feat: improve generation analytics and maintenance

README.md

@@ -134,7 +134,8 @@ npm run build
 | GET | `/api/generations/jobs/{job_id}` | 查询生成任务事件流 |
 | GET | `/api/generations/{story_id}/jobs` | 查询故事生成与重试历史 |
 | GET | `/api/generations/{story_id}/provider-stats` | 查询 Provider 调用聚合指标 |
-| GET | `/api/generations/provider-analytics` | 查询当前用户跨故事 Provider 运营摘要 |
+| GET | `/api/generations/ops-summary` | 查询最近任务运行概览、失败摘要和超时阈值 |
+| GET | `/api/generations/provider-analytics` | 查询当前用户跨故事 Provider 运营摘要，支持 `days` / `capability` 筛选 |
 | GET | `/api/audio/{story_id}/status` | 查询音频缓存状态，不触发生成 |
 | DELETE | `/api/audio/{story_id}/cache` | 清理故事音频缓存 |
 | GET | `/api/stories` | 故事列表 |
@@ -164,4 +165,4 @@ npm run build
 
 ## 当前取舍
 
-仓库只保留一个 Docker Compose 入口：`docker-compose.yml`。生产部署、HA 演练、旧 Claude 原型和历史归档已从主仓库移除，避免干扰当前求职演示主线。
+仓库只保留一个 Docker Compose 入口：`docker-compose.yml`。生产部署、HA 演练、旧 Claude 原型和历史归档已从主仓库移除，避免干扰当前求职演示主线。音频缓存默认按 `STORY_AUDIO_CACHE_TTL_DAYS=30` 做后台清理，Celery beat 会每日执行一次 prune；生成任务默认按 `GENERATION_JOB_STALE_MINUTES=60` 判定卡住，后台会定时自动收敛为失败态，避免故事长期显示“永远在跑”。

admin-frontend/src/views/MyStories.vue

@@ -1,5 +1,5 @@
 <script setup lang="ts">
-import { computed, ref, onMounted } from 'vue'
+import { computed, ref, onMounted, watch } from 'vue'
 import { useRouter } from 'vue-router'
 import { api } from '../api/client'
 import BaseButton from '../components/ui/BaseButton.vue'
@@ -45,6 +45,8 @@ interface GenerationProviderStat {
 }
 
 interface GenerationProviderAnalytics {
+  window_days: number | null
+  capability: string | null
   total_calls: number
   successful_calls: number
   failed_calls: number
@@ -53,14 +55,43 @@ interface GenerationProviderAnalytics {
   job_count: number
   story_count: number
   by_provider: GenerationProviderStat[]
+  failure_reasons: Array<{
+    reason: string
+    count: number
+  }>
+}
+
+interface GenerationRecentFailure {
+  job_id: string
+  story_id: number | null
+  story_title: string | null
+  output_mode: string
+  current_step: string
+  error_message: string | null
+  failure_label: string
+  updated_at: string
+}
+
+interface GenerationOpsSummary {
+  window_hours: number
+  stale_threshold_minutes: number
+  active_jobs: number
+  stale_running_jobs: number
+  failed_jobs: number
+  degraded_jobs: number
+  asset_retry_jobs: number
+  recent_failures: GenerationRecentFailure[]
 }
 
 const router = useRouter()
 const stories = ref<StoryItem[]>([])
 const providerAnalytics = ref<GenerationProviderAnalytics | null>(null)
+const opsSummary = ref<GenerationOpsSummary | null>(null)
 const loading = ref(true)
 const error = ref('')
 const showCreateModal = ref(false)
+const selectedWindow = ref<'7' | '30' | 'all'>('30')
+const selectedCapability = ref<'all' | 'text' | 'image' | 'tts' | 'storybook'>('all')
 const readableCount = computed(() =>
   stories.value.filter((story) => isReadableGenerationStatus(story.generation_status)).length,
 )
@@ -74,15 +105,30 @@ const providerSuccessRate = computed(() => {
   )
 })
 const topProvider = computed(() => providerAnalytics.value?.by_provider[0] ?? null)
+const topFailureReason = computed(() => providerAnalytics.value?.failure_reasons[0] ?? null)
+
+function buildProviderAnalyticsPath() {
+  const params = new URLSearchParams()
+  if (selectedWindow.value !== 'all') {
+    params.set('days', selectedWindow.value)
+  }
+  if (selectedCapability.value !== 'all') {
+    params.set('capability', selectedCapability.value)
+  }
+  const query = params.toString()
+  return `/api/generations/provider-analytics${query ? `?${query}` : ''}`
+}
 
 async function fetchStories() {
   try {
-    const [storyList, analytics] = await Promise.all([
+    const [storyList, analytics, ops] = await Promise.all([
       api.get<StoryItem[]>('/api/stories'),
-      api.get<GenerationProviderAnalytics>('/api/generations/provider-analytics'),
+      api.get<GenerationProviderAnalytics>(buildProviderAnalyticsPath()),
+      api.get<GenerationOpsSummary>('/api/generations/ops-summary'),
     ])
     stories.value = storyList
     providerAnalytics.value = analytics
+    opsSummary.value = ops
   } catch (e) {
     error.value = e instanceof Error ? e.message : '加载失败'
   } finally {
@@ -123,6 +169,27 @@ function formatCost(value?: number | null) {
   return typeof value === 'number' ? `$${value.toFixed(4)}` : '$0.0000'
 }
 
+function formatOutputMode(value: string) {
+  switch (value) {
+    case 'storybook':
+      return '绘本'
+    case 'asset_retry':
+      return '资源重试'
+    case 'asset_generation':
+      return '资源生成'
+    default:
+      return '故事'
+  }
+}
+
+function setWindow(value: '7' | '30' | 'all') {
+  selectedWindow.value = value
+}
+
+function setCapability(value: 'all' | 'text' | 'image' | 'tts' | 'storybook') {
+  selectedCapability.value = value
+}
+
 onMounted(() => {
   fetchStories()
   if (router.currentRoute.value.query.openCreate) {
@@ -130,6 +197,10 @@ onMounted(() => {
     router.replace({ query: { ...router.currentRoute.value.query, openCreate: undefined } })
   }
 })
+
+watch([selectedWindow, selectedCapability], () => {
+  void fetchStories()
+})
 </script>
 
 <template>
@@ -213,6 +284,18 @@ onMounted(() => {
             <p class="mt-2 text-sm leading-6 text-gray-500">
               生成、资源补全和失败恢复留下的供应商调用轨迹。
             </p>
+            <div class="mt-4 flex flex-wrap gap-2">
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedWindow === '7' ? 'border-gray-900 bg-gray-900 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setWindow('7')">最近 7 天</button>
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedWindow === '30' ? 'border-gray-900 bg-gray-900 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setWindow('30')">最近 30 天</button>
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedWindow === 'all' ? 'border-gray-900 bg-gray-900 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setWindow('all')">全部</button>
+            </div>
+            <div class="mt-3 flex flex-wrap gap-2">
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedCapability === 'all' ? 'border-emerald-600 bg-emerald-600 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setCapability('all')">全部能力</button>
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedCapability === 'text' ? 'border-emerald-600 bg-emerald-600 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setCapability('text')">文本</button>
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedCapability === 'image' ? 'border-emerald-600 bg-emerald-600 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setCapability('image')">图片</button>
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedCapability === 'tts' ? 'border-emerald-600 bg-emerald-600 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setCapability('tts')">语音</button>
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedCapability === 'storybook' ? 'border-emerald-600 bg-emerald-600 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setCapability('storybook')">绘本</button>
+            </div>
           </div>
           <div class="grid grid-cols-2 gap-3 sm:grid-cols-4 lg:min-w-[520px]">
             <div class="rounded-lg border border-gray-100 bg-gray-50 px-3 py-3">
@@ -236,6 +319,70 @@ onMounted(() => {
         <p v-if="topProvider" class="mt-4 text-sm text-gray-500">
           当前样本中最前面的能力组合是 {{ topProvider.capability }} / {{ topProvider.adapter }}，成功 {{ topProvider.success_count }} 次，失败 {{ topProvider.failure_count }} 次。
         </p>
+        <p v-if="topFailureReason" class="mt-2 text-sm text-rose-600">
+          最常见失败原因：{{ topFailureReason.reason }}（{{ topFailureReason.count }} 次）
+        </p>
+      </BaseCard>
+
+      <BaseCard
+        v-if="opsSummary"
+        class="mb-8"
+        padding="lg"
+      >
+        <div class="flex flex-col gap-5 lg:flex-row lg:items-center lg:justify-between">
+          <div>
+            <h2 class="text-xl font-bold text-gray-800">任务运行概览</h2>
+            <p class="mt-2 text-sm leading-6 text-gray-500">
+              最近 {{ opsSummary.window_hours }} 小时的任务健康度，运行超过
+              {{ opsSummary.stale_threshold_minutes }} 分钟会被视为卡住。
+            </p>
+          </div>
+          <div class="grid grid-cols-2 gap-3 sm:grid-cols-4 lg:min-w-[520px]">
+            <div class="rounded-lg border border-gray-100 bg-gray-50 px-3 py-3">
+              <div class="text-xs text-gray-500">运行中</div>
+              <div class="mt-1 text-lg font-semibold text-gray-800">{{ opsSummary.active_jobs }}</div>
+            </div>
+            <div class="rounded-lg border border-gray-100 bg-gray-50 px-3 py-3">
+              <div class="text-xs text-gray-500">超时待收敛</div>
+              <div class="mt-1 text-lg font-semibold" :class="opsSummary.stale_running_jobs ? 'text-amber-600' : 'text-gray-800'">
+                {{ opsSummary.stale_running_jobs }}
+              </div>
+            </div>
+            <div class="rounded-lg border border-gray-100 bg-gray-50 px-3 py-3">
+              <div class="text-xs text-gray-500">最近失败</div>
+              <div class="mt-1 text-lg font-semibold" :class="opsSummary.failed_jobs ? 'text-rose-600' : 'text-gray-800'">
+                {{ opsSummary.failed_jobs }}
+              </div>
+            </div>
+            <div class="rounded-lg border border-gray-100 bg-gray-50 px-3 py-3">
+              <div class="text-xs text-gray-500">资源任务</div>
+              <div class="mt-1 text-lg font-semibold text-gray-800">{{ opsSummary.asset_retry_jobs }}</div>
+            </div>
+          </div>
+        </div>
+        <p v-if="opsSummary.degraded_jobs" class="mt-4 text-sm text-amber-600">
+          最近 {{ opsSummary.window_hours }} 小时有 {{ opsSummary.degraded_jobs }} 个任务以降级完成收尾。
+        </p>
+        <div v-if="opsSummary.recent_failures.length" class="mt-4 space-y-3">
+          <div
+            v-for="failure in opsSummary.recent_failures"
+            :key="failure.job_id"
+            class="rounded-lg border border-rose-100 bg-rose-50 px-4 py-3"
+          >
+            <div class="flex flex-wrap items-center justify-between gap-3">
+              <div class="text-sm font-semibold text-gray-800">
+                {{ failure.story_title || `${formatOutputMode(failure.output_mode)}任务` }}
+              </div>
+              <div class="text-xs text-gray-500">{{ formatDate(failure.updated_at) }}</div>
+            </div>
+            <div class="mt-1 text-xs text-rose-600">
+              {{ failure.failure_label }} · {{ failure.error_message || '请打开任务轨迹查看原因' }}
+            </div>
+          </div>
+        </div>
+        <p v-else class="mt-4 text-sm text-emerald-600">
+          最近 {{ opsSummary.window_hours }} 小时没有失败任务，当前链路比较稳定。
+        </p>
       </BaseCard>
 
