diff --git a/cmd/orchestra/main.go b/cmd/orchestra/main.go index 7ee83ce..33724a5 100644 --- a/cmd/orchestra/main.go +++ b/cmd/orchestra/main.go @@ -49,6 +49,15 @@ func main() { } coordinator := &orchestrator.Coordinator{Store: s, Worktrees: orchestrator.GitWorktrees{Root: root, Repo: repo}, Adapters: orchestrator.AdapterFactory{Herdrs: adapters}} rt.OnLease = func(e domain.Event) error { return coordinator.Start(context.Background(), e) } + hard := 0.75 + if v, parseErr := strconv.ParseFloat(os.Getenv("ORCHESTRA_OCCUPANCY_HARD"), 64); parseErr == nil && v > 0 && v < 1 { + hard = v + } + go func() { + if monitorErr := coordinator.Monitor(context.Background(), hard, 30*time.Second); monitorErr != nil { + log.Printf("orchestrator monitor: %v", monitorErr) + } + }() } } mux := http.NewServeMux() diff --git a/internal/orchestrator/orchestrator.go b/internal/orchestrator/orchestrator.go index a897dc2..876c5ec 100644 --- a/internal/orchestrator/orchestrator.go +++ b/internal/orchestrator/orchestrator.go @@ -14,6 +14,7 @@ import ( "os/exec" "path/filepath" "sync" + "time" ) type Worktrees interface { @@ -68,6 +69,61 @@ type Coordinator struct { sessions map[string]herdr.Session } +// Monitor performs conservative hard-threshold rotation. The adapter owns +// the handoff creation; the coordinator only publishes its content address +// and frees the lease for pickup by the router. +func (c *Coordinator) Monitor(ctx context.Context, hard float64, interval time.Duration) error { + if interval <= 0 { + interval = 30 * time.Second + } + t := time.NewTicker(interval) + defer t.Stop() + for { + select { + case <-ctx.Done(): + return ctx.Err() + case <-t.C: + c.rotate(ctx, hard) + } + } +} +func (c *Coordinator) rotate(ctx context.Context, hard float64) { + c.mu.Lock() + sessions := make(map[string]herdr.Session, len(c.sessions)) + for id, s := range c.sessions { + sessions[id] = s + } + c.mu.Unlock() + for taskID, session := range sessions { + task, ok := c.Store.Task(taskID) + if !ok || task.State != domain.StateLeased { + continue + } + a, err := c.Adapters.Adapter(session.Harness) + if err != nil { + continue + } + occupancy, err := a.Occupancy(session) + if err != nil || occupancy < hard { + continue + } + ref, err := a.Release(ctx, session) + if err != nil { + continue + } + if ref == "" { + continue + } + b, _ := json.Marshal(map[string]string{"handoff_ref": ref, "reason": "threshold"}) + e := domain.Event{ID: domain.NewID(), Type: "TaskReleased", TaskID: taskID, Version: task.Version + 1, Payload: b} + if c.Store.Append(e) == nil { + c.mu.Lock() + delete(c.sessions, taskID) + c.mu.Unlock() + } + } +} + func (c *Coordinator) Start(ctx context.Context, e domain.Event) error { if e.Type != "TaskLeased" { return nil diff --git a/progress.md b/progress.md index f54839b..dfb2966 100644 --- a/progress.md +++ b/progress.md @@ -7,22 +7,22 @@ Updated: 2026-07-26 `go test ./...` passes, but the implementation is still a tested substrate/router prototype rather than a functioning unattended multi-harness orchestra. The following gaps were verified against `orchestra-spec (1).md` and the current code: - **Harness execution is wired for configured Git/Codex deployments.** The router invokes the coordinator; `ORCHESTRA_REPO` + `ORCHESTRA_WORKTREE_ROOT` enable Git worktree creation, configured herdr socket adapters, session creation, and optional bootstrap. Other harness types and monitoring callbacks remain pending. -- **Rotation is not implemented.** There is no turn-boundary callback, herdr event subscription, occupancy-triggered rotation, milestone/thrash trigger, or split-then-close coordinator. +- **Rotation is partially implemented.** The configured coordinator polls adapter occupancy, releases above `ORCHESTRA_OCCUPANCY_HARD` (default 75%), and publishes a handoff-backed `TaskReleased` event. Turn-boundary callbacks, milestone/thrash triggers, and split-then-close safety remain pending. - **Lifecycle API payloads are invalid.