Merge pull request #94 from acd19ml/develop

feat(Metrics&Alert): 完成告警规则调整记录

Author: Erickw87

client/src/api/index.ts

@@ -2,6 +2,7 @@ import axios from 'axios'
 
 // 创建 axios 实例
 const api = axios.create({
+  // baseURL: (import.meta as any).env?.VITE_API_BASE_URL || 'http://10.210.10.33:8080',
   timeout: 10000,
   headers: {
     'Content-Type': 'application/json'

client/src/mock/services.ts

@@ -1845,27 +1845,9 @@ loadServiceAlertStatus()
  */
 export const serviceVersionAlertStatusMap: Record<string, Record<string, ServiceAlertStatus>> = {}
 
-const saveServiceVersionAlertStatus = () => {
-  try {
-    localStorage.setItem('serviceVersionAlertStatusMap', JSON.stringify(serviceVersionAlertStatusMap))
-    console.log('服务版本告警状态已保存到 localStorage')
-  } catch (error) {
-    console.error('保存服务版本告警状态失败:', error)
-  }
-}
+const saveServiceVersionAlertStatus = () => {}
 
-const loadServiceVersionAlertStatus = () => {
-  try {
-    const data = localStorage.getItem('serviceVersionAlertStatusMap')
-    if (data) {
-      const parsed = JSON.parse(data)
-      Object.assign(serviceVersionAlertStatusMap, parsed)
-      console.log('已从 localStorage 加载服务版本告警状态')
-    }
-  } catch (error) {
-    console.error('从 localStorage 加载服务版本告警状态失败:', error)
-  }
-}
+const loadServiceVersionAlertStatus = () => {}
 
 /**
  * 根据告警状态更新服务版本状态
@@ -1919,8 +1901,7 @@ export const clearServiceVersionAlertStatus = (serviceName: string, version?: st
   console.log(`已清除服务 ${serviceName} ${version ? '版本 ' + version : '所有版本'} 的告警状态`)
 }
 
-// 页面加载时恢复服务版本告警状态
-loadServiceVersionAlertStatus()
+// 页面加载时不再从 localStorage 恢复服务版本告警状态（禁用持久化）
 
 // ==================== 发布任务状态管理 ====================
 // 管理服务的发布任务状态，用于显示发布指示器

client/src/views/ChangeLogView.vue

@@ -84,11 +84,31 @@
 import { ref, computed, onMounted, watch } from 'vue'
 import { useAppStore, type ChangeItem, type AlarmChangeItem } from '@/stores/app'
 import { mockApi } from '@/mock/api'
-import type { DeploymentChangelogResponse, DeploymentChangelogItem, AlertRuleChangelogResponse, AlertRuleChangeItem } from '@/mock/services'
+import { apiService } from '@/api'
+import type { DeploymentChangelogResponse } from '@/mock/services'
 import ChangeCard from '@/components/ChangeCard.vue'
 import AlarmChangeCard from '@/components/AlarmChangeCard.vue'
 import { ArrowLeft, Loading } from '@element-plus/icons-vue'
 
+interface AlertRuleChangeValue {
+  name: string
+  old: string
+  new: string
+}
+
+interface AlertRuleChangeItem {
+  name: string
+  editTime: string
+  scope: string
+  values: AlertRuleChangeValue[]
+  reason: string
+}
+
+interface AlertRuleChangelogResponse {
+  items: AlertRuleChangeItem[]
+  next?: string
+}
+
 const appStore = useAppStore()
 
 const activeTab = ref('service')
@@ -160,7 +180,7 @@ const transformAlertRuleChangelogToAlarmChangeItems = (changelogData: AlertRuleC
     const serviceName = item.scope?.startsWith('service:') ? item.scope.slice('service:'.length) + '服务' : '全局服务'
     
     // 构建变更描述
-    const changeDescription = item.values.map(value => {
+  const changeDescription = item.values.map((value) => {
       return `${value.name}: ${value.old} -> ${value.new}`
     }).join(', ')
     
@@ -200,21 +220,21 @@ const loadDeploymentChangelog = async (start?: string, limit?: number) => {
   }
 }
 
-// 加载告警规则变更记录
+// 加载告警规则变更记录（使用真实 API）
 const loadAlertRuleChangelog = async (start?: string, limit?: number) => {
   if (alertRuleLoading.value) return // 防止重复加载
   
   try {
     alertRuleLoading.value = true
     error.value = null
     
-    const response = await mockApi.getAlertRuleChangelog(start, limit)
-    alertRuleChangelog.value = response
+    const response = await apiService.getAlertRuleChangelog(start, limit ?? 10)
+    alertRuleChangelog.value = response.data
     
