When things go wrong with OpenShift or OpenShift Container Storage, we sometimes use must-gather or ocs-must-gather to collect logfiles and other documents to diagnose issues. While those tools are useful in that they can show us a lot about the state of the system(s), they don't offer anything to help process that information.
must-triage is a very small utility to aid in identifying problems recorded in must-gather output.
Currently, must-triage knows about the following potential issues:
- Unparseable YAML files
- Pods not in ready state
- Operators who haven't reported success
- YAML files which contain a top-level error field
- Empty JSON files
- Unparseable JSON files
- Ceph health not
HEALTH_OK - Logfiles which mention a panic
Teaching it to identify new issues is not difficult; PRs warmly welcomed!
pip install must-triage
must-triage /path/to/must-gather/output