       <!-- 故事网格 -->

backend/.env.example

@@ -48,6 +48,8 @@ STORYBOOK_PROVIDERS=["storybook_primary"]
 TEXT_MODEL=gemini-2.0-flash
 IMAGE_MODEL=nano-banana
 IMAGE_RESOLUTION=1K
+STORY_AUDIO_CACHE_TTL_DAYS=30
+GENERATION_JOB_STALE_MINUTES=60
 # TTS_MODEL=speech-2.6-turbo (MiniMax) / zh-CN-XiaoxiaoNeural (Edge)
 
 # [API 密钥池]

backend/app/api/stories.py

@@ -4,7 +4,7 @@ import json
 import uuid
 from typing import AsyncGenerator
 
-from fastapi import APIRouter, Depends, Response
+from fastapi import APIRouter, Depends, Query, Response
 from sqlalchemy.ext.asyncio import AsyncSession
 from sse_starlette.sse import EventSourceResponse
 
@@ -19,6 +19,7 @@ from app.schemas.story_schemas import (
     GenerateRequest,
     GenerationJobDetailResponse,
     GenerationJobSummaryResponse,
+    GenerationOpsSummaryResponse,
     GenerationProviderAnalyticsResponse,
     GenerationProviderStatsResponse,
     GenerationRequest,
@@ -36,6 +37,7 @@ from app.services import story_service
 from app.services.generation_jobs import (
     get_generation_job_detail,
     get_story_provider_stats,
+    get_user_generation_ops_summary,
     get_user_provider_analytics,
     list_story_generation_jobs,
 )
@@ -86,16 +88,36 @@ async def get_generation_job(
     return await get_generation_job_detail(db, job_id=job_id, user_id=user.id)
 
 
+@router.get(
+    "/generations/ops-summary",
+    response_model=GenerationOpsSummaryResponse,
+)
+async def get_generation_ops_summary(
+    hours: int = Query(default=24, ge=1, le=168),
+    user: User = Depends(require_user),
+    db: AsyncSession = Depends(get_db),
+):
+    """Get a compact recent operations summary for generation workflows."""
+    return await get_user_generation_ops_summary(db, user_id=user.id, hours=hours)
+
+
 @router.get(
     "/generations/provider-analytics",
     response_model=GenerationProviderAnalyticsResponse,
 )
 async def get_generation_provider_analytics(
+    days: int | None = Query(default=None, ge=1, le=365),
+    capability: str | None = Query(default=None),
     user: User = Depends(require_user),
     db: AsyncSession = Depends(get_db),
 ):
     """Get provider call stats aggregated across the user's generation history."""
-    return await get_user_provider_analytics(db, user_id=user.id)
+    return await get_user_provider_analytics(
+        db,
+        user_id=user.id,
+        days=days,
+        capability=capability,
+    )
 
 
 @router.get(
@@ -117,11 +139,19 @@ async def list_generation_jobs(
 )
 async def get_generation_provider_stats(
     story_id: int,
+    days: int | None = Query(default=None, ge=1, le=365),
+    capability: str | None = Query(default=None),
     user: User = Depends(require_user),
     db: AsyncSession = Depends(get_db),
 ):
     """Get provider call stats aggregated from generation job events."""
-    return await get_story_provider_stats(db, story_id=story_id, user_id=user.id)
+    return await get_story_provider_stats(
+        db,
+        story_id=story_id,
+        user_id=user.id,
+        days=days,
+        capability=capability,
+    )
 
 
 @router.get("/generations/{story_id}", response_model=StoryDetailResponse)

backend/app/core/celery_app.py

@@ -49,6 +49,14 @@ celery_app.conf.update(
             "task": "app.tasks.memory.prune_memories_task",
             "schedule": crontab(minute="0", hour="3"),  # Daily at 03:00
         },
+        "prune_story_audio_cache": {
+            "task": "app.tasks.audio_cache.prune_story_audio_cache_task",
+            "schedule": crontab(minute="30", hour="3"),  # Daily at 03:30
+        },
+        "prune_stale_generation_jobs": {
+            "task": "app.tasks.generation_maintenance.prune_stale_generation_jobs_task",
+            "schedule": crontab(minute="*/30"),
+        },
     },
 )
 

backend/app/core/config.py

@@ -62,12 +62,20 @@ class Settings(BaseSettings):
         False,
         description="Enable local deterministic demo providers for portfolio demos",
     )
-    story_audio_cache_dir: str = Field(
-        "storage/audio",
-        description="Directory for cached story audio files",
-    )
-
-    # Celery (Redis)
+    story_audio_cache_dir: str = Field(
+        "storage/audio",
+        description="Directory for cached story audio files",
+    )
+    story_audio_cache_ttl_days: int = Field(
+        30,
+        description="TTL in days before cached story audio is pruned",
+    )
+    generation_job_stale_minutes: int = Field(
+        60,
+        description="Minutes before a running generation job is considered stale",
+    )
+
+    # Celery (Redis)
     celery_broker_url: str = Field("redis://localhost:6379/0")
     celery_result_backend: str = Field("redis://localhost:6379/0")
 

backend/app/schemas/story_schemas.py

@@ -220,21 +220,33 @@ class GenerationProviderStatResponse(BaseModel):
     estimated_cost_usd: float = 0.0
 
 
+class GenerationProviderFailureReasonResponse(BaseModel):
+    """Aggregated failed provider call reason."""
+
+    reason: str
+    count: int
+
+
 class GenerationProviderStatsResponse(BaseModel):
     """Provider call stats aggregated from generation job events."""
 
     story_id: int
+    window_days: int | None = None
+    capability: str | None = None
     total_calls: int
     successful_calls: int
     failed_calls: int
     avg_latency_ms: float | None = None
     estimated_cost_usd: float = 0.0
     by_provider: list[GenerationProviderStatResponse] = Field(default_factory=list)
+    failure_reasons: list[GenerationProviderFailureReasonResponse] = Field(default_factory=list)
 
 
 class GenerationProviderAnalyticsResponse(BaseModel):
     """Provider call stats aggregated across one user's generation history."""
 
