Software Engineering Wiki
Reference notes for infrastructure and platform engineering.
Start from the symptom
- Debug a workloadPod status, events and container logs, in that order.
- Chase a slow requestBreak the timing down by stage before blaming the app.
- Find the bottleneckCheck utilisation, then saturation, then errors.
- Trace a network problemWalk the interface, the address and the route.
- Work out why a name failsAsk the authoritative server, then work back.
- Fix a stuck deployRead sync status and health separately.
- Review a Git changeCompare the working tree, the index and the last commit.
- Diagnose a dead serviceUnit state and exit code, then the journal.
Topics
Kubernetes
Containers and IaC
Observability
Security and identity
Languages
Snippets
Using these pages
Each page opens with the check to run first, then widens into reference. Heading links are stable, so a section can be bookmarked. Examples come from different environments and versions: substitute your own names and paths, and confirm what a command touches before running it.