完善中央与边缘部署、远程写入与监控文档

- 增加中央与边缘完整配置和部署脚本
- 引入 VictoriaMetrics 数据源与 remote_write 故障排查说明
- 新增 edge-agent 配置脚本、ONVIF 自建 exporter 与 ping 监控示例

Co-authored-by: Cursor <cursoragent@cursor.com>

edge-agent/onvif-exporter/Dockerfile

@@ -0,0 +1,12 @@
+# 多阶段构建：在镜像内编译，无需本机安装 Go
+FROM golang:1.21-alpine AS builder
+WORKDIR /app
+COPY go.mod ./
+COPY main.go ./
+RUN go mod tidy && CGO_ENABLED=0 GOOS=linux go build -o /onvif-exporter .
+
+FROM alpine:3.19
+RUN apk add --no-cache ca-certificates
+COPY --from=builder /onvif-exporter /onvif-exporter
+EXPOSE 9600
+ENTRYPOINT ["/onvif-exporter"]

edge-agent/onvif-exporter/README.md

@@ -0,0 +1,34 @@
+# ONVIF Exporter（自建）
+
+本目录为自建的 ONVIF 探测容器，供边缘节点可选使用。通过 ONVIF `GetDeviceInformation` 探测设备是否在线，并暴露 Prometheus 指标。
+
+## 指标
+
+- `onvif_device_up`：1=可达，0=不可达（标签：instance, location, model, device_type）
+- `onvif_probe_duration_seconds`：探测耗时（秒）
+
+## 配置
+
+- 从 **config/onvif-targets.json** 读取设备列表（与 `targets.csv` 中 onvif 行一致，由 `config/update-configs.sh` 生成）。
+- 环境变量：`TARGETS_FILE`（默认 `/config/targets.json`）、`EXPORTER_PORT`（默认 9600）。
+
+## 构建与运行
+
+在边缘节点目录下启用 ONVIF 并构建、启动：
+
+```bash
+cd edge-agent
+docker compose --profile onvif up -d --build
+```
+
+或仅构建镜像：
+
+```bash
+docker build -t onvif-exporter:local ./onvif-exporter
+```
+
+## 依赖
+
+- Go 1.21（仅构建时需要，Dockerfile 内已包含）
+- [github.com/use-go/onvif](https://github.com/use-go/onvif)（ONVIF 协议）
+- [prometheus/client_golang](https://github.com/prometheus/client_golang)（指标暴露）

edge-agent/onvif-exporter/go.mod

@@ -0,0 +1,24 @@
+module github.com/distributed-prometheus/onvif-exporter
+
+go 1.21
+
+require (
+	github.com/prometheus/client_golang v1.19.0
+	github.com/use-go/onvif v0.0.9
+)
+
+require (
+	github.com/beorn7/perks v1.0.1 // indirect
+	github.com/beevik/etree v1.1.0 // indirect
+	github.com/cespare/xxhash/v2 v2.2.0 // indirect
+	github.com/elgs/gostrgen v0.0.0-20161222160715-9d61ae07eeae // indirect
+	github.com/gofrs/uuid v3.2.0+incompatible // indirect
+	github.com/juju/errors v0.0.0-20220331221717-b38fca44723b // indirect
+	github.com/prometheus/client_model v0.6.0 // indirect
+	github.com/prometheus/common v0.50.0 // indirect
+	github.com/prometheus/procfs v0.13.0 // indirect
+	github.com/rs/zerolog v1.26.1 // indirect
+	golang.org/x/net v0.21.0 // indirect
+	golang.org/x/sys v0.17.0 // indirect
+	google.golang.org/protobuf v1.32.0 // indirect
+)

