feat: Phase 2-5 监控中心全栈实现 — 指标浏览器/SNMP/站点拨测/中间件

Phase 2 指标监控:
- 修复 /metrics 路由(原指向 AlertManager → 现指向 MetricsExplorer)
- 新建 MetricsHandler (4 API): targets/list/query/names
- 新建 MetricsExplorer 前端(170行): 目标选择器+指标分类树+多指标叠加SVG图表

Phase 3 SNMP设备:
- 新建 SNMPHandler (9 API): devices CRUD/templates/stats
- SNMPMonitor 接入真实API(428行): 设备表格+添加对话框+删除

Phase 4 站点拨测:
- 新建 SiteHandler (8 API): tasks CRUD/results/stats
- SiteMonitor 接入真实API(542行): 任务表+添加对话框+拨测类型网格
- 新增 siteApi 到 api.ts

Phase 5 中间件采集:
- Agent 新增 collector/middleware.go: Redis INFO + MySQL handshake 采集
- Agent CLI 新增 -redis/-redis-auth/-mysql/-mysql-user/-mysql-pass
- MiddlewareMonitor 接入真实API(301行): 实例卡片+指标面板

基础设施:
- AutoMigrate 新增 23 张监控表
- 前后端容器构建部署验证通过
This commit is contained in:
attacker committed 2026-06-18 13:02:19 +00:00
1 parent 94dbc89b9b
commit 2e71371257
15 files changed
+2616 -137

No files matched your search

+132
View File
@@ -0,0 +1,132 @@
package handlers
import (
"ezops/internal/models"
"net/http"
"strconv"
"time"
"github.com/gin-gonic/gin"
"gorm.io/gorm"
)
// MetricsHandler 指标浏览器 handler // Logan.Li
type MetricsHandler struct {
db *gorm.DB
}
// NewMetricsHandler 创建指标 handler
func NewMetricsHandler(db *gorm.DB) *MetricsHandler {
return &MetricsHandler{db: db}
}
// ListTargets 列出监控目标(按类型过滤)
// GET /api/v1/metrics/targets?type=host&status=online
func (h *MetricsHandler) ListTargets(c *gin.Context) {
var targets []models.MonitorTarget
q := h.db.Model(&models.MonitorTarget{})
if t := c.Query("type"); t != "" {
q = q.Where("type = ?", t)
}
if s := c.Query("status"); s != "" {
q = q.Where("status = ?", s)
}
if kw := c.Query("keyword"); kw != "" {
q = q.Where("name LIKE ? OR ip LIKE ?", "%"+kw+"%", "%"+kw+"%")
}
// 分页
page, _ := strconv.Atoi(c.DefaultQuery("page", "1"))
pageSize, _ := strconv.Atoi(c.DefaultQuery("page_size", "50"))
var total int64
q.Count(&total)
q.Offset((page - 1) * pageSize).Limit(pageSize).Order("name ASC").Find(&targets)
if targets == nil {
targets = []models.MonitorTarget{}
}
// 附带最新指标值
type TargetWithMetrics struct {
models.MonitorTarget
LatestMetrics map[string]float64 `json:"latest_metrics"`
}
result := make([]TargetWithMetrics, len(targets))
for i, t := range targets {
result[i] = TargetWithMetrics{MonitorTarget: t, LatestMetrics: make(map[string]float64)}
// 查询该目标最近 10 分钟内的最新指标
var histories []models.MetricHistory
h.db.Where("target_id = ? AND timestamp > ?", t.ID, now()-600).
