Add Summary metrics and configurable service-name for backward compatibility

Author: fridrik01

beacon/blockchain/blob_fetcher_metrics.go

@@ -45,41 +45,36 @@ func NewBlobFetcherMetrics(factory metrics.Factory) *BlobFetcherMetrics {
 	return &BlobFetcherMetrics{
 		RetriesTotal: factory.NewCounter(
 			metrics.CounterOpts{
-				Subsystem: "blob_fetcher",
-				Name:      "retries_total",
-				Help:      "Number of times a blob request was retried after failure",
+				Name: "beacon_kit_blob_fetcher_retries_total",
+				Help: "Number of times a blob request was retried after failure",
 			},
 			nil,
 		),
 		RequestsExpiredTotal: factory.NewCounter(
 			metrics.CounterOpts{
-				Subsystem: "blob_fetcher",
-				Name:      "requests_expired_total",
-				Help:      "Number of blob fetch requests that expired before completion",
+				Name: "beacon_kit_blob_fetcher_requests_expired_total",
+				Help: "Number of blob fetch requests that expired before completion",
 			},
 			[]string{"reason"},
 		),
 		RequestsCompletedTotal: factory.NewCounter(
 			metrics.CounterOpts{
-				Subsystem: "blob_fetcher",
-				Name:      "requests_completed_total",
-				Help:      "Number of blob fetch requests that completed successfully",
+				Name: "beacon_kit_blob_fetcher_requests_completed_total",
+				Help: "Number of blob fetch requests that completed successfully",
 			},
 			nil,
 		),
 		RequestsQueuedTotal: factory.NewCounter(
 			metrics.CounterOpts{
-				Subsystem: "blob_fetcher",
-				Name:      "requests_queued_total",
-				Help:      "Number of new blob fetch requests added to the queue",
+				Name: "beacon_kit_blob_fetcher_requests_queued_total",
+				Help: "Number of new blob fetch requests added to the queue",
 			},
 			nil,
 		),
 		QueueDepth: factory.NewGauge(
 			metrics.GaugeOpts{
-				Subsystem: "blob_fetcher",
-				Name:      "queue_depth",
-				Help:      "Current depth of the blob fetcher queue",
+				Name: "beacon_kit_blob_fetcher_queue_depth",
+				Help: "Current depth of the blob fetcher queue",
 			},
 			nil,
 		),

beacon/blockchain/metrics.go

@@ -25,17 +25,16 @@ import (
 
 	"github.com/berachain/beacon-kit/observability/metrics"
 	"github.com/berachain/beacon-kit/primitives/math"
-	"github.com/prometheus/client_golang/prometheus"
 )
 
