Implement initial draft

.gitignore

@@ -0,0 +1,12 @@
+# Built binaries
+
+bin/*
+!bin/.gitkeep
+
+# Goreleaser artifacts
+
+dist/*
+
+# VS Code
+
+.vscode

.goreleaser.yml

@@ -0,0 +1,30 @@
+env:
+  - CGO_ENABLED=0
+
+builds:
+  - main: ./cmd
+    targets:
+      - darwin_amd64
+      - darwin_arm64
+      - linux_386
+      - linux_amd64
+      - linux_arm
+      - linux_arm64
+      - windows_386
+      - windows_amd64
+
+archives:
+  - id: zip
+    format: zip
+    name_template: "{{ .ProjectName }}_{{ .Os }}_{{ .Arch }}"
+    files:
+      - none*
+
+checksum:
+  name_template: 'checksums.txt'
+
+signs:
+  - artifacts: checksum
+
+release:
+  prerelease: auto

Makefile

@@ -0,0 +1,14 @@
+default: build
+
+.PHONY: build
+build:
+	CGO_ENABLED=0 go build -o ./bin/node-healthchecker github.com/flashbots/node-healthchecker/cmd
+
+.PHONY: snapshot
+snapshot:
+	goreleaser release --snapshot --rm-dist
+
+.PHONY: release
+release:
+	@rm -rf ./dist
+	GITHUB_TOKEN=$$( gh auth token ) goreleaser release

cmd/main.go

@@ -0,0 +1,93 @@
+package main
+
+import (
+	"fmt"
+	"os"
+	"strings"
+
+	"github.com/urfave/cli/v2"
+	"go.uber.org/zap"
+	"go.uber.org/zap/zapcore"
+)
+
+var (
+	version = "development"
+)
+
+func main() {
+	var logFormat, logLevel string
+
+	app := &cli.App{
+		Name:    "node-healthchecker",
+		Usage:   "Aggregates complex health-checks of a blockchain node into a simple http endpoint",
+		Version: version,
+
+		Action: func(c *cli.Context) error {
+			return cli.ShowAppHelp(c)
+		},
+
+		Flags: []cli.Flag{
+			&cli.StringFlag{
+				Destination: &logLevel,
+				EnvVars:     []string{"LOG_LEVEL"},
+				Name:        "log-level",
+				Usage:       "logging level",
+				Value:       "info",
+			},
+
+			&cli.StringFlag{
+				Destination: &logFormat,
+				EnvVars:     []string{"LOG_MODE"},
+				Name:        "log-mode",
+				Usage:       "logging mode",
+				Value:       "prod",
+			},
+		},
+
+		Before: func(ctx *cli.Context) error {
+			err := setupLogger(logLevel, logFormat)
+			if err != nil {
+				fmt.Fprintf(os.Stderr, "failed to configure the logging: %s\n", err)
+			}
+			return err
+		},
+
+		Commands: []*cli.Command{
+			CommandServe(),
+		},
+	}
+
+	defer func() {
+		zap.L().Sync()
+	}()
+	if err := app.Run(os.Args); err != nil {
+		zap.L().Error("Failed with error", zap.Error(err))
+	}
+}
+
+func setupLogger(level, mode string) error {
+	var config zap.Config
+	switch strings.ToLower(mode) {
+	case "dev":
+		config = zap.NewDevelopmentConfig()
+	case "prod":
+		config = zap.NewProductionConfig()
+	default:
+		return fmt.Errorf("invalid log-mode '%s'", mode)
+	}
+	config.EncoderConfig.EncodeTime = zapcore.ISO8601TimeEncoder
+
+	logLevel, err := zap.ParseAtomicLevel(level)
+	if err != nil {
+		return fmt.Errorf("invalid log-level '%s': %w", level, err)
+	}
+	config.Level = logLevel
+
+	l, err := config.Build()
+	if err != nil {
+		return fmt.Errorf("failed to build the logger: %w", err)
+	}
+	zap.ReplaceGlobals(l)
+
+	return nil
+}

