grafana/k6 · error

metric %q is not a histogram

Error message

metric %q is not a histogram

What it means

The OTel registry found an instrument cached under this metric name, but its dynamic type is not Float64Histogram — another instrument kind (counter/gauge) already occupies this name in the histograms map.

Source

Thrown at internal/output/opentelemetry/registry.go:61

	c, err := r.meter.Float64Counter(name, opts...)
	if err != nil {
		return nil, fmt.Errorf("failed to create counter for %q: %w", name, err)
	}

	r.logger.Debugf("registered counter metric %q", name)

	r.counters.Store(name, c)
	return c, nil
}

func (r *registry) getOrCreateHistogram(name, unit string) (otelMetric.Float64Histogram, error) {
	if histogram, ok := r.histograms.Load(name); ok {
		if v, ok := histogram.(otelMetric.Float64Histogram); ok {
			return v, nil
		}

		return nil, fmt.Errorf("metric %q is not a histogram", name)
	}

	opts := []otelMetric.Float64HistogramOption{}
	if unit != "" {
		opts = append(opts, otelMetric.WithUnit(unit))
	}

	h, err := r.meter.Float64Histogram(name, opts...)
	if err != nil {
		return nil, fmt.Errorf("failed to create histogram for %q: %w", name, err)
	}

	r.logger.Debugf("registered histogram metric %q", name)

	r.histograms.Store(name, h)
	return h, nil
}

View on GitHub (pinned to 01ffac6f24)

Solutions

1. Rename the metric to avoid a counter/gauge/histogram name collision
2. Ensure Trend metrics don't share names with Counter/Gauge metrics

When it happens

Trigger: Thrown at internal/output/opentelemetry/registry.go:61 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of grafana/k6@01ffac6f24 (2026-08-18). Data as JSON: /api/errors/5faea9c73fa75370. Report an issue: GitHub.