Documentation Index

Fetch the complete documentation index at: https://docs.kindling.systems/llms.txt

Use this file to discover all available pages before exploring further.

​

Production foundations acceptance

Use this checklist to treat the Operator Controls and Recovery slice as complete enough to move on to broader multi-node reliability testing.

​

Control-plane backup and disaster recovery

• docs/control-plane-backup-and-dr.md is the team’s agreed contract for what is in Postgres vs volumes vs host-local state.
• Operators can describe RPO/RTO for the Postgres tier using their own backup tooling (Kindling does not ship scheduled control-plane backups).
• A documented restore ordering exists: Postgres → DSN files → kindling serve → validation (see Backup and restore).
• Volume backup is not mistaken for control-plane backup (see Secrets and volumes).

​

Audit logging (cluster-global admin)

• cluster_audit_events exists after migrate and receives rows for:
  • server.drain / server.activate
  • cluster.settings.update (persisted keys changed listed in details)
  • auth.provider.update (no secrets in details)
• Design doc docs/cluster-audit-events.md matches the implemented action names and redaction rules.
• Operators can query recent events via SQL (read API/UI deferred per design).

​

Runbooks

• Operations runbook links install, upgrade, backup, drain, and audit inspection.
• Backup and restore distinguishes Postgres procedures from volume procedures.
• Production Setup references the new operate docs in its documentation map.

​

Optional drill

• Restore Postgres to a non-production environment and confirm Kindling starts and passes the multi-server validation checklist in docs/high-availability.md.