// Package daemon is archied's resident loop: poll GitHub for labelled // issues, enqueue them, and process tasks through their routed workflows. // State lives in the store; the daemon is restartable at any point. package daemon import ( "context" "encoding/json" "errors" "fmt" "log/slog" "os" "strings" "sync" "time" "github.com/samcharles93/archie-core/internal/agentexec" "github.com/samcharles93/archie-core/internal/config" "github.com/samcharles93/archie-core/internal/container" "github.com/samcharles93/archie-core/internal/domain/binding" "github.com/samcharles93/archie-core/internal/domain/curator" "github.com/samcharles93/archie-core/internal/domain/mapping" "github.com/samcharles93/archie-core/internal/domain/storecontract" "github.com/samcharles93/archie-core/internal/domain/workflow" "github.com/samcharles93/archie-core/internal/domain/workintake" "github.com/samcharles93/archie-core/internal/eventbus" "github.com/samcharles93/archie-core/internal/events" "github.com/samcharles93/archie-core/internal/forge" agentnats "github.com/samcharles93/archie-core/internal/infrastructure/agenttransport/nats" "github.com/samcharles93/archie-core/internal/logging" "github.com/samcharles93/archie-core/internal/plugin" "github.com/samcharles93/archie-core/internal/storage" "github.com/samcharles93/archie-core/internal/taskrun" "github.com/samcharles93/archie-core/internal/tools" "github.com/samcharles93/archie-core/internal/worktree" ) // TaskBus is the messaging the daemon needs: announce discovered work, // collect it back, and ask a worker to run a task. // // Declared here rather than taking eventbus.Bus because a domain defines the // smallest interface required to do its work -- the daemon never subscribes // or creates reply inboxes. type TaskBus interface { PublishUnique(ctx context.Context, subject, idempotencyKey string, payload []byte) error Fetch(ctx context.Context) (eventbus.Message, error) Request(ctx context.Context, subject string, payload []byte) ([]byte, error) } // WorktreeGrantIssuer gives one container dispatch the narrow authority to // publish its already-prepared branch. The daemon owns grant lifetime; the // transport implementation owns token mechanics. type WorktreeGrantIssuer interface { Issue(task *workflow.Task) (token string, revoke func(), err error) } // StateStoreGrantIssuer gives one container dispatch a State Store credential // scoped to that task's own workflow.Store RPCs (Update, Transition, // InsertEvent) -- never the daemon's own administrative credential. The // daemon owns grant lifetime (issue before Acquire, revoke on Release); the // transport implementation (staterpc.GrantIssuer) owns token mechanics. type StateStoreGrantIssuer interface { Issue(task *workflow.Task) (token string, revoke func(), err error) } // NATSEndpoint is the broker address and credential archied connected with at // startup. It is runtime state rather than configuration: embedded mode // generates both values, and external secret references are resolved once. type NATSEndpoint struct { URL string Token string } // StateStoreEndpoint is the State Store gRPC target (and the bearer token the // daemon presents, when one is configured) that containerEnv injects as // STATE_STORE_URL / STATE_STORE_TOKEN so archie-agent authenticates to the // same remote archie-state-store service the daemon dials // (docs/prds/state-store-contract.md §6, §9, §10). After the in-process // serving path is deleted (docs/prds/state-store-contract.md §12 step 7) this // is always the configured [services.state].target, never a daemon-owned // listener. type StateStoreEndpoint struct { URL string Token string } type Daemon struct { // Cfg publishes the running configuration. Read through Cfg() so a // reload swaps the published snapshot atomically. See config.Holder. Cfg *config.Holder // ConnectedNATS is the endpoint and credential the daemon's own client // connected with at startup. Container env is built from this, not from // live config or environment, so reload cannot point new containers at a // broker the daemon is not publishing on. The zero value is invalid in // production composition and remains useful only to fail closed in tests. ConnectedNATS NATSEndpoint // ConnectedStateStore is the State Store gRPC target (and token) the // daemon's own client dialed at startup. containerEnv injects these as // STATE_STORE_URL / STATE_STORE_TOKEN so archie-agent reaches the same // remote archie-state-store service; it is always the configured // [services.state].target after the in-process serving path is deleted. ConnectedStateStore StateStoreEndpoint Store storecontract.TaskStore // Mappings persists payload field mappings (docs/prds/payload-field-mapping.md). // Used by the binding dispatch loop to resolve capture bodies against // the mapping a binding names. Optional: nil disables the binding // dispatch loop (legacy behaviour). Mappings storecontract.MappingStore // Bindings persists playbook bindings (docs/prds/playbook-binding.md). // Optional: nil disables the binding dispatch loop (legacy behaviour). Bindings storecontract.BindingStore // BindingDispatcher is the dispatch-time helper surface for bindings: // list undispatched captures, look up armed bindings by source for the // matcher, and write the at-most-once dedup ledger row. Split from // BindingStore to keep the CRUD interface under the interfacebloat // limit. Optional: nil disables the dispatch loop. BindingDispatcher storecontract.BindingDispatcher // BindingTaskCreator enqueues a task triggered by a binding. Split // off TaskLifecycle so the lifecycle surface stays narrow and the // dispatch loop does not acquire the full task-creation contract. // Optional: nil disables the dispatch loop. BindingTaskCreator storecontract.BindingTaskCreator Forge forge.Forge Trees *worktree.Manager Bus *events.Bus Log *slog.Logger // Tasks is the NATS task distribution bus: the poller publishes // discovered work over it and runViaAgent requests execution through it. // NATS startup is mandatory (there is no broker-off execution mode), so // production composition always sets it. Nil appears only in tests and // is handled fail-closed at the publish/run sites. Tasks TaskBus WorktreeGrants WorktreeGrantIssuer // StateStoreGrants issues per-task State Store credentials for container // env (see containerEnv). Required whenever ConnectedStateStore.URL is // set -- acquireTaskContainer parks the task rather than fall back to // forwarding the daemon's own administrative token. StateStoreGrants StateStoreGrantIssuer // TaskRunReadyTimeout bounds how long runViaAgent retries an initial // taskrun request that fails with nats.ErrNoResponders, giving a // freshly spawned archie-agent container time to connect to NATS, set // up its task-scoped core-NATS subscription, and become reachable before the daemon // gives up and parks the task. ContainerPool.Acquire returns as soon // as Docker has issued the start syscall, not once that setup // finishes, so without this bound the very first request after a // container spawn fails deterministically. Zero uses // defaultTaskRunReadyTimeout. TaskRunReadyTimeout time.Duration // TaskRunRetryBackoff is the delay between retry attempts within // TaskRunReadyTimeout. Zero uses defaultTaskRunRetryBackoff. TaskRunRetryBackoff time.Duration // AgentStatus records the version/install-type the most recently // completed archie-agent task response reported about itself, for // releaseupdate.Report.Verify to check