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

+277
View File
@@ -0,0 +1,277 @@
package siteprobe
import (
"context"
"crypto/tls"
"fmt"
"net"
"net/http"
"os/exec"
"regexp"
"strconv"
"strings"
"time"
"ezops/model"
)
// ProbeOutcome 单次拨测结果 // Logan.Li
type ProbeOutcome struct {
Status model.SiteResultStatus
HTTPStatus int
ResponseTime float64
TotalTime float64
RemoteIP string
ErrorMessage string
// HTTPS 证书信息 // Logan.Li
SSLCertExpiry *time.Time
SSLCertIssuer string
SSLCertSubject string
SSLDaysRemaining int
}
// Execute 按任务类型执行拨测 // Logan.Li
func Execute(ctx context.Context, task model.SiteTask) ProbeOutcome {
timeout := time.Duration(task.Timeout) * time.Second
if timeout <= 0 {
timeout = 10 * time.Second
}
ctx, cancel := context.WithTimeout(ctx, timeout)
defer cancel()
taskType := strings.ToLower(string(task.TaskType))
if taskType == "" {
taskType = inferTaskType(task)
}
switch taskType {
case "ssl":
return probeSSL(ctx, task)
case "tcp":
return probeTCP(ctx, task.URL)
case "dns":
return probeDNS(ctx, task.URL, task.DNSResolver)
case "ping":
return probePing(ctx, task.URL, task.Timeout)
case "http", "https":
return probeHTTP(ctx, task)
default:
return probeHTTP(ctx, task)
}
}
// InferTaskTypePublic 推断任务类型(供列表修复存量数据)// Logan.Li
func InferTaskTypePublic(task model.SiteTask) string {
return inferTaskType(task)
}
func inferTaskType(task model.SiteTask) string {
name := strings.ToLower(task.Name)
switch {
case strings.Contains(name, "tcp"):
return "tcp"
case strings.Contains(name, "dns"):
return "dns"
case strings.Contains(name, "ping"):
return "ping"
case strings.Contains(name, "ssl"):
return "ssl"
default:
return "http"
}
}
func probeHTTP(ctx context.Context, task model.SiteTask) ProbeOutcome {
start := time.Now()
url := task.URL
if !strings.HasPrefix(url, "http://") && !strings.HasPrefix(url, "https://") {
url = "http://" + url
}
transport := &http.Transport{
TLSClientConfig: &tls.Config{InsecureSkipVerify: !task.VerifySSL},
}
client := &http.Client{
Timeout: time.Until(deadlineFromCtx(ctx)),
Transport: transport,
}
if !task.FollowRedirect {
client.CheckRedirect = func(req *http.Request, via []*http.Request) error {
return http.ErrUseLastResponse
}
}
method := strings.ToUpper(string(task.Method))
if method == "" {
method = http.MethodGet
}
req, err := http.NewRequestWithContext(ctx, method, url, nil)
if err != nil {
return failOutcome(start, err.Error())
}
resp, err := client.Do(req)
elapsed := msSince(start)
if err != nil {
if ctx.Err() == context.DeadlineExceeded {
return ProbeOutcome{Status: model.SiteResultTimeout, TotalTime: elapsed, ResponseTime: elapsed, ErrorMessage: "请求超时"}
}
return failOutcome(start, err.Error())
}
defer resp.Body.Close()
statusOK := resp.StatusCode == task.ExpectedStatus
if task.ExpectedStatus == 0 {
statusOK = resp.StatusCode >= 200 && resp.StatusCode < 400
}
if !statusOK {
return ProbeOutcome{
Status: model.SiteResultFail,
HTTPStatus: resp.StatusCode,
ResponseTime: elapsed,
TotalTime: elapsed,
ErrorMessage: fmt.Sprintf("状态码 %d 不符合期望 %d", resp.StatusCode, task.ExpectedStatus),
}
}
return ProbeOutcome{
Status: model.SiteResultSuccess,
HTTPStatus: resp.StatusCode,
ResponseTime: elapsed,
TotalTime: elapsed,
}
}
func probeTCP(ctx context.Context, target string) ProbeOutcome {
start := time.Now()
