I have the same problem on a debian wood. It seems that the apache main process dies, but some child processes and the SSL apache process where still alive.
I've written a small php script that checks the apache logfile for the error, kills all remaining httpd processes and restarts apache. Its not a nice solution, but it works for me until i find the resaon why apache fails.
Code:
<?php
exec('grep "returned a Fatal error..." /var/log/apache/error.log > /tmp/apache_error_test.log');
clearstatcache();
if(filesize("/tmp/apache_error_test.log") > 0) {
// killing apache processes
exec("killall apache");
// make a backup of the error logfile
exec("cp /var/log/apache/error.log /var/log/apache/fatal_error.log");
// empty the error log
exec("cat /dev/null > /var/log/apache/error.log");
sleep(5);
// starting apache websrver
exec("/etc/init.d/apache start");
}
?>
I run this script as cronjob.
Recent comments
19 hours 25 min ago
22 hours 20 min ago
23 hours 34 min ago
1 day 57 min ago
1 day 2 hours ago
1 day 4 hours ago
1 day 5 hours ago
1 day 21 hours ago
1 day 22 hours ago
2 days 1 hour ago