an update claim against. Nil // disables observation -- runViaAgent skips the Observe call rather // than reporting the agent as unverifiable via a zero value, which // would read as a checked "unknown" rather than "never wired". AgentStatus *AgentStatus // ContainerPool manages Docker container lifecycle. Nil means autonomous // execution is unavailable and process parks tasks; every runnable task gets // a fresh container. ContainerPool *container.Pool // Storage is the pluggable storage backend for container mounts. // A runnable task requires it; acquireTaskContainer parks on nil. Storage storage.Backend // CapabilityHost owns validated plugin manifests and cross-family // lifecycle. Typed capability registries remain in their domain packages; // the host never exposes daemon internals or an untyped service locator. CapabilityHost *plugin.Host // Curators is the curator engine family registry (epic archie-core-yp9). // The runtime loop (archie-core-89x) and reference curators (i7i, gs8) // are driven through it; nil when the family is not composed. Curators *curator.Registry // Identities holds per-identity runner state for multi-identity mode. // When non-empty, Run() starts one goroutine per identity instead of // using the single-identity Forge/Trees/Cfg.Repos path. Identities []*IdentityRunner // Guardrails is the tool-call guardrail engine, wired by the composition // root. When non-nil, tool successes and failures are recorded and // warnings/hard-stops are issued per the configured thresholds. Nil // means guardrails are disabled (backward compatible). Guardrails *tools.GuardrailEngine // KindWorkflows and LabelWorkflows are the resolved kind/label -> // workflow-name routing bindings loaded at startup (WorkflowRoutingFile, // WorkflowLabelsFile, PlaybookDirs). They travel in taskrun.Request so // the archie-agent process that calls workflow.Route applies the same // bindings the daemon loaded; nil means built-in defaults. KindWorkflows workflow.KindWorkflows LabelWorkflows workflow.LabelWorkflows // ToolRegistry is the central tool registry, wired by the composition // root. MCP-discovered tools and built-in tools are registered here // and passed as CaptureTools in agent requests. Nil means no dynamic // tool discovery (backward compatible). ToolRegistry *tools.Registry // TaskLogs persists each task's own log output (including a sandboxed // container's, which otherwise disappears at AutoRemove) and mirrors it // live onto the dashboard feed, wired by the composition root. Nil // disables task logging (backward compatible); every TaskRegistry // method is nil-receiver-safe, so process() calls it unconditionally. TaskLogs *logging.TaskRegistry // running holds a cancel function for every task currently executing, // so /stop can reach work already in flight. Its zero value is ready // to use. running runningTasks } // IdentityRunner bundles identity-specific state for a single agent // identity running within a multi-identity daemon. Each identity gets its // own forge client, worktree manager, repo list, and config -- but shares // the store, NATS connection, container pool, and event bus with siblings. type IdentityRunner struct { Name string Forge forge.Forge Trees *worktree.Manager Repos []config.Repo // Cfg is the identity-scoped config subset (forge, models, dispatch, // budgets, etc.). Cfg config.IdentityConfig Log *slog.Logger } // NewIdentityRunner constructs an IdentityRunner from an identity config // and a pre-built forge client. The caller owns forge and trees lifecycle; // IdentityRunner just holds references. func NewIdentityRunner(ctx context.Context, idCfg config.IdentityConfig, fg forge.Forge, trees *worktree.Manager, log *slog.Logger) (*IdentityRunner, error) { if idCfg.Name == "" { return nil, fmt.Errorf("identity name is required") } return &IdentityRunner{ Name: idCfg.Name, Forge: fg, Trees: trees, Repos: idCfg.Repos, Cfg: idCfg, Log: log.With("identity", idCfg.Name), }, nil } // emit publishes an observability event; safe on a nil bus. func (d *Daemon) emit(e events.Event) { if d.Bus != nil { d.Bus.Publish(e) } } // Startup runs crash recovery and access verification once. func (d *Daemon) Startup(ctx context.Context) error { d.cleanupExpiredStorage(ctx) n, err := d.Store.RecoverStale(ctx) if err != nil { return err } if n > 0 { d.Log.Info("re-queued tasks left running by a previous daemon", "count", n) } if err := d.Forge.AcceptInvitations(ctx); err != nil { d.Log.Warn("invitation sweep failed", "err", err) } for _, r := range d.Cfg.Get().Repos { if err := d.Forge.VerifyPush(ctx, r.Owner, r.Name); err != nil { d.Log.Warn("repo not pushable -- tasks from it will fail", "repo", r.FullName(), "err", err) } } return nil } // Run polls until the context ends. When Identities is non-empty, each // identity gets its own goroutine with its own forge client, repo list, // and poll interval -- failure isolation between identities (one identity's // forge outage doesn't block another's poll cycle). func (d *Daemon) Run(ctx context.Context) error { if len(d.Identities) > 0 { return d.runIdentities(ctx) } interval := d.Cfg.Get().PollInterval.Std() ticker := time.NewTicker(interval) defer ticker.Stop() for { d.Cycle(ctx) select { case <-ctx.Done(): return ctx.Err() case <-ticker.C: // PollInterval may have been reloaded; reset the ticker so a // change takes effect on the next cycle. if iv := d.Cfg.Get().PollInterval.Std(); iv != interval { interval = iv ticker.Reset(iv) } } } } // runIdentities starts one goroutine per identity plus one shared // maintenance-and-drain loop. // // Polling is identity-scoped: each identity goroutine polls only its own // repos via its own forge client and enqueues under its own name. It never // drains or reconciles. Maintenance and draining are store-wide, not // identity-scoped, so they run once in a single shared loop -- that keeps // the global max_concurrency cap and the per-repo serialization intact // across identities (N per-identity dispatchers would multiply the cap and // split the same-repo lock), and it ensures reconcilePRs and storage // cleanup actually run in multi-identity mode. func (d *Daemon) runIdentities(ctx context.Context) error { ctx, cancel := context.WithCancel(ctx) defer cancel() var wg sync.WaitGroup // One poll loop per identity. Polling is the only identity-scoped // work; it never executes tasks. for _, id := range d.Identities { wg.Add(1) go func(id *IdentityRunner) { defer wg.Done() interval := d.Cfg.Get().PollInterval.Std() if id.Cfg.PollInterval > 0 { interval = id.Cfg.PollInterval.Std() } ticker := time.NewTicker(interval) defer ticker.Stop() for { d.pollForIdentity(ctx, id) select { case <-ctx.Done(): return case <-ticker.C: // The identity's own override is frozen (identity config is // startup-built); the root fallback may have been reloaded. iv := d.Cfg.Get().PollInterval.Std() if id.Cfg.PollInterval > 0 { iv = id.Cfg.PollInterval.Std() } if iv != interval { interval = iv ticker.Reset(iv) } } } }(id) } // One shared maintenance-and-drain loop. wg.Go(func() { interval := d.Cfg.Get().PollInterval.Std() ticker := time.NewTicker(interval) defer ticker.Stop() for { d.maintainAndDrain(ctx) select { case <-ctx.Done(): return case <-ticker.C: // PollInterval may have been reloaded; reset the ticker so a // change takes effect on the next cycle. if iv := d.Cfg.Get().PollInterval.Std(); iv != interval { interval = iv ticker.Reset(iv) } } } }) wg.Wait() return ctx.Err() } // pollForIdentity polls one