+    window_days: int | None = None
+    capability: str | None = None
     total_calls: int
     successful_calls: int
     failed_calls: int
@@ -243,6 +255,33 @@ class GenerationProviderAnalyticsResponse(BaseModel):
     job_count: int
     story_count: int
     by_provider: list[GenerationProviderStatResponse] = Field(default_factory=list)
+    failure_reasons: list[GenerationProviderFailureReasonResponse] = Field(default_factory=list)
+
+
+class GenerationRecentFailureResponse(BaseModel):
+    """One recent failed generation task for operations summary."""
+
+    job_id: str
+    story_id: int | None = None
+    story_title: str | None = None
+    output_mode: str
+    current_step: str
+    error_message: str | None = None
+    failure_label: str
+    updated_at: datetime
+
+
+class GenerationOpsSummaryResponse(BaseModel):
+    """Recent generation health summary for one user."""
+
+    window_hours: int
+    stale_threshold_minutes: int
+    active_jobs: int
+    stale_running_jobs: int
+    failed_jobs: int
+    degraded_jobs: int
+    asset_retry_jobs: int
+    recent_failures: list[GenerationRecentFailureResponse] = Field(default_factory=list)
 
 
 class AchievementItem(BaseModel):

backend/app/services/generation_jobs.py

@@ -2,14 +2,19 @@
 
 from __future__ import annotations
 
+from datetime import datetime, timedelta, timezone
 from typing import Any
 
 from fastapi import HTTPException
-from sqlalchemy import desc, distinct, func, select
+from sqlalchemy import desc, select
 from sqlalchemy.ext.asyncio import AsyncSession
 
+from app.core.config import settings
+from app.core.logging import get_logger
 from app.db.models import GenerationJob, GenerationJobEvent, Story
 
+logger = get_logger(__name__)
+
 
 def _story_snapshot(story: Story | None) -> dict[str, Any]:
     if story is None:
@@ -68,6 +73,7 @@ def _job_progress(job: GenerationJob) -> dict[str, Any]:
         "asset_generation_completed": (100, "资源已完成"),
         "asset_retry_completed": (100, "资源重试完成"),
         "generation_completed": (100, "生成完成"),
+        "generation_stale_failed": (100, "任务超时已收敛"),
     }
     percent, label = progress_map.get(job.current_step, (10, "生成处理中"))
     return {
@@ -77,6 +83,27 @@ def _job_progress(job: GenerationJob) -> dict[str, Any]:
     }
 
 
+def _normalize_datetime(value: datetime) -> datetime:
+    if value.tzinfo is None:
+        return value.replace(tzinfo=timezone.utc)
+    return value.astimezone(timezone.utc)
+
+
+def _is_stale_job(job: GenerationJob, *, stale_after_minutes: int) -> bool:
+    cutoff = datetime.now(timezone.utc) - timedelta(minutes=stale_after_minutes)
+    return job.status == "running" and _normalize_datetime(job.updated_at) <= cutoff
+
+
+def _failure_label(job: GenerationJob) -> str:
+    if job.current_step == "generation_stale_failed":
+        return "任务超时"
+    if job.output_mode == "asset_retry":
+        return "资源重试失败"
+    if job.output_mode == "asset_generation":
+        return "资源生成失败"
+    return "生成失败"
+
+
 async def create_generation_job(
     db: AsyncSession,
     *,
@@ -266,16 +293,64 @@ async def list_story_generation_jobs(
     return [generation_job_to_summary(job) for job in jobs]
 
 
+async def get_active_story_generation_job(
+    db: AsyncSession,
+    *,
+    story_id: int,
+    user_id: str,
+) -> GenerationJob | None:
+    """Return the most recent running job for a story, if any."""
+
+    result = await db.execute(
+        select(GenerationJob)
+        .where(
+            GenerationJob.story_id == story_id,
+            GenerationJob.user_id == user_id,
+            GenerationJob.status == "running",
+        )
+        .order_by(desc(GenerationJob.updated_at), desc(GenerationJob.id))
+        .limit(1)
+    )
+    return result.scalar_one_or_none()
+
+
+async def ensure_no_active_story_generation_job(
+    db: AsyncSession,
+    *,
+    story_id: int,
+    user_id: str,
+) -> None:
+    """Prevent duplicate asset work while a story already has a running job."""
+
+    active_job = await get_active_story_generation_job(db, story_id=story_id, user_id=user_id)
+    if active_job is None:
+        return
+
+    progress = _job_progress(active_job)
+    raise HTTPException(
+        status_code=409,
+        detail=(
+            f"当前故事已有运行中的任务（{progress['progress_label']}），"
+            "请等待当前任务完成后再试。"
+        ),
+    )
+
+
 def _as_float(value: Any) -> float | None:
     if isinstance(value, int | float):
         return float(value)
     return None
 
 
-def _aggregate_provider_events(events: list[GenerationJobEvent]) -> dict[str, Any]:
+def _aggregate_provider_events(
+    events: list[GenerationJobEvent],
+    *,
+    capability: str | None = None,
+) -> dict[str, Any]:
     """Aggregate provider telemetry from provider call events."""
 
     by_key: dict[tuple[str, str], dict[str, Any]] = {}
+    failure_reasons: dict[str, int] = {}
     total_latency = 0.0
     latency_count = 0
     total_cost = 0.0
@@ -284,13 +359,16 @@ def _aggregate_provider_events(events: list[GenerationJobEvent]) -> dict[str, An
 
     for event in events:
         metadata = event.event_metadata or {}
-        capability = str(metadata.get("capability") or "unknown")
+        event_capability = str(metadata.get("capability") or "unknown")
+        if capability is not None and event_capability != capability:
+            continue
+
         adapter = str(metadata.get("adapter") or "unknown")
-        key = (capability, adapter)
+        key = (event_capability, adapter)
         bucket = by_key.setdefault(
             key,
             {
-                "capability": capability,
+                "capability": event_capability,
                 "adapter": adapter,
                 "call_count": 0,
                 "success_count": 0,
@@ -318,6 +396,8 @@ def _aggregate_provider_events(events: list[GenerationJobEvent]) -> dict[str, An
         else:
             bucket["failure_count"] += 1
             failed_calls += 1
+            reason = str(metadata.get("error") or "unknown_error")
+            failure_reasons[reason] = failure_reasons.get(reason, 0) + 1
 
     by_provider = []
     for bucket in by_key.values():
@@ -349,67 +429,243 @@ def _aggregate_provider_events(events: list[GenerationJobEvent]) -> dict[str, An
         "avg_latency_ms": round(total_latency / latency_count, 2) if latency_count else None,
         "estimated_cost_usd": round(total_cost, 6),
         "by_provider": by_provider,
+        "failure_reasons": [
+            {"reason": reason, "count": count}
+            for reason, count in sorted(
+                failure_reasons.items(),
+                key=lambda item: (-item[1], item[0]),
+            )
+        ],
     }
 
 
+def _provider_events_query(
+    *,
+    user_id: str,
+    story_id: int | None = None,
+    days: int | None = None,
+):
+    query = (
+        select(GenerationJobEvent)
+        .join(GenerationJob, GenerationJobEvent.job_id == GenerationJob.id)
+        .where(
+            GenerationJob.user_id == user_id,
+            GenerationJobEvent.event_type.in_(
+                ["provider_call_succeeded", "provider_call_failed"]
+            ),
+        )
+    )
+
+    if story_id is not None:
+        query = query.where(GenerationJob.story_id == story_id)
+
+    if days is not None:
+        cutoff = datetime.now(timezone.utc) - timedelta(days=days)
+        query = query.where(GenerationJobEvent.created_at >= cutoff)
+
+    return query.order_by(GenerationJobEvent.id)
+
+
 async def get_story_provider_stats(
     db: AsyncSession,
     *,
     story_id: int,
     user_id: str,
+    days: int | None = None,
+    capability: str | None = None,
 ) -> dict[str, Any]:
     """Aggregate provider call telemetry from all user-owned jobs for one story."""
 
     events = (
         await db.execute(
-            select(GenerationJobEvent)
-            .join(GenerationJob, GenerationJobEvent.job_id == GenerationJob.id)
-            .where(
-                GenerationJob.story_id == story_id,
-                GenerationJob.user_id == user_id,
-                GenerationJobEvent.event_type.in_(
-                    ["provider_call_succeeded", "provider_call_failed"]
-                ),
+            _provider_events_query(
+                user_id=user_id,
+                story_id=story_id,
+                days=days,
             )
-            .order_by(GenerationJobEvent.id)
         )
     ).scalars().all()
 
-    return {"story_id": story_id, **_aggregate_provider_events(events)}
+    return {
+        "story_id": story_id,
+        "window_days": days,
+        "capability": capability,
+        **_aggregate_provider_events(events, capability=capability),
+    }
 
 
 async def get_user_provider_analytics(
     db: AsyncSession,
     *,
     user_id: str,
+    days: int | None = None,
+    capability: str | None = None,
 ) -> dict[str, Any]:
     """Aggregate provider telemetry across all stories owned by one user."""
 
