crowdsecurity/crowdsec · error · ErrInvalidMetricsLevel

%w: %s

Error message

%w: %s

What it means

RegisterMetrics maps a user-supplied metrics level string (e.g. full, standard, none) to a set of Prometheus collectors via a switch. An unrecognized level string falls through to default and returns this error wrapped around ErrInvalidMetricsLevel, aborting metrics registration.

Source

Thrown at pkg/metrics/metrics.go:53

	case MetricsLevelAggregated:
		prometheus.MustRegister(GlobalParserHits, GlobalParserHitsOk, GlobalParserHitsKo,
			GlobalCsInfo, GlobalParsingHistogram, GlobalPourHistogram,
			BucketsUnderflow, BucketsCanceled, BucketsInstantiation, BucketsOverflow,
			LapiRouteHits,
			BucketsCurrentCount,
			CacheMetrics, RegexpCacheMetrics, NodesWlHitsOk, NodesWlHits,
			PapiOrdersReceived, PapiInvalidOrdersReceived, PapiLastPullTimestamp, PapiPollErrors)
	case MetricsLevelFull:
		prometheus.MustRegister(GlobalParserHits, GlobalParserHitsOk, GlobalParserHitsKo,
			NodesHits, NodesHitsOk, NodesHitsKo,
			GlobalCsInfo, GlobalParsingHistogram, GlobalPourHistogram,
			LapiRouteHits, LapiMachineHits, LapiBouncerHits, LapiNilDecisions, LapiNonNilDecisions, LapiResponseTime,
			BucketsPour, BucketsUnderflow, BucketsCanceled, BucketsInstantiation, BucketsOverflow, BucketsCurrentCount,
			GlobalActiveDecisions, GlobalAlerts, GlobalMachinesLastHeartbeatTimestamp, NodesWlHitsOk, NodesWlHits,
			CacheMetrics, RegexpCacheMetrics,
			PapiOrdersReceived, PapiInvalidOrdersReceived, PapiLastPullTimestamp, PapiPollErrors)
	default:
		return fmt.Errorf("%w: %s", ErrInvalidMetricsLevel, metricsLevel)
	}
	return nil
}

View on GitHub (pinned to 909b515798)

Solutions

1. Set prometheus.level in the config to one of the supported values: none, standard, or full.
2. Check for typos, trailing whitespace, and correct casing in the config value.
3. Consult the current version's documentation, as the accepted level names may have changed between releases.

Example fix

// before (config.yaml)
prometheus:
  level: verbose
// after
prometheus:
  level: full

var validLevels = map[string]bool{"none": true, "standard": true, "full": true}
level := strings.TrimSpace(cfg.Prometheus.Level)
if !validLevels[level] {
    return fmt.Errorf("prometheus.level %q must be one of: none, standard, full", level)
}

if err := metrics.RegisterMetrics(level); err != nil {
    if errors.Is(err, metrics.ErrInvalidMetricsLevel) {
        logger.Fatalf("unsupported prometheus level %q: use none|standard|full", level)
    }
    return err
}

When it happens

Trigger: registerPrometheus calls RegisterMetrics with a prometheus_level config value that matches none of the switch cases (typo, wrong case, or unsupported level).

Common situations: config.yaml has prometheus: level: verbose instead of full/standard/none; upgrading crowdsec where a level name was renamed; hand-edited config with trailing whitespace or wrong casing.

Understand the failure class

Background: Invalid enum value errors: "Unknown type", "Invalid scope", "must be one of" — when a string is not on the library's allowed list — this error's family across 23 libraries.

Related errors

• path must start with /
• on_challenge hooks are only valid in-band, not under outofba
• empty cti key
• no listen_uri or listen_socket specified
• no configuration paths provided

AI-assisted analysis of crowdsecurity/crowdsec@909b515798 (2026-09-06). Data as JSON: /api/errors/78da8b6300dc72f5. Report an issue: GitHub.