identity's repos via its own forge client, // enqueuing any discovered issues under that identity's name. It never // drains or reconciles -- those are store-wide and run in the shared // maintainAndDrain loop. func (d *Daemon) pollForIdentity(ctx context.Context, id *IdentityRunner) { cfg := configForIdentity(d.Cfg.Get(), id.Cfg) for _, repo := range id.Repos { issues := d.pollIssuesWithConfig(ctx, id.Forge, cfg, repo) for _, is := range issues { labels := strings.Join(is.Labels, ",") if d.Tasks != nil { d.pollNATS(ctx, id.Forge, cfg, repo, is, labels, id.Name) } } } } // maintainAndDrain runs the store-wide passes of a multi-identity cycle: // storage cleanup, PR reconciliation, then task dispatch. A single shared // loop runs this so the global concurrency cap and per-repo serialization // hold across identities. func (d *Daemon) maintainAndDrain(ctx context.Context) { d.cleanupExpiredStorage(ctx) d.reconcilePRs(ctx) d.dispatchBindings(ctx) if d.Tasks != nil { d.drainNATS(ctx) } } // pollIssuesWithConfig discovers work for one repo using the given forge // client and dispatch config. Single-identity mode passes d.Forge/d.Cfg; // identity poll loops pass the identity's own forge and configForIdentity, // so each identity polls with its own bot user, label, and trigger. func (d *Daemon) pollIssuesWithConfig(ctx context.Context, fg forge.Forge, cfg config.Config, repo config.Repo) []forge.Issue { switch cfg.Dispatch.Trigger { case "label": // Empty-label rule shared with workintake.MatchesDispatch: a label // trigger needs a non-empty label or it would match every open issue. if workintake.RequiresLabel(cfg.Dispatch.Trigger) && cfg.Label == "" { d.Log.Error("label trigger configured with an empty label; refusing to poll (an empty label matches every open issue)", "repo", repo.FullName()) return nil } issues, err := fg.IssuesWithLabel(ctx, repo.Owner, repo.Name, cfg.Label) if err != nil { d.Log.Error("label poll failed", "repo", repo.FullName(), "err", err) return nil } return issues case "either": return d.pollEitherWithConfig(ctx, fg, cfg, repo) default: // "assignee" (default) issues, err := fg.AssignedIssues(ctx, repo.Owner, repo.Name, cfg.BotUser) if err != nil { d.Log.Error("poll failed", "repo", repo.FullName(), "err", err) return nil } return issues } } func (d *Daemon) pollEitherWithConfig(ctx context.Context, fg forge.Forge, cfg config.Config, repo config.Repo) []forge.Issue { seen := map[int]bool{} var out []forge.Issue assigned, err := fg.AssignedIssues(ctx, repo.Owner, repo.Name, cfg.BotUser) if err != nil { d.Log.Error("assigned poll failed", "repo", repo.FullName(), "err", err) } else { for _, is := range assigned { seen[is.Number] = true out = append(out, is) } } if workintake.RequiresLabel(cfg.Dispatch.Trigger) && cfg.Label == "" { d.Log.Error("either trigger configured with an empty label; skipping the label poll (an empty label matches every open issue)", "repo", repo.FullName()) return out } labelled, err := fg.IssuesWithLabel(ctx, repo.Owner, repo.Name, cfg.Label) if err != nil { d.Log.Error("label poll failed", "repo", repo.FullName(), "err", err) } else { for _, is := range labelled { if !seen[is.Number] { out = append(out, is) } } } return out } // Cycle is one poll-and-drain pass: enqueue newly assigned issues, // reconcile open PRs, run the binding dispatch loop, then process // queued tasks concurrently up to containers.max_concurrency. func (d *Daemon) Cycle(ctx context.Context) { d.cleanupExpiredStorage(ctx) d.poll(ctx) d.reconcilePRs(ctx) d.dispatchBindings(ctx) if d.Tasks != nil { d.drainNATS(ctx) } } func (d *Daemon) cleanupExpiredStorage(ctx context.Context) { ttl := d.Cfg.Get().Containers.VolumeTTL.Std() if ttl <= 0 { return } if d.Storage != nil { n, err := d.Storage.CleanupExpired(ctx, ttl) if err != nil { d.Log.Warn("persistent volume cleanup failed", "err", err) } if n > 0 { d.Log.Info("expired persistent volumes removed", "count", n) } } } // bindingDispatchBatchLimit caps how many undispatched captures a single // dispatchBindings pass will fetch from the store. The cap is defensive // (the captures table is also pruned by retention/maxEvents) and bounds a // single cycle's worst-case latency on a backlog. const bindingDispatchBatchLimit = 100 // dispatchBindings walks authenticated captures whose source has at // least one armed binding, evaluates each binding's matcher against // the capture, resolves fields through the binding's mapping, and // enqueues a task per matched (binding, capture) pair. // // The binding_dispatches ledger is the at-most-once guarantee across // cycles and daemon restarts: RecordDispatch returns // storecontract.ErrAlreadyDispatched on a duplicate (binding_id, capture_id) // pair, and dispatchOneBinding treats that as a normal "another cycle // raced us" outcome rather than an error. // // Nil Bindings, BindingDispatcher or BindingTaskCreator disables the // loop (legacy behaviour: daemons built before t2db.4 Phase E had no // playbook bindings, and each field's own doc comment promises this // degrade-not-panic contract independently). func (d *Daemon) dispatchBindings(ctx context.Context) { if d.Bindings == nil || d.BindingDispatcher == nil || d.BindingTaskCreator == nil { return } // Gather armed sources first so the captures query filters down to // only relevant rows -- the table-level view over captured_events // is otherwise unbounded, and a backlog of unrelated sources // should never be scanned on a cycle. bindings, err := d.Bindings.ListBindings(ctx) if err != nil { d.Log.Warn("list bindings failed", "error", err) return } sources := make([]string, 0, len(bindings)) for _, b := range bindings { if b.Status == binding.StatusArmed && b.Matcher.Source != "" { sources = append(sources, b.Matcher.Source) } } if len(sources) == 0 { return } captures, err := d.BindingDispatcher.ListUndispatchedCaptures(ctx, sources, bindingDispatchBatchLimit) if err != nil { d.Log.Warn("list undispatched captures failed", "error", err) return } for _, c := range captures { armed, err := d.BindingDispatcher.ArmedBindingsForSource(ctx, c.Source) if err != nil { d.Log.Warn("armed bindings lookup", "source", c.Source, "error", err) continue } for _, b := range armed { if !b.Matcher.Matches(c.Source, c.Authenticated) { continue } d.dispatchOneBinding(ctx, b, c) } } } // dispatchOneBinding handles a single (binding, capture) match: resolve // fields through the binding's mapping, enqueue a task, and record the // dispatch in the dedup ledger. Each step degrades independently -- // a mapping lookup failure logs and returns, a required-field // resolution failure records a binding_dispatch_failure event and // returns without enqueueing, an enqueue failure logs and returns, and // a RecordDispatch returning ErrAlreadyDispatched is treated as a // benign race winner (another cycle already created the task). // // Owner/repo resolution currently falls back to the single configured // repo (when exactly one is configured); binding.Owner/Repo is a // follow-up so multi-repo deployments don't have to live with the // ambiguity. func (d *Daemon) dispatchOneBinding(ctx context.Context, b binding.Binding, c storecontract.CapturedEvent) { if d.Mappings == nil { d.Log.Warn("binding dispatch: mapping store unavailable", "binding", b.ID) return } m, err := d.Mappings.GetMapping(ctx, b.MappingID) if err != nil { d.Log.Warn("binding dispatch: get mapping", "binding", b.ID, "mapping", b.MappingID, "error", err) return } if m == nil { d.Log.Warn("binding