fix: 修正轮询配置多匹配逻辑，支持同卡匹配多个配置并按 priority 合并 interval

核心变更：
- MatchConfig 改为 MatchConfigs，返回所有匹配配置
- MergedTaskIntervals 按 task type 合并各配置，选取最高优先级（非 nil 且最小 priority 值）
- hasAnyEnabledInterval 过滤所有 interval 均为 NULL 的配置
- calcInitialDelay 重构为纯函数，接收 interval 参数
- 移除 getEnabledTaskTypes 和 getIntervalByTaskType（被 MergedTaskIntervals 替代）
- scheduler.go 新增心跳 key + 顶层 panic recovery + Init 完成守卫
- initializer.go 批量失败日志升级为 Error，逐条检查 Pipeline 命令错误
- 数据迁移：禁用 id=29 的轮询配置（所有 interval 均为 NULL）

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

internal/polling/config_manager.go

@@ -27,6 +27,12 @@ type PollingConfigManager struct {
 	refreshOnce sync.Once
 }
 
+// TaskTypeInterval 某任务类型的轮询间隔信息
+type TaskTypeInterval struct {
+	Interval int // 秒
+	Priority int // 来源配置的 priority
+}
+
 // NewPollingConfigManager 创建配置管理器
 func NewPollingConfigManager(configStore *postgres.PollingConfigStore, redisClient *redis.Client, logger *zap.Logger) *PollingConfigManager {
 	return &PollingConfigManager{
@@ -81,16 +87,79 @@ func (m *PollingConfigManager) Start(ctx context.Context) {
 
 // MatchConfig 按优先级匹配第一个符合条件的轮询配置
 // 配置按 priority ASC 排序（DB 层保证），数字越小优先级越高
+// MatchConfigs()[0] 的前向兼容包装
 func (m *PollingConfigManager) MatchConfig(card *model.IotCard) *model.PollingConfig {
+	configs := m.MatchConfigs(card)
+	if len(configs) == 0 {
+		return nil
+	}
+	return configs[0]
+}
+
+// MatchConfigs 按 priority ASC 遍历所有启用的配置，返回所有满足匹配条件且至少有一个非 NULL interval 的配置
+// 无匹配时返回空切片
+func (m *PollingConfigManager) MatchConfigs(card *model.IotCard) []*model.PollingConfig {
 	m.mu.RLock()
 	defer m.mu.RUnlock()
 
+	var result []*model.PollingConfig
 	for _, cfg := range m.configs {
-		if matchConfigConditions(cfg, card) {
-			return cfg
+		if matchConfigConditions(cfg, card) && hasAnyEnabledInterval(cfg) {
+			result = append(result, cfg)
 		}
 	}
-	return nil
+	return result
+}
+
+// hasAnyEnabledInterval 检查配置是否有至少一个非 NULL 且大于 0 的轮询间隔
+func hasAnyEnabledInterval(cfg *model.PollingConfig) bool {
+	return (cfg.RealnameCheckInterval != nil && *cfg.RealnameCheckInterval > 0) ||
+		(cfg.CarddataCheckInterval != nil && *cfg.CarddataCheckInterval > 0) ||
+		(cfg.PackageCheckInterval != nil && *cfg.PackageCheckInterval > 0) ||
+		(cfg.ProtectCheckInterval != nil && *cfg.ProtectCheckInterval > 0) ||
+		(cfg.CardStatusCheckInterval != nil && *cfg.CardStatusCheckInterval > 0)
+}
+
+// MergedTaskIntervals 对所有匹配配置按 priority 遍历，对每种 task type 选取最高优先级的非 nil interval
+// 返回 map: taskType → TaskTypeInterval{Interval, Priority}
+func (m *PollingConfigManager) MergedTaskIntervals(card *model.IotCard) map[string]TaskTypeInterval {
+	configs := m.MatchConfigs(card)
+	merged := make(map[string]TaskTypeInterval)
+
+	for _, cfg := range configs {
+		priority := int(cfg.Priority)
+		if cfg.RealnameCheckInterval != nil && *cfg.RealnameCheckInterval > 0 {
+			taskType := constants.TaskTypePollingRealname
+			if _, exists := merged[taskType]; !exists {
+				merged[taskType] = TaskTypeInterval{Interval: *cfg.RealnameCheckInterval, Priority: priority}
+			}
+		}
+		if cfg.CarddataCheckInterval != nil && *cfg.CarddataCheckInterval > 0 {
+			taskType := constants.TaskTypePollingCarddata
+			if _, exists := merged[taskType]; !exists {
+				merged[taskType] = TaskTypeInterval{Interval: *cfg.CarddataCheckInterval, Priority: priority}
+			}
+		}
+		if cfg.PackageCheckInterval != nil && *cfg.PackageCheckInterval > 0 {
+			taskType := constants.TaskTypePollingPackage
+			if _, exists := merged[taskType]; !exists {
+				merged[taskType] = TaskTypeInterval{Interval: *cfg.PackageCheckInterval, Priority: priority}
+			}
+		}
+		if cfg.ProtectCheckInterval != nil && *cfg.ProtectCheckInterval > 0 {
+			taskType := constants.TaskTypePollingProtect
+			if _, exists := merged[taskType]; !exists {
+				merged[taskType] = TaskTypeInterval{Interval: *cfg.ProtectCheckInterval, Priority: priority}
+			}
+		}
+		if cfg.CardStatusCheckInterval != nil && *cfg.CardStatusCheckInterval > 0 {
+			taskType := constants.TaskTypePollingCardStatus
+			if _, exists := merged[taskType]; !exists {
+				merged[taskType] = TaskTypeInterval{Interval: *cfg.CardStatusCheckInterval, Priority: priority}
+			}
+		}
+	}
+	return merged
 }
 