cmd/serve.go

@@ -0,0 +1,62 @@
+package main
+
+import (
+	"time"
+
+	"github.com/flashbots/node-healthchecker/healthchecker"
+	"github.com/urfave/cli/v2"
+)
+
+func CommandServe() *cli.Command {
+	var cfg healthchecker.Config
+
+	return &cli.Command{
+		Name:  "serve",
+		Usage: "run the healthcheck server",
+
+		Flags: []cli.Flag{
+
+			// Serving
+
+			&cli.StringFlag{
+				Category:    "Serving:",
+				Destination: &cfg.ServeAddress,
+				Name:        "serve-address",
+				Usage:       "respond to health-checks at the address of `host:port`",
+				Value:       "0.0.0.0:8080",
+			},
+
+			&cli.DurationFlag{
+				Category:    "Serving:",
+				Destination: &cfg.Timeout,
+				Name:        "timeout",
+				Usage:       "healthcheck(s) timeout `duration`",
+				Value:       time.Second,
+			},
+
+			// Monitoring
+
+			&cli.StringFlag{
+				Category:    "Monitoring:",
+				Destination: &cfg.MonitorGethURL,
+				Name:        "monitor-geth-url",
+				Usage:       "monitor geth sync-status via RPC at specified `URL`",
+			},
+
+			&cli.StringFlag{
+				Category:    "Monitoring:",
+				Destination: &cfg.MonitorLighthouseURL,
+				Name:        "monitor-lighthouse-url",
+				Usage:       "monitor lighthouse sync-status via RPC at specified `URL`",
+			},
+		},
+
+		Action: func(ctx *cli.Context) error {
+			h, err := healthchecker.New(&cfg)
+			if err != nil {
+				return err
+			}
+			return h.Serve()
+		},
+	}
+}

go.mod

@@ -0,0 +1,16 @@
+module github.com/flashbots/node-healthchecker
+
+go 1.21
+
+require (
+	github.com/google/uuid v1.3.1
+	github.com/urfave/cli/v2 v2.25.7
+	go.uber.org/zap v1.26.0
+)
+
+require (
+	github.com/cpuguy83/go-md2man/v2 v2.0.2 // indirect
+	github.com/russross/blackfriday/v2 v2.1.0 // indirect
+	github.com/xrash/smetrics v0.0.0-20201216005158-039620a65673 // indirect
+	go.uber.org/multierr v1.11.0 // indirect
+)

go.sum

@@ -0,0 +1,24 @@
+github.com/cpuguy83/go-md2man/v2 v2.0.2 h1:p1EgwI/C7NhT0JmVkwCD2ZBK8j4aeHQX2pMHHBfMQ6w=
+github.com/cpuguy83/go-md2man/v2 v2.0.2/go.mod h1:tgQtvFlXSQOSOSIRvRPT7W67SCa46tRHOmNcaadrF8o=
+github.com/davecgh/go-spew v1.1.1 h1:vj9j/u1bqnvCEfJOwUhtlOARqs3+rkHYY13jYWTU97c=
+github.com/davecgh/go-spew v1.1.1/go.mod h1:J7Y8YcW2NihsgmVo/mv3lAwl/skON4iLHjSsI+c5H38=
+github.com/google/uuid v1.3.1 h1:KjJaJ9iWZ3jOFZIf1Lqf4laDRCasjl0BCmnEGxkdLb4=
+github.com/google/uuid v1.3.1/go.mod h1:TIyPZe4MgqvfeYDBFedMoGGpEw/LqOeaOT+nhxU+yHo=
+github.com/pmezard/go-difflib v1.0.0 h1:4DBwDE0NGyQoBHbLQYPwSUPoCMWR5BEzIk/f1lZbAQM=
+github.com/pmezard/go-difflib v1.0.0/go.mod h1:iKH77koFhYxTK1pcRnkKkqfTogsbg7gZNVY4sRDYZ/4=
+github.com/russross/blackfriday/v2 v2.1.0 h1:JIOH55/0cWyOuilr9/qlrm0BSXldqnqwMsf35Ld67mk=
+github.com/russross/blackfriday/v2 v2.1.0/go.mod h1:+Rmxgy9KzJVeS9/2gXHxylqXiyQDYRxCVz55jmeOWTM=
+github.com/stretchr/testify v1.8.1 h1:w7B6lhMri9wdJUVmEZPGGhZzrYTPvgJArz7wNPgYKsk=
+github.com/stretchr/testify v1.8.1/go.mod h1:w2LPCIKwWwSfY2zedu0+kehJoqGctiVI29o6fzry7u4=
+github.com/urfave/cli/v2 v2.25.7 h1:VAzn5oq403l5pHjc4OhD54+XGO9cdKVL/7lDjF+iKUs=
+github.com/urfave/cli/v2 v2.25.7/go.mod h1:8qnjx1vcq5s2/wpsqoZFndg2CE5tNFyrTvS6SinrnYQ=
+github.com/xrash/smetrics v0.0.0-20201216005158-039620a65673 h1:bAn7/zixMGCfxrRTfdpNzjtPYqr8smhKouy9mxVdGPU=
+github.com/xrash/smetrics v0.0.0-20201216005158-039620a65673/go.mod h1:N3UwUGtsrSj3ccvlPHLoLsHnpR27oXr4ZE984MbSER8=
+go.uber.org/goleak v1.2.0 h1:xqgm/S+aQvhWFTtR0XK3Jvg7z8kGV8P4X14IzwN3Eqk=
+go.uber.org/goleak v1.2.0/go.mod h1:XJYK+MuIchqpmGmUSAzotztawfKvYLUIgg7guXrwVUo=
+go.uber.org/multierr v1.11.0 h1:blXXJkSxSSfBVBlC76pxqeO+LN3aDfLQo+309xJstO0=
+go.uber.org/multierr v1.11.0/go.mod h1:20+QtiLqy0Nd6FdQB9TLXag12DsQkrbs3htMFfDN80Y=
+go.uber.org/zap v1.26.0 h1:sI7k6L95XOKS281NhVKOFCUNIvv9e0w4BF8N3u+tCRo=
+go.uber.org/zap v1.26.0/go.mod h1:dtElttAiwGvoJ/vj4IwHBS/gXsEu/pZ50mUIRWuG0so=
+gopkg.in/yaml.v3 v3.0.1 h1:fxVm/GzAzEWqLHuvctI91KS9hhNmmWOoWu0XTYJS7CA=
+gopkg.in/yaml.v3 v3.0.1/go.mod h1:K4uyk7z7BCEPqu6E+C64Yfv1cQ7kz7rIZviUmN+EgEM=

