gastownhall/beads · error

failed to create dependency: %w

Error message

failed to create dependency: %w

What it means

Wraps a failure from tx.AddDependency while persisting a cooked formula's dependency edges (DepBlocks/DepWaitsFor etc.) inside the single atomic transaction that creates issues, labels, and deps. When any single dependency insert fails the whole cook --persist transaction rolls back, leaving no partial subgraph. The wrapped cause carries the storage-level reason (duplicate edge, missing target issue, type not allowed, driver error).

Source

Thrown at cmd/bd/cook.go:926

			return fmt.Errorf("checking custom types: %w", err)
		}

		// Create all issues
		if err := tx.CreateIssues(ctx, issues, actor); err != nil {
			return fmt.Errorf("failed to create issues: %w", err)
		}

		// Add labels
		for _, l := range labels {
			if err := tx.AddLabel(ctx, l.issueID, l.label, actor); err != nil {
				return fmt.Errorf("failed to add label %s to %s: %w", l.label, l.issueID, err)
			}
		}

		// Add dependencies
		for _, dep := range deps {
			if err := tx.AddDependency(ctx, dep, actor); err != nil {
				return fmt.Errorf("failed to create dependency: %w", err)
			}
		}

		return nil
	})

	if err != nil {
		return nil, err
	}

	return &cookFormulaResult{
		ProtoID: protoID,
		Created: len(issues),
	}, nil
}

// collectDependencies collects blocking dependencies from depends_on, needs, and waits_for fields.
// This is the shared implementation used by both DB-persisted and in-memory subgraph cooking.

View on GitHub (pinned to 71377f2769)

Solutions

1. Read the wrapped cause: if it's a duplicate/unknown-target dependency, fix the formula's depends_on/needs/waits_for so each edge references an existing step ID and isn't redundant with waits_for
2. Validate the formula (bd mol / formula lint) before persisting so step IDs and dependency references resolve
3. Re-run the command — the transaction rolled back cleanly, so no cleanup is needed; retry after fixing the formula
4. If the cause is a driver/storage error, check database connectivity and Dolt health before retrying

Example fix

# before (formula.yaml)
depends_on: [build, build]   # duplicate edge
# after
depends_on: [build]

// Pre-validate the formula before persisting
for _, step := range f.Steps {
    for _, depID := range append(append([]string{}, step.DependsOn...), step.Needs...) {
        if _, ok := idMapping[depID]; !ok {
            return fmt.Errorf("step %q references unknown dependency %q", step.ID, depID)
        }
    }
}

if err := persistCookFormula(ctx, s, f); err != nil {
    var cause error
    errors.As(err, &cause) // inspect wrapped storage error
    return fmt.Errorf("cook persist rolled back, formula unchanged: %w", err)
}

When it happens

Trigger: Running `bd cook <formula> --persist` (via persistCookFormula) where a step's depends_on/needs/waits_for produces an AddDependency call that storage rejects — e.g. the target issue was not among the created issues, a duplicate edge on the same (source,target) pair, or a driver-level constraint failure.

Common situations: A formula YAML whose waits_for spawner overlaps depends_on in a way the dedupe logic (GH#3783) doesn't cover; a custom/registered step type that maps to an issue ID missing from idMapping; Dolt driver errors or constraint violations during the transaction.

Related errors

• delete %s: %w
• failed to link digest to root: %w
• ErrTransaction
• db: DependencySQLRepository.Insert: dep must not be nil
• db: DependencySQLRepository.Insert: IssueID must not be empt

AI-assisted analysis of gastownhall/beads@71377f2769 (2026-08-30). Data as JSON: /api/errors/11d9968fdf5384d5. Report an issue: GitHub.