Setting up a LAMP stack

Requirements

The requirements for a server running Fedora Core can range somewhat more widely than those recommended in the release notes, depending on the use case. A small, private server with light use can easily handle its expected load with even less memory/storage/speed than the release notes necessarily call for. On the other hand, a server designed to withstand huge amounts of traffic can benefit from multiple processors, several gigabytes of RAM, and large arrays of disk drives. As a server administrator, you'll need to use your best judgment for your use case.

Hardware aside, this article assumes nothing more than a fresh Fedora Core installation and shell access as root (all commands listed here assume you execute them as root). Configuring for specific uses (setting up Virtual Hosts, installing web applications or creating websites, or learning PHP) are beyond the scope of this article, which only covers setup.

Doing the work

The best way to install a LAMP stack is to install each letter in sequence; to spell out the word LAMP, in effect.

Linux

If you've already got Fedora Core installed, then you've accomplished this step. If you haven't, check out Install Solutions for more details on installing and configuring Fedora Core.

Apache

1. Installing Apache means installing the program which will handle requests to your web server. For most people, Apache itself is the web server, terminologically speaking. In your Fedora Core installation, depending on the choices you made during the install process, you may or may not have Apache already installed. Just use the following command to check:

# rpm -q httpd

2. If it tells you a version number, then it's installed and we can progress to the next step. Otherwise, we need to install it:

# yum install httpd

3. After installation, the service must be configured to start automatically.

# /sbin/chkconfig httpd on

4. To start the server process immediately:

# /sbin/service httpd start

5. Finally, it's important to make sure that port 80 is remotely accessible, by configuring iptables to open the port (and to leave it open by default from the outside). The following commands will remember the rule for you across reboots and apply it immediately:

# echo '-A RH-Firewall-1-INPUT -m state --state NEW -m tcp -p tcp --dport 80 -j ACCEPT' >> /etc/sysconfig/iptables

# /sbin/service iptables restart

You can put in the hostname or IP address of the server in the address bar of a web browser (whether on the same host or elsewhere in the network) to test the installation. You'll see a welcome page if everything was successful. Otherwise, refer to troubleshooting.

MySQL

1. Installing and running the database component of the LAMP stack, called MySQL, is at least as easy. The following command will install the database and the server process.

# yum install mysql mysql-server

2. The server process must likewise be running and accessible.

# /sbin/chkconfig  mysqld on

# echo '-A RH-Firewall-1-INPUT -m state --state NEW -m tcp -p tcp --dport 3306 -j ACCEPT' >> /etc/sysconfig/iptables

# /sbin/service iptables restart

# /sbin/service mysqld start

3. To test that everything is working okay so far, run this command to print out some information and check its status:

 # mysqladmin version status

PHP (or Perl or Python)

PHP interprets the scripts of your favorite web site or application, running on top of the other parts of the LAMP stack.

1. The following command will install PHP, along with the needed stuff to glue it together with the web server and the database.

# yum install php php-mysql

2. PHP is pretty much ready to run after that, although you may need to restart the web server to make sure it loads the PHP module.

# /sbin/service httpd restart

If you'd like to use another 'P' in the place of PHP (or install alongside), the steps are pretty similar. Both Perl and Python come with ways to access MySQL and modules to allow them to run smoothly and quickly inside of Apache.


3. To run Python on your web server, run the following. (Python is already installed on all Fedora Core installations, so it's not needed to install that as well.)

# yum install mod_python MySQL-python

4. For Perl, the following will get you started:

# yum install perl mod_perl perl-DBD-mysql

Don't forget to restart the web server after either (or both) of these commands to make sure the relevant modules will be loaded by Apache.

Troubleshooting

Definitively testing the whole thing is pretty easy. You can create a small test PHP file and try to access it.

1. Create the test script:

# echo "<?php phpinfo(); ?>" > /var/www/html/index.php

Then access the web server by putting the address in any web browser that can access the server. If your server's IP address is 192.168.1.100, put that in the address bar. By going to that address, you should see a long detailed page describing PHP's configuration and status in detail.

You can skip down to the MySQL section by putting a jump in the address bar to it. (If you used http://192.168.1.100/ to access the page, use http://192.168.1.100/#module_mysql to skip down to the relevant part.) You should see various options about MySQL enumerated.


2. If all this is successful, you should delete the script you just created.

# rm -f /var/www/html/index.php

Now it's time to install your favorite web application (phpBB, mediawiki, squirrelmail/roundcube) or learn PHP.

Bugs

Since I don't personally use Perl for web development, I'm not sure what exactly is needed to use it with a database.

Disclaimer

We test this stuff on our own machines, really we do. But you may run into problems, if you do, come to #fedora on irc.freenode.net