 // Metrics is a struct that contains metrics for the blockchain service.
 type Metrics struct {
-	StateTransitionDuration               metrics.Histogram
+	StateTransitionDuration               metrics.Summary
 	RebuildPayloadForRejectedBlockSuccess metrics.Counter
 	RebuildPayloadForRejectedBlockFailure metrics.Counter
 	OptimisticPayloadBuildSuccess         metrics.Counter
 	OptimisticPayloadBuildFailure         metrics.Counter
-	StateRootVerificationDuration         metrics.Histogram
+	StateRootVerificationDuration         metrics.Summary
 	FailedToGetBlockLogs                  metrics.Counter
 	FailedToEnqueueDeposits               metrics.Counter
 }
@@ -44,69 +43,61 @@ type Metrics struct {
 // Metric names are kept identical to cosmos-sdk/telemetry output for Grafana compatibility.
 func NewMetrics(factory metrics.Factory) *Metrics {
 	return &Metrics{
-		StateTransitionDuration: factory.NewHistogram(
-			metrics.HistogramOpts{
-				Subsystem: "beacon_blockchain",
-				Name:      "state_transition_duration",
-				Help:      "Time taken to process state transition in seconds",
-				Buckets:   prometheus.ExponentialBucketsRange(0.001, 10, 10),
+		StateTransitionDuration: factory.NewSummary(
+			metrics.SummaryOpts{
+				Name:       "beacon_kit_beacon_blockchain_state_transition_duration",
+				Help:       "Time taken to process state transition in seconds",
+				Objectives: metrics.QuantilesP50P90P99,
 			},
 			nil,
 		),
 		RebuildPayloadForRejectedBlockSuccess: factory.NewCounter(
 			metrics.CounterOpts{
-				Subsystem: "blockchain",
-				Name:      "rebuild_payload_for_rejected_block_success",
-				Help:      "Number of successful payload rebuilds for rejected blocks",
+				Name: "beacon_kit_blockchain_rebuild_payload_for_rejected_block_success",
+				Help: "Number of successful payload rebuilds for rejected blocks",
 			},
 			[]string{"slot"},
 		),
 		RebuildPayloadForRejectedBlockFailure: factory.NewCounter(
 			metrics.CounterOpts{
-				Subsystem: "blockchain",
-				Name:      "rebuild_payload_for_rejected_block_failure",
-				Help:      "Number of failed payload rebuilds for rejected blocks",
+				Name: "beacon_kit_blockchain_rebuild_payload_for_rejected_block_failure",
+				Help: "Number of failed payload rebuilds for rejected blocks",
 			},
 			[]string{"slot", "error"},
 		),
 		OptimisticPayloadBuildSuccess: factory.NewCounter(
 			metrics.CounterOpts{
-				Subsystem: "blockchain",
-				Name:      "optimistic_payload_build_success",
-				Help:      "Number of successful optimistic payload builds",
+				Name: "beacon_kit_blockchain_optimistic_payload_build_success",
+				Help: "Number of successful optimistic payload builds",
 			},
 			[]string{"slot"},
 		),
 		OptimisticPayloadBuildFailure: factory.NewCounter(
 			metrics.CounterOpts{
-				Subsystem: "blockchain",
-				Name:      "optimistic_payload_build_failure",
-				Help:      "Number of failed optimistic payload builds",
+				Name: "beacon_kit_blockchain_optimistic_payload_build_failure",
+				Help: "Number of failed optimistic payload builds",
 			},
 			[]string{"slot", "error"},
 		),
-		StateRootVerificationDuration: factory.NewHistogram(
-			metrics.HistogramOpts{
-				Subsystem: "blockchain",
-				Name:      "state_root_verification_duration",
-				Help:      "Time taken to verify state root in seconds",
-				Buckets:   prometheus.ExponentialBucketsRange(0.001, 10, 10),
+		StateRootVerificationDuration: factory.NewSummary(
+			metrics.SummaryOpts{
+				Name:       "beacon_kit_blockchain_state_root_verification_duration",
+				Help:       "Time taken to verify state root in seconds",
+				Objectives: metrics.QuantilesP50P90P99,
 			},
 			nil,
 		),
 		FailedToGetBlockLogs: factory.NewCounter(
 			metrics.CounterOpts{
-				Subsystem: "execution_deposit",
-				Name:      "failed_to_get_block_logs",
-				Help:      "Number of times failed to read deposits from execution layer block logs",
+				Name: "beacon_kit_execution_deposit_failed_to_get_block_logs",
+				Help: "Number of times failed to read deposits from execution layer block logs",
 			},
 			[]string{"block_num"},
 		),
 		FailedToEnqueueDeposits: factory.NewCounter(
 			metrics.CounterOpts{
-				Subsystem: "execution_deposit",
-				Name:      "failed_to_enqueue_deposits",
-				Help:      "Number of times failed to enqueue deposits to storage",
+				Name: "beacon_kit_execution_deposit_failed_to_enqueue_deposits",
+				Help: "Number of times failed to enqueue deposits to storage",
 			},
 			[]string{"block_num"},
 		),

beacon/validator/metrics.go

@@ -25,16 +25,17 @@ import (
 
 	"github.com/berachain/beacon-kit/observability/metrics"
 	"github.com/berachain/beacon-kit/primitives/math"
-	"github.com/prometheus/client_golang/prometheus"
 )
 
