RH124 Ch 16: Analysing Servers and Getting Support

When something breaks, work from symptoms to cause instead of guessing.

Quick health check - uptime / top — load versus nproc CPUs; free -h — memory and swap; df -h — full disks. - systemctl --failed — units that did not start; systemctl status unit shows the error and exit code. - journalctl -p err -b — errors since boot; journalctl -u unit for one service. - Status 203/EXEC means the program in ExecStart could not be run (missing, not executable, wrong path).

Web console (Cockpit): dnf install cockpit; systemctl enable --now cockpit.socket, then browse to https://host:9090. It shows logs, services, storage, networking and a terminal — handy, but the exam expects command-line skills.

Getting help from Red Hat - The Customer Portal has the knowledge base, documentation and support cases. - sos report gathers configuration and logs into one archive for a support case: sos report --batch (all plugins) or -o kernel,networking (selected ones); --tmp-dir chooses where it goes. It may contain sensitive data — review or use sos clean before sending. - Red Hat Insights (insights-client --register) analyses registered systems and recommends fixes for known problems.

Exam habit: the exam has no support line. For each failed task, read the error, check the log, fix the cause, then verify — the same loop you would use before calling support.