feat: Add metrics module

config/config.go

@@ -9,16 +9,17 @@ import (
 )
 
 type Config struct {
-	DB             DbConfig             `toml:"db"`
-	JWT            JWTConfig            `toml:"jwt"`
-	Log            LogConfig            `toml:"log"`
-	API            APIConfig            `toml:"api"`
-	Node           NodeConfig           `toml:"node"`
-	MessageService MessageServiceConfig `toml:"messageService"`
-	MessageState   MessageStateConfig   `toml:"messageState"`
-	Gateway        GatewayConfig        `toml:"gateway"`
-	RateLimit      RateLimitConfig      `toml:"rateLimit"`
-	Trace          metrics.TraceConfig  `toml:"tracing"`
+	DB             DbConfig               `toml:"db"`
+	JWT            JWTConfig              `toml:"jwt"`
+	Log            LogConfig              `toml:"log"`
+	API            APIConfig              `toml:"api"`
+	Node           NodeConfig             `toml:"node"`
+	MessageService MessageServiceConfig   `toml:"messageService"`
+	MessageState   MessageStateConfig     `toml:"messageState"`
+	Gateway        GatewayConfig          `toml:"gateway"`
+	RateLimit      RateLimitConfig        `toml:"rateLimit"`
+	Trace          *metrics.TraceConfig   `toml:"tracing"`
+	Metrics        *metrics.MetricsConfig `toml:"metrics"`
 }
 
 type NodeConfig struct {
@@ -137,11 +138,7 @@ func DefaultConfig() *Config {
 			Cfg:   gatewayTypes.Config{},
 		},
 		RateLimit: RateLimitConfig{Redis: ""},
-		Trace: metrics.TraceConfig{
-			JaegerEndpoint:       "",
-			ProbabilitySampler:   1.0,
-			JaegerTracingEnabled: false,
-			ServerName:           "venus-messenger",
-		},
+		Trace:     metrics.DefaultTraceConfig(),
+		Metrics:   metrics.DefaultMetricsConfig(),
 	}
 }

docs/zh/metrics-config-desc.md

@@ -0,0 +1,74 @@
+# metrics 配置及使用说明
+
+## 配置
+
+`Metrics` 基本的配置样例如下：
+```toml
+[Metrics]
+  # 是否开启metrics指标统计，默认为false
+  Enabled = false
+
+  [Metrics.Exporter]
+    # 指标导出器类型，目前可选：prometheus或graphite，默认为prometheus
+    Type = "prometheus"
+
+    [Metrics.Exporter.Prometheus]
+      # multiaddr
+      EndPoint = "/ip4/0.0.0.0/tcp/4568"
+      # 命名规范: "a_b_c", 不能带"-"
+      Namespace = "messager01" 
+      # 指标注册表类型，可选：default（默认，会附带程序运行的环境指标）或 define（自定义）
+      RegistryType = "define"
+      # prometheus 服务路径
+      Path = "/debug/metrics"
+      # 上报周期，单位为 秒（s）
+      ReportingPeriod = 10
+
+    [Metrics.Exporter.Graphite]
+      # 命名规范: "a_b_c", 不能带"-"
+      Namespace = "messager01" 
+      # graphite exporter 收集器服务地址
+      Host = "127.0.0.1"
+      # graphite exporter 收集器服务监听端口
+      Port = 4568
+      # 上报周期，单位为 秒（s）
+      ReportingPeriod = 10
+```
+## 导出器
+
+目前可以选择两类导出器（`exporter`）：`Prometheus exporter` 或 `Graphite exporter`，默认是前者。
+
+如果配置 `Prometheus exporter`，则在 `venus-messager` 服务启动时会附带启动 `Prometheus exporter` 的监听服务，可以通过以下方式快速查看指标：
+
+
+```bash
+ $  curl http://localhost:4568/debug/metrics
+   # HELP messager01_chain_head_stable_dur_s Duration of chain head stabilization
+   # TYPE messager01_chain_head_stable_dur_s histogram
+   messager01_chain_head_stable_dur_s_bucket{le="8"} 0
+   messager01_chain_head_stable_dur_s_bucket{le="9"} 11
+   messager01_chain_head_stable_dur_s_bucket{le="10"} 27
+   messager01_chain_head_stable_dur_s_bucket{le="12"} 43
+   messager01_chain_head_stable_dur_s_bucket{le="14"} 48
+   messager01_chain_head_stable_dur_s_bucket{le="16"} 49
+   messager01_chain_head_stable_dur_s_bucket{le="18"} 49
+   messager01_chain_head_stable_dur_s_bucket{le="20"} 49
+   messager01_chain_head_stable_dur_s_bucket{le="25"} 49
+   messager01_chain_head_stable_dur_s_bucket{le="30"} 49
+   messager01_chain_head_stable_dur_s_bucket{le="60"} 50
+   messager01_chain_head_stable_dur_s_bucket{le="+Inf"} 50
+   messager01_chain_head_stable_dur_s_sum 503.99999999999983
+   messager01_chain_head_stable_dur_s_count 50
+   # HELP messager01_chain_head_stable_s Delay of chain head stabilization
+   # TYPE messager01_chain_head_stable_s gauge
+   messager01_chain_head_stable_s 9
+   ... ...
+```
+> 如果遇到错误 `curl: (56) Recv failure: Connection reset by peer`, 请使用本机 `ip` 地址, 如下所示:
+```bash
+$  curl http://<ip>:4568/debug/metrics
+```
+
+如果配置 `Graphite exporter`，需要先启动 `Graphite exporter` 的收集器服务， `venus-messager` 服务启动时将指标上报给收集器。服务启动参考 [Graphite exporter](https://github.com/prometheus/graphite_exporter) 中的说明。
+
+`Graphite exporter` 和 `Prometheus exporter` 自身都不带图形界面的，如果需要可视化监控及更高阶的图表分析，请到 `venus-docs` 项目中查找关于 `Prometheus+Grafana` 的说明文档。