host, port, err := splitHostPort(target, "80")
if err != nil {
return failOutcome(start, err.Error())
}
d := net.Dialer{Timeout: time.Until(deadlineFromCtx(ctx))}
conn, err := d.DialContext(ctx, "tcp", net.JoinHostPort(host, port))
elapsed := msSince(start)
if err != nil {
if ctx.Err() == context.DeadlineExceeded {
return ProbeOutcome{Status: model.SiteResultTimeout, TotalTime: elapsed, ResponseTime: elapsed, ErrorMessage: "TCP 连接超时"}
}
return failOutcome(start, err.Error())
}
defer conn.Close()
remote := ""
if addr := conn.RemoteAddr(); addr != nil {
remote = addr.String()
}
return ProbeOutcome{
Status: model.SiteResultSuccess,
ResponseTime: elapsed,
TotalTime: elapsed,
RemoteIP: remote,
}
}
func probeDNS(ctx context.Context, target, resolver string) ProbeOutcome {
start := time.Now()
domain := normalizeDomain(target)
if domain == "" {
return failOutcome(start, "域名不能为空")
}
r := &net.Resolver{}
if resolver != "" {
r = &net.Resolver{
PreferGo: true,
Dial: func(ctx context.Context, network, address string) (net.Conn, error) {
d := net.Dialer{Timeout: time.Until(deadlineFromCtx(ctx))}
return d.DialContext(ctx, network, resolver+":53")
},
}
}
addrs, err := r.LookupHost(ctx, domain)
elapsed := msSince(start)
if err != nil {
if ctx.Err() == context.DeadlineExceeded {
return ProbeOutcome{Status: model.SiteResultTimeout, TotalTime: elapsed, ResponseTime: elapsed, ErrorMessage: "DNS 解析超时"}
}
return failOutcome(start, err.Error())
}
return ProbeOutcome{
Status: model.SiteResultSuccess,
ResponseTime: elapsed,
TotalTime: elapsed,
RemoteIP: strings.Join(addrs, ","),
}
}
func probePing(ctx context.Context, target string, timeoutSec int) ProbeOutcome {
start := time.Now()
host := normalizeDomain(target)
if host == "" {
return failOutcome(start, "主机不能为空")
}
if timeoutSec <= 0 {
timeoutSec = 5
}
cmd := exec.CommandContext(ctx, "ping", "-c", "1", "-W", strconv.Itoa(timeoutSec), host)
out, err := cmd.CombinedOutput()
elapsed := msSince(start)
if err != nil {
msg := strings.TrimSpace(string(out))
if msg == "" {
msg = err.Error()
}
if ctx.Err() == context.DeadlineExceeded {
return ProbeOutcome{Status: model.SiteResultTimeout, TotalTime: elapsed, ResponseTime: elapsed, ErrorMessage: "PING 超时"}
}
return ProbeOutcome{Status: model.SiteResultFail, TotalTime: elapsed, ResponseTime: elapsed, ErrorMessage: msg}
}
if m := regexp.MustCompile(`time[=<]([\d.]+)\s*ms`).FindStringSubmatch(string(out)); len(m) == 2 {
if v, parseErr := strconv.ParseFloat(m[1], 64); parseErr == nil {
elapsed = v
}
}
return ProbeOutcome{Status: model.SiteResultSuccess, ResponseTime: elapsed, TotalTime: elapsed, RemoteIP: host}
}
func splitHostPort(raw, defaultPort string) (string, string, error) {
raw = strings.TrimSpace(raw)
for _, prefix := range []string{"ssl://", "tcp://", "http://", "https://"} {
raw = strings.TrimPrefix(raw, prefix)
}
if host, port, err := net.SplitHostPort(raw); err == nil {
return host, port, nil
}
if raw == "" {
return "", "", fmt.Errorf("TCP 地址不能为空")
}
return raw, defaultPort, nil
}
func normalizeDomain(raw string) string {
raw = strings.TrimSpace(raw)
raw = strings.TrimPrefix(raw, "http://")
raw = strings.TrimPrefix(raw, "https://")
if host, _, err := net.SplitHostPort(raw); err == nil {
return host
}
return strings.TrimSuffix(raw, "/")
}
func failOutcome(start time.Time, msg string) ProbeOutcome {
