What causes "failed to create flow worker: %w" in vxcontrol/pentagi?

Check the wrapped cause from LoadFlowWorker logs to find which component failed.

failed to create flow worker: %w

Error message

failed to create flow worker: %w

What it means

CreateFlow wraps a failure of LoadFlowWorker (which assembles flowProviderWorkers and spins up the worker) with this message. The flow row may have been created in the DB, but the in-memory worker could not be constructed, so no worker is registered in fc.flows.

Source

Thrown at backend/pkg/controller/flows.go:189

			db:     fc.db,
			cfg:    fc.cfg,
			docker: fc.docker,
			provs:  fc.provs,
			subs:   fc.subs,
			flowProviderControllers: flowProviderControllers{
				mlc:  fc.mlc,
				aslc: fc.aslc,
				alc:  fc.alc,
				slc:  fc.slc,
				tlc:  fc.tlc,
				vslc: fc.vslc,
				tclc: fc.tclc,
				sc:   fc.sc,
			},
		},
	})
	if err != nil {
		return nil, fmt.Errorf("failed to create flow worker: %w", err)
	}

	fc.flows[fw.GetFlowID()] = fw

	return fw, nil
}

func (fc *flowController) CreateAssistant(
	ctx context.Context,
	userID int64,
	flowID int64,
	input string,
	useAgents bool,
	prvname provider.ProviderName,
	prvtype provider.ProviderType,
	functions *tools.Functions,
	resources []database.UserResource,
) (AssistantWorker, error) {

View on GitHub (pinned to ea665308ba)

Solutions

Check the wrapped cause from LoadFlowWorker logs to find which component failed.
Delete or clean up the orphaned flow row if the worker cannot be created, then retry CreateFlow.
Verify all storage backends (DB, screenshot store, vector store) are healthy.
Reduce creation concurrency or scale the DB connection pool if under load.

Example fix

// before
fw, err := LoadFlowWorker(ctx, flow, flowWorkerCtx{...})
if err != nil {
	return nil, fmt.Errorf("failed to create flow worker: %w", err)
}
// after
fw, err := LoadFlowWorker(ctx, flow, flowWorkerCtx{...})
if err != nil {
	_ = fc.db.DeleteFlow(ctx, flow.GetFlowID()) // avoid orphaned flow rows
	return nil, fmt.Errorf("failed to create flow worker: %w", err)
}

Defensive patterns

Strategy: try-catch

Validate before calling

// ensure backing stores are healthy before creating flows
if err := db.PingContext(ctx); err != nil { /* abort flow creation */ }
// storage/vector/screenshot services health-checked at boot

Type guard

func isCreateWorkerError(err error) bool {
	return err != nil && strings.Contains(err.Error(), "failed to create flow worker")
}

Try / catch

fw, err := LoadFlowWorker(ctx, flow, flowWorkerCtx{...})
if err != nil {
	// roll back the persisted flow row to avoid orphans
	_ = fc.db.DeleteFlow(ctx, flow.GetFlowID())
	return nil, fmt.Errorf("failed to create flow worker: %w", err)
}

Prevention

Health-check all worker dependencies (DB, screenshot store, vector store) before accepting flow creation.
Roll back the DB flow row when worker creation fails.
Unwrap and log the root cause to identify the failing component quickly.
Limit concurrent flow creation to avoid resource exhaustion.

When it happens

Trigger: Calling CreateFlow when any component of flowWorkerCtx fails to initialize its worker — e.g., any of the log fetchers (msg/agent/search/term/vector-store/tool-call/screenshot) errors for the freshly created flowID.

Common situations: DB degraded right after flow insert; storage components misconfigured (screenshots/vector store); resource exhaustion (goroutines/connections) under heavy flow creation load.

Related errors

AI-assisted analysis of vxcontrol/pentagi@ea665308ba (2026-09-01). Data as JSON: /api/errors/cc7c6ea8fadf92c2. Report an issue: GitHub.