Apache
Note: Apache on Fedora and CentOS may be called httpd
Errors
Use either of the below commands to check the apache2 configuration:
apache2 -t apache2ctl -t apache2ctl -configtest #and the status /etc/init.d/apache2 status /etc/init.d/httpd status
Mime Types
A common issue is being only able to either download or view a PDF. This needs to either be added in the /etc/mime.types with:
application/pdf .pdf
Or the .htaccess file for that domain or /etc/apache2/mods-enabled/mime.conf with:
AddType application/pdf .pdf
New web server configuration
New files of configuration for Apache web server were not built due to errors in configuration templates. The detailed error message was e-mailed to you, so please check the e-mail, fix the errors, and click here to retry generating configuration or New web server configuration for some domains within the marked subscriptions was not built due to errors in configuration templates. The detailed error message was e-mailed to you, so please check the e-mail, fix the errors, and click here to retry generating configuration.
apache2: bad user name ${APACHE_RUN_USER} ("Apache2 bad user name $ apache run user") or
apache2: Syntax error on line X of /etc/apache2/apache2.conf: Syntax error on line X of /etc/apache2/conf.d/zz010_psa_httpd.conf: Syntax error on line X of /opt/psa/admin/conf/vhosts_bootstrap.conf: Could not open configuration file
Fixs:
If it is complaining about a user name, check if this was an FTP or Email account, or a User under Customers or Users. You can sometimes suspend the offending domain to temporarily resolve the issue.
If when clicking the domain you get this sort of error below, see this Parallels KB article.
Internal error: Unable to load object of type SysUser with id=1: SysUser: unable to select: no such row in the table Message Unable to load object of type SysUser with id=1: SysUser: unable to select: no such row in the table File common_func.php3 Type PleskFatalException
If you have removed a domain or SSL and do:
/etc/init.d/apache2 restart
You may get an error in /etc/apache2/apache2.conf
vim /etc/apache2/apache2.conf
Check the line in the error. It may then reference:
/etc/apache2/conf.d/zz010_psa_httpd.conf
Which references:
/opt/psa/admin/conf/vhosts_bootstrap.conf
Which will want to include the following type of file:
/var/www/vhosts/domain.com/conf/XXXXXXXXXXX.XXXXXXXX_httpd.include /opt/psa/var/certificates/cert*
/var/www/vhosts/domain.com/conf
Run:
/usr/local/psa/admin/bin/httpdmng --reconfigure-all
If this does not work, run:
/usr/local/psa/bin/domain --remove domain.com
And then you may need to remove the include entries for the missing domain in:
/opt/psa/admin/conf/vhosts_bootstrap.conf
You MAY be able to remove some of the httpd.include files from here so it might not include them in the Apache configuration, however this has not been tried.
If this does not work and you are getting the following error:
New files of configuration for Apache web server were not built due to errors in configuration templates: Syntax error on line XXX of /opt/psa/admin/conf/generated/13589366030.21908000_server.include: SSLCertificateFile: file '/opt/psa/var/certificates/certl8w1mkz' does not exist or is empty
ERR (3): Apache config (13589358830.79743100) generation failed: Syntax error on line 131 of /opt/psa/admin/conf/generated/13589358830.79743100_server.include: SSLCertificateFile: file '/opt/psa/var/certificates/certl8w1mkz' does not exist or is empty
Check this and then if that does not work do this:
ls -lah /opt/psa/var/certificates/ vim cert-xxXx00
Copy the contents of that file and input it into the one it says is missing:
vim certl8w1mkz :wq /etc/init.d/apache2 restart
This can also be caused due to the MySQL bug seen here as well. Do not upgrade Plesk to 10.4.4 on Ubuntu 8.04.
Alternate fixes:
http://innovationframes.com/phpbb/viewtopic.php?f=33&t=274
http://kb.parallels.com/en/111605
Rebuild Plesk Apache httpd.conf configuration file
SSL server IP/port conflict
This will normally appear in either /var/log/apache2/error.log or /var/log/syslog. It can be caused by Plesk's webmail clients horde.webmail:443 or atmail.webmail:443 conflicting with an SSL assigned to a domain.com:443
This can cause Apache to get stuck in a loop on startup and sometimes use 100% CPU usage. Either get another IP address, remove and re-add the domain/SSL, do not use webmail or configure it to use a different port.