healthchecker/check_geth.go

@@ -0,0 +1,83 @@
+package healthchecker
+
+import (
+	"bytes"
+	"context"
+	"encoding/json"
+	"errors"
+	"fmt"
+	"io"
+	"net/http"
+)
+
+func (h *Healthchecker) checkGeth(ctx context.Context, url string) error {
+	// https://ethereum.org/en/developers/docs/apis/json-rpc/#eth_syncing
+
+	const query = `{"jsonrpc":"2.0","method":"eth_syncing","params":[],"id":0}`
+
+	type isNotSyncing struct {
+		Result bool // `json:"result"`
+	}
+
+	type isSyncing struct {
+		Result struct {
+			CurrentBlock string // `json:"currentBlock"`
+			HighestBlock string // `json:"highestBlock"`
+		} // `json:"result"`
+	}
+
+	//--------------------------------------------------------------------------
+
+	req, err := http.NewRequestWithContext(
+		ctx,
+		http.MethodGet,
+		url,
+		bytes.NewReader([]byte(query)),
+	)
+	if err != nil {
+		return err
+	}
+	req.Header.Set("accept", "application/json")
+	req.Header.Set("content-type", "application/json")
+
+	res, err := http.DefaultClient.Do(req)
+	if err != nil {
+		return err
+	}
+	defer res.Body.Close()
+
+	body, err := io.ReadAll(res.Body)
+	if err != nil {
+		return err
+	}
+
+	if res.StatusCode != http.StatusOK {
+		return fmt.Errorf(
+			"unexpected HTTP status '%d': %s",
+			res.StatusCode,
+			string(body),
+		)
+	}
+
+	var status isNotSyncing
+	if err := json.Unmarshal(body, &status); err != nil {
+		var status isSyncing
+		if err2 := json.Unmarshal(body, &status); err2 != nil {
+			return fmt.Errorf(
+				"failed to parse JSON body '%s': %w",
+				string(body),
+				errors.Join(err, err2),
+			)
+		}
+		return fmt.Errorf(
+			"geth is still syncing (current: %s, highest: %s)",
+			status.Result.CurrentBlock,
+			status.Result.HighestBlock,
+		)
+	}
+	if status.Result { // i.e. it's syncing
+		return errors.New("geth is (still) syncing")
+	}
+
+	return nil
+}