firewall configuration and hit Run Tool:

Some Control Panels you may wish to install after come with a firewall. In that case you may wish to disable the CentOS firewall. If you won’t be using control panel you’ll likely want to leave it enabled and configure it to your needs (but then you likely shouldn’t use another firewall later on as it could conflict with CentOS firewall).

SELinux is a security extension that provides additional security. In my opinion you normally don’t need it to configure a secure system, it can cause more problems than advantages. Therefore I disable it for this walk through. But certainly read up on it and enable it if you feel it benefits your server.

Then leave the Setup Agent by selecting Exit:

Then log in as root and reboot the system so that your changes can be applied:

shutdown -r now

Firstly, login to your server through root SSH – If you need an SSH client you can download Putty here.

Enter the commands below one by one:

2.1: Login to your WHM

2.2: Scroll down your navigation to the bottom, you will need to find the ‘Plugins’ section (or Add-Ons depending on your cPanel/WHM version).

2.3: Locate ‘Fantastico De Luxe Admin’ and open it

2.4: Now just follow the instructions to complete the installation of Fantastico.

2.5: Once the installation is complete, you can access the Fantastico client interface through any of your cPanel accounts.

If you receive the “Source Guardian Error” when you visit the Fantastico client interface, run the command:

If you can’t see the Fantastico option within your cPanel accounts then:

1. Go to WHM

2. Edit the ‘Default Features List’

3. Select to activate Fantastico

A large percentage of unmanaged VPS customers are website developers, they use our service as a testing environment whilst they work on developing their scripts / software.

The important question is, why is a VPS important for a website developer?

A VPS (Virtual Private Server) offers totally dedicated server resources to you.

In the early stages of a website development project, the scripting can be messy and quite resource hungry. A shared web hosting provider needs to monitor their services for things like this. Due to the fact that shared web hosting will be sharing a server with hundreds of other customers websites, everyone has to play fair and use their fair share of server resources. A shared web hosting provider will kick you straight off of their servers if your script is taking up high server resources, or in a worst case scenario your code may even loop! This can cause the entire server to crash due to high CPU load. The host and it’s customers won’t be very happy if you loop their server.

You won’t have this problem with a VPS. If your script is using a high amount of server resources this effects nobody else but you. If your script loops, you can just reboot the VPS and fix the faulty coding. You won’t have complaints from your web host.

Improved server resource monitoring.

If you’re developing a website script, in most situations you’d want it to use as little server resources as possible. When the script goes live and the website has a lot of visitors, the more resource demanding your script is, the more money the website owner will need to spend on hosting it. For example, if you script is quite resource demanding and you have 100 visitors on your site at the same time, every time the visitor loads a new page, your server is processing multiple apache connections along with querying a database, etc. The more resource friendly the script is, will make the difference on whether you require an entire dedicated server or just a VPS to host your website. There can be quite a large price difference between the two hosting solutions, and when you are paying renewal fees constantly to keep the site online, it can be quite costly.

When developing your website script on a VPS, you will have full access to monitor your servers resource usage. This will help you with improving it’s resource efficiency.

Full control over the configuration of your server.

All scripts require the server to be set up differently. They require different Apache modules to be installed for parts of the code to operate correctly. A shared web hosting provider will have their servers set up in a way to provide the best overall service and security for their customers. They will have a lot of restrictions on their Apache configuration, this stops people using scripts to carry out certain actions, which may degrade the service for other customers sharing the server with you. If the shared web hosting provider doesn’t allow a certain mod to be enabled on their server, your script may not work correctly, and you cannot complete your development project.

With a VPS you have FULL control over how you configure the server. You can change the Apache configuration at any time to enable or disable certain modules.

No limits to how many databases or other features you can create.

You will find that shared web hosting providers often allocate certain packages with feature limits. They may only allow 1 database to be created on a certain package. This can be very restrictive for a website developer.

A VPS has no limits to how many databases, domains, email addresses or other features you can have. You are in total control of your own server, you can manage it however you like to suit your development project.

I hope you found this useful for your future development projects. There are plenty of VPS hosting providers about to suit any requirement, good luck in your hunt. If you have any questions, please leave a comment.

You will need to visit http://www.webmin.com/download.html and copy the link for their latest RPM format Webmin download.

Download this link either to your computer and then upload it to your server, or by using wget, apt-get or similar depending on your linux operating system. For this example I will be using wget:

Enter the command, and then press enter:

Next we need to kick off the installation. You need to use the command:

Note: remember to change the file name depending on the file name of the version you are installing!

HELP! I’ve just received an error! – If you receive the error “/usr/bin/perl is needed by webmin-1.510-1.noarch” then you will need to install Perl – Check out STEP 2A further down this article to find out how!

You can now login to Webmin by visiting http://youriphere:10000/ – You need to log in with your servers root login information (E.G. Username: root – Password: thisisaverysecurepassword).

And success! Webmin appears to be working correctly!

Perl can be installed easily using yum if you have it installed on your server:

When it asks you if it’s OK to download the Perl packages, type “Y” and press enter:

And then go back to STEP 2.