     events = (
         await db.execute(
-            select(GenerationJobEvent)
-            .join(GenerationJob, GenerationJobEvent.job_id == GenerationJob.id)
+            _provider_events_query(
+                user_id=user_id,
+                days=days,
+            )
+        )
+    ).scalars().all()
+    filtered_event_job_ids = {
+        event.job_id
+        for event in events
+        if capability is None
+        or str((event.event_metadata or {}).get("capability") or "unknown") == capability
+    }
+    filtered_story_ids = {
+        event.story_id
+        for event in events
+        if event.story_id is not None
+        and (
+            capability is None
+            or str((event.event_metadata or {}).get("capability") or "unknown") == capability
+        )
+    }
+
+    return {
+        "window_days": days,
+        "capability": capability,
+        **_aggregate_provider_events(events, capability=capability),
+        "job_count": len(filtered_event_job_ids),
+        "story_count": len(filtered_story_ids),
+    }
+
+
+async def get_user_generation_ops_summary(
+    db: AsyncSession,
+    *,
+    user_id: str,
+    hours: int = 24,
+    recent_failure_limit: int = 5,
+) -> dict[str, Any]:
+    """Summarize recent generation health for one user."""
+
+    stale_after_minutes = settings.generation_job_stale_minutes
+    recent_cutoff = datetime.now(timezone.utc) - timedelta(hours=hours)
+
+    running_jobs = (
+        await db.execute(
+            select(GenerationJob)
             .where(
                 GenerationJob.user_id == user_id,
-                GenerationJobEvent.event_type.in_(
-                    ["provider_call_succeeded", "provider_call_failed"]
-                ),
+                GenerationJob.status == "running",
             )
-            .order_by(GenerationJobEvent.id)
+            .order_by(desc(GenerationJob.updated_at), desc(GenerationJob.id))
         )
     ).scalars().all()
 
-    job_count, story_count = (
+    recent_jobs = (
         await db.execute(
-            select(
-                func.count(GenerationJob.id),
-                func.count(distinct(GenerationJob.story_id)),
-            ).where(GenerationJob.user_id == user_id)
+            select(GenerationJob, Story.title)
+            .outerjoin(Story, Story.id == GenerationJob.story_id)
+            .where(
+                GenerationJob.user_id == user_id,
+                GenerationJob.updated_at >= recent_cutoff,
+            )
+            .order_by(desc(GenerationJob.updated_at), desc(GenerationJob.id))
         )
-    ).one()
+    ).all()
+
+    recent_failures: list[dict[str, Any]] = []
+    failed_jobs = 0
+    degraded_jobs = 0
+    asset_retry_jobs = 0
+
+    for job, story_title in recent_jobs:
+        if job.status == "failed":
+            failed_jobs += 1
+            if len(recent_failures) < recent_failure_limit:
+                recent_failures.append(
+                    {
+                        "job_id": job.id,
+                        "story_id": job.story_id,
+                        "story_title": story_title,
+                        "output_mode": job.output_mode,
+                        "current_step": job.current_step,
+                        "error_message": job.error_message,
+                        "failure_label": _failure_label(job),
+                        "updated_at": job.updated_at,
+                    }
+                )
+        elif job.status == "degraded_completed":
+            degraded_jobs += 1
+
+        if job.output_mode in {"asset_retry", "asset_generation"}:
+            asset_retry_jobs += 1
 
     return {
-        **_aggregate_provider_events(events),
-        "job_count": job_count,
-        "story_count": story_count,
+        "window_hours": hours,
+        "stale_threshold_minutes": stale_after_minutes,
+        "active_jobs": len(running_jobs),
+        "stale_running_jobs": sum(
+            1 for job in running_jobs if _is_stale_job(job, stale_after_minutes=stale_after_minutes)
+        ),
+        "failed_jobs": failed_jobs,
+        "degraded_jobs": degraded_jobs,
+        "asset_retry_jobs": asset_retry_jobs,
+        "recent_failures": recent_failures,
+    }
+
+
+async def mark_stale_generation_jobs(
+    db: AsyncSession,
+    *,
+    stale_after_minutes: int | None = None,
+) -> dict[str, int]:
+    """Mark long-running generation jobs as failed so they no longer appear stuck forever."""
+
+    threshold = stale_after_minutes or settings.generation_job_stale_minutes
+    running_jobs = (
+        await db.execute(
+            select(GenerationJob)
+            .where(GenerationJob.status == "running")
+            .order_by(GenerationJob.updated_at, GenerationJob.id)
+        )
+    ).scalars().all()
+
+    marked_stale = 0
+
+    for job in running_jobs:
+        if not _is_stale_job(job, stale_after_minutes=threshold):
+            continue
+
+        story = None
+        if job.story_id is not None:
+            story = (
+                await db.execute(
+                    select(Story).where(
+                        Story.id == job.story_id,
+                        Story.user_id == job.user_id,
+                    )
+                )
+            ).scalar_one_or_none()
+
+        await finish_generation_job(
+            db,
+            job=job,
+            story=story,
+            status="failed",
+            current_step="generation_stale_failed",
+            error_message=f"Generation job exceeded {threshold} minutes without progress.",
+            message="Generation job was marked failed after exceeding the stale threshold.",
+            metadata={"stale_after_minutes": threshold},
+        )
+        marked_stale += 1
+        logger.warning(
+            "generation_job_marked_stale",
+            job_id=job.id,
+            story_id=job.story_id,
+            output_mode=job.output_mode,
+            stale_after_minutes=threshold,
+        )
+
+    return {
+        "running": len(running_jobs),
+        "marked_stale": marked_stale,
+        "stale_after_minutes": threshold,
     }

backend/app/services/story_service.py

@@ -2,6 +2,7 @@
 
 import asyncio
 from dataclasses import dataclass
+from datetime import datetime, timedelta, timezone
 from typing import Literal
 
 from fastapi import HTTPException
@@ -9,6 +10,7 @@ from sqlalchemy import desc, select
 from sqlalchemy.ext.asyncio import AsyncSession
 from sqlalchemy.orm import joinedload
 
+from app.core.config import settings
 from app.core.logging import get_logger
 from app.db.models import ChildProfile, Story, StoryUniverse
 from app.schemas.story_schemas import (
@@ -32,6 +34,7 @@ from app.services.audio_storage import (
 )
 from app.services.generation_jobs import (
     create_generation_job,
+    ensure_no_active_story_generation_job,
     finish_generation_job,
     record_generation_event,
 )
@@ -1369,6 +1372,7 @@ async def retry_story_assets(
     db: AsyncSession,
 ) -> Story:
     """Retry selected assets through one workflow-level endpoint."""
+    await ensure_no_active_story_generation_job(db, story_id=story_id, user_id=user_id)
     requested_assets = list(dict.fromkeys(assets))
     job = await create_generation_job(
         db,
@@ -1443,6 +1447,7 @@ async def generate_story_cover(
     db: AsyncSession,
 ) -> str:
     """Generate cover image for an existing story."""
+    await ensure_no_active_story_generation_job(db, story_id=story_id, user_id=user_id)
     job = await create_generation_job(
         db,
         user_id=user_id,
@@ -1495,6 +1500,7 @@ async def generate_story_audio(
     db: AsyncSession,
 ) -> bytes:
     """Generate audio for a story."""
+    await ensure_no_active_story_generation_job(db, story_id=story_id, user_id=user_id)
     job = await create_generation_job(
         db,
         user_id=user_id,
@@ -1597,6 +1603,50 @@ async def clear_story_audio_cache(
     return await get_story_audio_status(story_id, user_id, db)
 
 
+async def prune_story_audio_cache(db: AsyncSession) -> dict[str, int]:
+    """Prune expired audio cache files and repair story metadata."""
+
+    ttl_days = max(1, settings.story_audio_cache_ttl_days)
+    cutoff = datetime.now(timezone.utc) - timedelta(days=ttl_days)
+    result = await db.execute(select(Story).where(Story.audio_path.is_not(None)))
+    stories = result.scalars().all()
+
+    scanned = 0
+    pruned = 0
+    repaired = 0
+
+    for story in stories:
+        scanned += 1
+        metadata = get_audio_cache_metadata(story.audio_path)
+
+        if not metadata.exists:
+            story.audio_path = None
+            if story.audio_status == StoryAssetStatus.READY.value:
+                sync_story_status(story, audio_status=StoryAssetStatus.NOT_REQUESTED)
+            repaired += 1
+            continue
+
+        if metadata.updated_at and metadata.updated_at < cutoff:
+            delete_audio_cache(story.audio_path)
+            story.audio_path = None
+            sync_story_status(
+                story,
+                audio_status=StoryAssetStatus.NOT_REQUESTED,
+                last_error=None,
+            )
+            pruned += 1
+
+    await db.commit()
+    logger.info(
+        "story_audio_cache_pruned",
+        scanned=scanned,
+        pruned=pruned,
+        repaired=repaired,
+        ttl_days=ttl_days,
+    )
+    return {"scanned": scanned, "pruned": pruned, "repaired": repaired}
+
+
 async def get_story_achievements(
     story_id: int,
     user_id: str,

