Guides
Cloud troubleshooting and incident practice
Short, practical guides on working through cloud failures, running practice for a team and showing what an engineer can do.
- DraftHow to prepare for an SRE interview with hands-on incident practiceSRE interviews test how you work through a failing system. How to practise incident response on a live cloud account before the interview does it for you.
- DraftAWS troubleshooting practice: a 30-minute drill you can run every weekA weekly 30-minute drill for AWS troubleshooting practice: one broken system, a clock, and a short written review of how you worked.
- DraftRunning a cloud gameday for your team without breaking productionHow to run a cloud gameday for an engineering team: isolated accounts, a realistic failure, a clock, and a debrief that leads to something you change.
- DraftHow to assess cloud skills beyond certificationsCertifications show what an engineer knows about a cloud. How to assess what they can do with it: hands-on tasks, observed work and a readable record.
- GuideWhy messages vanish: a field guide to dead-letter handlingMessages that disappear after retries usually have nowhere to go. What a dead-letter queue does in SQS, and the checks to run when one seems to do nothing.
- GuideDNS failover that never fails over: what to check firstWhen Route 53 failover never moves traffic to the standby, the cause is usually in how health is evaluated. The first checks to run, in order.
- GuidePublic bucket exposure: how to find and close it in five minutesHow an S3 bucket ends up readable by anyone, how to confirm which bucket is exposed, and the controls that close public access account-wide.
- DraftWriting an incident timeline your manager can readAn incident timeline that reads in two minutes: what to record during the incident, the structure that works afterwards, and the words to leave out.