Your browser is blocking local storage, so your progress will be lost when you close or reload this page.

Find out what a running system is doing and why it fails.

Topics

• Logging FundamentalsExpected level: Working

  Write application logs that help diagnose problems.

  Not reviewed
• Log AnalysisExpected level: Proficient

  Read and interpret application logs efficiently.

  Not reviewed
• Troubleshooting MethodologyExpected level: Proficient

  Investigate problems in running systems using logs, configuration and evidence.

  Not reviewed
• Following a Change Through a SystemExpected level: Working

  Trace data from UI/API through backend and database layers.

  Not reviewed
• Root Cause AnalysisExpected level: Working

  Find out why a problem happened as well as how to fix it.

  Not reviewed
• Monitoring ConceptsExpected level: Awareness

  Health, performance and availability monitoring.

  Not reviewed
• Incident BasicsExpected level: Awareness

  Production support and what is expected of you during an incident.

  Not reviewed

Self-assessment

You should be able to do each of these without help.

• Add a log statement at an appropriate level that will be useful in production. It should say what happened and include the identifiers needed to find the case, without logging passwords or personal data.Related: Logging Fundamentals
• Find the relevant entries in a log file and follow a request through them. Filter by a correlation id or timestamps, show the lines around a match, and watch a live log while you reproduce the problem.Related: Log Analysis
• Trace one value from the browser through the API and into the database. Check each layer in turn and name the first one where the value is wrong.Related: Following a Change Through a System
• Investigate a failure using logs and configuration before asking for help. Keep a short log of symptom, evidence, hypothesis, test and result, so that you can say what you checked and found.Related: Troubleshooting Methodology, Log Analysis
• Explain the difference between a symptom and a root cause. Take a failure you have seen and name both, and say how you could confirm the cause.Related: Root Cause Analysis, Troubleshooting Methodology
• Say how to raise an incident in your team and what you should and should not do during one. Know where the runbook and on-call contact are.Related: Incident Basics