edge-agent/onvif-exporter/main.go

@@ -0,0 +1,122 @@
+package main
+
+import (
+	"encoding/json"
+	"log"
+	"net/http"
+	"os"
+	"strings"
+	"time"
+
+	"github.com/prometheus/client_golang/prometheus"
+	"github.com/prometheus/client_golang/prometheus/promhttp"
+	"github.com/use-go/onvif"
+	"github.com/use-go/onvif/device"
+)
+
+const (
+	defaultTargetsFile = "/config/targets.json"
+	defaultPort        = "9600"
+)
+
+type targetGroup struct {
+	Targets []string          `json:"targets"`
+	Labels  map[string]string `json:"labels"`
+}
+
+func main() {
+	targetsFile := os.Getenv("TARGETS_FILE")
+	if targetsFile == "" {
+		targetsFile = defaultTargetsFile
+	}
+	port := os.Getenv("EXPORTER_PORT")
+	if port == "" {
+		port = defaultPort
+	}
+
+	up := prometheus.NewGaugeVec(prometheus.GaugeOpts{
+		Name: "onvif_device_up",
+		Help: "1 if ONVIF device is reachable, 0 otherwise",
+	}, []string{"instance", "location", "model", "device_type"})
+	duration := prometheus.NewGaugeVec(prometheus.GaugeOpts{
+		Name: "onvif_probe_duration_seconds",
+		Help: "ONVIF probe duration in seconds",
+	}, []string{"instance", "location", "model", "device_type"})
+
+	reg := prometheus.NewRegistry()
+	reg.MustRegister(up, duration)
+
+	go func() {
+		ticker := time.NewTicker(60 * time.Second)
+		defer ticker.Stop()
+		for ; true; <-ticker.C {
+			probe(targetsFile, up, duration)
+		}
+	}()
+	// 启动时立即探测一次
+	probe(targetsFile, up, duration)
+
+	http.Handle("/metrics", promhttp.HandlerFor(reg, promhttp.HandlerOpts{EnableOpenMetrics: true}))
+	http.HandleFunc("/health", func(w http.ResponseWriter, _ *http.Request) { w.WriteHeader(http.StatusOK) })
+	log.Printf("ONVIF exporter listening on :%s", port)
+	if err := http.ListenAndServe(":"+port, nil); err != nil {
+		log.Fatal(err)
+	}
+}
+
+func probe(targetsFile string, up, duration *prometheus.GaugeVec) {
+	data, err := os.ReadFile(targetsFile)
+	if err != nil {
+		log.Printf("read targets file: %v", err)
+		return
+	}
+	var groups []targetGroup
+	if err := json.Unmarshal(data, &groups); err != nil {
+		log.Printf("parse targets: %v", err)
+		return
+	}
+	for _, g := range groups {
+		if len(g.Targets) == 0 {
+			continue
+		}
+		ip := strings.TrimSpace(g.Targets[0])
+		port := "80"
+		if p, ok := g.Labels["onvif_port"]; ok && p != "" {
+			port = strings.TrimSpace(p)
+		}
+		user := strings.TrimSpace(g.Labels["username"])
+		pass := strings.TrimSpace(g.Labels["password"])
+		location := strings.TrimSpace(g.Labels["location"])
+		model := strings.TrimSpace(g.Labels["model"])
+		deviceType := strings.TrimSpace(g.Labels["device_type"])
+		instance := ip + ":" + port
+		labels := prometheus.Labels{
+			"instance":    instance,
+			"location":    location,
+			"model":       model,
+			"device_type": deviceType,
+		}
+
+		start := time.Now()
+		err := probeONVIF(ip, port, user, pass)
+		elapsed := time.Since(start).Seconds()
+		duration.With(labels).Set(elapsed)
+		if err != nil {
+			log.Printf("onvif probe %s: %v", instance, err)
+			up.With(labels).Set(0)
+		} else {
+			up.With(labels).Set(1)
+		}
+	}
+}
+
+func probeONVIF(ip, port, username, password string) error {
+	xaddr := ip + ":" + port
+	params := onvif.DeviceParams{Xaddr: xaddr, Username: username, Password: password}
+	dev, err := onvif.NewDevice(params)
+	if err != nil {
+		return err
+	}
+	_, err = dev.CallMethod(device.GetDeviceInformation{})
+	return err
+}