hermeticormus System diagnosis
A system issue is happening. Diagnose and fix it on a Debian/Ubuntu Linux machine using a systematic, hypothesis-driven loop. Gather evidence before changing anything; confirm before applying a fix.
Diagnostic steps
- Describe the problem: what is happening, and when did it start?
- Gather information:
- System logs:
journalctl -xe - Specific service logs:
systemctl status <service>andjournalctl -u <service> -xe - Resource usage:
top,htop,df -h - Network:
ip addr,ping,ss -tlnp
- System logs:
- Identify patterns: is it consistent or intermittent?
- Check recent changes: what changed before the issue appeared?
- Research: look for known issues.
- Test hypotheses: verify the likely cause before acting.
- Implement the fix: apply the solution.
- Verify: confirm the issue is resolved (
systemctl is-active <service>, reproduce the original symptom).
Be systematic and thorough. Form a hypothesis from the evidence rather than guessing, and document findings for future reference.
See full content at https://github.com/HermeticOrmus/linux-sysadmin-skills.