First check if the mod_rewrite module is loaded in apache. To do that, type the following command and you should get the output as shown below:

[rootuser@server ~]# cat /etc/httpd/conf/httpd.conf | grep rewrite
LoadModule rewrite_module modules/mod_rewrite.so

If you do not use virtual hosts, the next step refers to your httpd.conf file. If you do use virtual hosts, the next step refers to the vhost conf file.

Use your favorite editor to edit your vhost configuration file and check for all instances of the line that contains: ‘AllowOverride’ and see if it looks something like this:

AllowOverride Indexes FileInfo

The key here is ‘FileInfo’ if it isn’t there, add it. Now reload apache by issuing:

service httpd reload

If the RewriteEngine has been called in either your vhost file (or httpd.conf) file or the .htaccess file, you should now see the Internal Server Error go away.

This was a pretty weird error and I couldn’t find enough documentation about ‘FileInfo’ on the internet (not even on redhat’s website). I was lucky to come across one instance of it

On another note, it is safer not to use AllowOverride All as it increases the surface area for an attacker to compromise your website. Always only allow modules that are needed. Also, for better performance, always call and configure the modules in the apache configuration files rather than the .htaccess. htaccess is called at page load but the modules are loaded and configured when the webserver has been started.

Let me know if you have found a similar problems and if you have found a better way of fixing it!

Related posts:

1. Setting up Split Brain DNS in Windows Server 2003
2. OCS 2007 Protocol Stack Error Event ID 14517 When Starting Services
3. Symantec Backup Exec Error 0xe00084af (All Versions)
4. Exchange 2007 Recreate OWA folders

#!/bin/bash

# This is a script to identify user processes that have been running

# longer than a pre set number of days.

　

# **** BEGIN CONFIGURABLE PARAMETERS ****

# WARNING!! The following values should NOT be NULL.

# The script will not work with NULL values.

# Path to the logfile

LOG=/var/log/prockill

# List of groups to which users belong to

# The format should be ‘group1| group2| group3| group4′

GROUP=’everyone| utmp| 2000′

# Set the maximum age of a process before it is killed or email notification is sent

# If you want to notify users after 5 days and kill processes after 8 days then

# AGEKILL=’1-| 2-| 3-| 4-| 5-| 6-| 7-| 8-’

# AGEMAIL=’1-| 2-| 3-| 4-| 5-’

# AGE=3

AGEKILL=’1-| 2-| 3-| 4-| 5-| 6-| 7-| 8-| 9-| 10-| 11-| 12-| 13-| 14-| 15-| 16-| 17-| 18-| 19-| 20-| 21-| 22-| 23-| 24-’

AGEEMAIL=’1-| 2-| 3-| 4-| 5-| 6-| 7-| 8-| 9-| 10-| 11-| 12-| 13-| 14-| 15-| 16-| 17-| 18-| 19-| 20-’

AGE=3

# Email Address to send notifications to

MAIL=user@email.com

# Exceptions: processes that you dont want to kill

EXCP=’ldap’

# **** END CONFIGURABLE PARAMETERS ****

　

# **** USER NOTIFICATION & PROCESS KILL ****

# Check if users have been notified

if [ -f /tmp/prockill.notified ]

then

# **** PROCESS KILL ****

if [ "`find /tmp/prockill.notified -mtime +$AGE`" != "" ] # Checking if the file was created x days ago

then # If file was created x days ago, its been x days since the users were notified.

# Demarc start of log

echo “**** Start List Processes Killed on `date` ****” >> $LOG

# Logging list of processes that match our conditions

ps -A -o user,pid,etime,group,args | egrep “$GROUP” | egrep -v “$EXCP” | grep “-” | egrep -v “$AGEKILL” >> $LOG

# Demarc start of log

echo “**** End List Processes Killed on `date` ****” >> $LOG

# Issuing SIGTERM to give processes a chance to exit gracefully

kill -15 `ps -A -o user,pid,etime,group,args | egrep “$GROUP” | egrep -v “$EXCP” | grep “-” | egrep -v “$AGEKILL” | cut -c10-15` > /dev/null

# Wait for 30 seconds to allow processes to terminate

echo “Waiting for processess to cleanly exit…”

sleep 30s

# Issuing SIGKILL to force all misbehaving processes to die.

kill -9 `ps -A -o user,pid,etime,group,args | egrep “$GROUP” | egrep -v “$EXCP” | grep “-” | egrep -v “$AGEKILL” | cut -c10-15` > /dev/null

rm -rf /tmp/prockill.notified # Clear the notified flag

exit 1

fi

else

# **** USER NOTIFICATION ****

# Get list of users and number of processes owned by them.

ps -A -o user,pid,etime,group | egrep “$GROUP” | egrep -v “$EXCP” | grep “-” | egrep -v “$AGEEMAIL” | cut -d’ ‘ -f1 | uniq -c >> /tmp/prockill.eml

# Send email only if the filesize is greater than 0

if [ -s /tmp/prockill.eml ]

then

# Log notification.

echo “**** Start List Notified Processes on `date` ****” >> $LOG

echo “The following users have the indicated number of proceses that will be killed in 24 hours:”

cat /tmp/prockill.eml >> $LOG

echo “**** End List Notified Processes on `date` ****” >> $LOG

# Email Process list

mail -s “$HOSTNAME – Old Process Report” $MAIL < /tmp/prockill.eml

rm -rf /tmp/prockill.eml

touch /tmp/prockill.notified # Set notified flag

exit 1

else

# Log notification

echo “**** Start List Notified Processes on `date` ****” >> $LOG

echo “No old processes to kill!” >> $LOG

echo “**** End List Notified Processes on `date` ****” >> $LOG

exit 1

fi

fi

Related posts:

1. Active Directory Password Expiry Reminder Email
2. Active Directory Audit Script