 // matchConfigConditions 检查卡是否满足配置的匹配条件

internal/polling/initializer.go

@@ -157,7 +157,8 @@ func (p *PollingInitializer) run(ctx context.Context) {
 		}
 
 		if initErr := p.initBatch(ctx, cards); initErr != nil {
-			p.logger.Warn("批量初始化失败", zap.Error(initErr))
+			p.logger.Error("批量初始化失败，该批次卡未入轮询队列",
+				zap.Uint("batch_start_id", lastID), zap.Error(initErr))
 		}
 
 		lastID = cards[len(cards)-1].ID
@@ -206,16 +207,24 @@ func (p *PollingInitializer) initBatch(ctx context.Context, cards []*model.IotCa
 		if cmdCount == 0 {
 			return
 		}
-		if _, execErr := pipe.Exec(ctx); execErr != nil {
-			p.logger.Warn("Pipeline flush 失败，继续下一批", zap.Error(execErr))
+		cmds, execErr := pipe.Exec(ctx)
+		if execErr != nil {
+			p.logger.Error("Pipeline flush 失败，部分卡可能未入队", zap.Error(execErr))
+		}
+		// 逐条检查，记录具体失败命令（不中断批次）
+		for _, cmd := range cmds {
+			if cmdErr := cmd.Err(); cmdErr != nil && cmdErr != redis.Nil {
+				p.logger.Warn("Pipeline 单条命令失败",
+					zap.String("cmd", cmd.Name()), zap.Error(cmdErr))
+			}
 		}
 		pipe = p.redis.Pipeline()
 		cmdCount = 0
 	}
 
 	for _, card := range cards {
-		cfg := p.configMgr.MatchConfig(card)
-		if cfg == nil {
+		intervals := p.configMgr.MergedTaskIntervals(card)
+		if len(intervals) == 0 {
 			skippedCards++
 			continue
 		}
@@ -224,37 +233,10 @@ func (p *PollingInitializer) initBatch(ctx context.Context, cards []*model.IotCa
 		shardID := int(card.ID) % p.queueMgr.shardCount
 		cardIDStr := fmt.Sprintf("%d", card.ID)
 
