What causes "journal: close dependency removals from %s: %w" in gastownhall/beads?

Verify connection health and retry; this is often transient

journal: close dependency removals from %s: %w

Error message

journal: close dependency removals from %s: %w

What it means

This error is returned when rows.Close() fails after successfully iterating dependency-removal rows in dependencyEdgesInTableForIssueIDsInTx. Closing a rows handle releases database resources, and a close failure usually signals an already-broken connection. The library surfaces it because leaking cursors on journal tables can exhaust server resources.

Source

Thrown at internal/storage/issueops/journal_dependencies.go:108

			if optionalBlockedTable(table) && isTableNotExistError(err) {
				continue
			}
			return nil, fmt.Errorf("journal: query dependency removals from %s: %w", table, err)
		}
		for rows.Next() {
			var edge journalDependencyEdge
			if err := rows.Scan(&edge.source, &edge.target, &edge.kind, &edge.metadata); err != nil {
				_ = rows.Close()
				return nil, fmt.Errorf("journal: scan dependency removal from %s: %w", table, err)
			}
			byKey[dependencyEdgeKey(edge)] = edge
		}
		if err := rows.Err(); err != nil {
			_ = rows.Close()
			return nil, fmt.Errorf("journal: iterate dependency removals from %s: %w", table, err)
		}
		if err := rows.Close(); err != nil {
			return nil, fmt.Errorf("journal: close dependency removals from %s: %w", table, err)
		}
	}
	return sortedDependencyEdges(byKey), nil
}

func dependencyEdgeKey(edge journalDependencyEdge) string {
	return edge.source + "\x00" + edge.target + "\x00" + edge.kind + "\x00" + edge.metadata
}

func sortedDependencyEdges(byKey map[string]journalDependencyEdge) []journalDependencyEdge {
	edges := make([]journalDependencyEdge, 0, len(byKey))
	for _, edge := range byKey {
		edges = append(edges, edge)
	}
	sort.Slice(edges, func(i, j int) bool {
		if edges[i].source != edges[j].source {
			return edges[i].source < edges[j].source
		}

View on GitHub (pinned to 71377f2769)

Solutions

Verify connection health and retry; this is often transient
Check the wrapped driver error for ErrBadConn and let the pool replace the connection
Ensure connection pool settings (MaxLifetime) are shorter than any network idle timeouts

Defensive patterns

Strategy: retry

Validate before calling

// no pre-call validation possible; verify pool health instead
if db == nil { return errors.New("nil DB handle before dependency rebuild") }

Try / catch

if err != nil && errors.Is(err, driver.ErrBadConn) {
    // let sql.DB replace the connection and retry once
}

Prevention

Set conn MaxIdleTime/MaxLifetime below network idle timeouts
Avoid holding transactions across long computations
Let database/sql pool manage connections rather than caching handles

When it happens

Trigger: RecordDependencyRemovalsForTableInTx or dependencyEdgesForIssueIDsInTx call paths where the database connection has gone bad by the time rows.Close() is called after iteration completes.

Common situations: Network interruptions to a remote Dolt server; connection-pool recycling a broken connection; driver ErrBadConn surfacing at close time.

Related errors

AI-assisted analysis of gastownhall/beads@71377f2769 (2026-08-30). Data as JSON: /api/errors/2d62f073326b538a. Report an issue: GitHub.