What causes "couldn't scan metric status row: %w" in SigNoz/signoz?

Inspect the wrapped scan error for which column mismatched

SigNoz/signoz · error · model.InternalError

couldn't scan metric status row: %w

Error message

couldn't scan metric status row: %w

What it means

Internal error from scanning a row of the metric-status query into (metric name, labels JSON, last-received timestamp millis). Indicates the result columns don't match the expected three-column shape/types, or a NULL in a non-nullable target.

Source

Thrown at pkg/query-service/app/clickhouseReader/reader.go:3423

			"couldn't query clickhouse for received metrics status: %w", err,
		))
	}
	defer rows.Close()

	var result *model.MetricStatus

	if rows.Next() {

		result = &model.MetricStatus{}
		var labelsJson string

		err := rows.Scan(
			&result.MetricName,
			&labelsJson,
			&result.LastReceivedTsMillis,
		)
		if err != nil {
			return nil, model.InternalError(fmt.Errorf(
				"couldn't scan metric status row: %w", err,
			))
		}

		err = json.Unmarshal([]byte(labelsJson), &result.LastReceivedLabels)
		if err != nil {
			return nil, model.InternalError(fmt.Errorf(
				"couldn't unmarshal metric labels json: %w", err,
			))
		}
	}

	return result, nil
}

func isColumn(tableStatement, attrType, field, datType string) bool {
	name := fmt.Sprintf("`%s`", utils.GetClickhouseColumnNameV2(attrType, datType, field))
	return strings.Contains(tableStatement, fmt.Sprintf("%s ", name))

View on GitHub (pinned to 5069bf80b0)

Solutions

Inspect the wrapped scan error for which column mismatched
Run the status query manually and check for NULLs in the three selected columns
Scan into sql.Null* types where NULLs are possible
Re-run migrations to align schema

Example fix

// before
&result.LastReceivedTsMillis,

// after (if ts can be NULL)
var ts sql.NullInt64
... scan &ts ...
if ts.Valid { result.LastReceivedTsMillis = ts.Int64 }

Defensive patterns

Strategy: fallback

Try / catch

if err := rows.Scan(&name, &labels, &ts); err != nil {
    // log and skip this row instead of failing the whole request
    continue
}

Prevention

Prefer sql.Null* scans for nullable columns
Keep single ingester version writing time_series metadata

When it happens

Trigger: time_series table returns rows where last_received_ts or labels JSON is NULL, or column order/type changed between SigNoz versions.

Common situations: Schema drift after upgrade; rows written by an older ingester lacking label JSON; partial writes during ingestion incidents.

Related errors

AI-assisted analysis of SigNoz/signoz@5069bf80b0 (2026-08-28). Data as JSON: /api/errors/efa56cb49d569fb3. Report an issue: GitHub.