elapsed := msSince(start)
return ProbeOutcome{
Status: model.SiteResultFail,
ResponseTime: elapsed,
TotalTime: elapsed,
ErrorMessage: msg,
}
}
func msSince(start time.Time) float64 {
return float64(time.Since(start).Milliseconds())
}
func deadlineFromCtx(ctx context.Context) time.Time {
if d, ok := ctx.Deadline(); ok {
return d
}
return time.Now().Add(10 * time.Second)
}
+108
View File
@@ -0,0 +1,108 @@
package siteprobe
import (
"context"
"crypto/tls"
"crypto/x509"
"fmt"
"net"
"net/url"
"strings"
"time"
"ezops/model"
)
// parseSSLTarget 解析 SSL 拨测地址,支持 example.com:443 / https://example.com // Logan.Li
func parseSSLTarget(raw string) (host, port string, err error) {
raw = strings.TrimSpace(raw)
if raw == "" {
return "", "", fmt.Errorf("SSL 地址不能为空")
}
lower := strings.ToLower(raw)
// 兼容 https:example.com(单斜杠写法)
if strings.HasPrefix(lower, "https:") && !strings.HasPrefix(lower, "https://") {
raw = "https://" + strings.TrimSpace(raw[len("https:"):])
lower = strings.ToLower(raw)
}
if strings.HasPrefix(lower, "http://") || strings.HasPrefix(lower, "https://") || strings.HasPrefix(lower, "ssl://") {
u, parseErr := url.Parse(raw)
if parseErr != nil || u.Host == "" {
return "", "", fmt.Errorf("SSL 地址格式无效: %s", raw)
}
if h, p, splitErr := net.SplitHostPort(u.Host); splitErr == nil {
return h, p, nil
}
defPort := "443"
if u.Scheme == "http" {
defPort = "80"
}
return u.Hostname(), defPort, nil
}
// 去掉路径后缀,如 example.com:443/foo
if idx := strings.Index(raw, "/"); idx >= 0 {
raw = raw[:idx]
}
return splitHostPort(raw, "443")
}
// probeSSL 纯 TLS 握手检测证书(不发起 HTTP 请求)// Logan.Li
func probeSSL(ctx context.Context, task model.SiteTask) ProbeOutcome {
start := time.Now()
host, port, err := parseSSLTarget(task.URL)
if err != nil {
return failOutcome(start, err.Error())
}
addr := net.JoinHostPort(host, port)
d := &net.Dialer{Timeout: time.Until(deadlineFromCtx(ctx))}
conn, err := tls.DialWithDialer(d, "tcp", addr, &tls.Config{
ServerName: host,
InsecureSkipVerify: !task.VerifySSL,
})
elapsed := msSince(start)
if err != nil {
if ctx.Err() == context.DeadlineExceeded {
return ProbeOutcome{Status: model.SiteResultTimeout, TotalTime: elapsed, ResponseTime: elapsed, ErrorMessage: "TLS 握手超时"}
}
return failOutcome(start, fmt.Sprintf("TLS 握手失败: %s", err.Error()))
}
defer conn.Close()
remote := ""
if addr := conn.RemoteAddr(); addr != nil {
remote = addr.String()
}
out := ProbeOutcome{
Status: model.SiteResultSuccess,
ResponseTime: elapsed,
TotalTime: elapsed,
RemoteIP: remote,
}
applyCertsToOutcome(&out, conn.ConnectionState().PeerCertificates)
return out
}
// applyCertsToOutcome 从 TLS 证书链填充到期信息 // Logan.Li
func applyCertsToOutcome(out *ProbeOutcome, certs []*x509.Certificate) {
if len(certs) == 0 {
return
}
cert := certs[0]
expiry := cert.NotAfter
days := int(time.Until(expiry).Hours() / 24)
out.SSLCertExpiry = &expiry
out.SSLDaysRemaining = days
out.SSLCertIssuer = cert.Issuer.CommonName
if out.SSLCertIssuer == "" {
out.SSLCertIssuer = cert.Issuer.String()
}
out.SSLCertSubject = cert.Subject.CommonName
if out.SSLCertSubject == "" {
out.SSLCertSubject = cert.Subject.String()
}
}
// IsSSLProbeTask 是否为 SSL 证书拨测任务 // Logan.Li
func IsSSLProbeTask(task model.SiteTask) bool {
return strings.EqualFold(string(task.TaskType), "ssl")
}
+37
View File
@@ -0,0 +1,37 @@
package siteprobe
