jaegertracing/jaeger · error

error reading service_names from storage: %w

Error message

error reading service_names from storage: %w

What it means

GetServices lists distinct service names from the service_names table; if the gocql iterator returns an error when closed (the query failed during or after scanning), the driver error is wrapped with this message so the caller knows the service-name listing read failed in storage.

Source

Thrown at internal/storage/v1/cassandra/spanstore/service_names.go:97

	// it simply means we might write the same service name twice.
	inCache := c.Get(key)
	if inCache == nil {
		c.Put(key, key)
	}
	return inCache != nil
}

// GetServices returns all services traced by Jaeger
func (s *ServiceNamesStorage) GetServices() ([]string, error) {
	iter := s.session.Query(s.QueryStmt).Iter()

	var service string
	var services []string
	for iter.Scan(&service) {
		services = append(services, service)
	}
	if err := iter.Close(); err != nil {
		err = fmt.Errorf("error reading service_names from storage: %w", err)
		return nil, err
	}
	return services, nil
}

View on GitHub (pinned to 806f444784)

Solutions

1. Retry GetServices after checking cluster health; it is a cheap read-only call.
2. Inspect the wrapped driver error and Cassandra logs for the root cause.
3. Tune Cassandra read timeouts/retry policy if timeouts are common.
4. Verify the service_names table exists and is readable (permissions, schema version).

services, err := reader.GetServices(ctx)
if err != nil {
    if strings.Contains(err.Error(), "error reading service_names") {
        // transient read failure: retry with backoff
    }
    return err
}

When it happens

Trigger: Calling GetServices on the service/operation names storage while the underlying SELECT from service_names fails — Cassandra read timeout, node unavailability, or connection error during the scan.

Common situations: Cluster under load (timeouts on the small but frequently-hit service_names table); network issues; node restarts; stale schema state after failed migrations.

Related errors

• error reading operation_names from storage: %w
• error reading %s from storage: %w
• invalid version
• failed to Exec query '%s': %w
• failed to acquire resource lock due to cassandra error: %w

AI-assisted analysis of jaegertracing/jaeger@806f444784 (2026-09-01). Data as JSON: /api/errors/7181d963fe9f587b. Report an issue: GitHub.