Level
Practitioner
Reading time
300 min
Entries
36 lessons
You get
Certificate (70% to pass)
/ Courses

Managing a Server Safely

For people responsible for keeping a server running - not just building it. Learn the operational habits that turn a fragile box into a dependable one and keep small problems from becoming outages.

42 quiz questions

Read the first lesson

The whole of lesson one, as written. The rest of the course reads the same way.

Every server outage traces back to one of three sources. Naming them is the first step to managing them, because each one needs a different defence.

The first source is you, the operator. A mistyped rm -rf, a firewall rule that locks out SSH, a config reload with a syntax error at 2am. Across most small teams, operator mistakes cause more downtime than attackers do, because you touch the box every day and attackers do not.

The second source is attackers. Exposed services, reused passwords, an unpatched CVE. A fresh server on a public IP starts seeing automated SSH brute force attempts within minutes of coming online.

The third source is entropy: the slow drift that breaks a machine nobody touched. Disks fill with logs, a TLS certificate expires, a package upgrade changes a default, a cron job silently stops firing. Entropy is the one people forget, because there is no villain and no keystroke to blame.

Blast radiushow much breaks when one thing goes wrong
Driftthe gap between how a system was set up and how it actually runs now
Why it matters

Safe management is not one skill. You reduce operator error with process, attackers with hardening and patching, and entropy with monitoring. If you only cover one of the three, the other two still take you down.

Lesson 1 of 36.

Syllabus

  1. What managing safely means

    6 lessons · module test

    1. 01Three things break a server: you, attackers, entropy
    2. 02Blast radius: think before you type
    3. 03Least privilege as the default
    4. 04Reversibility: never make a one-way change
    5. 05The server inventory you must keep
    6. 06Documentation is part of the job, not after it
  2. Users, roles and access

    6 lessons · module test

    1. 07Why nobody logs in as root
    2. 08sudo and privilege separation
    3. 09Per-user accounts and SSH keys
    4. 10Turning off password login
    5. 11Groups and file permissions
    6. 12Offboarding: revoke access and rotate keys
  3. Change management without downtime

    6 lessons · module test

    1. 13The change you did not test
    2. 14Staging that mirrors production
    3. 15Maintenance windows and announcements
    4. 16Config in version control
    5. 17Every change ships with a rollback
    6. 18The change log
  4. Monitoring and alerting

    6 lessons · module test

    1. 19The four golden signals
    2. 20Metrics vs logs vs traces
    3. 21Thresholds, and why static ones fail
    4. 22Alerts people will act on (no fatigue)
    5. 23External uptime checks
    6. 24Where logs go and how long they stay
  5. Backups and a simple incident plan

    6 lessons · module test

    1. 25Backups for the server admin (the short version)
    2. 26The one-page incident plan
    3. 27Roles during an incident
    4. 28The severity ladder
    5. 29Communicating with users and stakeholders
    6. 30The blameless post-mortem
  6. Ports, services and the safe baseline

    6 lessons · module test

    1. 31The service inventory
    2. 32Choosing a port without collisions
    3. 33Firewall default-deny
    4. 34Exposed vs internal services
    5. 35Patch cadence and unattended upgrades
    6. 36The safe baseline checklist
Enroll
$99