Systems don't announce when something is about to go wrong. Good administrators watch their servers and networks all the time, so they spot trouble early, and they have a calm, repeatable way to find the cause when something breaks.
You'll start with monitoring, using tools like Prometheus and Grafana to collect and display metrics. Then you'll look at Linux performance troubleshooting, especially disk I/O problems, and learn to read system logs with journalctl. Finally you'll practice network troubleshooting with everyday tools such as ping, tracert, and nslookup.
Watch the lessons in order, then continue on to Configuration Management & IaC.
How to track down disk I/O performance problems on a Linux server.
Using journalctl to read and filter systemd logs when you are troubleshooting a Linux system.
The everyday command-line tools for finding out where a network connection is failing.