** The release, complete, and block endpoints all emit `{"source":"api"}`, while validation requires `handoff_ref` or `reason`, `report_ref`, and `blocker` respectively. The documented lifecycle endpoints therefore cannot complete successfully. - **Provider integrations are partially wired.** JSONL watching and optional Gitea webhook/poll loops now start from environment configuration, but terminal reflection, provider health, cancellation, and delivery fan-out remain absent. - **CAS references are not content-verified at event append.** Lifecycle events check that referenced files exist, but do not verify that the file content hashes to the supplied reference. - **Replay bypasses event validation.** Startup replay unmarshals and applies events without validating the event envelope, payload schema, or sequence/version invariants. - **Snapshots are written but never loaded or used for replay acceleration.** Startup always replays the complete event log. - **Task creation projection is incomplete.** `parent`, `due`, `inherent_priority`, and `estimate` are defined in the domain model but are not projected from `TaskCreated` payloads. -- **Occupancy support is incomplete relative to the spec.** Native readers exist, but Codex active-session discovery, opencode server/SSE plus fallback, and coordinator monitoring are not implemented. +- **Occupancy support is incomplete relative to the spec.** Native readers and configured hard-threshold monitoring exist, but Codex active-session discovery, opencode server/SSE plus fallback, and turn-boundary monitoring are not implemented. - **Authorization is mostly enforced at HTTP ingress.** Lifecycle and approval handlers now call `AuthorizeEvent`; an absent surface still defaults to full-control Web, and non-HTTP/event-bus integrations remain unwired. The first pass closed the store/API defects (lifecycle defaults, CAS content verification, validated replay, snapshot loading, and projection of task metadata) and added optional Gitea webhook/poll wiring. The remaining server-side gaps are below. ### Remaining server-side gaps -- **The orchestration coordinator is operational but incomplete.** It is constructed from deployment configuration, resolves a worktree, invokes `herdr.Adapter.Lease`, bootstraps handoffs, and blocks failed starts. Session mappings are in-memory and turn/lifecycle monitoring is still pending. -- **Rotation is still absent.** No adapter turn-boundary callback, occupancy trigger, milestone/thrash trigger, handoff save/validate flow, split-then-close sequence, or lease transfer coordinator is wired into the server. The existing occupancy readers and anchor validator are standalone library primitives. +- **The orchestration coordinator is operational but incomplete.** It is constructed from deployment configuration, resolves a worktree, invokes `herdr.Adapter.Lease`, bootstraps handoffs, monitors occupancy, and rotates through handoff references. Session mappings are in-memory and turn-boundary monitoring is still pending. +- **Rotation remains incomplete.** Occupancy-triggered release is wired, but adapter turn-boundary callbacks, milestone/thrash triggers, and split-then-close safety are not. - **Harness discovery and registration are not operational.** Static herdr configuration and socket clients exist, but startup does not create adapters, ping configured herdrs, discover active Codex sessions, subscribe to opencode SSE, or run the required fallback/TTL monitoring loop. - **Provider ingestion is only partially wired.** JSONL and Gitea are available when configured, but there is no provider lifecycle management, cancellation, error health projection, terminal-state reflection, or provider fan-out. - **Lifecycle event contracts remain incomplete.** Validation does not enforce the spec's `expected_version`, `ttl`, `anchor_sha`, `receipt`, or optional `handoff_ref` relationships, and the HTTP API does not validate actor/surface authorization at the event construction site. Completion without a report currently creates a generated placeholder artifact rather than requiring the stop-hook/wrapper receipt described by the spec.