After finally installing Centreon 2.4 on CentOS 6.3, all was up and running, one issue I had left. No email notifications were sent when a host was down although everything seemed to be correctly configured from the GUI (contacts, notification period, etc).
Looking at /var/log/maillog gave no hints.
But checking /usr/local/nagios/var/nagios.log was helpful.
 HOST NOTIFICATION: Supervisor;Centreon2;DOWN;host-notify-by-email;CRITICAL - 192.168.189.59: Host unreachable @ 192.168.189.36. rta nan, lost 100%  Warning: Attempting to execute the command "/usr/bin/printf "%b" "***** centreon Notification *****\n\nType:PROBLEM\nHost: Centreon2\nState: DOWN\nAddress: 192.168.189.59\nInfo: CRITICAL - 192.168.189.59: Host unreachable @ 192.168.189.36. rta nan, lost 100%\nDate/Time: 17-02-2013" | -s "Host DOWN alert for Centreon2!" firstname.lastname@example.org" resulted in a return code of 127. Make sure the script or binary you are trying to execute actually exists...
There is no command before the -s and the pipe. This results in email delivery failure.
This can be fixed from the GUI.
In Centreon go to Configuration -> Commands, select Notifications from the left.
Two commands need to be edited: host-notify-by-email and service-notify-by-email.
The string /bin/mail has to be inserted after the pipe. Then, export Nagios settings. Of course this fix implies that running /bin/mail from the command line results in successful email delivery.