-		if cfg.RealnameCheckInterval != nil && *cfg.RealnameCheckInterval > 0 {
-			nextCheck := calculateNextCheckTime(card.LastRealNameCheckAt, *cfg.RealnameCheckInterval, now)
-			pipe.ZAdd(ctx, constants.RedisPollingShardQueueKey(shardID, constants.TaskTypePollingRealname), redis.Z{
-				Score: float64(nextCheck.Unix()), Member: cardIDStr,
-			})
-			cmdCount++
-		}
-		if cfg.CarddataCheckInterval != nil && *cfg.CarddataCheckInterval > 0 {
-			nextCheck := calculateNextCheckTime(card.LastDataCheckAt, *cfg.CarddataCheckInterval, now)
-			pipe.ZAdd(ctx, constants.RedisPollingShardQueueKey(shardID, constants.TaskTypePollingCarddata), redis.Z{
-				Score: float64(nextCheck.Unix()), Member: cardIDStr,
-			})
-			cmdCount++
-		}
-		if cfg.PackageCheckInterval != nil && *cfg.PackageCheckInterval > 0 {
-			nextCheck := calculateNextCheckTime(card.LastDataCheckAt, *cfg.PackageCheckInterval, now)
-			pipe.ZAdd(ctx, constants.RedisPollingShardQueueKey(shardID, constants.TaskTypePollingPackage), redis.Z{
-				Score: float64(nextCheck.Unix()), Member: cardIDStr,
-			})
-			cmdCount++
-		}
-		if cfg.ProtectCheckInterval != nil && *cfg.ProtectCheckInterval > 0 {
-			nextCheck := calculateNextCheckTime(card.LastProtectCheckAt, *cfg.ProtectCheckInterval, now)
-			pipe.ZAdd(ctx, constants.RedisPollingShardQueueKey(shardID, constants.TaskTypePollingProtect), redis.Z{
-				Score: float64(nextCheck.Unix()), Member: cardIDStr,
-			})
-			cmdCount++
-		}
-		if cfg.CardStatusCheckInterval != nil && *cfg.CardStatusCheckInterval > 0 {
-			nextCheck := calculateNextCheckTime(card.LastCardStatusCheckAt, *cfg.CardStatusCheckInterval, now)
-			pipe.ZAdd(ctx, constants.RedisPollingShardQueueKey(shardID, constants.TaskTypePollingCardStatus), redis.Z{
+		for taskType, info := range intervals {
+			lastCheckAt := lastCheckAtByTaskType(card, taskType)
+			nextCheck := calculateNextCheckTime(lastCheckAt, info.Interval, now)
+			pipe.ZAdd(ctx, constants.RedisPollingShardQueueKey(shardID, taskType), redis.Z{
 				Score: float64(nextCheck.Unix()), Member: cardIDStr,
 			})
 			cmdCount++
@@ -294,6 +276,22 @@ func (p *PollingInitializer) initBatch(ctx context.Context, cards []*model.IotCa
 	return nil
 }
 
+// lastCheckAtByTaskType 根据 task type 返回对应的上次检查时间字段
+func lastCheckAtByTaskType(card *model.IotCard, taskType string) *time.Time {
+	switch taskType {
+	case constants.TaskTypePollingRealname:
+		return card.LastRealNameCheckAt
+	case constants.TaskTypePollingCarddata, constants.TaskTypePollingPackage:
+		return card.LastDataCheckAt
+	case constants.TaskTypePollingProtect:
+		return card.LastProtectCheckAt
+	case constants.TaskTypePollingCardStatus:
+		return card.LastCardStatusCheckAt
+	default:
+		return nil
+	}
+}
+
 // calculateNextCheckTime 计算下次检查时间
 func calculateNextCheckTime(lastCheckAt *time.Time, intervalSeconds int, now time.Time) time.Time {
 	if lastCheckAt == nil {

internal/polling/lifecycle_service.go

@@ -9,7 +9,6 @@ import (
 
 	"github.com/break/junhong_cmp_fiber/internal/model"
 	"github.com/break/junhong_cmp_fiber/internal/store/postgres"
-	"github.com/break/junhong_cmp_fiber/pkg/constants"
 )
 
 // PollingLifecycleService 卡生命周期轮询管理服务
@@ -134,68 +133,23 @@ func (s *PollingLifecycleService) shouldEnqueue(ctx context.Context, card *model
 
 // enqueueCard 按匹配配置将卡入队分片队列
 func (s *PollingLifecycleService) enqueueCard(ctx context.Context, card *model.IotCard) {
-	cfg := s.configMgr.MatchConfig(card)
-	if cfg == nil {
+	intervals := s.configMgr.MergedTaskIntervals(card)
+	if len(intervals) == 0 {
 		return
 	}
-	taskTypes := getEnabledTaskTypes(cfg)
-	for _, taskType := range taskTypes {
-		if err := s.queueMgr.Requeue(ctx, card.ID, taskType, calcInitialDelay(card, cfg, taskType)); err != nil {
+	for taskType, info := range intervals {
+		if err := s.queueMgr.Requeue(ctx, card.ID, taskType, calcInitialDelay(info.Interval)); err != nil {
 			s.logger.Warn("卡入队失败",
 				zap.Uint("card_id", card.ID), zap.String("task_type", taskType), zap.Error(err))
 		}
 	}
 }
 