Order("timestamp DESC").Limit(10).Find(&histories)
for _, mh := range histories {
if _, ok := result[i].LatestMetrics[mh.MetricName]; !ok {
result[i].LatestMetrics[mh.MetricName] = mh.Value
}
}
}
c.JSON(http.StatusOK, models.Response{Code: 200, Data: gin.H{
"list": result,
"total": total,
}})
}
// ListMetrics 列出系统中所有指标元数据
// GET /api/v1/metrics/list
func (h *MetricsHandler) ListMetrics(c *gin.Context) {
var metrics []models.MonitorMetric
h.db.Order("category ASC, name ASC").Find(&metrics)
if metrics == nil {
metrics = []models.MonitorMetric{}
}
c.JSON(http.StatusOK, models.Response{Code: 200, Data: metrics})
}
// QueryValues 查询指标时序数据
// GET /api/v1/metrics/query?target_id=N&metric_name=cpu_usage&from=0&to=0&limit=100
func (h *MetricsHandler) QueryValues(c *gin.Context) {
var histories []models.MetricHistory
q := h.db.Model(&models.MetricHistory{})
if tid := c.Query("target_id"); tid != "" {
id, _ := strconv.ParseUint(tid, 10, 64)
q = q.Where("target_id = ?", id)
}
if mn := c.Query("metric_name"); mn != "" {
q = q.Where("metric_name = ?", mn)
}
if from := c.Query("from"); from != "" {
ts, _ := strconv.ParseInt(from, 10, 64)
q = q.Where("timestamp >= ?", ts)
}
if to := c.Query("to"); to != "" {
ts, _ := strconv.ParseInt(to, 10, 64)
q = q.Where("timestamp <= ?", ts)
}
limit, _ := strconv.Atoi(c.DefaultQuery("limit", "200"))
q.Order("timestamp ASC").Limit(limit).Find(&histories)
if histories == nil {
histories = []models.MetricHistory{}
}
c.JSON(http.StatusOK, models.Response{Code: 200, Data: histories})
}
// ListMetricNames 列出所有可用的指标名称(去重)
// GET /api/v1/metrics/names
func (h *MetricsHandler) ListMetricNames(c *gin.Context) {
var names []string
h.db.Model(&models.MetricHistory{}).
Distinct("metric_name").
Order("metric_name ASC").
Pluck("metric_name", &names)
if names == nil {
names = []string{}
}
c.JSON(http.StatusOK, models.Response{Code: 200, Data: names})
}
func now() int64 {
return time.Now().Unix()
}
+166
View File
@@ -0,0 +1,166 @@
package handlers
import (
"ezops/internal/models"
"net/http"
"strconv"
"github.com/gin-gonic/gin"
"gorm.io/gorm"
)
// SiteHandler 站点拨测管理 handler // Logan.Li
type SiteHandler struct {
db *gorm.DB
}
// NewSiteHandler 创建站点拨测 handler
func NewSiteHandler(db *gorm.DB) *SiteHandler {
return &SiteHandler{db: db}
}
// ========== Site Task CRUD ==========
// ListTasks 列出拨测任务
// GET /api/v1/site/tasks?page=1&page_size=20&type=&status=
func (h *SiteHandler) ListTasks(c *gin.Context) {
var tasks []models.SiteTask
q := h.db.Model(&models.SiteTask{})
if t := c.Query("type"); t != "" {
q = q.Where("type = ?", t)
}
if s := c.Query("status"); s != "" {
q = q.Where("status = ?", s)
}
if kw := c.Query("keyword"); kw != "" {
q = q.Where("name LIKE ? OR url LIKE ?", "%"+kw+"%", "%"+kw+"%")
}
page, _ := strconv.Atoi(c.DefaultQuery("page", "1"))
pageSize, _ := strconv.Atoi(c.DefaultQuery("page_size", "20"))
var total int64
q.Count(&total)
q.Offset((page - 1) * pageSize).Limit(pageSize).Order("name ASC").Find(&tasks)
if tasks == nil {
tasks = []models.SiteTask{}
}
c.JSON(http.StatusOK, models.Response{Code: 200, Data: gin.H{"list": tasks, "total": total}})
}
// CreateTask 创建拨测任务
// POST /api/v1/site/tasks
func (h *SiteHandler) CreateTask(c *gin.Context) {
var task models.SiteTask
if err := c.ShouldBindJSON(&task); err != nil {
c.JSON(http.StatusBadRequest, models.Response{Code: 400, Message: err.Error()})
return
}
if err := h.db.Create(&task).Error; err != nil {