dispatch: mapping missing", "binding", b.ID, "mapping", b.MappingID) return } values, failures := mapping.Resolve(m.Fields, []byte(c.Body)) if hasBlockingFailure(m.Fields, failures) { _, _ = d.Store.InsertEvent(ctx, events.Event{ Kind: "binding_dispatch_failure", Detail: fmt.Sprintf("binding %d (capture %d): required field failed", b.ID, c.ID), Data: map[string]any{ "binding_id": b.ID, "binding_version": b.Version, "capture_id": c.ID, "failures": failures, }, }) return } owner, repo, ok := d.resolveBindingRepo(b) if !ok { return // resolveBindingRepo already logged a warning } title := fmt.Sprintf("binding %s/%d from %s", b.Name, b.Version, c.Source) body := renderBindingBody(values, c) task, err := d.BindingTaskCreator.EnqueueBindingTask(ctx, owner, repo, title, body, b.Workflow, "", b.ID, b.Version) if err != nil { d.Log.Warn("binding dispatch: enqueue", "binding", b.ID, "capture", c.ID, "error", err) return } if err := d.BindingDispatcher.RecordDispatch(ctx, b.ID, int64(b.Version), c.ID, task.ID); err != nil { if errors.Is(err, storecontract.ErrAlreadyDispatched) { // Another cycle or daemon instance raced us and already // recorded the dispatch. The duplicate task remains // queued (delete-on-races is its own can of worms), // but the ledger prevents the same (binding, capture) // pair from ever being dispatched again. return } d.Log.Warn("binding dispatch: record", "binding", b.ID, "capture", c.ID, "error", err) return } } // hasBlockingFailure reports whether any of the failures belongs to a // required field. The mapping package reports failures only by field // name and reason -- the Field's Required flag is on the Field // declaration, so the dispatch loop re-derives it from the full // Field list. Failures on optional fields are warnings, not blocks: // the workflow still receives a partial value map. func hasBlockingFailure(fields []mapping.Field, failures []mapping.Failure) bool { if len(failures) == 0 { return false } required := make(map[string]bool, len(fields)) for _, f := range fields { if f.Required { required[f.Name] = true } } for _, fail := range failures { if required[fail.FieldName] { return true } } return false } // renderBindingBody builds the task body from the values a mapping // resolved, one "key=value" pair per line. Empty maps still produce // a body that records the source capture id so the workflow has // provenance to hand back to the operator. func renderBindingBody(values map[string]any, c storecontract.CapturedEvent) string { if len(values) == 0 { return fmt.Sprintf("(no fields resolved from capture %d)", c.ID) } parts := make([]string, 0, len(values)) for k, v := range values { parts = append(parts, fmt.Sprintf("%s=%v", k, v)) } return strings.Join(parts, "\n") } // resolveBindingRepo picks the owner/repo a binding-spawned task // operates against. A binding that pins Owner/Repo (binding.Binding's // Owner/Repo fields, validated as a complete pair or not set at all) // dispatches to that target unconditionally -- multi-repo deployments // use this to disambiguate. An unpinned binding falls back to the // single-configured-repo behaviour that predates the pin: with exactly // one configured repo the choice is unambiguous; with zero or more than // one the loop refuses to dispatch (logged below) so a binding cannot // silently fire against the wrong target. func (d *Daemon) resolveBindingRepo(b binding.Binding) (string, string, bool) { if b.Owner != "" && b.Repo != "" { return b.Owner, b.Repo, true } repos := d.Cfg.Get().Repos if len(repos) == 1 { return repos[0].Owner, repos[0].Name, true } d.Log.Warn("binding dispatch: cannot resolve target repo", "hint", "configure exactly one [[repos]] entry, or pin the binding to a specific owner/repo", "configured_repos", len(repos)) return "", "", false } // drainNATS processes tasks from NATS, falling back to SQLite ClaimNext for // requeued tasks (waiting_human approval, retry-parked) that didn't come // through a NATS publish. func (d *Daemon) drainNATS(ctx context.Context) { dispatcher := newTaskDispatcher(d.Cfg.Get().Containers.MaxConcurrency, d.allowConcurrentForTask) for ctx.Err() == nil { msg, err := d.Tasks.Fetch(ctx) if err != nil && !errors.Is(err, eventbus.ErrNoMessage) { d.Log.Error("nats fetch failed", "err", err) break } if err == nil { d.submitNATSTask(ctx, dispatcher, msg) continue } // ErrNoMessage means the stream is drained. That is precisely when // the SQLite fallback must run: requeued tasks (waiting_human // approval, retry-parked) never went through a NATS publish, so // nothing will ever arrive on the stream to trigger them. task, claimErr := d.Store.ClaimNext(ctx) if claimErr != nil { d.Log.Error("sqlite claim failed", "err", claimErr) break } if task == nil { break } dispatcher.Submit(ctx, task, d.process) } dispatcher.Wait() } // submitNATSTask decodes a fetched message and queues it for processing. A // message that cannot be decoded is acked rather than redelivered: it will // never decode on a retry, so leaving it unacked would block the queue. func (d *Daemon) submitNATSTask(ctx context.Context, dispatcher *taskDispatcher, msg eventbus.Message) { envelope, err := workintake.DecodeTask(msg.Data()) if err != nil { d.Log.Error("nats decode failed", "subject", msg.Subject(), "err", err) _ = msg.Ack() return } task := &workflow.Task{Owner: envelope.Owner, Repo: envelope.Repo, Identity: envelope.Identity} dispatcher.Submit(ctx, task, func(ctx context.Context, _ *workflow.Task) { d.processNATSTask(ctx, msg) }) } // taskDispatcher bounds task execution globally while preserving the default // one-task-per-repository safety rule. Submit order is retained within a repo; // tasks for different repos may run concurrently. Repos for which // allowConcurrent reports true opt out of the same-repo serialization // entirely -- the global slot limit is the only bound on their concurrency. type taskDispatcher struct { slots chan struct{} allowConcurrent func(task *workflow.Task) bool mu sync.Mutex repoTail map[string]chan struct{} wg sync.WaitGroup } func newTaskDispatcher(maxConcurrency int, allowConcurrent func(task *workflow.Task) bool) *taskDispatcher { var slots chan struct{} if maxConcurrency > 0 { slots = make(chan struct{}, maxConcurrency) } if allowConcurrent == nil { allowConcurrent = func(*workflow.Task) bool { return false } } return &taskDispatcher{ slots: slots, allowConcurrent: allowConcurrent, repoTail: make(map[string]chan struct{}), } } func (d *taskDispatcher) Submit( ctx context.Context, task *workflow.Task, process func(context.Context, *workflow.Task), ) { repo := task.Owner + "/" + task.Repo var previous, done chan struct{} if !d.allowConcurrent(task) { done = make(chan struct{}) d.mu.Lock() previous = d.repoTail[repo] d.repoTail[repo] = done d.mu.Unlock() } d.wg.Go(func() { if previous != nil { <-previous } if d.slots != nil { d.slots <- struct{}{} defer func() { <-d.slots }() } if done != nil { defer func() { close(done) d.mu.Lock() if d.repoTail[repo] == done { delete(d.repoTail, repo) } d.mu.Unlock() }() } process(ctx, task) }) } func (d *taskDispatcher) Wait() { d.wg.Wait() } func (d *Daemon) poll(ctx context.Context) { for _, repo := range d.Cfg.Get().Repos { issues := d.pollIssues(ctx, repo) for _, is := range issues { labels := strings.Join(is.Labels, ",") if d.Tasks != nil { d.pollNATS(ctx, d.Forge, d.Cfg.Get(), repo, is, labels, "") } } } } // pollNATS publishes discovered issues to NATS (new flow). func (d *Daemon) pollNATS(ctx context.Context, fg forge.Forge, cfg