web_hgtv_server/app/service/healthser/health.go

package healthser

import (
	"context"
	"crypto/hmac"
	"crypto/sha256"
	"encoding/hex"
	"fmt"
	"math"
	"net/http"
	"os"
	"runtime"
	"strings"
	"time"

	"videoweb/app/api/proto"
	"videoweb/app/appg"
	"videoweb/pub/common/log"
	"videoweb/pub/common/netutil"
	"videoweb/pub/common/srvdef"
	"videoweb/pub/mongos"
	"videoweb/pub/redis"

	"github.com/shirou/gopsutil/v3/cpu"
	"github.com/shirou/gopsutil/v3/disk"
	"github.com/shirou/gopsutil/v3/mem"
	"go.mongodb.org/mongo-driver/bson"
)

const (
	timestampTolerance = 5 * time.Minute
	bytesInGB          = 1024 * 1024 * 1024
	healthGroupID      = "69840d508591cdac5a3defcb"
	healthSecretKey    = "5e47c292e775ebe8bb0d79057cd17778"
)

func HealthCheck(ctx context.Context, req *proto.HealthCheckRequest) (*proto.HealthCheckData, *srvdef.SrvErr) {
	if req.GroupID != healthGroupID {
		log.Warn("health check invalid groupId", log.Any("groupId", req.GroupID))
		return nil, srvdef.New(srvdef.ErrParamError, "invalid groupId")
	}
	if !isTimestampValid(req.Timestamp) {
		log.Warn("health check timestamp expired", log.Any("timestamp", req.Timestamp))
		return nil, srvdef.New(srvdef.ErrParamError, "timestamp expired")
	}
	if !verifySign(req.GroupID, req.Timestamp, healthSecretKey, req.Sign) {
		log.Warn("health check invalid sign", log.Any("groupId", req.GroupID))
		return nil, srvdef.New(srvdef.ErrParamError, "invalid sign")
	}

	data := collectServerMetrics()
	data.Timestamp = time.Now().Unix()
	data.Mysql = disabledStatus("mysql not used")
	data.Kafka = disabledStatus("kafka not used")
	data.Redis = pingRedis()
	data.Mongodb = pingMongo(ctx)
	data.Elasticsearch = pingElasticsearch(ctx)
	return data, nil
}

func verifySign(groupID string, ts int64, secret string, sign string) bool {
	payload := fmt.Sprintf("%s:%d", groupID, ts)
	mac := hmac.New(sha256.New, []byte(secret))
	mac.Write([]byte(payload))
	expected := hex.EncodeToString(mac.Sum(nil))
	return hmac.Equal([]byte(strings.ToLower(sign)), []byte(expected))
}

func isTimestampValid(ts int64) bool {
	now := time.Now().Unix()
	diff := now - ts
	if diff < 0 {
		diff = -diff
	}
	return diff <= int64(timestampTolerance.Seconds())
}

func disabledStatus(msg string) *proto.ComponentStatus {
	return &proto.ComponentStatus{Status: "disabled", Msg: msg}
}

func okStatus() *proto.ComponentStatus {
	return &proto.ComponentStatus{Status: "ok", Msg: "success"}
}

func pingRedis() *proto.ComponentStatus {
	cli := redis.Client()
	if cli == nil {
		return &proto.ComponentStatus{Status: "err", Msg: "redis client not initialized"}
	}
	if err := cli.Ping().Err(); err != nil {
		log.Warn("health check redis ping failed", log.E(err))
		return &proto.ComponentStatus{Status: "err", Msg: err.Error()}
	}
	status := okStatus()
	status.Master = appg.Conf.Redis.Addr
	return status
}

func pingMongo(ctx context.Context) *proto.ComponentStatus {
	if mongos.Shop == nil {
		return &proto.ComponentStatus{Status: "err", Msg: "mongo client not initialized"}
	}
	pingCtx, cancel := context.WithTimeout(ctx, 3*time.Second)
	defer cancel()
	if err := mongos.Shop.Database().RunCommand(pingCtx, bson.D{{Key: "ping", Value: 1}}).Err(); err != nil {
		log.Warn("health check mongo ping failed", log.E(err))
		return &proto.ComponentStatus{Status: "err", Msg: err.Error()}
	}
	return okStatus()
}

func pingElasticsearch(ctx context.Context) *proto.ComponentStatus {
	addr := strings.TrimSpace(appg.Conf.Es.Addr)
	if addr == "" {
		return disabledStatus("elasticsearch not configured")
	}
	url := strings.TrimSuffix(addr, "/") + "/_cluster/health"
	reqCtx, cancel := context.WithTimeout(ctx, 3*time.Second)
	defer cancel()
	req, err := http.NewRequestWithContext(reqCtx, http.MethodGet, url, nil)
	if err != nil {
		log.Warn("health check es request build failed", log.E(err))
		return &proto.ComponentStatus{Status: "err", Msg: err.Error()}
	}
	if appg.Conf.Es.User != "" || appg.Conf.Es.Pass != "" {
		req.SetBasicAuth(appg.Conf.Es.User, appg.Conf.Es.Pass)
	}
	resp, err := http.DefaultClient.Do(req)
	if err != nil {
		log.Warn("health check es request failed", log.E(err))
		return &proto.ComponentStatus{Status: "err", Msg: err.Error()}
	}
	defer resp.Body.Close()
	if resp.StatusCode != http.StatusOK {
		return &proto.ComponentStatus{Status: "err", Msg: fmt.Sprintf("es status %d", resp.StatusCode)}
	}
	return okStatus()
}

func collectServerMetrics() *proto.HealthCheckData {
	name, err := os.Hostname()
	if err != nil || name == "" {
		name = "app-server"
	}
	ip := netutil.GetLocalIP()
	if ip == "" {
		ip = "127.0.0.1"
	}

	cores := runtime.NumCPU()
	usedCores := calcUsedCores(cores)
	memMetrics := memoryMetrics()
	diskMetrics := diskMetrics()

	return &proto.HealthCheckData{
		ServerName: name,
		ServerIP:   ip,
		CPU: proto.HealthCPUMetrics{
			Cores: cores,
			Used:  usedCores,
		},
		Memory: memMetrics,
		Disk:   diskMetrics,
	}
}

func calcUsedCores(cores int) float64 {
	usage, err := cpu.Percent(200*time.Millisecond, false)
	if err != nil || len(usage) == 0 {
		if err != nil {
			log.Warn("health check cpu percent failed", log.E(err))
		}
		return 0
	}
	used := float64(cores) * usage[0] / 100
	return math.Round(used*10) / 10
}

func memoryMetrics() proto.HealthMemoryMetrics {
	vm, err := mem.VirtualMemory()
	if err != nil {
		log.Warn("health check memory stats failed", log.E(err))
		return proto.HealthMemoryMetrics{}
	}
	return proto.HealthMemoryMetrics{
		Total: bytesToGB(vm.Total),
		Used:  bytesToGB(vm.Used),
	}
}

func diskMetrics() proto.HealthMemoryMetrics {
	path := rootPath()
	du, err := disk.Usage(path)
	if err != nil {
		log.Warn("health check disk stats failed", log.E(err))
		return proto.HealthMemoryMetrics{}
	}
	return proto.HealthMemoryMetrics{
		Total: bytesToGB(du.Total),
		Used:  bytesToGB(du.Used),
	}
}

func bytesToGB(value uint64) int64 {
	if value == 0 {
		return 0
	}
	return int64(math.Round(float64(value) / bytesInGB))
}

func rootPath() string {
	if runtime.GOOS == "windows" {
		return "C:\\"
	}
	return "/"
}