What causes "failed to get queue names: %w" in hibiken/asynq?

Check Redis connectivity from the exporter (redis-cli ping) and fix REDIS address/credentials.

failed to get queue names

Error message

failed to get queue names: %w

What it means

QueueMetricsCollector.collectQueueInfo calls Inspector.Queues() to list all queue names and wraps any underlying Redis/transport error with this message. Collect propagates it, failing the whole Prometheus scrape for this collector since queue info cannot be gathered.

Solutions

Check Redis connectivity from the exporter (redis-cli ping) and fix REDIS address/credentials.
Unwrap the %w-wrapped cause to identify the underlying error (connection refused, auth, timeout).
Make the Inspector's RedisConnOpt point to the correct, reachable Redis instance.
Add retry/backoff in the collector or handle per-scrape errors in Prometheus instrumentation.

Example fix

// before
inspector := asynq.NewInspector(asynq.RedisClientOpt{Addr: ""})
// after
inspector := asynq.NewInspector(asynq.RedisClientOpt{Addr: "localhost:6379", Password: os.Getenv("REDIS_PASSWORD")})
if _, err := inspector.Queues(); err != nil {
    log.Fatalf("redis unreachable: %v", err)
}

Defensive patterns

Strategy: try-catch

Validate before calling

insp := asynq.NewInspector(redisOpt)
if _, err := insp.Queues(); err != nil {
    log.Fatalf("redis unreachable for metrics: %v", err)
}

Try / catch

if infos, err := qmc.collectQueueInfo(); err != nil {
    log.Printf("metrics collection failed: %v", err) // keep scrape alive
    return prometheus.NoopCollector
}

Prevention

Health-check Redis connectivity before registering the metrics collector.
Use connection retry/timeouts in RedisClientOpt.
Alert on Redis reachability separately so scrapes failing is a symptom, not the first signal.
Verify exporter REDIS address/password match the asynq server's.

When it happens

Trigger: Prometheus scrape invokes Collect → collectQueueInfo; Inspector.Ques()/Queues() fails due to Redis connection failure, auth failure, closed connection, or the underlying client not initialized.

Common situations: Redis down or restarted when Prometheus scrapes; wrong REDIS address/password in the metrics exporter config; network partition between exporter and Redis; using an Inspector built with invalid RedisConnOpt.

Understand the failure class

Background: Database query failed: Internal Server Error 500s wrapping SQL, Prisma, and connection failures — what to check first — this error's family across 16 libraries.

Related errors

AI-assisted analysis of hibiken/asynq@d135f1439b (2026-09-07). Data as JSON: /api/errors/816e22808baf6ce4. Report an issue: GitHub.

Appendix: source

Thrown at x/metrics/metrics.go:28

)

// Namespace used in fully-qualified metrics names.
const namespace = "asynq"

// QueueMetricsCollector gathers queue metrics.
// It implements prometheus.Collector interface.
//
// All metrics exported from this collector have prefix "asynq".
type QueueMetricsCollector struct {
	inspector *asynq.Inspector
}

// collectQueueInfo gathers QueueInfo of all queues.
// Since this operation is expensive, it must be called once per collection.
func (qmc *QueueMetricsCollector) collectQueueInfo() ([]*asynq.QueueInfo, error) {
	qnames, err := qmc.inspector.Queues()
	if err != nil {
		return nil, fmt.Errorf("failed to get queue names: %w", err)
	}
	infos := make([]*asynq.QueueInfo, len(qnames))
	for i, qname := range qnames {
		qinfo, err := qmc.inspector.GetQueueInfo(qname)
		if err != nil {
			return nil, fmt.Errorf("failed to get queue info: %w", err)
		}
		infos[i] = qinfo
	}
	return infos, nil
}

// Descriptors used by QueueMetricsCollector
var (
	tasksQueuedDesc = prometheus.NewDesc(
		prometheus.BuildFQName(namespace, "", "tasks_enqueued_total"),
		"Number of tasks enqueued; broken down by queue and state.",
		[]string{"queue", "state"}, nil,

View on GitHub (pinned to d135f1439b)