     // 转换数据格式
-    alarmChangeItems.value = transformAlertRuleChangelogToAlarmChangeItems(response.items)
+    alarmChangeItems.value = transformAlertRuleChangelogToAlarmChangeItems(response.data.items)
     
-    console.log('告警规则变更记录加载成功:', response)
+    console.log('告警规则变更记录加载成功:', response.data)
   } catch (err) {
     error.value = '加载告警规则变更记录失败'
     console.error('加载告警规则变更记录失败:', err)

client/vite.config.ts

@@ -18,7 +18,7 @@ export default defineConfig({
   server: {
     proxy: {
       '/v1': {
-        target: 'http://127.0.0.1:8080',
+        target: 'http://10.210.10.33:8080',
         changeOrigin: true,
         secure: false,
       }

cmd/zeroops/Dockerfile

@@ -1,11 +1,11 @@
-FROM golang:1.24-alpine AS builder
+FROM docker.m.daocloud.io/library/golang:1.24-alpine AS builder
 WORKDIR /src
 COPY go.mod go.sum ./
 RUN go mod download
 COPY . .
 RUN CGO_ENABLED=0 GOOS=linux GOARCH=amd64 go build -o /out/zeroops ./cmd/zeroops
 
-FROM gcr.io/distroless/base-debian12
+FROM gcr.m.daocloud.io/distroless/base-debian12:nonroot
 WORKDIR /app
 COPY --from=builder /out/zeroops /app/zeroops
 # 复制配置文件目录

cmd/zeroops/main.go

@@ -6,7 +6,7 @@ import (
 	"strconv"
 	"time"
 
-	"github.com/fox-gonic/fox"
+	"github.com/gin-gonic/gin"
 	alertapi "github.com/qiniu/zeroops/internal/alerting/api"
 	adb "github.com/qiniu/zeroops/internal/alerting/database"
 	"github.com/qiniu/zeroops/internal/alerting/service/healthcheck"
@@ -96,22 +96,24 @@ func main() {
 	go rem.Start(ctx, alertCh)
 
 	// start Prometheus anomaly detection scheduler
-	promInterval := parseDuration(cfg.Alerting.Prometheus.SchedulerInterval, 6*time.Hour)
+	promInterval := parseDuration(cfg.Alerting.Prometheus.SchedulerInterval, 5*time.Minute)
 	promStep := parseDuration(cfg.Alerting.Prometheus.QueryStep, time.Minute)
 	promRange := parseDuration(cfg.Alerting.Prometheus.QueryRange, 6*time.Hour)
 	promCfg := healthcheck.NewPrometheusConfigFromApp(&cfg.Alerting.Prometheus)
-	promClient := healthcheck.NewPrometheusClient(promCfg)
+	anomalyDetectClient := healthcheck.NewAnomalyDetectClient(promCfg)
 	go healthcheck.StartPrometheusScheduler(ctx, healthcheck.PrometheusDeps{
-		DB:               alertDB,
-		PrometheusClient: promClient,
-		Interval:         promInterval,
-		QueryStep:        promStep,
-		QueryRange:       promRange,
-		RulesetBase:      cfg.Alerting.Ruleset.APIBase,
-		RulesetTimeout:   parseDuration(cfg.Alerting.Ruleset.APITimeout, 10*time.Second),
+		DB:                  alertDB,
+		AnomalyDetectClient: anomalyDetectClient,
+		Interval:            promInterval,
+		QueryStep:           promStep,
+		QueryRange:          promRange,
+		RulesetBase:         cfg.Alerting.Ruleset.APIBase,
+		RulesetTimeout:      parseDuration(cfg.Alerting.Ruleset.APITimeout, 10*time.Second),
 	})
 
-	router := fox.New()
+	router := gin.New()
+	router.Use(gin.Logger())
+	router.Use(gin.Recovery())
 	router.Use(middleware.Authentication)
 	alertapi.NewApiWithConfig(router, cfg)
 	if err := serviceManagerSrv.UseApi(router); err != nil {