c.JSON(http.StatusInternalServerError, models.Response{Code: 500, Message: err.Error()})
return
}
c.JSON(http.StatusOK, models.Response{Code: 200, Data: task, Message: "创建成功"})
}
// UpdateTask 更新拨测任务
// PUT /api/v1/site/tasks/:id
func (h *SiteHandler) UpdateTask(c *gin.Context) {
id := c.Param("id")
var task models.SiteTask
if err := h.db.First(&task, id).Error; err != nil {
c.JSON(http.StatusNotFound, models.Response{Code: 404, Message: "任务不存在"})
return
}
if err := c.ShouldBindJSON(&task); err != nil {
c.JSON(http.StatusBadRequest, models.Response{Code: 400, Message: err.Error()})
return
}
h.db.Save(&task)
c.JSON(http.StatusOK, models.Response{Code: 200, Data: task, Message: "更新成功"})
}
// DeleteTask 删除拨测任务
// DELETE /api/v1/site/tasks/:id
func (h *SiteHandler) DeleteTask(c *gin.Context) {
id := c.Param("id")
if err := h.db.Delete(&models.SiteTask{}, id).Error; err != nil {
c.JSON(http.StatusInternalServerError, models.Response{Code: 500, Message: err.Error()})
return
}
c.JSON(http.StatusOK, models.Response{Code: 200, Message: "删除成功"})
}
// BatchDeleteTasks 批量删除
// DELETE /api/v1/site/tasks/batch-delete
func (h *SiteHandler) BatchDeleteTasks(c *gin.Context) {
var req models.IDsRequest
if err := c.ShouldBindJSON(&req); err != nil {
c.JSON(http.StatusBadRequest, models.Response{Code: 400, Message: err.Error()})
return
}
h.db.Where("id IN ?", req.IDs).Delete(&models.SiteTask{})
c.JSON(http.StatusOK, models.Response{Code: 200, Message: "批量删除成功"})
}
// ========== Site Results ==========
// ListResults 查询拨测结果
// GET /api/v1/site/results?task_id=N&from=&to=&limit=100
func (h *SiteHandler) ListResults(c *gin.Context) {
var results []models.SiteResult
q := h.db.Model(&models.SiteResult{})
if tid := c.Query("task_id"); tid != "" {
q = q.Where("task_id = ?", tid)
}
if from := c.Query("from"); from != "" {
q = q.Where("created_at >= ?", from)
}
if to := c.Query("to"); to != "" {
q = q.Where("created_at <= ?", to)
}
limit, _ := strconv.Atoi(c.DefaultQuery("limit", "100"))
q.Order("created_at DESC").Limit(limit).Find(&results)
if results == nil {
results = []models.SiteResult{}
}
c.JSON(http.StatusOK, models.Response{Code: 200, Data: results})
}
// ========== Site Statistics ==========
// GetSiteStats 获取拨测概览统计
// GET /api/v1/site/stats
func (h *SiteHandler) GetSiteStats(c *gin.Context) {
var total, normal, abnormal int64
h.db.Model(&models.SiteTask{}).Count(&total)
h.db.Model(&models.SiteTask{}).Where("status = ?", "active").Count(&normal)
h.db.Model(&models.SiteTask{}).Where("status = ?", "error").Count(&abnormal)
// Calculate availability rate from recent results
var availRate float64
if total > 0 {
var successCount int64
h.db.Model(&models.SiteResult{}).
Where("created_at > datetime('now', '-1 hour') AND status = 'success'").
Count(&successCount)
var totalRecent int64
h.db.Model(&models.SiteResult{}).
Where("created_at > datetime('now', '-1 hour')").
Count(&totalRecent)
if totalRecent > 0 {
availRate = float64(successCount) / float64(totalRecent) * 100
}
}
c.JSON(http.StatusOK, models.Response{Code: 200, Data: gin.H{
"total": total,
"normal": normal,
"abnormal": abnormal,
"avail_rate": availRate,
}})
}
+143
View File
@@ -0,0 +1,143 @@
package handlers
import (
"ezops/internal/models"
"net/http"
"strconv"
"github.com/gin-gonic/gin"
"gorm.io/gorm"
)
// SNMPHandler SNMP 设备管理 handler // Logan.Li
type SNMPHandler struct {
db *gorm.DB
}
// NewSNMPHandler 创建 SNMP handler
func NewSNMPHandler(db *gorm.DB) *SNMPHandler {
return &SNMPHandler{db: db}
