I didn't like the existing typo3 nagios monitoring solution so I created my own one.
This script uses the directorys used when nagios core is built from source.
Copy getTypo3.sh and createNewClient.sh to /usr/local/sbin/, the other files to /tmp/ and create the folder /backup.
I've used the pipe "|" as delimiter in the typo3 file, which sed will then use to create newlines.
Add your client urls to the CLIENTS list in getTypo3.sh and run it once manually to create the txt file. After the getTypo3.sh script ran there should be a new .txt file in /tmp/ which should look like this :
Now you should be able to run createNewClient.sh. It creates the server.cfg, creates the checks and also adds the commands to the commands.cfg. If you have added the Client URL already, you can discard that output and just restart the nagios server.
I know this is not the normal way for nagios to monitor hosts and services, but our developers prefered this over the normal typo3 nagios check.
Each of those checks will count the number of 0(OK), 1(WARNING) and 2(CRITICAL).
Im running the getTypo3.sh via a cronjob.
If you have questions, feel free to ask.
Todo: add clientURL to getTypo3.sh script