Unable to start/restart Apache
If the following error appears after issuing /etc/init.d/apache2 start command
* Starting web server apache2 (98)Address already in use: make_sock: could not bind to address 0.0.0.0:80 no listening sockets available, shutting down Unable to open logs
It is normally what they call "hogging a port". You can use the following commands below to kill it and start it again. The issue as caused by apache hogging the HTTP port.
/etc/init.d/apache2 stop kill -15 process *wait 5-15 seconds* kill -9 apache2 killall apache2 /etc/init.d/apache2 start
If it can continually be stopped and shows this when starting:
* Starting web server apache2 98)Address already in use: make_sock: could not bind to address 0.0.0.0:443 no listening sockets available, shutting down Unable to open logs
Ensure /etc/apache2/ports.conf only has this in it:
NameVirtualHost *:80
Listen 80
<IfModule mod_ssl.c>
Listen 443
</IfModule>
<IfModule mod_gnutls.c>
Listen 443
</IfModule>
If you get:
Service /etc/init.d/apache2 failed to start websrvmng: /opt/psa/admin/bin/apache_control_adapter execution failed: (20014)Internal error: Error retrieving pid file /var/run/apache2.pid Remove it before continuing if it is corrupted.
Do this:
cd /var/run mv apache2.pid apache2.pid.old /etc/init.d/apache2 stop /etc/init.d/apache2 start
Vhost Configuration Syntax
The below error indicates there may be spacing, closing and spelling issues with the vhosts configuration file or you have broken the apache configuration as in the header above.
Syntax error on line 1177 of /etc/apache2/apache2.conf: Syntax error on line 151 of /etc/apache2/conf.d/zz010_psa_httpd.conf: Syntax error on line 86 of /var/www/vhosts/domain.com/conf/httpd.include: Syntax error on line 1 of /var/www/vhosts/domain.com/conf/vhost.conf: /var/www/vhosts/domain.com/conf/vhost.conf:8: Less Than Greater Than was not closed. \n/var/www/vhosts/domain.com/conf/vhost.conf:1: Less Than Greater Than was not closed. [fail]
vim /var/www/vhosts/domain.com/conf/vhost.conf /opt/psa/admin/sbin/websrvmng -av /etc/init.d/apache2 start * Starting web server apache2
Normal Apache Virtual Host configuration
<VirtualHost X.X.X.X:80> ServerName "domain.co.uk:80" ServerAlias "ipv4.domain.co.uk" ServerAlias "www.domain.co.uk" UseCanonicalName Off
Cannot upload to website
Check the following location:
tail -f /var/www/vhosts/yourdomain.com/statistics/logs/error.log
If you see:
mod_fcgid: HTTP request length 131100 (so far) exceeds MaxRequestLen (131072)
go into:
vim /etc/apache2/mods-available/fcgid.conf #add the below between the <IfModule> and </IfModule> tags. FcgidMaxRequestLen 1073741824 (or cgid for CGI)
It should appear in white if correct. Restart Apache using /etc/init.d/apache2 restart.
Creating Virtual Hosts from scratch
This is using an Ubuntu 12.04, Apache 2.2.22 server as an example.
cd /var/www mkdir sites ; cd sites ; mkdir yourdomain.co.uk cd /etc/apache2/sites-available vim domain.co.uk
Insert the following:
<VirtualHost *:80>
ServerAdmin [email protected]
DocumentRoot "/var/www/sites/domain.co.uk"
ServerName domain.co.uk
ServerAlias www.domain.co.uk
#for subdomains just put the ServerAlias the same as ServerName
ErrorLog "/var/log/apache2/domain.co.uk-error_log"
CustomLog "/var/log/apache2/domain.co.uk--access_log" common
</VirtualHost>
Note, if you have RewriteEngine On inside your vhost (Virtual Host), restart Apache and it says this:
"Invalid command 'RewriteEngine', perhaps misspelled or defined by a module not included in the server configuration"
This means you need to do:
a2enmod rewrite /etc/init.d/apache2 restart
a2ensite domain.co.uk [ -f /etc/init.d/httpd ] && service httpd restart ; [ -f /etc/init.d/apache2 ] && service apache2 restart apache2ctl -t
To disable a site, do a2dissite domain.co.uk
Extension Modules
Before installing, use apt-get update
Search using apt-cache search <module name>
After installing, use /etc/init.d/apache2 restart
C Compiler (gcc)
apt-get install build-essential y
Crypt:SSLEay, Net:SSL and SSLEay
These are commonly needed alongside OpenSSL and ModPerl.