go.mod

@@ -3,6 +3,7 @@ module github.com/filecoin-project/venus-messager
 go 1.17
 
 require (
+	contrib.go.opencensus.io/exporter/graphite v0.0.0-20200424223504-26b90655e0ce // indirect
 	github.com/acarl005/stripansi v0.0.0-20180116102854-5a71ef0e047d
 	github.com/fatih/color v1.13.0
 	github.com/filecoin-project/go-address v0.0.6
@@ -15,7 +16,7 @@ require (
 	github.com/gbrlsnchs/jwt/v3 v3.0.1
 	github.com/google/uuid v1.3.0
 	github.com/hunjixin/automapper v0.0.0-20191127090318-9b979ce72ce2
-	github.com/ipfs-force-community/metrics v1.0.1-0.20211228055608-9462dc86e157
+	github.com/ipfs-force-community/metrics v1.0.1-0.20220719063006-2c54bb379466
 	github.com/ipfs-force-community/venus-common-utils v0.0.0-20210924063144-1d3a5b30de87
 	github.com/ipfs-force-community/venus-gateway v1.6.0
 	github.com/ipfs/go-cid v0.1.0
@@ -36,6 +37,8 @@ require (
 	modernc.org/mathutil v1.1.1
 )
 
+require go.opencensus.io v0.23.0
+
 require (
 	contrib.go.opencensus.io/exporter/jaeger v0.2.1 // indirect
 	contrib.go.opencensus.io/exporter/prometheus v0.4.0 // indirect
@@ -163,10 +166,7 @@ require (
 	github.com/subosito/gotenv v1.2.0 // indirect
 	github.com/uber/jaeger-client-go v2.25.0+incompatible // indirect
 	github.com/ugorji/go/codec v1.2.4 // indirect
-	github.com/warpfork/go-testmark v0.9.0 // indirect
 	github.com/whyrusleeping/go-logging v0.0.1 // indirect
-	github.com/xlab/c-for-go v0.0.0-20200718154222-87b0065af829 // indirect
-	go.opencensus.io v0.23.0 // indirect
 	go.uber.org/atomic v1.9.0 // indirect
 	go.uber.org/dig v1.12.0 // indirect
 	go.uber.org/multierr v1.8.0 // indirect