Day-2 operations
Preview. These pages are for operators of P6 on Kubernetes (Kubernetes and OpenShift), the first path planned for Stable. Facts come from the manifests,
setup.shand the code as of 2026-10-13. Where something is not documented anywhere in the repository, the page says so rather than guessing. Single-VM and Compose notes appear only where they differ.
| Page | Covers |
|---|---|
| Upgrades | Re-running setup.sh, image pinning, rollout behaviour, version skew, rollback |
| Redis backup and recovery | What lives in Redis, what a loss costs, which secrets to back up separately, key rotation |
| Observability | Health endpoints, the JSON metrics, probes, log lines, the audit stream |
| Sizing and scaling | Resource defaults, workers and turns per worker, replicas, sandbox capacity |
| Hardening | A checklist per deployment path, from the security model |
| Troubleshooting | Symptom, cause, check and fix |
Other deployment paths​
The pages above are written for Kubernetes. For the single VM and the sandbox service, the same day-2 tasks are on their own pages:
| Task | Single VM | Sandbox service |
|---|---|---|
| Upgrade | Upgrading an existing VM | Upgrading |
| Check health and status | Checking it | Status and the audit lines |
| Redis and its data | Redis persistence | Uninstalling (what is kept, what is deleted) |
| Reboots and host outages | Reboots | A KVM-host outage |
| Troubleshooting | The units' start failures | Reproducing a member's issue |
| Limits | What round one does not claim | Limits operators must know |
Gaps you should know about​
These are properties of the current deployment, not documentation omissions:
- Single points of failure. Redis, the relay, the control plane and (by default) the supervisor each run one replica, and there is no PodDisruptionBudget or anti-affinity, so a node drain can take them all at once.
- No tested backup or rollback procedure exists in the repository.
- Sessions have no TTL and per-session Redis memory has not been measured, while Redis is limited to 256Mi.
/metricsis JSON, not Prometheus format, and the relay and sandboxes have no health endpoints.- A drain protects only short turns (about 15 to 20 seconds), although the pod's grace period is 120 seconds.
- No capacity numbers are published. The worker and turn defaults are trial values, not measurements.