What causes "applyGraph: read existing deps for %s: %w" in gastownhall/beads?

Check database connectivity and retry the bd command (transient driver errors often clear).

applyGraph: read existing deps for %s: %w

Error message

applyGraph: read existing deps for %s: %w

What it means

Wraps a failure from depRepo.ListByIssueIDs (regular dependencies table) while graphHasPath lazily loads a node's existing outgoing deps during dependency-graph cycle detection. The underlying store/query error is chained as %w. Thrown during ApplyGraph when the walk visits a node not yet in the dep cache.

Source

Thrown at internal/storage/domain/issue.go:1533

		}
		if seen[id] {
			return false, nil
		}
		seen[id] = true
		for _, next := range adj[id] {
			found, err := visit(next)
			if err != nil || found {
				return found, err
			}
		}
		deps, ok := depCache[id]
		if !ok {
			regular, err := u.depRepo.ListByIssueIDs(ctx, []string{id}, DepListOpts{
				Direction:     DepDirectionOut,
				UseWispsTable: false,
			})
			if err != nil {
				return false, fmt.Errorf("applyGraph: read existing deps for %s: %w", id, err)
			}
			wisp, err := u.depRepo.ListByIssueIDs(ctx, []string{id}, DepListOpts{
				Direction:     DepDirectionOut,
				UseWispsTable: true,
			})
			if err != nil {
				return false, fmt.Errorf("applyGraph: read existing wisp deps for %s: %w", id, err)
			}
			deps = append(regular.Outgoing[id], wisp.Outgoing[id]...)
			depCache[id] = deps
		}
		for _, dep := range deps {
			if !followExistingDep(dep.Type) {
				continue
			}
			found, err := visit(dep.DependsOnID)
			if err != nil || found {
				return found, err

View on GitHub (pinned to 71377f2769)

Solutions

Check database connectivity and retry the bd command (transient driver errors often clear).
Inspect the wrapped %w error for the concrete storage error; fix storage-level cause (e.g. restart dolt server).
Verify deps table schema matches the expected version (bd doctor / driver migration).
Reduce graph size or raise context timeout if deadline-exceeded.

Defensive patterns

Strategy: retry

Validate before calling

// ensure the issue id is resolvable before applying the graph
if id == "" {
    return fmt.Errorf("graph node id is empty")
}
if err := ctx.Err(); err != nil {
    return err
}

Try / catch

if err != nil {
    var retryable bool
    if errors.Is(err, context.DeadlineExceeded) || errors.Is(err, driver.ErrUnavailable) {
        retryable = true
    }
    return fmt.Errorf("applyGraph: read existing deps for %s: %w (retryable=%v)", id, err, retryable)
}

Prevention

Keep DB connection healthy before bulk graph operations (bd doctor).
Use adequate context timeouts for large dependency graphs.
Retry transient driver errors with backoff.
Keep bd version and DB schema in sync.

When it happens

Trigger: ApplyGraph cycle-check walk reaches a node whose outgoing deps are not cached and the non-wisp ListByIssueIDs query fails (DB unavailable, bad connection, storage driver error, corrupted deps table, context canceled).

Common situations: Dolt/database process down mid-command; network partition to remote Dolt; ctx deadline exceeded on very large dependency graphs; table schema drift after version upgrade.

Related errors

AI-assisted analysis of gastownhall/beads@71377f2769 (2026-08-30). Data as JSON: /api/errors/f9e59cb9ac9b8ffa. Report an issue: GitHub.