configs/alerting/rules.json

@@ -6,7 +6,7 @@
       "expr": "histogram_quantile(0.98, sum(rate(http_latency_seconds_bucket{}[2m])) by (service, service_version, le))",
       "op": ">",
       "severity": "P0",
-      "watch_time": "5 minutes",
+      "watch_time": "5m",
       "metas": [
         { "labels": { "service": "storage-service", "service_version": "1.0.0" }, "threshold": 1000 },
         { "labels": { "service": "queue-service", "service_version": "1.0.0" }, "threshold": 1000 },
@@ -19,7 +19,7 @@
       "expr": "histogram_quantile(0.98, sum(rate(http_latency_seconds_bucket{}[2m])) by (service, service_version, le))",
       "op": ">",
       "severity": "P1",
-      "watch_time": "4 minutes",
+      "watch_time": "4m",
       "metas": [
         { "labels": { "service": "storage-service", "service_version": "1.0.0" }, "threshold": 500 },
         { "labels": { "service": "queue-service", "service_version": "1.0.0" }, "threshold": 500 },
@@ -31,7 +31,7 @@
         "description":"HTTP error rate by service P0",
         "op":">",
         "severity":"P0",
-        "watch_time":"5 minutes",
+        "watch_time":"5m",
         "expr":"sum(rate(http_latency_seconds_count{\"http.status_code\"=~\"4..|5..\", \"http.route\"!=\"/metrics\"}[2m])) by (service, service_version) / sum(rate(http_latency_seconds_count{\"http.route\"!=\"/metrics\"}[2m])) by (service, service_version)",
         "metas":[
             {"labels":{"service":"storage-service","service_version":"1.0.0"},"threshold":5},
@@ -44,7 +44,7 @@
         "description":"HTTP error rate by service P1",
         "op":">",
         "severity":"P1",
-        "watch_time":"5 minutes",
+        "watch_time":"5m",
         "expr":"sum(rate(http_latency_seconds_count{\"http.status_code\"=~\"4..|5..\", \"http.route\"!=\"/metrics\"}[2m])) by (service, service_version) / sum(rate(http_latency_seconds_count{\"http.route\"!=\"/metrics\"}[2m])) by (service, service_version)",
         "metas":[
             {"labels":{"service":"storage-service","service_version":"1.0.0"},"threshold":3},

docker-compose.override.yml renamed to docker-compose.override.yml

@@ -315,6 +315,49 @@ curl -X POST http://localhost:8080/v1/integrations/alertmanager/webhook \
   }'
 ```
 
+### 4. 获取告警规则变更记录
+
+用于查询统一化告警规则的变更记录（阈值、观察窗口等），支持按时间游标分页。
+
+**请求：**
+```http
+GET /v1/changelog/alertrules?start={start}&limit={limit}
+```
+
+**查询参数：**
+
+| 参数名 | 类型 | 必填 | 说明 |
+|--------|------|------|------|
+| start | string | 否 | 游标时间（ISO 8601）。第一页可不传；翻页使用上次响应的 `next` |
+| limit | integer | 是 | 返回数量，范围 1-100 |
+
+分页说明：按 `change_time` 倒序返回，`start` 为上界（`<= start`）。响应中的 `next` 为当前页最后一条的 `editTime`。
+
+**响应示例：**
+```json
+{
+  "items": [
+    {
+      "name": "http_request_latency_p98_seconds_P1",
+      "editTime": "2024-01-03T03:00:00Z",
+      "scope": "",
+      "values": [
+        {"name": "threshold", "old": "10", "new": "15"}
+      ],
+      "reason": "Update"
+    }
+  ],
+  "next": "2024-01-03T03:00:00Z"
+}
+```
+
+**状态码：**
+- `200 OK`: 成功
+- `400 Bad Request`: 参数错误
+- `401 Unauthorized`: 认证失败
+- `500 Internal Server Error`: 服务器内部错误
+
 ## 版本历史
 
+- **v1.1** (2025-10-07): 新增 `GET /v1/changelog/alertrules`
 - **v1.0** (2025-09-11): 初始版本，支持基础的告警列表和详情查询

docs/alerting/api.md

@@ -13,7 +13,7 @@
 
 ## 数据表设计
 
-### 1) talert_issues（告警问题表）
+### 1) alert_issues（告警问题表）
 
 存储告警问题的主要信息。
 
@@ -23,7 +23,7 @@
 | state | enum(Closed, Open) | 问题状态 |
 | level | varchar(32) | 告警等级：如 P0/P1/Px |
 | alert_state | enum(Pending, Restored, AutoRestored, InProcessing) | 处理状态 |
-| title | varchar(255) | 告警标题 |
+| title | varchar(255) | 告警标题 
 | labels | json | 标签，格式：[{key, value}] |
 | alert_since | TIMESTAMP(6) | 告警发生时间 |
 | resolved_at | TIMESTAMP(6) | 告警结束时间 |
@@ -64,8 +64,6 @@
 | labels | text | labels 的 JSON 字符串表示（规范化后） |
 | old_threshold | numeric | 旧阈值（可空） |
 | new_threshold | numeric | 新阈值（可空） |
-| old_watch | interval | 旧观察窗口（可空） |
-| new_watch | interval | 新观察窗口（可空） |
 
 
 **索引建议：**