import "testing"
// SSL 地址解析单测 // Logan.Li
func TestParseSSLTarget(t *testing.T) {
cases := []struct {
raw string
host string
port string
wantFail bool
}{
{"example.com:443", "example.com", "443", false},
{"example.com", "example.com", "443", false},
{"https://example.com", "example.com", "443", false},
{"https://example.com:8443", "example.com", "8443", false},
{"https://example.com/path", "example.com", "443", false},
{"https:example.com", "example.com", "443", false},
{"ssl://example.com:443", "example.com", "443", false},
{"", "", "", true},
}
for _, c := range cases {
host, port, err := parseSSLTarget(c.raw)
if c.wantFail {
if err == nil {
t.Fatalf("expected error for %q", c.raw)
}
continue
}
if err != nil {
t.Fatalf("parseSSLTarget(%q): %v", c.raw, err)
}
if host != c.host || port != c.port {
t.Fatalf("parseSSLTarget(%q) = %s:%s, want %s:%s", c.raw, host, port, c.host, c.port)
}
}
}
+118
View File
@@ -0,0 +1,118 @@
package siteprobe
import (
"context"
"log"
"sync"
"time"
"ezops/model"
"gorm.io/gorm"
)
// Worker 站点拨测后台采集器 // Logan.Li
type Worker struct {
db *gorm.DB
stopCh chan struct{}
wg sync.WaitGroup
}
// NewWorker 创建拨测 worker
func NewWorker(db *gorm.DB) *Worker {
return &Worker{db: db, stopCh: make(chan struct{})}
}
// Start 启动拨测循环
func (w *Worker) Start(intervalSec int) {
if intervalSec < 5 {
intervalSec = 10
}
w.wg.Add(1)
go func() {
defer w.wg.Done()
ticker := time.NewTicker(time.Duration(intervalSec) * time.Second)
defer ticker.Stop()
log.Printf("[SiteProbe] 拨测引擎启动,扫描间隔 %ds", intervalSec)
w.runOnce()
for {
select {
case <-ticker.C:
w.runOnce()
case <-w.stopCh:
log.Printf("[SiteProbe] 拨测引擎停止")
return
}
}
}()
}
// Stop 停止拨测
func (w *Worker) Stop() {
close(w.stopCh)
w.wg.Wait()
}
// RunTaskNow 立即执行单个任务拨测 // Logan.Li
func (w *Worker) RunTaskNow(taskID uint) error {
var task model.SiteTask
if err := w.db.First(&task, taskID).Error; err != nil {
return err
}
w.probeTask(task)
return nil
}
func (w *Worker) runOnce() {
var tasks []model.SiteTask
w.db.Where("status = ?", model.SiteTaskStatusRunning).Find(&tasks)
now := time.Now()
for _, task := range tasks {
if task.Frequency <= 0 {
task.Frequency = 60
}
if task.LastCheckAt != nil && now.Sub(*task.LastCheckAt) < time.Duration(task.Frequency)*time.Second {
continue
}
w.probeTask(task)
}
}
func (w *Worker) probeTask(task model.SiteTask) {
if task.TaskType == "" {
task.TaskType = model.SiteTaskType(inferTaskType(task))
}
outcome := Execute(context.Background(), task)
now := time.Now()
result := model.SiteResult{
TaskID: task.ID,
Status: outcome.Status,
HTTPStatusCode: outcome.HTTPStatus,
ResponseTime: outcome.ResponseTime,
TotalTime: outcome.TotalTime,
RemoteIP: outcome.RemoteIP,
ErrorMessage: outcome.ErrorMessage,
SSLCertExpiry: outcome.SSLCertExpiry,
SSLCertIssuer: outcome.SSLCertIssuer,
SSLCertSubject: outcome.SSLCertSubject,
SSLDaysRemaining: outcome.SSLDaysRemaining,
CheckedAt: now,
}
w.db.Create(&result)
lastResult := "fail"
if outcome.Status == model.SiteResultSuccess {
lastResult = "success"
}
avg := outcome.ResponseTime
if task.AvgResponse > 0 {
avg = (task.AvgResponse + outcome.ResponseTime) / 2
}
w.db.Model(&model.SiteTask{}).Where("id = ?", task.ID).Updates(map[string]interface{}{
"task_type": task.TaskType,
"last_result": lastResult,
"last_check_at": now,
"avg_response": avg,
})
}