refactor(ops): 优化任务心跳和组件刷新机制

后端改动:
- 添加 ops_job_heartbeats.last_result 字段记录任务执行结果
- 优化告警评估器统计信息(规则数/事件数/邮件数)
- 统一各定时任务的心跳记录格式

前端改动:
- 重构 OpsConcurrencyCard 使用父组件统一控制刷新节奏
- 移除独立的 5 秒刷新定时器,改用 refreshToken 机制
- 修复 TypeScript 类型错误

backend/internal/repository/ops_repo_metrics.go

@@ -296,9 +296,10 @@ INSERT INTO ops_job_heartbeats (
   last_error_at,
   last_error,
   last_duration_ms,
+  last_result,
   updated_at
 ) VALUES (
-  $1,$2,$3,$4,$5,$6,NOW()
+  $1,$2,$3,$4,$5,$6,$7,NOW()
 )
 ON CONFLICT (job_name) DO UPDATE SET
   last_run_at = COALESCE(EXCLUDED.last_run_at, ops_job_heartbeats.last_run_at),
@@ -312,6 +313,10 @@ ON CONFLICT (job_name) DO UPDATE SET
     ELSE COALESCE(EXCLUDED.last_error, ops_job_heartbeats.last_error)
   END,
   last_duration_ms = COALESCE(EXCLUDED.last_duration_ms, ops_job_heartbeats.last_duration_ms),
+  last_result = CASE
+    WHEN EXCLUDED.last_success_at IS NOT NULL THEN COALESCE(EXCLUDED.last_result, ops_job_heartbeats.last_result)
+    ELSE ops_job_heartbeats.last_result
+  END,
   updated_at = NOW()`
 
 	_, err := r.db.ExecContext(
@@ -323,6 +328,7 @@ ON CONFLICT (job_name) DO UPDATE SET
 		opsNullTime(input.LastErrorAt),
 		opsNullString(input.LastError),
 		opsNullInt(input.LastDurationMs),
+		opsNullString(input.LastResult),
 	)
 	return err
 }
@@ -340,6 +346,7 @@ SELECT
   last_error_at,
   last_error,
   last_duration_ms,
+  last_result,
   updated_at
 FROM ops_job_heartbeats
 ORDER BY job_name ASC`
@@ -359,6 +366,8 @@ ORDER BY job_name ASC`
 		var lastError sql.NullString
 		var lastDuration sql.NullInt64
 
+		var lastResult sql.NullString
+
 		if err := rows.Scan(
 			&item.JobName,
 			&lastRun,
@@ -366,6 +375,7 @@ ORDER BY job_name ASC`
 			&lastErrorAt,
 			&lastError,
 			&lastDuration,
+			&lastResult,
 			&item.UpdatedAt,
 		); err != nil {
 			return nil, err
@@ -391,6 +401,10 @@ ORDER BY job_name ASC`
 			v := lastDuration.Int64
 			item.LastDurationMs = &v
 		}
+		if lastResult.Valid {
+			v := lastResult.String
+			item.LastResult = &v
+		}
 
 		out = append(out, &item)
 	}

backend/internal/service/ops_aggregation_service.go

@@ -4,6 +4,7 @@ import (
 	"context"
 	"database/sql"
 	"errors"
+	"fmt"
 	"log"
 	"strings"
 	"sync"
@@ -235,11 +236,13 @@ func (s *OpsAggregationService) aggregateHourly() {
 	successAt := finishedAt
 	hbCtx, hbCancel := context.WithTimeout(context.Background(), 2*time.Second)
 	defer hbCancel()
+	result := truncateString(fmt.Sprintf("window=%s..%s", start.Format(time.RFC3339), end.Format(time.RFC3339)), 2048)
 	_ = s.opsRepo.UpsertJobHeartbeat(hbCtx, &OpsUpsertJobHeartbeatInput{
 		JobName:        opsAggHourlyJobName,
 		LastRunAt:      &runAt,
 		LastSuccessAt:  &successAt,
 		LastDurationMs: &dur,
+		LastResult:     &result,
 	})
 }
 
@@ -331,11 +334,13 @@ func (s *OpsAggregationService) aggregateDaily() {
 	successAt := finishedAt
 	hbCtx, hbCancel := context.WithTimeout(context.Background(), 2*time.Second)
 	defer hbCancel()
+	result := truncateString(fmt.Sprintf("window=%s..%s", start.Format(time.RFC3339), end.Format(time.RFC3339)), 2048)
 	_ = s.opsRepo.UpsertJobHeartbeat(hbCtx, &OpsUpsertJobHeartbeatInput{
 		JobName:        opsAggDailyJobName,
 		LastRunAt:      &runAt,
 		LastSuccessAt:  &successAt,
 		LastDurationMs: &dur,
+		LastResult:     &result,
 	})
 }
 

backend/internal/service/ops_alert_evaluator_service.go

@@ -190,6 +190,13 @@ func (s *OpsAlertEvaluatorService) evaluateOnce(interval time.Duration) {
 		return
 	}
 