config.Config, repo config.Repo, is forge.Issue, labels, identity string) { // Read-only existence check prevents rediscovering an existing task. existing, err := d.Store.TaskByIssue(ctx, repo.Owner, repo.Name, is.Number) if err != nil { d.Log.Error("task lookup failed", "repo", repo.FullName(), "issue", is.Number, "err", err) return } if existing != nil { return } // Classify and encode here: the envelope, its routing kind and its // subject belong to the work-intake domain, not to the transport. parsed := workintake.SplitLabels(labels) task := workintake.TaskEnvelope{ Owner: repo.Owner, Repo: repo.Name, Number: is.Number, Title: is.Title, Body: is.Body, Labels: parsed, Identity: identity, Kind: workintake.KindForLabels(parsed), } if err := d.PublishTask(ctx, task); err != nil { d.Log.Error("task publish failed", "repo", repo.FullName(), "issue", is.Number, "err", err) return } d.acknowledge(ctx, fg, cfg, repo, is) } // PublishTask validates, encodes and publishes a task envelope. The // idempotency key means rediscovering the same issue on a later poll does not // enqueue the work twice. It is the single enqueue path shared by the poller // and the forge webhook intake, so neither can drift in how a discovered issue // becomes a task. func (d *Daemon) PublishTask(ctx context.Context, task workintake.TaskEnvelope) error { if d.Tasks == nil { return fmt.Errorf("publish task %s: no task bus configured", task.Ref()) } if err := task.Kind.Validate(); err != nil { return fmt.Errorf("publish task %s: %w", task.Ref(), err) } payload, err := task.Encode() if err != nil { return err } return d.Tasks.PublishUnique(ctx, task.Subject(), task.IdempotencyKey(), payload) } // acknowledge posts the pickup reaction and queued event // using fg -- the forge client that owns repo (identity-scoped or root) -- // and the ack reaction from cfg, the dispatch config that owns this poll. func (d *Daemon) acknowledge(ctx context.Context, fg forge.Forge, cfg config.Config, repo config.Repo, is forge.Issue) { d.Log.Info("issue queued", "repo", repo.FullName(), "issue", is.Number, "title", is.Title) if ack := cfg.Dispatch.AckReaction; ack != "" { if err := fg.React(ctx, repo.Owner, repo.Name, is.Number, ack); err != nil { d.Log.Warn("ack reaction failed", "issue", is.Number, "err", err) } } d.emit(events.Event{ Kind: events.KindTaskQueued, Repo: repo.FullName(), Issue: is.Number, Detail: is.Title, }) } // processNATSTask decodes a NATS message, writes it to SQLite, claims it, // and runs the workflow. The message is ack'd on terminal (park is a valid // outcome). Nak on transient errors so NATS redelivers. func (d *Daemon) processNATSTask(ctx context.Context, msg eventbus.Message) { tm, err := workintake.DecodeTask(msg.Data()) if err != nil { d.Log.Error("nats decode failed", "err", err) if err := msg.Ack(); err != nil { d.Log.Warn("ack failed", "err", err) } // bad message, don't retry return } inserted, err := d.Store.EnqueueIssue(ctx, tm.Owner, tm.Repo, tm.Number, tm.Title, tm.Body, strings.Join(tm.Labels, ","), tm.Identity) if err != nil { d.Log.Error("nats enqueue failed", "err", err) if err := msg.Nak(); err != nil { d.Log.Warn("nats nak failed", "err", err) } return } if !inserted { // Already tracked -- dedup. Ack and move on. if err := msg.Ack(); err != nil { d.Log.Warn("ack failed", "err", err) } return } task, err := d.Store.ClaimByIssue(ctx, tm.Owner, tm.Repo, tm.Number) if err != nil { d.Log.Error("nats claim failed", "err", err) if err := msg.Nak(); err != nil { d.Log.Warn("nats nak failed", "err", err) } return } if task == nil { // Claimed by another consumer, or task is not queued. Ack. if err := msg.Ack(); err != nil { d.Log.Warn("ack failed", "err", err) } return } d.process(ctx, task) if err := msg.Ack(); err != nil { d.Log.Warn("ack failed", "err", err) } } // pollIssues discovers work for one repo according to the root dispatch // config, using the root forge client. Identity poll loops use // pollIssuesWithConfig with their own forge and config instead. func (d *Daemon) pollIssues(ctx context.Context, repo config.Repo) []forge.Issue { return d.pollIssuesWithConfig(ctx, d.Forge, d.Cfg.Get(), repo) } // closeResolvedIssue closes the forge issue behind a finished task. // // A chat task's issue number is synthetic and matches no forge issue, so it // is skipped. Failure is logged rather than returned: the task's own state is // already correct, and the next reconcile pass will not retry, so a warning // is the honest outcome -- the issue simply stays open. func (d *Daemon) closeResolvedIssue(ctx context.Context, fg forge.Forge, task *workflow.Task, comment string) { if !task.IsForgeBacked() { return } if err := fg.CloseIssue(ctx, task.Owner, task.Repo, task.IssueNumber, comment); err != nil { d.Log.Warn("closing resolved issue failed; it stays open and will be re-polled", "repo", task.Owner+"/"+task.Repo, "issue", task.IssueNumber, "err", err) } } // reconcilePRs moves pr_open tasks to merged/rejected from GitHub state // and cleans up their worktrees. func (d *Daemon) reconcilePRs(ctx context.Context) { tasks, err := d.Store.OpenPRs(ctx) if err != nil { d.Log.Error("reconcile query failed", "err", err) return } for _, t := range tasks { fg := d.forgeFor(&t) trees := d.treesFor(&t) // PRState/PR reconciliation is valid regardless of Source -- a // chat-sourced task can still open a real PR against a real // branch. Only the issue-label call (tied to the synthetic // issue number) is forge-only. state, err := fg.PRState(ctx, t.Owner, t.Repo, t.PRNumber) if err != nil { d.Log.Warn("PR state check failed", "pr", t.PRNumber, "err", err) continue } // OpenPRs returns a deliberately narrow projection of a pr_open task // (id/owner/repo/issue/pr/status/source/identity) that does not carry // the attempt, so the outcome event below reads it from the task's own // row. A failed lookup costs attribution and never the merge handling: // the status transition and the worktree cleanup still run. OpenPRs // carries attempt, so these events are attributed without a second // read per open PR on every reconcile tick. attempt := t.Attempt switch state { case "merged": _ = d.Store.Transition(ctx, t.ID, workflow.StatusPROpen, workflow.StatusMerged, "") _ = trees.Cleanup(t.Owner, t.Repo, t.IssueNumber) // Close the issue ourselves rather than relying on the forge // noticing a "Closes #N" in the PR body. Nothing else closes it: // LinkBranch is sidebar linkage on Gitea and a no-op on GitHub. // Left open the issue stays labelled and assigned, the next poll // re-enqueues it, and once the dashboard's Clear removes the task // row that is a second implementation and a second PR for work // already merged. d.closeResolvedIssue(ctx, fg, &t, fmt.Sprintf( "Closed by #%d, merged by archie.", t.PRNumber, )) d.Log.Info("PR merged", "repo", t.Owner+"/"+t.Repo, "pr", t.PRNumber) d.emit(events.Event{ Kind: events.KindPRMerged, TaskID: t.ID, Attempt: attempt, Repo: t.Owner + "/" + t.Repo, Issue: t.IssueNumber, Data: map[string]any{"pr": t.PRNumber}, }) case "closed": _ = d.Store.Transition(ctx, t.ID, workflow.StatusPROpen, workflow.StatusRejected, "PR closed without merge") _ = trees.Cleanup(t.Owner, t.Repo, t.IssueNumber) d.Log.Info("PR rejected", "repo", t.Owner+"/"+t.Repo, "pr", t.PRNumber) d.emit(events.Event{ Kind: events.KindPRRejected, TaskID: t.ID, Attempt: attempt, Repo: t.Owner + "/" + t.Repo, Issue: t.IssueNumber, Data: map[string]any{"pr": t.PRNumber}, }) } } } func (d *Daemon) process(ctx context.Context, task *workflow.Task) { // Register before any work starts so the task is stoppable for its // whole life, including the slow setup -- clone, worktree prepare, // image pull -- which is exactly when someone realises they asked for // the wrong thing. Every caller reaches execution through here, so // this is the one place that has to do it. ctx, finished := d.running.begin(ctx, task.ID, task.Identity) defer finished() defer d.openTaskLog(task)() trees := d.treesFor(task) repo, ok := d.repoFor(task) if !ok { d.parkRunningTask(ctx, task.ID, "repo no longer in config") return } if d.ContainerPool == nil { const reason = "managed agent container pool is unavailable; refusing to run this task on the host" d.Log.Error("task parked: managed worker unavailable", "task", task.ID, "hint", "enable containers and make the archie-agent image available, then retry the task") if err := d.Store.Transition(ctx, task.ID, workflow.StatusRunning, workflow.StatusParked, reason); err != nil { d.Log.Warn("managed worker park transition failed", "task", task.ID, "err", err) return } d.recordPark(ctx, task.ID, reason) d.emit(events.Event{ Kind: events.KindParked, TaskID: task.ID, Attempt: task.Attempt, Repo: repo.FullName(), Issue: task.IssueNumber, Detail: reason, }) return } if d.Tasks == nil { const reason = "agent task transport is unavailable; refusing to run this task on the host" d.Log.Error("task parked: agent task transport unavailable", "task", task.ID) if err := d.Store.Transition(ctx, task.ID, workflow.StatusRunning, workflow.StatusParked, reason); err != nil { d.Log.Warn("agent task transport park transition failed", "task", task.ID, "err", err) return } d.recordPark(ctx, task.ID, reason) d.emit(events.Event{ Kind: events.KindParked, TaskID: task.ID, Attempt: task.Attempt, Repo: repo.FullName(), Issue: task.IssueNumber, Detail: reason, }) return } workDir := trees.Dir(task.Owner, task.Repo, task.IssueNumber) d.Log.Info("processing task", "repo", repo.FullName(), "issue", task.IssueNumber, "attempt", task.Attempt) // Clone the worktree before Docker mount setup. The container // binds /data/worktree to workDir on the host -- the directory // must exist before Acquire is called. var branch string var err error // Every task gets an independent full clone. The former // PreparePersistent path shared objects with a per-repo bare cache; // go-git has no --dissociate, so a shared cache would stay a live // dependency of each worktree and expiring one would corrupt running // tasks. repo.PersistentStorage still governs the container volume. _, branch, err = trees.Prepare(ctx, task.Owner, task.Repo, repo.Base, task.IssueNumber, task.Title, task.Body, task.Labels) if err != nil { d.Log.Error("worktree prepare failed", "err", err) d.parkRunningTask(ctx, task.ID, "worktree prepare failed: "+err.Error()) return } task.Branch = branch if err := d.Store.Update(ctx, task); err != nil { d.Log.Warn("task branch not persisted", "task", task.ID, "err", err) } ctr, revokeStateStoreGrant, ok := d.acquireTaskContainer(ctx, task, repo, workDir) if !ok { return } defer d.ContainerPool.Release(ctx, ctr) defer revokeStateStoreGrant() // Hand the whole task to archie-agent in one NATS round trip. archie-agent // proxies Store/Forge/worktree-push calls back to archied over storerpc/ // forgerpc/worktreerpc -- by the time runViaAgent returns, the task's // terminal state already landed via those calls. The task's identity // travels in the taskrun request; the agent scopes its forgerpc and // worktreerpc clients to that identity, and the daemon registered one // server pair per identity (plus the root pair), so a container-mode // task is always served by its own forge client and worktree manager. d.runViaAgent(ctx, task, repo) // Teardown storage after workflow completes. The Docker backend is a // no-op; future backends (temp volumes, NFS leases) use this hook. if d.Storage != nil { _ = d.Storage.Teardown(ctx, storage.TaskRef{ WorktreeDir: workDir, Ecosystem: repo.Ecosystem, PersistentStorage: repo.PersistentStorage, Owner: task.Owner, Repo: task.Repo, }) } d.cleanupTerminalTaskWorktree(ctx, task, trees) } func (d *Daemon) cleanupTerminalTaskWorktree(ctx context.Context, task *workflow.Task, trees *worktree.Manager) { if d.Store == nil || trees == nil || task == nil { return } cleanupCtx, cancel := context.WithTimeout(context.WithoutCancel(ctx), 10*time.Second) defer cancel() latest, err := d.Store.TaskByID(cleanupCtx, task.ID) if err != nil { d.Log.Warn("terminal worktree cleanup task lookup failed", "task", task.ID, "err", err) return } if latest == nil { return } switch latest.Status { case workflow.StatusMerged, workflow.StatusRejected, workflow.StatusClosedWontDo: if latest.PRNumber == 0 { if err := trees.Cleanup(task.Owner, task.Repo, task.IssueNumber); err != nil { d.Log.Warn("terminal worktree cleanup failed", "task", task.ID, "err", err) } } } } // openTaskLog opens task's log sink for the lifetime of one process() call // and returns the function to close it. That lifetime covers preparation, // the full-task container handoff, and teardown. A failure to open is not // fatal to the task -- it means this // attempt's own output won't be recoverable if something goes wrong, which // is worse than the pre-existing state, not equal to it, so the run // proceeds rather than parking over a logging problem. func (d *Daemon) openTaskLog(task *workflow.Task) func() { if err := d.TaskLogs.Open(task.ID, task.Attempt); err != nil { d.Log.Warn("task log sink unavailable", "task", task.ID, "attempt", task.Attempt, "err", err) } return func() { if err := d.TaskLogs.Close(task.ID); err != nil { d.Log.Warn("task log sink close failed", "task", task.ID, "err", err) } } } // acquireTaskContainer prepares and acquires the sandbox container for a // task, reporting false when it could not -- having already parked the // task and logged why. The caller releases the container it returns. func (d *Daemon) acquireTaskContainer( ctx context.Context, task *workflow.Task, repo config.Repo, workDir string, ) (*container.Container, func(), bool) { park := func(reason string, err error) { d.Log.Error(reason, "err", err) d.parkRunningTask(ctx, task.ID, reason+": "+err.Error()) } // Write task.json -- the container's boot-time brief. if err := container.WriteTaskJSON(workDir, container.TaskPayload{ ID: task.ID, Owner: task.Owner, Repo: task.Repo, Number: task.IssueNumber, Title: task.Title, Body: task.Body, Labels: strings.Split(task.Labels, ","), Workflow: task.Workflow, Branch: task.Branch, Plan: task.Plan, }); err != nil { park("task.json write failed", err) return nil, nil, false } // Guard: Storage may be nil if the daemon was wired incorrectly. In // normal operation, Storage is always set when ContainerPool is set. if d.Storage == nil { d.Log.Error("storage backend is nil -- cannot acquire container") d.parkRunningTask(ctx, task.ID, "storage backend not configured") return nil, nil, false } mounts, err := d.Storage.Setup(ctx, storage.TaskRef{ WorktreeDir: workDir, Ecosystem: repo.Ecosystem, PersistentStorage: repo.PersistentStorage, Owner: task.Owner, Repo: task.Repo, }) if err != nil { park("storage setup failed", err) return nil, nil, false } stateStoreToken, revokeStateStoreGrant, err := d.stateStoreGrantToken(task) if err != nil { if terr := d.Storage.Teardown(ctx, storage.TaskRef{ WorktreeDir: workDir, Ecosystem: repo.Ecosystem, PersistentStorage: repo.PersistentStorage, Owner: task.Owner, Repo: task.Repo, }); terr != nil { d.Log.Warn("storage