Review production readiness · 0.5.x

Use ProductionReady=True as an operator-known configuration gate, then prove the controls that the operator cannot observe: cloud permissions, client access, alert delivery, restore usability, and your response process.

Meet the operator-known gate

For a Hardened cluster, ProductionReady=True requires the installed admission guardrails and the configured security posture to pass the controller’s checks. Depending on the configuration, those checks include:

non-static unseal and a usable provider identity; plugin-backed KMS also requires a matching declared KMS plugin
External or ACME TLS, with required integration and shared-cache conditions
self-initialization instead of a stored root-token bootstrap
acceptable network, audit storage, edge integration, and workload security settings

Read the condition before routing traffic:

verify

Inspect the production gate

kubectl -n <namespace> get openbaocluster <name> \
  -o jsonpath='{range .status.conditions[?(@.type=="ProductionReady")]}{.status}{" "}{.reason}{"\n"}{.message}{"\n"}{end}'

Prove the operating controls

Before production traffic

Set an availability objective and alert on Available, Degraded, OpenBaoSealed, and OpenBaoLeader.
Monitor storage capacity, PVC health, restart loops, reconciliation errors, certificate expiry, and unseal dependencies.
Run a scheduled backup, confirm the object in storage, and restore it into an isolated target.
Record the backup identity, retention owner, recovery point objective, and recovery time objective.
Test the human authentication path and the emergency-access custody process without using a root token for routine work.
Pin an upgrade strategy, review the compatibility matrix, and document the rollback boundary.
Confirm at least three voter replicas for a production Raft quorum and distribute them across failure domains.
Verify that a node drain respects the managed PodDisruptionBudget.

Capture a baseline

verify

Record cluster, workload, and Raft state

kubectl -n <namespace> get openbaocluster <name> -o yaml
kubectl -n <namespace> get pods,pvc,pdb -l openbao.org/cluster=<name> -o wide
kubectl -n <namespace> exec <pod-name> -- bao status
kubectl -n <namespace> exec <pod-name> -- bao operator raft list-peers

Keep this evidence with the service runbook. Recheck it after an upgrade, maintenance window, restore rehearsal, or security-boundary change.

The Raft command requires an authenticated OpenBao session with configuration read access. Use your approved interactive login path; do not pass a privileged token on the command line.

Next, configure backups and review the compatibility matrix.