+	rulesTotal := len(rules)
+	rulesEnabled := 0
+	rulesEvaluated := 0
+	eventsCreated := 0
+	eventsResolved := 0
+	emailsSent := 0
+
 	now := time.Now().UTC()
 	safeEnd := now.Truncate(time.Minute)
 	if safeEnd.IsZero() {
@@ -205,6 +212,7 @@ func (s *OpsAlertEvaluatorService) evaluateOnce(interval time.Duration) {
 		if rule == nil || !rule.Enabled || rule.ID <= 0 {
 			continue
 		}
+		rulesEnabled++
 
 		scopePlatform, scopeGroupID, scopeRegion := parseOpsAlertRuleScope(rule.Filters)
 
@@ -220,6 +228,7 @@ func (s *OpsAlertEvaluatorService) evaluateOnce(interval time.Duration) {
 			s.resetRuleState(rule.ID, now)
 			continue
 		}
+		rulesEvaluated++
 
 		breachedNow := compareMetric(metricValue, rule.Operator, rule.Threshold)
 		required := requiredSustainedBreaches(rule.SustainedMinutes, interval)
@@ -278,8 +287,11 @@ func (s *OpsAlertEvaluatorService) evaluateOnce(interval time.Duration) {
 				continue
 			}
 
+			eventsCreated++
 			if created != nil && created.ID > 0 {
-				s.maybeSendAlertEmail(ctx, runtimeCfg, rule, created)
+				if s.maybeSendAlertEmail(ctx, runtimeCfg, rule, created) {
+					emailsSent++
+				}
 			}
 			continue
 		}
@@ -289,11 +301,14 @@ func (s *OpsAlertEvaluatorService) evaluateOnce(interval time.Duration) {
 			resolvedAt := now
 			if err := s.opsRepo.UpdateAlertEventStatus(ctx, activeEvent.ID, OpsAlertStatusResolved, &resolvedAt); err != nil {
 				log.Printf("[OpsAlertEvaluator] resolve event failed (event=%d): %v", activeEvent.ID, err)
+			} else {
+				eventsResolved++
 			}
 		}
 	}
 
-	s.recordHeartbeatSuccess(runAt, time.Since(startedAt))
+	result := truncateString(fmt.Sprintf("rules=%d enabled=%d evaluated=%d created=%d resolved=%d emails_sent=%d", rulesTotal, rulesEnabled, rulesEvaluated, eventsCreated, eventsResolved, emailsSent), 2048)
+	s.recordHeartbeatSuccess(runAt, time.Since(startedAt), result)
 }
 
 func (s *OpsAlertEvaluatorService) pruneRuleStates(rules []*OpsAlertRule) {
@@ -585,32 +600,32 @@ func buildOpsAlertDescription(rule *OpsAlertRule, value float64, windowMinutes i
 	)
 }
 
-func (s *OpsAlertEvaluatorService) maybeSendAlertEmail(ctx context.Context, runtimeCfg *OpsAlertRuntimeSettings, rule *OpsAlertRule, event *OpsAlertEvent) {
+func (s *OpsAlertEvaluatorService) maybeSendAlertEmail(ctx context.Context, runtimeCfg *OpsAlertRuntimeSettings, rule *OpsAlertRule, event *OpsAlertEvent) bool {
 	if s == nil || s.emailService == nil || s.opsService == nil || event == nil || rule == nil {
-		return
+		return false
 	}
 	if event.EmailSent {
-		return
+		return false
 	}
 	if !rule.NotifyEmail {
-		return
+		return false
 	}
 
 	emailCfg, err := s.opsService.GetEmailNotificationConfig(ctx)
 	if err != nil || emailCfg == nil || !emailCfg.Alert.Enabled {
-		return
+		return false
 	}
 
 	if len(emailCfg.Alert.Recipients) == 0 {
-		return
+		return false
 	}
 	if !shouldSendOpsAlertEmailByMinSeverity(strings.TrimSpace(emailCfg.Alert.MinSeverity), strings.TrimSpace(rule.Severity)) {
-		return
+		return false
 	}
 
 	if runtimeCfg != nil && runtimeCfg.Silencing.Enabled {
 		if isOpsAlertSilenced(time.Now().UTC(), rule, event, runtimeCfg.Silencing) {
-			return
+			return false
 		}
 	}
 