teardown after state store grant failure failed", "err", terr) } park("state store grant failed", err) return nil, nil, false } ctr, err := d.ContainerPool.Acquire(ctx, mounts, d.containerEnv(task, stateStoreToken)) if err != nil { revokeStateStoreGrant() // Roll back the storage we just set up: the mounts were created // for this task but no container will use them. Leaking them until // a later TTL sweep is not acceptable on a backend that allocates // real resources (volumes, NFS leases). if terr := d.Storage.Teardown(ctx, storage.TaskRef{ WorktreeDir: workDir, Ecosystem: repo.Ecosystem, PersistentStorage: repo.PersistentStorage, Owner: task.Owner, Repo: task.Repo, }); terr != nil { d.Log.Warn("storage teardown after acquire failure failed", "err", terr) } park("container acquire failed", err) return nil, nil, false } return ctr, revokeStateStoreGrant, true } // parkRunningTask transitions a task from running to parked using a context // detached from caller cancellation, with a bounded timeout. // // When a task fails or is cancelled (e.g. via /stop or dashboard Stop), ctx is // already cancelled. Writing terminal state using the cancelled ctx fails // immediately in SQLite transactions and silently leaves the task row 'running' // indefinitely. Like container teardown in Pool.Release, terminal state // recording is cleanup on the way out of a task and must use a bounded context // that survives cancellation while preserving values. // // The transition remains guarded from StatusRunning: if the worker already // recorded a terminal state over storerpc, storecontract.ErrStaleTransition is returned // and ignored. Unexpected store errors are logged as warnings. func (d *Daemon) parkRunningTask(ctx context.Context, taskID int64, reason string) { if d.Store == nil { return } writeCtx, cancel := context.WithTimeout(context.WithoutCancel(ctx), 10*time.Second) defer cancel() if err := d.Store.Transition(writeCtx, taskID, workflow.StatusRunning, workflow.StatusParked, reason); err != nil { if !errors.Is(err, storecontract.ErrStaleTransition) { d.Log.Warn("terminal park transition failed", "task", taskID, "reason", reason, "err", err) } return } // Recorded only once the transition landed. An entry claiming the task // parked would be a lie when another actor moved it out of running first // -- the same race the parked event is withheld for. d.recordPark(ctx, taskID, reason) } // recordPark writes one park reason into the task attempt's own log. Every // place that parks a task calls it, including the two capability parks in // process() that write their transition directly instead of going through // parkRunningTask -- a park an operator cannot read the reason for is the same // defect wherever it happens. Nothing daemon-side used to write to a task log // at all: the container's system-log subscription was the only writer, so an // attempt that parked before any container produced output left an empty log // file and an unexplained park reason. // // A task with no open sink (a retry this instance never dispatched, or logging // unwired) silently drops the entry, which is TaskRegistry.Write's own // contract. func (d *Daemon) recordPark(ctx context.Context, taskID int64, reason string) { d.TaskLogs.Write(ctx, taskID, logging.Entry{ Time: time.Now(), Level: slog.LevelError.String(), Message: "task parked: " + reason, Fields: map[string]any{"component": "daemon", "task": taskID}, }) } // containerEnv returns the environment variables passed to agent containers. // runViaAgent publishes a full-task handoff to archie-agent over the // infrastructure-owned task subject and waits for its completion report. Every durable // side effect (Store transitions, Forge calls, git push) happens inside // archie-agent's workflow.Run, proxied back to this daemon over // storerpc/forgerpc/worktreerpc -- those RPC servers are the sole place a // terminal state gets written on success. runViaAgent only parks the task // itself when nothing else could have: the request never reached (or was // never answered by) an archie-agent, or archie-agent failed before its // own workflow.Run got a chance to record an outcome. func (d *Daemon) runViaAgent(ctx context.Context, task *workflow.Task, repo config.Repo) { if d.WorktreeGrants == nil { const reason = "worktree publication grants are unavailable" d.Log.Error(reason, "task", task.ID) d.parkRunningTask(ctx, task.ID, reason) return } grant, revoke, err := d.WorktreeGrants.Issue(task) if err != nil { d.Log.Error("worktree publication grant failed", "task", task.ID, "err", err) d.parkRunningTask(ctx, task.ID, "worktree publication grant failed: "+err.Error()) return } defer revoke() cfg := d.configFor(task) taskCfg := cfg.ForTask() d.captureAttemptConfig(ctx, task, taskCfg) req := taskrun.Request{ Task: task, Repo: repo, Cfg: taskCfg, Providers: agentexec.ProvidersFromConfig(cfg.Providers), MCPServers: cfg.Tools.MCPServers, WorktreeGrant: grant, KindWorkflows: d.KindWorkflows, LabelWorkflows: d.LabelWorkflows, } data, err := json.Marshal(req) if err != nil { d.Log.Error("taskrun encode failed", "task", task.ID, "err", err) d.parkRunningTask(ctx, task.ID, "taskrun encode failed: "+err.Error()) return } reply, err := d.requestTaskRun(ctx, task.ID, data) if err != nil { d.Log.Error("taskrun request failed", "task", task.ID, "err", err) d.parkRunningTask(ctx, task.ID, "taskrun request failed: "+err.Error()) return } var resp taskrun.Response if err := json.Unmarshal(reply, &resp); err != nil { d.Log.Error("taskrun decode response failed", "task", task.ID, "err", err) d.parkRunningTask(ctx, task.ID, "taskrun decode response failed: "+err.Error()) return } if resp.Error != "" { d.Log.Error("taskrun run failed", "task", task.ID, "err", resp.Error) d.parkRunningTask(ctx, task.ID, "taskrun run failed: "+resp.Error) return } // Reached only past every failure return above, so this is always a // fully decoded, error-free response -- a version is never recorded off // a request that never reached an agent, or a response describing a // parked/errored task. if d.AgentStatus != nil && resp.AgentVersion != "" { d.AgentStatus.Observe(resp.AgentVersion, resp.AgentInstallType) } d.Log.Info("taskrun complete", "task", task.ID, "status", resp.Status) } const ( // defaultTaskRunReadyTimeout is how long requestTaskRun retries a // nats.ErrNoResponders before giving up, when Daemon.TaskRunReadyTimeout // is unset. defaultTaskRunReadyTimeout = 20 * time.Second // defaultTaskRunRetryBackoff is the delay between retry attempts, when // Daemon.TaskRunRetryBackoff is unset. defaultTaskRunRetryBackoff = 250 * time.Millisecond ) func (d *Daemon) taskRunReadyTimeout() time.Duration { if d.TaskRunReadyTimeout > 0 { return d.TaskRunReadyTimeout } return defaultTaskRunReadyTimeout } func (d *Daemon) taskRunRetryBackoff() time.Duration { if d.TaskRunRetryBackoff > 0 { return d.TaskRunRetryBackoff } return defaultTaskRunRetryBackoff } // requestTaskRun publishes the taskrun request and retries while no // archie-agent has subscribed yet (nats.ErrNoResponders): the container // pool's Acquire returns as soon as Docker has issued the start syscall, // not once the spawned container has connected to core NATS and subscribed // to this per-task subject -- // a gap of hundreds of milliseconds to a few seconds that would otherwise // fail the very first request deterministically, every time, on every // task. Any error other than ErrNoResponders (encode failures, a context // that's already done, etc.) is returned