feat: 监控告警体系统一重构 + 自动化任务调度引擎 + 监控大屏拆分

告警中心
- 新增统一告警(alert_unified)、监控告警规则(monitor_alert_rule)、指标目录(alert_metric_catalog)
- 通知策略与消息分发(alert_notify_policy/dispatch/message)、静默规则
- 告警引擎按阈值/SNMP 拆分(alert_engine_monitor/snmp)

监控采集
- SNMP 常量/模板与历史指标(snmp templates/constants、snmp_metrics_history、snmp_collect)
- 中间件监控 Redis 采集与种子(middleware_monitor_redis/seed)
- 站点探测 pkg(siteprobe) + 站点视图(site_view)

自动化任务
- cron 调度器(cronscheduler)、脚本运行器(scriptrunner)、系统任务(systask)及运行记录(cron_job_run、system_task)

审计与凭证
- 审计设置(audit_setting) + 历史清理(auditcleaner)

前端
- 监控页面(host/middleware/site/snmp)拆分为子模块并大屏化
- 告警中心多 Tab(规则/指标目录/通知策略/静默)、任务管理组件重构、Agent CLI Markdown 渲染
- 移除旧脚本卡片/详情/日志组件,统一为 ScriptTable/ScriptCodeEditor/TerminalOutput

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
LoganandClaude committed 2026-07-01 23:44:45 +08:00
1 parent 092d0e4b9e
commit 471353e3d7
190 files changed
+16640 -3484

No files matched your search

+4 -104
View File
@@ -1,13 +1,10 @@
package controller
import (
"fmt"
"net/http"
"strconv"
"time"
"ezops/model"
"ezops/pkg/remoteexec"
"github.com/gin-gonic/gin"
)
@@ -294,113 +291,16 @@ func (h *AutomationHandler) ToggleCronJob(c *gin.Context) {
// @Router /cron-jobs/{id}/execute [post]
func (h *AutomationHandler) ExecuteCronJob(c *gin.Context) {
id, _ := strconv.Atoi(c.Param("id"))
var job model.CronJob
if err := h.db.First(&job, id).Error; err != nil {
execution, err := RunCronJobByID(h.db, uint(id))
if err != nil {
c.JSON(http.StatusNotFound, model.Response{
Code: 404,
Message: "定时任务不存在",
Message: err.Error(),
})
return
}
// Resolve the content to execute. For type=script, load the linked Script.
content := job.Content
scriptType := job.Type // shell | python | command | script
if scriptType == "command" {
scriptType = "shell" // one-liner commands run via bash
}
if job.Type == "script" && job.ScriptID != nil {
var script model.Script
if err := h.db.First(&script, *job.ScriptID).Error; err == nil {
content = script.Content
scriptType = script.Type
h.db.Model(&script).UpdateColumn("run_count", script.RunCount+1)
}
}
now := time.Now()
execution := model.TaskExecution{
TaskType: "cron",
TaskID: uint(id),
TaskName: job.Name,
Status: "running",
TargetType: job.TargetType,
Targets: job.Targets,
StartedAt: &now,
}
if err := h.db.Create(&execution).Error; err != nil {
c.JSON(http.StatusInternalServerError, model.Response{
Code: 500,
Message: "创建执行记录失败",
})
return
}
var output string
var exitCode int
var duration int
status := "success"
// 根据 TargetType 决定执行方式 // Logan.Li
if job.TargetType != "" && job.TargetType != "local" && job.Targets != "" {
// 远程目标执行 — 通过 SSH 并发执行
remoteResults, remoteErr := ExecuteCronOnTargets(h.db, job, scriptType, content, job.Timeout)
duration = int(time.Since(now).Milliseconds())
if remoteErr != nil {
output = fmt.Sprintf("远程执行调度失败: %v", remoteErr)
exitCode = -1
status = "failed"
} else {
// 格式化远程执行结果
output = remoteexec.FormatExecResults(remoteResults)
// 任意一台失败则整体标记为 failed
for _, r := range remoteResults {
if !r.Success {
status = "failed"
exitCode = -1
break
}
}
}
} else {
// 本地执行
output, exitCode, duration = runLocal(scriptType, content, job.Timeout)
if exitCode != 0 {
status = "failed"
}
}
finishedAt := time.Now()
h.db.Model(&execution).Updates(map[string]interface{}{
"status": status,
"output": output,
"duration": duration,
"finished_at": finishedAt,
})
// Update cron job statistics.
statsUpdates := map[string]interface{}{
"last_run_at": now,
"last_status": status,
"run_count": job.RunCount + 1,
}
if status == "success" {
statsUpdates["success_count"] = job.SuccessCount + 1
} else {
statsUpdates["fail_count"] = job.FailCount + 1
}
h.db.Model(&job).Updates(statsUpdates)
execution.Output = output
execution.Status = status
execution.Duration = duration
execution.FinishedAt = &finishedAt
msg := "执行成功"
if status == "failed" {
if execution.Status == "failed" {
msg = "执行失败"
}
c.JSON(http.StatusOK, model.Response{