@@ -639,6 +654,7 @@ func (s *OpsAlertEvaluatorService) maybeSendAlertEmail(ctx context.Context, runt
 	if anySent {
 		_ = s.opsRepo.UpdateAlertEventEmailSent(context.Background(), event.ID, true)
 	}
+	return anySent
 }
 
 func buildOpsAlertEmailBody(rule *OpsAlertRule, event *OpsAlertEvent) string {
@@ -806,7 +822,7 @@ func (s *OpsAlertEvaluatorService) maybeLogSkip(key string) {
 	log.Printf("[OpsAlertEvaluator] leader lock held by another instance; skipping (key=%q)", key)
 }
 
-func (s *OpsAlertEvaluatorService) recordHeartbeatSuccess(runAt time.Time, duration time.Duration) {
+func (s *OpsAlertEvaluatorService) recordHeartbeatSuccess(runAt time.Time, duration time.Duration, result string) {
 	if s == nil || s.opsRepo == nil {
 		return
 	}
@@ -814,11 +830,17 @@ func (s *OpsAlertEvaluatorService) recordHeartbeatSuccess(runAt time.Time, durat
 	durMs := duration.Milliseconds()
 	ctx, cancel := context.WithTimeout(context.Background(), 2*time.Second)
 	defer cancel()
+	msg := strings.TrimSpace(result)
+	if msg == "" {
+		msg = "ok"
+	}
+	msg = truncateString(msg, 2048)
 	_ = s.opsRepo.UpsertJobHeartbeat(ctx, &OpsUpsertJobHeartbeatInput{
 		JobName:        opsAlertEvaluatorJobName,
 		LastRunAt:      &runAt,
 		LastSuccessAt:  &now,
 		LastDurationMs: &durMs,
+		LastResult:     &msg,
 	})
 }
 

backend/internal/service/ops_cleanup_service.go

@@ -149,7 +149,7 @@ func (s *OpsCleanupService) runScheduled() {
 		log.Printf("[OpsCleanup] cleanup failed: %v", err)
 		return
 	}
-	s.recordHeartbeatSuccess(runAt, time.Since(startedAt))
+	s.recordHeartbeatSuccess(runAt, time.Since(startedAt), counts)
 	log.Printf("[OpsCleanup] cleanup complete: %s", counts)
 }
 
@@ -330,12 +330,13 @@ func (s *OpsCleanupService) tryAcquireLeaderLock(ctx context.Context) (func(), b
 	return release, true
 }
 
-func (s *OpsCleanupService) recordHeartbeatSuccess(runAt time.Time, duration time.Duration) {
+func (s *OpsCleanupService) recordHeartbeatSuccess(runAt time.Time, duration time.Duration, counts opsCleanupDeletedCounts) {
 	if s == nil || s.opsRepo == nil {
 		return
 	}
 	now := time.Now().UTC()
 	durMs := duration.Milliseconds()
+	result := truncateString(counts.String(), 2048)
 	ctx, cancel := context.WithTimeout(context.Background(), 2*time.Second)
 	defer cancel()
 	_ = s.opsRepo.UpsertJobHeartbeat(ctx, &OpsUpsertJobHeartbeatInput{
@@ -343,6 +344,7 @@ func (s *OpsCleanupService) recordHeartbeatSuccess(runAt time.Time, duration tim
 		LastRunAt:      &runAt,
 		LastSuccessAt:  &now,
 		LastDurationMs: &durMs,
+		LastResult:     &result,
 	})
 }
 

backend/internal/service/ops_port.go

@@ -235,6 +235,9 @@ type OpsUpsertJobHeartbeatInput struct {
 	LastErrorAt    *time.Time
 	LastError      *string
 	LastDurationMs *int64
+
+	// LastResult is an optional human-readable summary of the last successful run.
+	LastResult *string
 }
 
 type OpsJobHeartbeat struct {
@@ -245,6 +248,7 @@ type OpsJobHeartbeat struct {
 	LastErrorAt    *time.Time `json:"last_error_at"`
 	LastError      *string    `json:"last_error"`
 	LastDurationMs *int64     `json:"last_duration_ms"`
+	LastResult     *string    `json:"last_result"`
 
 	UpdatedAt time.Time `json:"updated_at"`
 }

backend/internal/service/ops_scheduled_report_service.go

@@ -177,6 +177,10 @@ func (s *OpsScheduledReportService) runOnce() {
 		return
 	}
 
+	reportsTotal := len(reports)
+	reportsDue := 0
+	sentAttempts := 0
+
 	for _, report := range reports {
 		if report == nil || !report.Enabled {
 			continue
@@ -184,14 +188,18 @@ func (s *OpsScheduledReportService) runOnce() {
 		if report.NextRunAt.After(now) {
 			continue
 		}
+		reportsDue++
 