Ubuntu:
apt-get install libnet-ssleay-perl libcrypt-ssleay-perl
CentOS:
yum install perl-Crypt-SSLeay.x86_64 perl-Net-SSLeay.x86_64
In case you need it for an additional module:
yum install php-pecl-apc.x86_64
libmcrypt
apt-get update apt-cache search mcrypt apt-get install <package name>
Examples of libraries are libmcrypt-dev, libmcrypt4, libtomcrypt-dev, libtomcrypt and libtomcrypt0
mcrypt
apt-get install php5-mcrypt
mod_headers and mod_expires
a2enmod expires a2enmod headers service apache2 restart
expires is needed to set expiration dates on files (for caching)
mysql-devel
Under Ubuntu 10 this has been renamed to libmysqlclient16-dev and libmysql++-dev (or 15)
pear
apt-get install php-pear
You may need php-pear.noarch on CentOS. Alternatively you may need the EPEL repository installing:
wget serverkb.co.uk/tools/epel.sh ; chmod +x epel.sh ; ./epel.sh ; yum search pear
Using Apache on Windows
As long as the IP address is not within IIS and no domains are assigned to it it should be possible. You may want this for tomcat statistics for example. It is not advised however as it is native to Linux.
httpd dead but subsys locked
Googling this message comes back with a lot of different information regarding it. It can happen if a database using MySQL comes under high load and overloads the web server. Restarting/starting Apache/httpd normally fixes this.
Manually create Virtual Hosts
Performance Tuning
As mentioned on the front page http://pingdom.com and http://slowcop.com are good places to start. For example, I improved the load speeds of this Wiki by enabling compression on the server by doing the following in SSH:
a2enmod deflate vim /etc/apache2/mods-enabled/deflate.conf
Add in the following underneath the </IfModule>
<Location /> # Insert filter SetOutputFilter DEFLATE # Netscape 4.x has some problems... BrowserMatch ^Mozilla/4 gzip-only-text/html # Netscape 4.06-4.08 have some more problems BrowserMatch ^Mozilla/4\.0[678] no-gzip # MSIE masquerades as Netscape, but it is fine # BrowserMatch \bMSIE !no-gzip !gzip-only-text/html # NOTE: Due to a bug in mod_setenvif up to Apache 2.0.48 # the above regex won't work. You can use the following # workaround to get the desired effect: BrowserMatch \bMSI[E] !no-gzip !gzip-only-text/html # Don't compress images SetEnvIfNoCase Request_URI \ \.(?:gif|jpe?g|png)$ no-gzip dont-vary </Location>
Then restart the web server: /etc/init.d/apache2 restart
If you do not use mod_perl or mod_python on your websites, disabling these may decrease memory usage.
Premature end of script headers
This can be an indication the server has run out of RAM. This could be due to the amount of traffic, specific actions on the server or a deliberate attack on the server. Check the server status logs to see if there are spikes where the load on the server increases.
Check for any suspicious network traffic destined for your server. You can install a reactive firewall that will log and block a large range of deliberate attacks against your server. It's unlikely to have any side effects but if you do notice any changes to functionality of your site the rules can be relaxed or disabled.
Install apf http://www.rfxn.com/projects/advanced-policy-firewall . If the firewall prevents anything from functioning just do /etc/init.d/apf stop
Scan your site for Malware (e.g. Maldet or NOD32).
It's possible the issue is simply the legitimate traffic to the site putting more load on the server than it can handle. If you are using Wordpress use caching plugins. We would strongly recommend totalcache or supercache as these can dramatically improve the performance of Wordpress sites. RSS Feeds are notorious for attracting traffic.
CGI scripts in arbitrary directories
UID of script is smaller than min_uid
You likely need to chown the files to the right user:group