backend/app/tasks/audio_cache.py

@@ -0,0 +1,29 @@
+"""Celery tasks for story audio cache maintenance."""
+
+import asyncio
+
+from app.core.celery_app import celery_app
+from app.core.logging import get_logger
+from app.db.database import _get_session_factory
+from app.services.story_service import prune_story_audio_cache
+
+logger = get_logger(__name__)
+
+
+@celery_app.task
+def prune_story_audio_cache_task():
+    """Daily task to prune expired story audio cache files."""
+    logger.info("prune_story_audio_cache_task_started")
+
+    async def _run():
+        session_factory = _get_session_factory()
+        async with session_factory() as session:
+            return await prune_story_audio_cache(session)
+
+    try:
+        result = asyncio.run(_run())
+        logger.info("prune_story_audio_cache_task_completed", **result)
+        return result
+    except Exception as exc:
+        logger.error("prune_story_audio_cache_task_failed", error=str(exc))
+        raise

backend/app/tasks/generation_maintenance.py

@@ -0,0 +1,30 @@
+"""Generation job maintenance tasks."""
+
+import asyncio
+
+from app.core.celery_app import celery_app
+from app.core.logging import get_logger
+from app.db.database import _get_session_factory
+from app.services.generation_jobs import mark_stale_generation_jobs
+
+logger = get_logger(__name__)
+
+
+@celery_app.task
+def prune_stale_generation_jobs_task():
+    """Periodically mark stale running generation jobs as failed."""
+
+    logger.info("prune_stale_generation_jobs_task_started")
+
+    async def _run():
+        session_factory = _get_session_factory()
+        async with session_factory() as session:
+            return await mark_stale_generation_jobs(session)
+
+    try:
+        result = asyncio.run(_run())
+        logger.info("prune_stale_generation_jobs_task_completed", **result)
+        return result
+    except Exception as exc:
+        logger.error("prune_stale_generation_jobs_task_failed", error=str(exc))
+        raise

backend/tests/test_audio_cache.py

@@ -0,0 +1,65 @@
+"""Story audio cache maintenance tests."""
+
+import os
+from datetime import datetime, timedelta, timezone
+from pathlib import Path
+
+import pytest
+from fastapi import HTTPException
+from sqlalchemy import select
+
+from app.core.config import settings
+from app.db.models import Story
+from app.services.generation_jobs import create_generation_job
+from app.services.story_service import generate_story_audio, prune_story_audio_cache
+
+pytestmark = pytest.mark.asyncio
+
+
+async def test_prune_story_audio_cache_removes_expired_audio(
+    db_session,
+    test_story,
+    mock_tts_provider,
+    monkeypatch,
+):
+    await generate_story_audio(test_story.id, test_story.user_id, db_session)
+
+    cached_audio_path = Path(settings.story_audio_cache_dir) / f"story-{test_story.id}.mp3"
+    assert cached_audio_path.is_file()
+
+    old_time = datetime.now(timezone.utc) - timedelta(days=10)
+    timestamp = old_time.timestamp()
+    os.utime(cached_audio_path, (timestamp, timestamp))
+    monkeypatch.setattr(settings, "story_audio_cache_ttl_days", 7)
+
+    result = await prune_story_audio_cache(db_session)
+
+    assert result == {"scanned": 1, "pruned": 1, "repaired": 0}
+    assert not cached_audio_path.exists()
+
+    story = (
+        await db_session.execute(select(Story).where(Story.id == test_story.id))
+    ).scalar_one()
+    assert story.audio_path is None
+    assert story.audio_status == "not_requested"
+    assert story.generation_status == "partial_ready"
+
+
+async def test_generate_story_audio_rejects_when_story_has_active_job(
+    db_session,
+    test_story,
+):
+    await create_generation_job(
+        db_session,
+        user_id=test_story.user_id,
+        output_mode="asset_retry",
+        input_type="audio",
+        request_payload={"story_id": test_story.id},
+        story_id=test_story.id,
+    )
+
+    with pytest.raises(HTTPException) as exc_info:
+        await generate_story_audio(test_story.id, test_story.user_id, db_session)
+
+    assert exc_info.value.status_code == 409
+    assert "已有运行中的任务" in str(exc_info.value.detail)

backend/tests/test_generation_jobs.py

@@ -1,5 +1,6 @@
 """Generation job tracking tests."""
 
+from datetime import datetime, timedelta, timezone
 from unittest.mock import AsyncMock, patch
 
 import pytest
@@ -12,7 +13,11 @@ from app.main import app
 from app.services.adapters import AdapterConfig
 from app.services.adapters.storybook.primary import Storybook, StorybookPage
 from app.services.adapters.text.models import StoryOutput
-from app.services.generation_jobs import create_generation_job, record_generation_event
+from app.services.generation_jobs import (
+    create_generation_job,
+    mark_stale_generation_jobs,
+    record_generation_event,
+)
 
 pytestmark = pytest.mark.asyncio
 
@@ -520,6 +525,7 @@ async def test_user_provider_analytics_aggregate_across_stories(
         assert data["failed_calls"] == 1
         assert data["avg_latency_ms"] == 60.0
         assert data["estimated_cost_usd"] == 0.013
+        assert data["failure_reasons"] == [{"reason": "timeout", "count": 1}]
         assert data["by_provider"] == [
             {
                 "capability": "image",
@@ -551,3 +557,249 @@ async def test_user_provider_analytics_aggregate_across_stories(
         ]
     finally:
         app.dependency_overrides.clear()
+
+
+async def test_provider_analytics_support_days_and_capability_filters(
+    db_session,
+    auth_token,
+    degraded_story_with_text,
+    test_story,
+):
+    async def override_get_db():
+        yield db_session
+
+    app.dependency_overrides[get_db] = override_get_db
+
+    image_job = await create_generation_job(
+        db_session,
+        user_id=degraded_story_with_text.user_id,
+        output_mode="asset_retry",
+        input_type="image",
+        request_payload={"assets": ["image"]},
+        story_id=degraded_story_with_text.id,
+    )
+    old_event = await record_generation_event(
+        db_session,
+        job=image_job,
+        story_id=degraded_story_with_text.id,
+        event_type="provider_call_failed",
+        status="failed",
+        metadata={
+            "capability": "image",
+            "adapter": "cqtai",
+            "strategy": "priority",
+            "latency_ms": 120,
+            "error": "timeout",
+        },
+    )
+    old_event.created_at = datetime.now(timezone.utc) - timedelta(days=10)
+    await db_session.commit()
+
+    tts_job = await create_generation_job(
+        db_session,
+        user_id=test_story.user_id,
+        output_mode="asset_retry",
+        input_type="audio",
+        request_payload={"assets": ["audio"]},
+        story_id=test_story.id,
+    )
+    await record_generation_event(
+        db_session,
+        job=tts_job,
+        story_id=test_story.id,
+        event_type="provider_call_succeeded",
+        status="succeeded",
+        metadata={
+            "capability": "tts",
+            "adapter": "edge_tts",
+            "strategy": "priority",
+            "latency_ms": 18,
+            "estimated_cost_usd": 0.003,
+        },
+    )
+
+    transport = ASGITransport(app=app)
+    try:
+        async with AsyncClient(transport=transport, base_url="http://test") as client:
+            client.cookies.set("access_token", auth_token)
+
+            response = await client.get("/api/generations/provider-analytics?days=7")
+            assert response.status_code == 200
+            data = response.json()
+            assert data["window_days"] == 7
+            assert data["total_calls"] == 1
+            assert data["job_count"] == 1
+            assert data["story_count"] == 1
+            assert data["failure_reasons"] == []
+
+            response = await client.get(
+                "/api/generations/provider-analytics?capability=image"
+            )
+            assert response.status_code == 200
+            data = response.json()
+            assert data["capability"] == "image"
+            assert data["total_calls"] == 1
+            assert data["failed_calls"] == 1
+            assert data["job_count"] == 1
+            assert data["story_count"] == 1
+            assert data["failure_reasons"] == [{"reason": "timeout", "count": 1}]
+
+            response = await client.get(
+                f"/api/generations/{degraded_story_with_text.id}/provider-stats?capability=image"
+            )
+            assert response.status_code == 200
+            data = response.json()
+            assert data["capability"] == "image"
+            assert data["failure_reasons"] == [{"reason": "timeout", "count": 1}]
+    finally:
+        app.dependency_overrides.clear()
+
+
+async def test_generation_ops_summary_exposes_running_stale_and_recent_failures(
+    db_session,
+    auth_token,
+    degraded_story_with_text,
+    test_story,
+):
+    async def override_get_db():
+        yield db_session
+
+    app.dependency_overrides[get_db] = override_get_db
+
+    running_job = await create_generation_job(
+        db_session,
+        user_id=test_story.user_id,
+        output_mode="story",
+        input_type="keywords",
+        request_payload={"data": "星星"},
+        story_id=test_story.id,
+    )
+    stale_job = await create_generation_job(
+        db_session,
+        user_id=degraded_story_with_text.user_id,
+        output_mode="asset_generation",
+        input_type="image",
+        request_payload={"story_id": degraded_story_with_text.id},
+        story_id=degraded_story_with_text.id,
+    )
+    failed_job = await create_generation_job(
+        db_session,
+        user_id=degraded_story_with_text.user_id,
+        output_mode="asset_retry",
+        input_type="image",
+        request_payload={"assets": ["image"]},
+        story_id=degraded_story_with_text.id,
+    )
+    degraded_job = await create_generation_job(
+        db_session,
+        user_id=test_story.user_id,
+        output_mode="storybook",
+        input_type="keywords",
+        request_payload={"data": "月亮"},
+        story_id=test_story.id,
+    )
+
+    stale_job.updated_at = datetime.now(timezone.utc) - timedelta(hours=3)
+    failed_job.status = "failed"
+    failed_job.current_step = "asset_retry_failed"
+    failed_job.error_message = "image timeout"
+    failed_job.updated_at = datetime.now(timezone.utc) - timedelta(hours=1)
+    degraded_job.status = "degraded_completed"
+    degraded_job.current_step = "generation_completed"
+    degraded_job.updated_at = datetime.now(timezone.utc) - timedelta(minutes=30)
+    running_job.updated_at = datetime.now(timezone.utc) - timedelta(minutes=10)
+    await db_session.commit()
+
+    transport = ASGITransport(app=app)
+    try:
+        async with AsyncClient(transport=transport, base_url="http://test") as client:
+            client.cookies.set("access_token", auth_token)
+
+            response = await client.get("/api/generations/ops-summary?hours=48")
+
+        assert response.status_code == 200
+        data = response.json()
+        assert data["window_hours"] == 48
+        assert data["active_jobs"] == 2
+        assert data["stale_running_jobs"] == 1
+        assert data["failed_jobs"] == 1
+        assert data["degraded_jobs"] == 1
+        assert data["asset_retry_jobs"] == 2
+        assert len(data["recent_failures"]) == 1
+        assert data["recent_failures"][0]["job_id"] == failed_job.id
+        assert data["recent_failures"][0]["story_title"] == degraded_story_with_text.title
+        assert data["recent_failures"][0]["failure_label"] == "资源重试失败"
+    finally:
+        app.dependency_overrides.clear()
+
+
+async def test_mark_stale_generation_jobs_marks_old_running_jobs_failed(
+    db_session,
+    degraded_story_with_text,
+):
+    stale_job = await create_generation_job(
+        db_session,
+        user_id=degraded_story_with_text.user_id,
+        output_mode="story",
+        input_type="keywords",
+        request_payload={"data": "超时任务"},
+        story_id=degraded_story_with_text.id,
+    )
+    stale_job.updated_at = datetime.now(timezone.utc) - timedelta(hours=2)
+    await db_session.commit()
+
+    result = await mark_stale_generation_jobs(db_session, stale_after_minutes=30)
+
+    assert result == {"running": 1, "marked_stale": 1, "stale_after_minutes": 30}
+
+    refreshed_job = (
+        await db_session.execute(select(GenerationJob).where(GenerationJob.id == stale_job.id))
+    ).scalar_one()
+    assert refreshed_job.status == "failed"
+    assert refreshed_job.current_step == "generation_stale_failed"
+    assert refreshed_job.error_message == "Generation job exceeded 30 minutes without progress."
+
+    events = (
+        await db_session.execute(
+            select(GenerationJobEvent)
+            .where(GenerationJobEvent.job_id == stale_job.id)
+            .order_by(GenerationJobEvent.id)
+        )
+    ).scalars().all()
+    assert events[-1].event_type == "generation_stale_failed"
+    assert events[-1].event_metadata["stale_after_minutes"] == 30
+
+
+async def test_retry_assets_rejects_when_story_has_active_job(
+    db_session,
+    auth_token,
+    degraded_story_with_text,
+):
+    async def override_get_db():
+        yield db_session
+
+    app.dependency_overrides[get_db] = override_get_db
+
+    await create_generation_job(
+        db_session,
+        user_id=degraded_story_with_text.user_id,
+        output_mode="asset_generation",
+        input_type="image",
+        request_payload={"story_id": degraded_story_with_text.id},
+        story_id=degraded_story_with_text.id,
+    )
+
+    transport = ASGITransport(app=app)
+    try:
+        async with AsyncClient(transport=transport, base_url="http://test") as client:
+            client.cookies.set("access_token", auth_token)
+
+            response = await client.post(
+                f"/api/generations/{degraded_story_with_text.id}/retry-assets",
+                json={"assets": ["image"]},
+            )
+
+        assert response.status_code == 409
+        assert "已有运行中的任务" in response.json()["detail"]
+    finally:
+        app.dependency_overrides.clear()

