How to fix common problems with Apache2

Problems loading a website are often blamed on the Internet connection, but even the most perfectly set-up network cannot help if there is no service to reply at your destination. One of the most popular HTTP servers used for this task is Apache2.

Much of Apache’s popularity can be attributed to its easy installation and use, but nonetheless, it is possible to run into problems with even the easiest of the software. If you’ve encountered an issue loading your web page, follow these simple troubleshooting methods outlined in this guide to attempt to get your web server back up and working again.

Make sure the service is running

The first step in troubleshooting any service is to check that it is running and able to function. A straightforward approach is to restart the service. On Ubuntu and Debian servers, use the following command.

sudo systemctl restart apache2

In CentOS and other Red Hat environments, the Apache2 service is called ‘httpd’, so use this command instead.

sudo systemctl restart httpd

The restart command output will vary depending on your distribution choice. Subjectively, Ubuntu usually gives the most helpful reply, as shown below.

* Restarting web server apache2             [ OK ]

CentOS and Debian probably do not say anything as long as the restart didn’t fail.

If your system couldn’t restart Apache2, you most likely got an error message. If you got a reply resembling either of the output examples below, the service is most likely not properly installed on your system, or some files are missing.

apache2: unrecognized service

Failed to restart apache2.service: Unit apache.service failed to load: No such file or directory.

If you see these errors, try installing the service again. On servers running Debian or Ubuntu, use the following command.

sudo apt-get install apache2

On CentOS, install httpd instead with the next command.

sudo yum install httpd

Once you are sure the web server is fully installed, restart the service using the same command you used before.

If you received a different error message, try obtaining more information on the state of your web service. Use one of the following commands that applies to your system.

sudo systemctl status apache2

sudo systemctl status httpd

The output from the status check will tell at least whether the service is running or stopped. Debian and CentOS systems usually show a more detailed report, including service uptime and a couple of log lines. Below is an example of a Debian status display; one on CentOS would be nearly identical. Ubuntu does not have this type of output by default.

● apache2.service - LSB: Apache2 web server
    Loaded: loaded (/etc/init.d/apache2)
    Active: active (running) since Fri 2015-07-31 10:44:26 EEST; 2h 14min ago
   Process: 9704 ExecStop=/etc/init.d/apache2 stop (code=exited, status=0/SUCCESS)
   Process: 9711 ExecStart=/etc/init.d/apache2 start (code=exited, status=0/SUCCESS)
    CGroup: /system.slice/apache2.service
            ├─9726 /usr/sbin/apache2 -k start
            ├─9730 /usr/sbin/apache2 -k start
            ├─9731 /usr/sbin/apache2 -k start
            ├─9732 /usr/sbin/apache2 -k start
            ├─9733 /usr/sbin/apache2 -k start
            ├─9734 /usr/sbin/apache2 -k start
            └─9747 /usr/sbin/apache2 -k start
 Jul 31 10:44:26 debian.example.com apache2[9711]: Starting web server: apache2.

The important part here is on the third line, active and running, which means the process should be working; if instead it shows active and stopped or failed, the process will most likely crash.

Next, check the processes by their name for either apache2 or httpd.

sudo ps aux | grep -E 'apache2|httpd'

For every mention of the searched keyword, grep will print out a line on which the keyword was found. This includes the search process itself, so you will see at least the grep command. If there is more than one line of output, all but the last are processes related to your web service. An example below displays apache2 processes running on Ubuntu.

root      1457  0.0  1.5 321908 16132 ?      Ss   Jul30  0:02 /usr/sbin/apache2 -k start
www-data  1461  0.0  2.8 326532 29172 ?      S    Jul30  0:00 /usr/sbin/apache2 -k start
www-data  1462  0.0  3.1 327480 32364 ?      S    Jul30  0:00 /usr/sbin/apache2 -k start
www-data  1463  0.0  2.9 326688 30260 ?      S    Jul30  0:00 /usr/sbin/apache2 -k start
www-data  1464  0.0  3.1 326496 32148 ?      S    Jul30  0:00 /usr/sbin/apache2 -k start
www-data  1465  0.0  2.7 326816 28040 ?      S    Jul30  0:00 /usr/sbin/apache2 -k start
www-data  1841  0.0  2.0 323132 21044 ?      S    Jul30  0:00 /usr/sbin/apache2 -k start
www-data  1871  0.0  2.2 323396 23280 ?      S    Jul30  0:00 /usr/sbin/apache2 -k start
user     11669  0.0  0.0  11744   928 pts/0  S+   15:32  0:00 grep --color=auto -E apache2|httpd

Since the service restart command didn’t work earlier, any processes grep might list have probably stopped functioning and need to be closed before the service can start again. On Debian and Ubuntu systems, all Apache2 processes are stopped using this command.

sudo killall apache2

On CentOS servers not only is the web service called something else but also the kill command functions little differently, use the following instead for the same result.

sudo kill -a httpd

After the kill command, you can run the process check again to confirm that no more zombies are left. Then, try to restart the service using these same commands as at the beginning of this guide.

sudo systemctl status apache2

sudo systemctl status httpd

This should get the web service running, provided that it has not been misconfigured.

