Skip to main contentCrate orca_control
Source - alerts
- AI alert pipeline integration for the control plane.
- api
- auth
- Bearer token authentication middleware with role-based access control.
- backup_scheduler
- Scheduled backup runner: spawns a background task that runs backups on a cron schedule.
- cleanup_scheduler
- Scheduled cleanup: docker system prune on all nodes + registry GC on master.
- cluster_api
- API endpoints for multi-node cluster management.
- cluster_state
- Cluster-wide state backed by Raft consensus.
- deploy_history
- In-memory deploy history for rollback support.
- health
- Background health checker for service instances.
- metrics
- Prometheus metrics endpoint.
- proto
- Generated gRPC types from
proto/orca.proto. - raft
- Raft consensus layer for multi-node cluster coordination.
- reconciler
- Reconciler: ensures actual running containers/wasm instances match desired service config.
- routes
- Routing table management for container and Wasm workloads.
- scheduler
- state
- Shared application state for the control plane.
- stats
- Background container stats collector.
- store
- Persistent cluster store backed by redb.
- topo_sort
- Topological sorting for service dependency ordering.
- watchdog
- Background watchdog that periodically reconciles degraded services.
- webhook
- Webhook handler for GitHub/Gitea/GitLab push events.
- webhook_invocations
- Webhook invocation tracking — per-service ring buffer of recent pushes,
plus the
GET /api/v1/webhooks/{id}/invocations handler that exposes it. - ws_handler
- WebSocket handler for agent↔master streaming communication.
- run_server
- Start the orca control plane (API server).
- run_server_with_acme
- Start the orca control plane with optional ACME hot-provisioning.