docs/planning/demo-validation-log.md

@@ -16,6 +16,8 @@
 - `./scripts/demo_smoke.sh` 已覆盖音频缓存状态查询。
 - Week 4 Demo 包装已完成：新增架构说明、Demo 包装文档、Week 4 sprint review，用户端和管理端绘本阅读器支持阅读位置恢复。
 - Week 4 最终回归通过：后端全量测试 85 passed，ruff 通过，用户端/管理端构建通过，`docker compose up -d --build` 和 `./scripts/demo_smoke.sh` 通过。
+- 继续优化后再次验证：Provider analytics 已支持时间窗口与能力筛选、失败原因摘要；音频缓存已加入 TTL 配置和后台 prune 任务。
+- 新一轮优化验证通过：新增 `GET /api/generations/ops-summary`，故事库已展示最近失败与卡住任务摘要；生成任务已支持 stale 自动收敛和重复资产任务保护。
 - 后端新增 `partial_ready`、`text_status` 与迁移 `0012_story_text_status` 后，`backend/.venv/bin/python -m pytest backend/tests -q` 通过，82 个测试通过。
 - `backend/.venv/bin/python -m ruff check backend/app backend/tests backend/alembic/versions/0012_add_story_text_status_and_partial_ready.py` 通过。
 - 用户端与管理端 `npm run build` 均通过。

docs/planning/week-2-to-4-execution-backlog.md

@@ -66,6 +66,11 @@ Week 2 已完成演示闭环、统一生成工作流、generation job/event、
 | W4-03 | Demo | 求职版 Demo 包装 | `docs/planning/demo-package.md` | P0 | Done |
 | W4-04 | QA | 全量回归与验证记录 | pytest、ruff、前端 build、Docker smoke | P0 | Done |
 | W4-05 | Product | 项目复盘与下一阶段路线 | `docs/planning/week-4-sprint-review.md` | P1 | Done |
+| W4-06 | Ops | Provider analytics 支持时间窗口与失败原因 | `days` / `capability` 筛选 + failure reason summary | P1 | Done |
+| W4-07 | Ops | 音频缓存后台清理 | TTL 配置 + Celery beat prune task | P1 | Done |
+| W4-08 | Ops | 任务运行概览与失败摘要 | `GET /api/generations/ops-summary` + 最近失败列表 | P1 | Done |
+| W4-09 | Workflow | 卡住任务自动收敛 | `GENERATION_JOB_STALE_MINUTES` + Celery beat stale job maintenance | P1 | Done |
+| W4-10 | Workflow | 防止重复资产任务 | 运行中故事拒绝重复封面/音频/资产重试请求 | P1 | Done |
 
 ---
 

docs/planning/week-4-sprint-review.md

@@ -34,6 +34,7 @@ DreamWeaver 已经具备求职演示所需的完整闭环：
 - generation job/event
 - Provider failover 和聚合指标
 - 跨故事 Provider analytics
+- 任务运行概览、最近失败摘要与卡住任务收敛
 - 前端生成轨迹和自动轮询形态
 
 ---
@@ -42,7 +43,7 @@ DreamWeaver 已经具备求职演示所需的完整闭环：
 
 最近一轮验证包括：
 
-- 后端全量测试：85 passed
+- 后端全量测试：91 passed
 - 后端 ruff：通过
 - 用户端生产构建：通过
 - 管理端生产构建：通过
@@ -56,10 +57,9 @@ DreamWeaver 已经具备求职演示所需的完整闭环：
 | Priority | Task | Why |
 | --- | --- | --- |
 | P0 | 将同步生成迁移到 Celery worker | 支持真实长任务、断点恢复和后台进度 |
-| P0 | Provider analytics 加入时间窗口和失败原因 | 让运营分析可用于成本与稳定性决策 |
-| P1 | 音频缓存过期策略与后台清理 | 控制磁盘占用和缓存生命周期 |
 | P1 | 生成任务取消与重试队列 | 防止重复任务和用户误触造成浪费 |
-| P1 | 监控告警与结构化 dashboard | 上线前需要可观测性闭环 |
+| P1 | 跨用户 / 跨环境 Provider dashboard | 当前已支持单用户摘要，后续要支持运营视角 |
+| P1 | 监控告警与结构化 dashboard | 目前已有故事库级概览，后续要接入更完整观测体系 |
 | P2 | 更细粒度叙事风格与音色策略 | 扩展体验，但不影响当前求职版主线 |
 