-// getEnabledTaskTypes 返回配置中启用的任务类型列表
-func getEnabledTaskTypes(cfg *model.PollingConfig) []string {
-	var types []string
-	if cfg.RealnameCheckInterval != nil && *cfg.RealnameCheckInterval > 0 {
-		types = append(types, constants.TaskTypePollingRealname)
-	}
-	if cfg.CarddataCheckInterval != nil && *cfg.CarddataCheckInterval > 0 {
-		types = append(types, constants.TaskTypePollingCarddata)
-	}
-	if cfg.PackageCheckInterval != nil && *cfg.PackageCheckInterval > 0 {
-		types = append(types, constants.TaskTypePollingPackage)
-	}
-	if cfg.ProtectCheckInterval != nil && *cfg.ProtectCheckInterval > 0 {
-		types = append(types, constants.TaskTypePollingProtect)
-	}
-	if cfg.CardStatusCheckInterval != nil && *cfg.CardStatusCheckInterval > 0 {
-		types = append(types, constants.TaskTypePollingCardStatus)
-	}
-	return types
-}
-
 // calcInitialDelay 计算新入队卡的初始检查时间（加随机抖动避免惊群）
 // 抖动范围 [0, max(interval/10, 2)) 秒；下界取 2 保证 rand.Intn 有实际随机性
 // （rand.Intn(1) 永远返回 0，对 interval≤10 的配置会导致零抖动）
-func calcInitialDelay(_ *model.IotCard, cfg *model.PollingConfig, taskType string) time.Time {
+func calcInitialDelay(interval int) time.Time {
 	now := time.Now()
-	var interval int
-	switch taskType {
-	case constants.TaskTypePollingRealname:
-		if cfg.RealnameCheckInterval != nil {
-			interval = *cfg.RealnameCheckInterval
-		}
-	case constants.TaskTypePollingCarddata:
-		if cfg.CarddataCheckInterval != nil {
-			interval = *cfg.CarddataCheckInterval
-		}
-	case constants.TaskTypePollingPackage:
-		if cfg.PackageCheckInterval != nil {
-			interval = *cfg.PackageCheckInterval
-		}
-	case constants.TaskTypePollingProtect:
-		if cfg.ProtectCheckInterval != nil {
-			interval = *cfg.ProtectCheckInterval
-		}
-	case constants.TaskTypePollingCardStatus:
-		if cfg.CardStatusCheckInterval != nil {
-			interval = *cfg.CardStatusCheckInterval
-		}
-	}
 	if interval <= 0 {
 		return now
 	}

internal/polling/scheduler.go

@@ -28,6 +28,7 @@ type Scheduler struct {
 
 	packageActivationHandler *PackageActivationHandler
 	dataResetHandler         *DataResetHandler
+	initializer              *PollingInitializer // 可选，nil 表示不守卫 Init 完成
 
 	stopChan chan struct{}
 	wg       sync.WaitGroup
@@ -103,9 +104,29 @@ func (s *Scheduler) SetStopResumeCallback(callback packagepkg.StopResumeCallback
 	}
 }
 
+// SetInitializer 注入初始化器（可选，在 Start 前调用）
+// Init 未完成时调度器跳过分片出队，消除启动期无效轮询噪音
+func (s *Scheduler) SetInitializer(init *PollingInitializer) {
+	s.initializer = init
+}
+
+// writeHeartbeat 写入调度器心跳，表明调度器存活
+func (s *Scheduler) writeHeartbeat(ctx context.Context) {
+	if err := s.redis.Set(ctx, constants.RedisPollingSchedulerHeartbeatKey(),
+		time.Now().Unix(), 2*s.cfg.ScheduleInterval).Err(); err != nil {
+		s.logger.Warn("写入调度器心跳失败", zap.Error(err))
+	}
+}
+
 // scheduleLoop 调度循环
 func (s *Scheduler) scheduleLoop(ctx context.Context) {
 	defer s.wg.Done()
+	defer func() {
+		if r := recover(); r != nil {
+			s.logger.Error("调度主循环发生 panic，调度已停止，需重启 Worker",
+				zap.Any("panic", r), zap.Stack("stack"))
+		}
+	}()
 