Check your server configuration

When starting a service fails with errors referring to files located in either /etc/apache2 or /etc/httpd/, the system had trouble reading the service configuration files. Apache2 comes with some handy tools for file integrity and syntax checks that can help in locating any typing mistakes or other irregularities in the configuration. Check the config files with Debian and Ubuntu servers by running the following command.

sudo apache2ctl -t

On CentOS machines, call httpd instead.

sudo httpd -t

The output will show any problems found with the configuration file. Or, if everything is in order, it simply prints out a confirmation like the one shown below.

Syntax OK

Another troubleshooting option for the web service is to show the parsed virtual host and run settings with commands for Debian/Ubuntu and CentOS, respectively.

sudo apache2ctl -S

sudo httpd -S

Below is an example of the command output from a CentOS system. Make sure the server and document roots point to the correct directories.

VirtualHost configuration:
ServerRoot: "/etc/httpd"
Main DocumentRoot: "/var/www/html"
Main ErrorLog: "/etc/httpd/logs/error_log"
Mutex proxy: using_defaults
Mutex authn-socache: using_defaults
Mutex default: dir="/run/httpd/" mechanism=default
Mutex mpm-accept: using_defaults
Mutex authdigest-opaque: using_defaults
Mutex proxy-balancer-shm: using_defaults
Mutex rewrite-map: using_defaults
Mutex authdigest-client: using_defaults
PidFile: "/run/httpd/httpd.pid"
Define: DUMP_VHOSTS
Define: DUMP_RUN_CFG
User: name="apache" id=48
Group: name="apache" id=48

If your server uses a custom virtual host configuration, like when hosting multiple websites on one server, check that each virtual host file has the correct domain name and points to the correct root directory. Debian and Ubuntu machines have a virtual host file by default, and it is stored in /etc/apache2/sites-enabled/. Open the file for editing.

sudo nano /etc/apache2/sites-enabled/000-default.conf

This file usually has some instructions on what each parameter means in the comments, which have been left out in the example below. However, as already mentioned, the important parts are ServerName and DocumentRoot.

<VirtualHost *:80>
   ServerName www.example.com
   ServerAdmin webmaster@localhost 
   DocumentRoot /var/www/html
   #LogLevel info ssl:warn      
   ErrorLog ${APACHE_LOG_DIR}/error.log
   CustomLog ${APACHE_LOG_DIR}/access.log combined
   #Include conf-available/serve-cgi-bin.conf
</VirtualHost>

CentOS and httpd do not have the same virtual host file set by default but instead use the httpd service configuration to store the default settings. Check the configuration file using the command below.

sudo vi /etc/httpd/conf/httpd.conf

Look for the same parameters, ServerName and DocumentRoot, and make sure they are correctly set.

If you made any changes to the configuration files, the service needs to be reloaded for the changes to take effect. Restarting the service does the job, but if you wish to avoid downtime on your web server, reload instead with one of the following commands.

sudo systemctl reload apache2

sudo systemctl reload httpd

Check Logs

When everything on the service side is working as expected, and you cannot find a fault, but the website still won’t load, it’s always a good time to dig through logs. Apache2 keeps two sets of logs: access and error. Depending on your choice of Linux distribution, you can find the logs stored at /var/log/log/apache2/ or /var/log/httpd. You can list all files in your web server’s log directory using the commands below.

sudo ls /var/log/apache2/

sudo ls /var/log/httpd/

The log lists will differ slightly as different systems name the logs a little differently. Ubuntu and Debian servers store the current uptime logs to access.log or error.log. Previous logs are marked with a running number, 1 being the latest, and older logs than that are also compressed. On CentOS and other Red Hat variants, the logs are named access_log and error_log. Older logs have their name appended with the date the log was written on, e.g., access_log-20150108.

An easy way to start reading the logs when you don’t necessarily know what you are looking for is to use the filtering app ‘grep’. Search for any errors using one of the commands below that correspond with your system’s web application name.

sudo grep -i -r error /var/log/apache2/

sudo grep -i -r error /var/log/httpd/

Ubuntu and Debian users might also need to check through the compressed log files. This can be done using ‘zgrep’ instead, but due to its limitations, you can only search one log at a time, for example, using the following command.

sudo zgrep error /var/log/apache2/error.log.2.gz

Not all errors logged by your web service necessarily mean something is wrong with your server, but look for repeating problems, like missing files in the example error below.

[Fri Jul 17 12:36:08.431065 2015] [:error] [pid 4649] [client 80.69.160.92] script '/var/www/html/index.php' not found or unable to start

If searching for errors does not yield results, you may also wish to enable more verbose logging. The configuration file’s parameter ‘LogLevel’ controls the log output amount. On Debian and Ubuntu systems, open your configuration file using the following command.

sudo nano /etc/apache2/apache2.conf

With CentOS and other Red Hat-based servers, use the command here instead.

sudo vi /etc/httpd/conf/httpd.conf

Find the LogLevel parameter and change it from the default ‘warn’ value to ‘debug’, then save the file and exit. Again, for any changes to be applied, the service needs to be reloaded. You can do this with one of the following commands that are appropriate for your system.

sudo systemctl reload apache2

sudo systemctl reload httpd

To generate new entries using the new log levels, try accessing your website again a couple of times, even if it doesn’t work. The more verbose logs should help narrow down any underlying issues.

Check other services

If your website still won’t load after all the troubleshooting with Apache, check other related services on your cloud server, such as databases, firewalls, and network connections, which also have their own troubleshooting instructions.

Do not hesitate to contact our support team. Remember to include any relevant information you may have discovered while troubleshooting, as every bit of detail will be useful in further investigations.