}
// ========== SNMP Device CRUD ==========
// ListDevices 列出 SNMP 设备
// GET /api/v1/snmp/devices?page=1&page_size=20&keyword=
func (h *SNMPHandler) ListDevices(c *gin.Context) {
var devices []models.SNMPDevice
q := h.db.Model(&models.SNMPDevice{})
if kw := c.Query("keyword"); kw != "" {
q = q.Where("name LIKE ? OR ip LIKE ?", "%"+kw+"%", "%"+kw+"%")
}
page, _ := strconv.Atoi(c.DefaultQuery("page", "1"))
pageSize, _ := strconv.Atoi(c.DefaultQuery("page_size", "20"))
var total int64
q.Count(&total)
q.Offset((page - 1) * pageSize).Limit(pageSize).Order("name ASC").Find(&devices)
if devices == nil {
devices = []models.SNMPDevice{}
}
c.JSON(http.StatusOK, models.Response{Code: 200, Data: gin.H{"list": devices, "total": total}})
}
// CreateDevice 创建 SNMP 设备
// POST /api/v1/snmp/devices
func (h *SNMPHandler) CreateDevice(c *gin.Context) {
var device models.SNMPDevice
if err := c.ShouldBindJSON(&device); err != nil {
c.JSON(http.StatusBadRequest, models.Response{Code: 400, Message: err.Error()})
return
}
if err := h.db.Create(&device).Error; err != nil {
c.JSON(http.StatusInternalServerError, models.Response{Code: 500, Message: err.Error()})
return
}
c.JSON(http.StatusOK, models.Response{Code: 200, Data: device, Message: "创建成功"})
}
// UpdateDevice 更新 SNMP 设备
// PUT /api/v1/snmp/devices/:id
func (h *SNMPHandler) UpdateDevice(c *gin.Context) {
id := c.Param("id")
var device models.SNMPDevice
if err := h.db.First(&device, id).Error; err != nil {
c.JSON(http.StatusNotFound, models.Response{Code: 404, Message: "设备不存在"})
return
}
if err := c.ShouldBindJSON(&device); err != nil {
c.JSON(http.StatusBadRequest, models.Response{Code: 400, Message: err.Error()})
return
}
h.db.Save(&device)
c.JSON(http.StatusOK, models.Response{Code: 200, Data: device, Message: "更新成功"})
}
// DeleteDevice 删除 SNMP 设备
// DELETE /api/v1/snmp/devices/:id
func (h *SNMPHandler) DeleteDevice(c *gin.Context) {
id := c.Param("id")
if err := h.db.Delete(&models.SNMPDevice{}, id).Error; err != nil {
c.JSON(http.StatusInternalServerError, models.Response{Code: 500, Message: err.Error()})
return
}
c.JSON(http.StatusOK, models.Response{Code: 200, Message: "删除成功"})
}
// BatchDeleteDevices 批量删除
// DELETE /api/v1/snmp/devices/batch-delete
func (h *SNMPHandler) BatchDeleteDevices(c *gin.Context) {
var req models.IDsRequest
if err := c.ShouldBindJSON(&req); err != nil {
c.JSON(http.StatusBadRequest, models.Response{Code: 400, Message: err.Error()})
return
}
h.db.Where("id IN ?", req.IDs).Delete(&models.SNMPDevice{})
c.JSON(http.StatusOK, models.Response{Code: 200, Message: "批量删除成功"})
}
// ========== SNMP Template CRUD ==========
// ListTemplates 列出 OID 模板
// GET /api/v1/snmp/templates
func (h *SNMPHandler) ListTemplates(c *gin.Context) {
var templates []models.SNMPTemplate
h.db.Order("name ASC").Find(&templates)
if templates == nil {
templates = []models.SNMPTemplate{}
}
c.JSON(http.StatusOK, models.Response{Code: 200, Data: templates})
}
// CreateTemplate 创建 OID 模板
// POST /api/v1/snmp/templates
func (h *SNMPHandler) CreateTemplate(c *gin.Context) {
var template models.SNMPTemplate
if err := c.ShouldBindJSON(&template); err != nil {
c.JSON(http.StatusBadRequest, models.Response{Code: 400, Message: err.Error()})
return
}
if err := h.db.Create(&template).Error; err != nil {
c.JSON(http.StatusInternalServerError, models.Response{Code: 500, Message: err.Error()})
return
}
c.JSON(http.StatusOK, models.Response{Code: 200, Data: template, Message: "创建成功"})