immediately without retrying, // since those don't mean "not ready yet". func (d *Daemon) requestTaskRun(ctx context.Context, taskID int64, data []byte) ([]byte, error) { subject := agentnats.SubjectForTask(taskID) deadline := time.Now().Add(d.taskRunReadyTimeout()) backoff := d.taskRunRetryBackoff() for { reply, err := d.Tasks.Request(ctx, subject, data) if err == nil { return reply, nil } if !errors.Is(err, eventbus.ErrNoResponders) || !time.Now().Before(deadline) { return nil, err } wait := min(backoff, time.Until(deadline)) if wait <= 0 { return nil, err } select { case <-ctx.Done(): return nil, ctx.Err() case <-time.After(wait): } } } func (d *Daemon) containerEnv(task *workflow.Task, stateStoreToken string) []string { var env []string // The endpoint the daemon's own client connected with at startup, not // the live config: a reloaded [nats] section must not point new // containers at a server the daemon is not publishing on. env = append(env, "NATS_URL="+d.ConnectedNATS.URL) if token := d.ConnectedNATS.Token; token != "" { env = append(env, "NATS_TOKEN="+token) } // The State Store gRPC target follows the same seam as NATS_URL/ // NATS_TOKEN above (docs/prds/state-store-contract.md §6). It is always // the configured [services.state].target after the in-process serving // path is deleted. stateStoreToken is this task's own scoped credential // (see stateStoreGrantToken) -- never d.ConnectedStateStore.Token, which // is the daemon's administrative credential and must never reach a // container. if d.ConnectedStateStore.URL != "" { env = append(env, "STATE_STORE_URL="+d.ConnectedStateStore.URL) if stateStoreToken != "" { env = append(env, "STATE_STORE_TOKEN="+stateStoreToken) } } // The agent process runs as root inside the container (no USER in the // Dockerfile, no userns-remap), so a commit it writes to the bind-mounted // worktree lands owned by UID 0 on the host. The daemon then reads those // same loose objects to push, running as its own non-root host user -- // archie-core#520's "permission denied" on .git/objects. Passing the // daemon's own UID/GID lets the agent chown the worktree back to it // before pushing. env = append(env, fmt.Sprintf("WORKTREE_UID=%d", os.Getuid())) env = append(env, fmt.Sprintf("WORKTREE_GID=%d", os.Getgid())) for _, p := range d.configFor(task).Providers { if p.APIKeyEnv != "" { if v := os.Getenv(p.APIKeyEnv); v != "" { env = append(env, p.APIKeyEnv+"="+v) } } } return env } // stateStoreGrantToken issues this task's scoped State Store credential, or // ("", no-op revoke, nil) when no State Store is configured. It fails closed // -- rather than falling back to the daemon's administrative token -- when a // State Store is configured but no StateStoreGrantIssuer is wired, which is // a composition bug, not a runtime condition to paper over. func (d *Daemon) stateStoreGrantToken(task *workflow.Task) (string, func(), error) { if d.ConnectedStateStore.URL == "" { return "", func() {}, nil } if d.StateStoreGrants == nil { return "", nil, fmt.Errorf("state store is configured but no task-scoped grant issuer is wired") } return d.StateStoreGrants.Issue(task) } func (d *Daemon) configFor(task *workflow.Task) config.Config { if id := d.identityFor(task); id != nil { return configForIdentity(d.Cfg.Get(), id.Cfg) } return d.Cfg.Get() } // captureAttemptConfig persists the effective task configuration this attempt // runs under, as a durable event on the attempt's own key. // // It is written where the configuration is materialised for the dispatch, so // the document is exactly what the run received. The published configuration // snapshot cannot answer this: it is the CURRENT configuration, replaced on // every publish, so a run that finished last week has no record anywhere else. // // Fail-open by construction: a configuration that could not be recorded is // worth a log line and nothing more. Parking a task over a reporting failure // would trade the run for the receipt. func (d *Daemon) captureAttemptConfig(ctx context.Context, task *workflow.Task, cfg config.TaskConfig) { if d.Store == nil { return } document, err := json.Marshal(cfg) if err != nil { d.Log.Warn("attempt configuration not captured", "task", task.ID, "attempt", task.Attempt, "err", err) return } event := events.Event{ At: time.Now().UTC(), Kind: events.KindConfigCaptured, TaskID: task.ID, Repo: task.Owner + "/" + task.Repo, Issue: task.IssueNumber, Workflow: task.Workflow, Attempt: task.Attempt, Data: map[string]any{ "schema": events.ConfigCapturedSchema, // Raw so the document is embedded once, as it was encoded, rather // than as a string a reader would have to decode again. "document": json.RawMessage(document), }, } id, err := d.Store.InsertEvent(ctx, event) if err != nil { d.Log.Warn("attempt configuration not captured", "task", task.ID, "attempt", task.Attempt, "err", err) return } // The assigned ID tells the event sink to broadcast without inserting a // duplicate row. event.ID = id d.emit(event) } func configForIdentity(root config.Config, identity config.IdentityConfig) config.Config { root.BotUser = identity.BotUser root.BotEmail = identity.BotEmail root.DiffCapLines = identity.DiffCapLines root.Forge = identity.Forge root.Dispatch = identity.Dispatch root.Models = identity.Models root.Providers = identity.Providers root.Budgets = identity.Budgets root.Notify = identity.Notify root.Repos = identity.Repos return root } // identityFor resolves the IdentityRunner that owns task, or nil for // single-identity deployments and forge-sourced tasks recorded // before multi-identity routing existed (task.Identity == ""). Callers // must fall back to the root d.Forge/d.Trees/d.Cfg when this returns nil. func (d *Daemon) identityFor(task *workflow.Task) *IdentityRunner { if task == nil || task.Identity == "" { return nil } for _, id := range d.Identities { if id.Name == task.Identity { return id } } return nil } // forgeFor returns the forge client that owns task: the identity's own // client when task.Identity names a configured identity, else the root // d.Forge. This is the safety boundary that keeps one identity's forge // token from being used against another identity's repos. func (d *Daemon) forgeFor(task *workflow.Task) forge.Forge { if id := d.identityFor(task); id != nil { return id.Forge } return d.Forge } // treesFor returns the worktree manager that owns task, mirroring forgeFor. func (d *Daemon) treesFor(task *workflow.Task) *worktree.Manager { if id := d.identityFor(task); id != nil { return id.Trees } return d.Trees } // repoFor resolves task's repo config from the owning identity's repo // list when task.Identity is set, else the root Cfg.Repos. func (d *Daemon) repoFor(t *workflow.Task) (config.Repo, bool) { repos := d.Cfg.Get().Repos if id := d.identityFor(t); id != nil { repos = id.Repos } for _, r := range repos { if r.Owner == t.Owner && r.Name == t.Repo { return r, true } } return config.Repo{}, false } // allowConcurrentForTask reports whether the task's owning repo has opted // into concurrent task dispatch (config.Repo.AllowConcurrent). The repo is // resolved through repoFor, which selects the identity-scoped repo list when // task.Identity names a configured identity -- so multi-identity concurrency // policy comes from the identity's own config, not the root daemon's. // Unknown repos default to the safe, serialized behavior. func (d *Daemon) allowConcurrentForTask(task *workflow.Task) bool { repo, ok := d.repoFor(task) return ok && repo.AllowConcurrent }