-		if err := s.runReport(ctx, report, now); err != nil {
+		attempts, err := s.runReport(ctx, report, now)
+		if err != nil {
 			s.recordHeartbeatError(runAt, time.Since(startedAt), err)
 			return
 		}
+		sentAttempts += attempts
 	}
 
-	s.recordHeartbeatSuccess(runAt, time.Since(startedAt))
+	result := truncateString(fmt.Sprintf("reports=%d due=%d send_attempts=%d", reportsTotal, reportsDue, sentAttempts), 2048)
+	s.recordHeartbeatSuccess(runAt, time.Since(startedAt), result)
 }
 
 type opsScheduledReport struct {
@@ -297,9 +305,9 @@ func (s *OpsScheduledReportService) listScheduledReports(ctx context.Context, no
 	return out
 }
 
-func (s *OpsScheduledReportService) runReport(ctx context.Context, report *opsScheduledReport, now time.Time) error {
+func (s *OpsScheduledReportService) runReport(ctx context.Context, report *opsScheduledReport, now time.Time) (int, error) {
 	if s == nil || s.opsService == nil || s.emailService == nil || report == nil {
-		return nil
+		return 0, nil
 	}
 	if ctx == nil {
 		ctx = context.Background()
@@ -310,11 +318,11 @@ func (s *OpsScheduledReportService) runReport(ctx context.Context, report *opsSc
 
 	content, err := s.generateReportHTML(ctx, report, now)
 	if err != nil {
-		return err
+		return 0, err
 	}
 	if strings.TrimSpace(content) == "" {
 		// Skip sending when the report decides not to emit content (e.g., digest below min count).
-		return nil
+		return 0, nil
 	}
 
 	recipients := report.Recipients
@@ -325,22 +333,24 @@ func (s *OpsScheduledReportService) runReport(ctx context.Context, report *opsSc
 		}
 	}
 	if len(recipients) == 0 {
-		return nil
+		return 0, nil
 	}
 
 	subject := fmt.Sprintf("[Ops Report] %s", strings.TrimSpace(report.Name))
 
+	attempts := 0
 	for _, to := range recipients {
 		addr := strings.TrimSpace(to)
 		if addr == "" {
 			continue
 		}
+		attempts++
 		if err := s.emailService.SendEmail(ctx, addr, subject, content); err != nil {
 			// Ignore per-recipient failures; continue best-effort.
 			continue
 		}
 	}
-	return nil
+	return attempts, nil
 }
 
 func (s *OpsScheduledReportService) generateReportHTML(ctx context.Context, report *opsScheduledReport, now time.Time) (string, error) {
@@ -650,7 +660,7 @@ func (s *OpsScheduledReportService) setLastRunAt(ctx context.Context, reportType
 	_ = s.redisClient.Set(ctx, key, strconv.FormatInt(t.UTC().Unix(), 10), 14*24*time.Hour).Err()
 }
 
-func (s *OpsScheduledReportService) recordHeartbeatSuccess(runAt time.Time, duration time.Duration) {
+func (s *OpsScheduledReportService) recordHeartbeatSuccess(runAt time.Time, duration time.Duration, result string) {
 	if s == nil || s.opsService == nil || s.opsService.opsRepo == nil {
 		return
 	}
@@ -658,11 +668,17 @@ func (s *OpsScheduledReportService) recordHeartbeatSuccess(runAt time.Time, dura
 	durMs := duration.Milliseconds()
 	ctx, cancel := context.WithTimeout(context.Background(), 2*time.Second)
 	defer cancel()
+	msg := strings.TrimSpace(result)
+	if msg == "" {
+		msg = "ok"
+	}
+	msg = truncateString(msg, 2048)
 	_ = s.opsService.opsRepo.UpsertJobHeartbeat(ctx, &OpsUpsertJobHeartbeatInput{
 		JobName:        opsScheduledReportJobName,
 		LastRunAt:      &runAt,
 		LastSuccessAt:  &now,
 		LastDurationMs: &durMs,
+		LastResult:     &msg,
 	})
 }
 

backend/migrations/039_ops_job_heartbeats_add_last_result.sql

@@ -0,0 +1,6 @@
+-- Add last_result to ops_job_heartbeats for UI job details.
+
+ALTER TABLE IF EXISTS ops_job_heartbeats
+    ADD COLUMN IF NOT EXISTS last_result TEXT;
+
+COMMENT ON COLUMN ops_job_heartbeats.last_result IS 'Last successful run result summary (human readable).';