}
// ========== SNMP Statistics ==========
// GetSNMPStats 获取 SNMP 概览统计
// GET /api/v1/snmp/stats
func (h *SNMPHandler) GetSNMPStats(c *gin.Context) {
var total, online int64
h.db.Model(&models.SNMPDevice{}).Count(&total)
h.db.Model(&models.SNMPDevice{}).Where("status = ?", "online").Count(&online)
c.JSON(http.StatusOK, models.Response{Code: 200, Data: gin.H{
"total": total,
"online": online,
"abnormal": total - online,
}})
}
+37
View File
@@ -465,6 +465,43 @@ func Register(r *gin.Engine, db *gorm.DB, log *zap.Logger, cfg *config.Config, c
aiopsGroup.DELETE("/chat-history", llmHandler.ClearChatHistory)
}
// 指标浏览器 — 监控指标元数据 + 时序查询 // Logan.Li
metricsHandler := handlers.NewMetricsHandler(db)
metricsGroup := authorized.Group("/metrics")
{
metricsGroup.GET("/targets", metricsHandler.ListTargets)
metricsGroup.GET("/list", metricsHandler.ListMetrics)
metricsGroup.GET("/query", metricsHandler.QueryValues)
metricsGroup.GET("/names", metricsHandler.ListMetricNames)
}
// SNMP 设备管理 — 网络设备/服务器硬件 // Logan.Li
snmpHandler := handlers.NewSNMPHandler(db)
snmpGroup := authorized.Group("/snmp")
{
snmpGroup.GET("/devices", snmpHandler.ListDevices)
snmpGroup.POST("/devices", snmpHandler.CreateDevice)
snmpGroup.PUT("/devices/:id", snmpHandler.UpdateDevice)
snmpGroup.DELETE("/devices/:id", snmpHandler.DeleteDevice)
snmpGroup.DELETE("/devices/batch-delete", snmpHandler.BatchDeleteDevices)
snmpGroup.GET("/templates", snmpHandler.ListTemplates)
snmpGroup.POST("/templates", snmpHandler.CreateTemplate)
snmpGroup.GET("/stats", snmpHandler.GetSNMPStats)
}
// 站点拨测 — HTTP/TCP/DNS/PING 主动监控 // Logan.Li
siteHandler := handlers.NewSiteHandler(db)
siteGroup := authorized.Group("/site")
{
siteGroup.GET("/tasks", siteHandler.ListTasks)
siteGroup.POST("/tasks", siteHandler.CreateTask)
siteGroup.PUT("/tasks/:id", siteHandler.UpdateTask)
siteGroup.DELETE("/tasks/:id", siteHandler.DeleteTask)
siteGroup.DELETE("/tasks/batch-delete", siteHandler.BatchDeleteTasks)
siteGroup.GET("/results", siteHandler.ListResults)
siteGroup.GET("/stats", siteHandler.GetSiteStats)
}
// 启动告警引擎(每 30 秒检查一次)
alertEngine := handlers.NewAlertEngine(db)
go func() {
+23
View File
@@ -107,6 +107,29 @@ func AutoMigrate(db *gorm.DB) error {
&models.LLMConfig{}, // 大语言模型配置(AI 智能助手) // Logan.Li
&models.ChatHistory{}, // 对话历史记录(按用户隔离) // Logan.Li
&models.MetricsKline{}, // 指标K线聚合(Redis 热数据 → MySQL) // Logan.Li
&models.MonitorTarget{}, // 监控目标(主机/中间件/设备/站点) // Logan.Li
&models.MonitorMetric{}, // 监控指标定义 // Logan.Li
&models.MetricValue{}, // 指标当前值 // Logan.Li
&models.MetricHistory{}, // 指标历史时序 // Logan.Li
&models.MetricTag{}, // 指标标签 // Logan.Li
&models.MonitorDashboard{},// 监控大屏配置 // Logan.Li
&models.DashboardPanel{}, // 大屏面板 // Logan.Li
&models.HostGroup{}, // 主机组 // Logan.Li
&models.SNMPDevice{}, // SNMP设备 // Logan.Li
&models.SNMPTemplate{}, // SNMP模板 // Logan.Li
&models.SNMPTemplateOID{}, // SNMP模板OID // Logan.Li
&models.SNMPMetric{}, // SNMP指标 // Logan.Li
&models.SNMPInterface{}, // SNMP接口 // Logan.Li
&models.SiteTask{}, // 站点拨测任务 // Logan.Li
&models.SiteResult{}, // 拨测结果 // Logan.Li
&models.SiteAlert{}, // 拨测告警 // Logan.Li
&models.AlertRuleV2{}, // 告警规则V2 // Logan.Li
&models.AlertCondition{}, // 告警条件 // Logan.Li
&models.AlertNotifyTemplate{}, // 告警通知模板 // Logan.Li
&models.AlertUpgrade{}, // 告警升级 // Logan.Li
&models.AlertUpgradeRecord{},// 告警升级记录 // Logan.Li
&models.AlertSilence{}, // 告警静默 // Logan.Li
&models.AlertEventV2{}, // 告警事件V2 // Logan.Li
}
var lastErr error