 // Metrics is a struct that contains metrics for the validator service.
 type Metrics struct {
 	// RequestBlockForProposalDuration tracks time to request block for proposal
-	RequestBlockForProposalDuration metrics.Histogram
+	// Using Summary for backward compatibility with cosmos-sdk/telemetry.
+	RequestBlockForProposalDuration metrics.Summary
 
 	// StateRootComputationDuration tracks time to compute state root
-	StateRootComputationDuration metrics.Histogram
+	// Using Summary for backward compatibility with cosmos-sdk/telemetry.
+	StateRootComputationDuration metrics.Summary
 
 	// FailedToRetrievePayload tracks failed payload retrievals
 	FailedToRetrievePayload metrics.Counter
@@ -44,29 +45,26 @@ type Metrics struct {
 // Metric names are kept identical to cosmos-sdk/telemetry output for Grafana compatibility.
 func NewMetrics(factory metrics.Factory) *Metrics {
 	return &Metrics{
-		RequestBlockForProposalDuration: factory.NewHistogram(
-			metrics.HistogramOpts{
-				Subsystem: "validator",
-				Name:      "request_block_for_proposal_duration",
-				Help:      "Time taken to request block for proposal in seconds",
-				Buckets:   prometheus.ExponentialBucketsRange(0.001, 10, 10),
+		RequestBlockForProposalDuration: factory.NewSummary(
+			metrics.SummaryOpts{
+				Name:       "beacon_kit_validator_request_block_for_proposal_duration",
+				Help:       "Time taken to request block for proposal in seconds",
+				Objectives: metrics.QuantilesP50P90P99,
 			},
 			nil,
 		),
-		StateRootComputationDuration: factory.NewHistogram(
-			metrics.HistogramOpts{
-				Subsystem: "validator",
-				Name:      "state_root_computation_duration",
-				Help:      "Time taken to compute state root in seconds",
-				Buckets:   prometheus.ExponentialBucketsRange(0.001, 10, 10),
+		StateRootComputationDuration: factory.NewSummary(
+			metrics.SummaryOpts{
+				Name:       "beacon_kit_validator_state_root_computation_duration",
+				Help:       "Time taken to compute state root in seconds",
+				Objectives: metrics.QuantilesP50P90P99,
 			},
 			nil,
 		),
 		FailedToRetrievePayload: factory.NewCounter(
 			metrics.CounterOpts{
-				Subsystem: "validator",
-				Name:      "failed_to_retrieve_payload",
-				Help:      "Number of times validator failed to retrieve payload",
+				Name: "beacon_kit_validator_failed_to_retrieve_payload",
+				Help: "Number of times validator failed to retrieve payload",
 			},
 			[]string{"slot", "error"},
 		),

config/config/config.go

@@ -82,6 +82,9 @@ type TelemetryConfig struct {
 	// When true, all beacon-kit services emit Prometheus metrics.
 	// When false, all metrics are no-op with zero runtime overhead.
 	Enabled bool `mapstructure:"enabled"`
+
+	// ServiceName defines the namespace for all Prometheus metrics.
+	ServiceName string `mapstructure:"service-name"`
 }
 
 // Config defines the server's top level configuration.
@@ -106,7 +109,8 @@ func DefaultConfig() *Config {
 			IAVLDisableFastNode: false,
 		},
 		Telemetry: TelemetryConfig{
-			Enabled: false,
+			Enabled:     false,
+			ServiceName: "beacond_node",
 		},
 	}
 }

config/config/config.toml.tpl

@@ -50,7 +50,7 @@ inter-block-cache = {{ .BaseConfig.InterBlockCache }}
 # IavlCacheSize set the size of the iavl tree cache (in number of nodes).
 iavl-cache-size = {{ .BaseConfig.IAVLCacheSize }}
 
-# IAVLDisableFastNode enables or disables the fast node feature of IAVL. 
+# IAVLDisableFastNode enables or disables the fast node feature of IAVL.
 # Default is false.
 iavl-disable-fastnode = {{ .BaseConfig.IAVLDisableFastNode }}
 
