The monitoring is done by 2 ways: The server/cluster part (RAM/CPU/HDD...) and via specialized tools for CatchChallenger.
The server monitoring tools like nagios/munin/zabbix, allow check the nominal serveur status. I don't will details it here because you have all the information on internet.
The specialized tools slow control the part specific to CatchChallenger:
- The status of each node checking the protocol (is all the game and login server is up)
- The backup
- The database replication
- The mirror status and control the content is exactly the official content
- This is coupled with automated tools which control the full connectivity by connecting on login server, select and connect on game server.
- The game server up and disconnected and their load in real time.
It need monitor the client setup or dmg, unpack it, control if dependency and datapack is correct as execpted
I use zabbix to monitor some page + test the smtp mail send (to control if new user can be registred).