 ---

docs/product/unified-generation-workflow-prd.md

@@ -64,6 +64,10 @@ DreamWeaver 当前同时支持普通故事生成、完整故事生成和绘本
 - Provider 调用已可按故事聚合为成功率、平均耗时、预估成本和 adapter 明细
 - generation job 响应已提供 `progress_percent`、`progress_label` 和 `is_terminal`，前端可直接用于进度条和轮询
 - 已新增跨故事 Provider 运营摘要 `GET /api/generations/provider-analytics`，故事库可展示总调用、成功率、平均耗时、预估成本和任务/故事覆盖数
+- 跨故事 Provider 运营摘要已支持按时间窗口和 capability 筛选，并聚合失败原因
+- 已新增任务运行概览 `GET /api/generations/ops-summary`，故事库可展示最近失败、运行中任务和超时待收敛任务
+- 重复资产任务已加入保护：同一故事存在运行中 job 时，不再重复触发封面、音频或统一资产重试
+- Celery beat 已支持定时收敛卡住的 generation job，避免任务长期停在 running
 - 用户端与管理端生成轨迹组件会在任务未终止时自动轮询，为后续后台 worker 进度流保留前端形态
 - `POST /api/generations` 响应已返回 `generation_job_id`，smoke 脚本会验证 job 查询与 story job history
 - 用户端与管理端的故事详情页、绘本阅读页已接入生成轨迹，展示生成/重试任务、关键事件、Provider 调用结果和聚合指标
@@ -74,7 +78,7 @@ DreamWeaver 当前同时支持普通故事生成、完整故事生成和绘本
 
 - 普通故事、完整生成、绘本生成已有统一外部入口，内部 workflow 仍可继续减少兼容层分支
 - 统一资产重试入口已覆盖普通故事封面、绘本缺失插图和故事音频，后续可继续扩展更细的资产级审计
-- 后台异步 worker 执行、断点续跑、跨时间窗口筛选和更完整的 Provider 运营分析仍属于后续生产化增强
+- 后台异步 worker 执行、断点续跑、跨用户/跨环境 Provider 分析，以及真正的取消/重试队列仍属于后续生产化增强
 
 ### What This Means
 

docs/technical/generation-job-state.md

@@ -19,9 +19,10 @@
 - `GET /api/generations/jobs/{job_id}`：查询单次生成/补全任务及其事件流。
 - `GET /api/generations/{story_id}/jobs`：查询某个故事或绘本的生成与重试历史。
 - `GET /api/generations/{story_id}/provider-stats`：按故事聚合 Provider 调用成功率、平均耗时、预估成本和 adapter 明细。
-- `GET /api/generations/provider-analytics`：按当前用户聚合跨故事 Provider 调用、任务数、故事数、成功率、平均耗时和预估成本。
+- `GET /api/generations/provider-analytics`：按当前用户聚合跨故事 Provider 调用、任务数、故事数、成功率、平均耗时和预估成本，并支持 `days` / `capability` 筛选。
+- `GET /api/generations/ops-summary`：按当前用户聚合最近任务健康度，包括运行中数量、超时阈值、卡住任务数和最近失败摘要。
 
-job 响应会返回 `progress_percent`、`progress_label` 和 `is_terminal`，用户端与管理端已经消费这些查询入口，在故事详情页和绘本阅读页展示最近任务、任务历史、事件时间线、进度条和 Provider 聚合指标；当任务未终止时，前端会自动轮询，为后台 worker 进度流预留体验形态。
+job 响应会返回 `progress_percent`、`progress_label` 和 `is_terminal`，用户端与管理端已经消费这些查询入口，在故事详情页和绘本阅读页展示最近任务、任务历史、事件时间线、进度条和 Provider 聚合指标；当任务未终止时，前端会自动轮询，为后台 worker 进度流预留体验形态。当前 analytics 还会聚合失败原因，便于快速解释“最近为什么失败”；ops summary 会额外把“哪些任务卡住了、最近哪些任务失败了”压缩成故事库首页能直接看的摘要。
 
 ## 现有状态模型
 
@@ -37,6 +38,12 @@ job 响应会返回 `progress_percent`、`progress_label` 和 `is_terminal`，
 
 这些字段足够支撑前端展示、smoke 检查、失败降级、资产重试和生成轨迹解释。
 
+## 当前维护策略
+
+- 音频缓存由 `STORY_AUDIO_CACHE_TTL_DAYS` 控制过期时间，Celery beat 会每日清理。
+- 生成任务由 `GENERATION_JOB_STALE_MINUTES` 控制卡住阈值，Celery beat 会每 30 分钟扫描一次，将超时运行中的任务标记为 `generation_stale_failed`。
+- 当某个故事已经有运行中的 job 时，封面补全、音频生成和统一资产重试会直接拒绝重复请求，避免用户连点造成重复成本。
+
 ## 什么时候需要落库 job
 
 如果后续进入真实生产化，需要扩展当前 job/event 模型：
@@ -52,7 +59,7 @@ job 响应会返回 `progress_percent`、`progress_label` 和 `is_terminal`，
 当前已有两层记录，未来可以继续扩展字段和事件颗粒度：
 
 - 将同步生成请求迁移到真正异步 worker 后，继续复用现有 job 查询和前端轮询进度条。
-- 将当前跨故事 provider 指标扩展为跨时间窗口、跨用户和失败原因维度的运营分析。
+- 将当前跨故事 provider 指标扩展为跨用户、跨环境和更细颗粒度的失败原因维度分析。
 
 ## 面试表达
 

frontend/src/types/generation.ts

@@ -42,15 +42,23 @@ export interface GenerationProviderStat {
 
 export interface GenerationProviderStats {
   story_id: number
+  window_days: number | null
+  capability: string | null
   total_calls: number
   successful_calls: number
   failed_calls: number
   avg_latency_ms: number | null
   estimated_cost_usd: number
   by_provider: GenerationProviderStat[]
+  failure_reasons: Array<{
+    reason: string
+    count: number
+  }>
 }
 
 export interface GenerationProviderAnalytics {
+  window_days: number | null
+  capability: string | null
   total_calls: number
   successful_calls: number
   failed_calls: number
@@ -59,4 +67,30 @@ export interface GenerationProviderAnalytics {
   job_count: number
   story_count: number
   by_provider: GenerationProviderStat[]
+  failure_reasons: Array<{
+    reason: string
+    count: number
+  }>
+}
+
+export interface GenerationRecentFailure {
+  job_id: string
+  story_id: number | null
+  story_title: string | null
+  output_mode: string
+  current_step: string
+  error_message: string | null
+  failure_label: string
+  updated_at: string
+}
+
+export interface GenerationOpsSummary {
+  window_hours: number
+  stale_threshold_minutes: number
+  active_jobs: number
+  stale_running_jobs: number
+  failed_jobs: number
+  degraded_jobs: number
+  asset_retry_jobs: number
+  recent_failures: GenerationRecentFailure[]
 }

frontend/src/views/MyStories.vue

@@ -1,5 +1,5 @@
 <script setup lang="ts">
-import { computed, onMounted, ref } from 'vue'
+import { computed, onMounted, ref, watch } from 'vue'
 import { useRouter } from 'vue-router'
 import { api } from '../api/client'
 import CreateStoryModal from '../components/CreateStoryModal.vue'
@@ -7,7 +7,7 @@ import BaseButton from '../components/ui/BaseButton.vue'
 import BaseCard from '../components/ui/BaseCard.vue'
 import EmptyState from '../components/ui/EmptyState.vue'
 import LoadingSpinner from '../components/ui/LoadingSpinner.vue'
-import type { GenerationProviderAnalytics } from '../types/generation'
+import type { GenerationOpsSummary, GenerationProviderAnalytics } from '../types/generation'
 import {
   getAssetStatusMeta,
   getGenerationStatusMeta,
@@ -39,9 +39,12 @@ interface StoryItem {
 const router = useRouter()
 const stories = ref<StoryItem[]>([])
 const providerAnalytics = ref<GenerationProviderAnalytics | null>(null)
+const opsSummary = ref<GenerationOpsSummary | null>(null)
 const loading = ref(true)
 const error = ref('')
 const showCreateModal = ref(false)
+const selectedWindow = ref<'7' | '30' | 'all'>('30')
+const selectedCapability = ref<'all' | 'text' | 'image' | 'tts' | 'storybook'>('all')
 
 const readableCount = computed(() =>
   stories.value.filter((story) => isReadableGenerationStatus(story.generation_status)).length,
@@ -57,15 +60,30 @@ const providerSuccessRate = computed(() => {
   )
 })
 const topProvider = computed(() => providerAnalytics.value?.by_provider[0] ?? null)
+const topFailureReason = computed(() => providerAnalytics.value?.failure_reasons[0] ?? null)
+
+function buildProviderAnalyticsPath() {
+  const params = new URLSearchParams()
+  if (selectedWindow.value !== 'all') {
+    params.set('days', selectedWindow.value)
+  }
+  if (selectedCapability.value !== 'all') {
+    params.set('capability', selectedCapability.value)
+  }
+  const query = params.toString()
+  return `/api/generations/provider-analytics${query ? `?${query}` : ''}`
+}
 
 async function fetchStories() {
   try {
-    const [storyList, analytics] = await Promise.all([
+    const [storyList, analytics, ops] = await Promise.all([
       api.get<StoryItem[]>('/api/stories'),
-      api.get<GenerationProviderAnalytics>('/api/generations/provider-analytics'),
+      api.get<GenerationProviderAnalytics>(buildProviderAnalyticsPath()),
+      api.get<GenerationOpsSummary>('/api/generations/ops-summary'),
     ])
     stories.value = storyList
     providerAnalytics.value = analytics
+    opsSummary.value = ops
   } catch (e) {
     error.value = e instanceof Error ? e.message : '加载失败'
   } finally {
@@ -106,6 +124,27 @@ function formatCost(value?: number | null) {
   return typeof value === 'number' ? `$${value.toFixed(4)}` : '$0.0000'
 }
 