@@ -66,3 +66,6 @@ iavl-disable-fastnode = {{ .BaseConfig.IAVLDisableFastNode }}
 # When false, all metrics are no-op with zero runtime overhead.
 # Default: false
 enabled = {{ .Telemetry.Enabled }}
+
+# ServiceName defines the namespace prefix for all Prometheus metrics.
+service-name = "{{ .Telemetry.ServiceName }}"

consensus/cometbft/service/metrics.go

@@ -24,64 +24,55 @@ import (
 	"time"
 
 	"github.com/berachain/beacon-kit/observability/metrics"
-	prominternal "github.com/prometheus/client_golang/prometheus"
 )
 
 // Metrics holds metrics for the CometBFT service.
-//
-// Metric names are kept identical to cosmos-sdk/telemetry output for Grafana compatibility.
 type Metrics struct {
 	// QueryCount tracks the number of ABCI queries received, labeled by query path.
 	QueryCount metrics.Counter
 
 	// QueryDuration tracks the time taken to process ABCI queries, labeled by query path.
-	QueryDuration metrics.Histogram
+	QueryDuration metrics.Summary
 
 	// PrepareProposalDuration tracks the time taken to prepare a proposal.
-	PrepareProposalDuration metrics.Histogram
+	PrepareProposalDuration metrics.Summary
 
 	// ProcessProposalDuration tracks the time taken to process a proposal.
-	ProcessProposalDuration metrics.Histogram
+	ProcessProposalDuration metrics.Summary
 }
 
-// NewMetrics creates a new Metrics instance using the provided factory.
-// The factory determines whether real Prometheus metrics or no-op metrics are created.
-//
-//nolint:mnd // magic numbers are histogram bucket ranges for timing metrics
+// NewMetrics creates a new Metrics instance using the provided factory. The factory determines
+// whether real Prometheus metrics or no-op metrics are created.
 func NewMetrics(factory metrics.Factory) *Metrics {
 	return &Metrics{
 		QueryCount: factory.NewCounter(
 			metrics.CounterOpts{
-				Subsystem: "comet",
-				Name:      "query_count",
-				Help:      "Total number of ABCI queries received",
+				Name: "beacon_kit_comet_query_count",
+				Help: "Total number of ABCI queries received",
 			},
 			[]string{"path"},
 		),
-		QueryDuration: factory.NewHistogram(
-			metrics.HistogramOpts{
-				Subsystem: "comet",
-				Name:      "query_duration",
-				Help:      "Time taken to process ABCI queries in seconds",
-				Buckets:   prominternal.ExponentialBucketsRange(0.001, 10, 10),
+		QueryDuration: factory.NewSummary(
+			metrics.SummaryOpts{
+				Name:       "beacon_kit_comet_query_duration",
+				Help:       "Time taken to process ABCI queries in seconds",
+				Objectives: metrics.QuantilesP50P90P99,
 			},
 			[]string{"path"},
 		),
-		PrepareProposalDuration: factory.NewHistogram(
-			metrics.HistogramOpts{
-				Subsystem: "runtime",
-				Name:      "prepare_proposal_duration",
-				Help:      "Time taken to prepare a proposal in seconds",
-				Buckets:   prominternal.ExponentialBucketsRange(0.001, 10, 10),
+		PrepareProposalDuration: factory.NewSummary(
+			metrics.SummaryOpts{
+				Name:       "beacon_kit_runtime_prepare_proposal_duration",
+				Help:       "Time taken to prepare a proposal in seconds",
+				Objectives: metrics.QuantilesP50P90P99,
 			},
 			nil,
 		),
-		ProcessProposalDuration: factory.NewHistogram(
-			metrics.HistogramOpts{
-				Subsystem: "runtime",
-				Name:      "process_proposal_duration",
-				Help:      "Time taken to process a proposal in seconds",
-				Buckets:   prominternal.ExponentialBucketsRange(0.001, 10, 10),
+		ProcessProposalDuration: factory.NewSummary(
+			metrics.SummaryOpts{
+				Name:       "beacon_kit_runtime_process_proposal_duration",
+				Help:       "Time taken to process a proposal in seconds",
+				Objectives: metrics.QuantilesP50P90P99,
 			},
 			nil,
 		),