A practical guide to operating, diagnosing, and securing Linux systems
This repository is an original Linux operations knowledge base built around a simple principle: observe first, change deliberately, and verify the result.
It focuses on the work that matters when supporting applications and servers—service failures, permissions, storage pressure, network reachability, logging, host security, performance, automation, and incident response. Examples use generic production scenarios and avoid pretending that one command fits every distribution or environment.
Important
Commands that change permissions, services, storage, networking, or security can interrupt a system. Review the relevant manual page and test in a safe environment before applying a change.
| Operate | Diagnose | Protect and improve |
|---|---|---|
| systemd services | Network failures | Host security |
| Processes and signals | Storage incidents | Safe automation |
| Users and permissions | Performance | Incident workflow |
Browse the complete guide index.
flowchart LR
A["System mental model"] --> B["Shell + filesystem"]
B --> C["Identity + processes"]
C --> D["Services + logs"]
D --> E["Networking + storage"]
E --> F["Security + performance"]
F --> G["Automation + incidents"]
linux-engineering-guide/
├── README.md
├── CONTRIBUTING.md
└── docs/
├── 01-foundations/
├── 02-shell/
├── 03-filesystem/
├── 04-permissions/
├── 05-processes/
├── 06-services/
├── 07-networking/
├── 08-storage/
├── 09-observability/
├── 10-security/
├── 11-performance/
├── 12-automation/
└── 13-troubleshooting/
- Begin with the operational scenario instead of memorizing commands.
- Run observation commands in a disposable Linux environment.
- Explain what each output proves—and what it does not prove.
- Practice the workflow with a deliberately introduced failure.
- Keep personal environment notes separate from reusable guidance.
- Original explanations and examples.
- Distribution-aware guidance with primary references.
- Safe, observable, and reversible operations.
- No copied course content, hidden provenance, or fabricated experience.
- No credentials, proprietary hostnames, or employer-specific data.
No license is granted yet. The original content remains all rights reserved until the repository owner selects a license explicitly.