+function formatOutputMode(value: string) {
+  switch (value) {
+    case 'storybook':
+      return '绘本'
+    case 'asset_retry':
+      return '资源重试'
+    case 'asset_generation':
+      return '资源生成'
+    default:
+      return '故事'
+  }
+}
+
+function setWindow(value: '7' | '30' | 'all') {
+  selectedWindow.value = value
+}
+
+function setCapability(value: 'all' | 'text' | 'image' | 'tts' | 'storybook') {
+  selectedCapability.value = value
+}
+
 onMounted(() => {
   void fetchStories()
 
@@ -114,6 +153,10 @@ onMounted(() => {
     router.replace({ query: { ...router.currentRoute.value.query, openCreate: undefined } })
   }
 })
+
+watch([selectedWindow, selectedCapability], () => {
+  void fetchStories()
+})
 </script>
 
 <template>
@@ -191,6 +234,18 @@ onMounted(() => {
             <p class="mt-2 text-sm leading-6 text-gray-500">
               最近生成和资源补全留下的供应商调用轨迹。
             </p>
+            <div class="mt-4 flex flex-wrap gap-2">
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedWindow === '7' ? 'border-gray-900 bg-gray-900 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setWindow('7')">最近 7 天</button>
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedWindow === '30' ? 'border-gray-900 bg-gray-900 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setWindow('30')">最近 30 天</button>
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedWindow === 'all' ? 'border-gray-900 bg-gray-900 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setWindow('all')">全部</button>
+            </div>
+            <div class="mt-3 flex flex-wrap gap-2">
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedCapability === 'all' ? 'border-emerald-600 bg-emerald-600 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setCapability('all')">全部能力</button>
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedCapability === 'text' ? 'border-emerald-600 bg-emerald-600 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setCapability('text')">文本</button>
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedCapability === 'image' ? 'border-emerald-600 bg-emerald-600 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setCapability('image')">图片</button>
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedCapability === 'tts' ? 'border-emerald-600 bg-emerald-600 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setCapability('tts')">语音</button>
+              <button type="button" class="rounded-lg border px-3 py-1.5 text-sm transition-colors" :class="selectedCapability === 'storybook' ? 'border-emerald-600 bg-emerald-600 text-white' : 'border-gray-200 bg-white text-gray-600 hover:border-gray-400'" @click="setCapability('storybook')">绘本</button>
+            </div>
           </div>
           <div class="grid grid-cols-2 gap-3 sm:grid-cols-4 lg:min-w-[520px]">
             <div class="rounded-lg border border-gray-100 bg-gray-50 px-3 py-3">
@@ -214,6 +269,70 @@ onMounted(() => {
         <p v-if="topProvider" class="mt-4 text-sm text-gray-500">
           当前样本中最前面的能力组合是 {{ topProvider.capability }} / {{ topProvider.adapter }}，成功 {{ topProvider.success_count }} 次，失败 {{ topProvider.failure_count }} 次。
         </p>
+        <p v-if="topFailureReason" class="mt-2 text-sm text-rose-600">
+          最常见失败原因：{{ topFailureReason.reason }}（{{ topFailureReason.count }} 次）
+        </p>
+      </BaseCard>
+
+      <BaseCard
+        v-if="opsSummary"
+        class="mb-8"
+        padding="lg"
+      >
+        <div class="flex flex-col gap-5 lg:flex-row lg:items-center lg:justify-between">
+          <div>
+            <h2 class="text-xl font-bold text-gray-800">任务运行概览</h2>
+            <p class="mt-2 text-sm leading-6 text-gray-500">
+              最近 {{ opsSummary.window_hours }} 小时的任务健康度，运行超过
+              {{ opsSummary.stale_threshold_minutes }} 分钟会被视为卡住。
+            </p>
+          </div>
+          <div class="grid grid-cols-2 gap-3 sm:grid-cols-4 lg:min-w-[520px]">
+            <div class="rounded-lg border border-gray-100 bg-gray-50 px-3 py-3">
+              <div class="text-xs text-gray-500">运行中</div>
+              <div class="mt-1 text-lg font-semibold text-gray-800">{{ opsSummary.active_jobs }}</div>
+            </div>
+            <div class="rounded-lg border border-gray-100 bg-gray-50 px-3 py-3">
+              <div class="text-xs text-gray-500">超时待收敛</div>
+              <div class="mt-1 text-lg font-semibold" :class="opsSummary.stale_running_jobs ? 'text-amber-600' : 'text-gray-800'">
+                {{ opsSummary.stale_running_jobs }}
+              </div>
+            </div>
+            <div class="rounded-lg border border-gray-100 bg-gray-50 px-3 py-3">
+              <div class="text-xs text-gray-500">最近失败</div>
+              <div class="mt-1 text-lg font-semibold" :class="opsSummary.failed_jobs ? 'text-rose-600' : 'text-gray-800'">
+                {{ opsSummary.failed_jobs }}
+              </div>
+            </div>
+            <div class="rounded-lg border border-gray-100 bg-gray-50 px-3 py-3">
+              <div class="text-xs text-gray-500">资源任务</div>
+              <div class="mt-1 text-lg font-semibold text-gray-800">{{ opsSummary.asset_retry_jobs }}</div>
+            </div>
+          </div>
+        </div>
+        <p v-if="opsSummary.degraded_jobs" class="mt-4 text-sm text-amber-600">
+          最近 {{ opsSummary.window_hours }} 小时有 {{ opsSummary.degraded_jobs }} 个任务以降级完成收尾。
+        </p>
+        <div v-if="opsSummary.recent_failures.length" class="mt-4 space-y-3">
+          <div
+            v-for="failure in opsSummary.recent_failures"
+            :key="failure.job_id"
+            class="rounded-lg border border-rose-100 bg-rose-50 px-4 py-3"
+          >
+            <div class="flex flex-wrap items-center justify-between gap-3">
+              <div class="text-sm font-semibold text-gray-800">
+                {{ failure.story_title || `${formatOutputMode(failure.output_mode)}任务` }}
+              </div>
+              <div class="text-xs text-gray-500">{{ formatDate(failure.updated_at) }}</div>
+            </div>
+            <div class="mt-1 text-xs text-rose-600">
+              {{ failure.failure_label }} · {{ failure.error_message || '请打开任务轨迹查看原因' }}
+            </div>
+          </div>
+        </div>
+        <p v-else class="mt-4 text-sm text-emerald-600">
+          最近 {{ opsSummary.window_hours }} 小时没有失败任务，当前链路比较稳定。
+        </p>
       </BaseCard>
 
       <div class="grid grid-cols-1 sm:grid-cols-2 lg:grid-cols-3 gap-6">

scripts/demo_smoke.sh

@@ -165,6 +165,16 @@ provider_analytics_json="$(get_json "$APP_URL/api/generations/provider-analytics
 assert_jq "$provider_analytics_json" '.total_calls >= 2 and .successful_calls >= 2 and .job_count >= 4 and .story_count >= 2 and (.by_provider | length) >= 1' "provider analytics should summarize calls across generated stories"
 echo "$provider_analytics_json" | jq '{total_calls,successful_calls,failed_calls,job_count,story_count,avg_latency_ms,estimated_cost_usd}'
 
+say "Checking filtered provider analytics"
+filtered_provider_analytics_json="$(get_json "$APP_URL/api/generations/provider-analytics?days=7&capability=text")"
+assert_jq "$filtered_provider_analytics_json" '.window_days == 7 and .capability == "text" and .total_calls >= 1' "filtered provider analytics should honor days/capability filters"
+echo "$filtered_provider_analytics_json" | jq '{window_days,capability,total_calls,successful_calls,failed_calls,failure_reasons}'
+
+say "Checking generation ops summary"
+ops_summary_json="$(get_json "$APP_URL/api/generations/ops-summary?hours=24")"
+assert_jq "$ops_summary_json" '.window_hours == 24 and .active_jobs >= 0 and .stale_running_jobs >= 0 and .failed_jobs >= 0 and .asset_retry_jobs >= 2' "generation ops summary should expose recent task health"
+echo "$ops_summary_json" | jq '{window_hours,stale_threshold_minutes,active_jobs,stale_running_jobs,failed_jobs,degraded_jobs,asset_retry_jobs,recent_failures}'
+
 say "Checking story list"
 list_json="$(get_json "$APP_URL/api/stories?limit=5")"
 assert_jq "$list_json" "map(.id) | index($story_id) != null" "story list should include generated story"