 	ticker := time.NewTicker(s.cfg.ScheduleInterval)
 	activationTicker := time.NewTicker(10 * time.Second)
@@ -123,6 +144,7 @@ func (s *Scheduler) scheduleLoop(ctx context.Context) {
 			s.logger.Info("调度循环收到 ctx 取消信号")
 			return
 		case <-ticker.C:
+			s.writeHeartbeat(ctx)
 			s.processShardSchedule(ctx)
 		case <-activationTicker.C:
 			s.processActivationTasks(ctx)
@@ -133,10 +155,16 @@ func (s *Scheduler) scheduleLoop(ctx context.Context) {
 // processShardSchedule 处理手动队列和分片定时队列（每 1 秒触发）
 // 使用 90% 的 tick 间隔作为超时，确保单分片 Redis 挂起时不阻塞下一个 tick
 func (s *Scheduler) processShardSchedule(ctx context.Context) {
+	// 手动队列不受 Init 影响（ConfigManager 已就绪即可）
 	for _, taskType := range allTaskTypes {
 		s.processManualQueue(ctx, taskType, s.cfg.MaxManualBatchSize)
 	}
 
+	// Init 未完成时跳过分片扫描，避免空轮询噪音
+	if s.initializer != nil && !s.initializer.IsCompleted() {
+		return
+	}
+
 	if s.queueMgr == nil {
 		return
 	}

internal/task/polling_base.go

@@ -102,20 +102,18 @@ func (b *PollingBase) requeueCard(ctx context.Context, cardID uint, taskType str
 		return b.queueMgr.Requeue(ctx, cardID, taskType, time.Now().Add(30*time.Second))
 	}
 
-	cfg := b.configMgr.MatchConfig(card)
-	if cfg == nil {
-		// 兜底：配置未加载（DB 暂时不可达）时延迟 30 秒重入队，防止卡从轮询永久消失
+	intervals := b.configMgr.MergedTaskIntervals(card)
+	if len(intervals) == 0 {
+		// 兜底：配置未加载时延迟 30 秒重入队，防止卡从轮询永久消失
 		b.logger.Warn("无匹配轮询配置，30秒后重入队（配置可能未加载）",
 			zap.Uint("card_id", cardID), zap.String("task_type", taskType))
 		return b.queueMgr.Requeue(ctx, cardID, taskType, time.Now().Add(30*time.Second))
 	}
-
-	interval := getIntervalByTaskType(cfg, taskType)
-	if interval <= 0 {
+	info, ok := intervals[taskType]
+	if !ok || info.Interval <= 0 {
 		b.logger.Debug("轮询间隔为 NULL，不入队", zap.Uint("card_id", cardID), zap.String("task_type", taskType))
 		return nil
 	}
-
-	nextCheckAt := time.Now().Add(time.Duration(interval) * time.Second)
+	nextCheckAt := time.Now().Add(time.Duration(info.Interval) * time.Second)
 	return b.queueMgr.Requeue(ctx, cardID, taskType, nextCheckAt)
 }

internal/task/polling_utils.go

@@ -7,38 +7,8 @@ import (
 	"github.com/bytedance/sonic"
 	"go.uber.org/zap"
 	"gorm.io/gorm"
-
-	"github.com/break/junhong_cmp_fiber/internal/model"
-	"github.com/break/junhong_cmp_fiber/pkg/constants"
 )
 
-// getIntervalByTaskType 从配置中提取指定任务类型的轮询间隔（秒）
-func getIntervalByTaskType(cfg *model.PollingConfig, taskType string) int {
-	switch taskType {
-	case constants.TaskTypePollingRealname:
-		if cfg.RealnameCheckInterval != nil && *cfg.RealnameCheckInterval > 0 {
-			return *cfg.RealnameCheckInterval
-		}
-	case constants.TaskTypePollingCarddata:
-		if cfg.CarddataCheckInterval != nil && *cfg.CarddataCheckInterval > 0 {
-			return *cfg.CarddataCheckInterval
-		}
-	case constants.TaskTypePollingPackage:
-		if cfg.PackageCheckInterval != nil && *cfg.PackageCheckInterval > 0 {
-			return *cfg.PackageCheckInterval
-		}
-	case constants.TaskTypePollingProtect:
-		if cfg.ProtectCheckInterval != nil && *cfg.ProtectCheckInterval > 0 {
-			return *cfg.ProtectCheckInterval
-		}
-	case constants.TaskTypePollingCardStatus:
-		if cfg.CardStatusCheckInterval != nil && *cfg.CardStatusCheckInterval > 0 {
-			return *cfg.CardStatusCheckInterval
-		}
-	}
-	return 0
-}
-
 // shortTaskType 从完整任务类型中提取简短名称（如 polling:carddata → carddata）
 func shortTaskType(fullTaskType string) string {
 	for i := len(fullTaskType) - 1; i >= 0; i-- {