fix(task_distributor): adjust high load wait parameters and improve timeout handling

- Increase high load wait interval from 60 to 120 seconds (2 minutes)
- Increase max retries from 10 to 60 to support up to 2 hours total wait time
- Improve timeout message to show actual wait duration in minutes
- Remove duplicate return statement in worker selection logic
- Update notification message to reflect new wait parameters (2 minutes check interval, 2 hours max wait)
- Clean up trailing whitespace in task_distributor.py
- Remove redundant error message from install.sh about missing/incorrect image versions
- Better handling of high load scenarios with clearer logging and user communication

Author: yyhuni

backend/apps/engine/services/task_distributor.py

@@ -156,10 +156,10 @@ def select_best_worker(self) -> Optional[WorkerNode]:
         # 降级策略：如果没有正常负载的，循环等待后重新检测
         if not scored_workers:
             if high_load_workers:
-                # 高负载等待参数（默认每 60 秒检测一次，最多 10 次）
-                high_load_wait = getattr(settings, 'HIGH_LOAD_WAIT_SECONDS', 60)
-                high_load_max_retries = getattr(settings, 'HIGH_LOAD_MAX_RETRIES', 10)
-                
+                # 高负载等待参数（每 2 分钟检测一次，最多等待 2 小时）
+                high_load_wait = getattr(settings, 'HIGH_LOAD_WAIT_SECONDS', 120)
+                high_load_max_retries = getattr(settings, 'HIGH_LOAD_MAX_RETRIES', 60)
+
                 # 开始等待前发送高负载通知
                 high_load_workers.sort(key=lambda x: x[1])
                 _, _, first_cpu, first_mem = high_load_workers[0]
@@ -170,51 +170,51 @@ def select_best_worker(self) -> Optional[WorkerNode]:
                     cpu=first_cpu,
                     mem=first_mem
                 )
-                
+
                 for retry in range(high_load_max_retries):
                     logger.warning(
                         "所有 Worker 高负载，等待 %d 秒后重试... (%d/%d)",
                         high_load_wait, retry + 1, high_load_max_retries
                     )
                     time.sleep(high_load_wait)
-                    
+
                     # 重新获取负载数据
                     loads = worker_load_service.get_all_loads(worker_ids)
-                    
+
                     # 重新评估
                     scored_workers = []
                     high_load_workers = []
-                    
+
                     for worker in workers:
                         load = loads.get(worker.id)
                         if not load:
                             continue
-                        
+
                         cpu = load.get('cpu', 0)
                         mem = load.get('mem', 0)
                         score = cpu * 0.7 + mem * 0.3
-                        
+
                         if cpu > 85 or mem > 85:
                             high_load_workers.append((worker, score, cpu, mem))
                         else:
                             scored_workers.append((worker, score, cpu, mem))
-                    
+
                     # 如果有正常负载的 Worker，跳出循环
                     if scored_workers:
                         logger.info("检测到正常负载 Worker，结束等待")
                         break
-                
-                # 超时或仍然高负载，选择负载最低的
+
+                # 超时后强制派发到负载最低的 Worker
                 if not scored_workers and high_load_workers:
                     high_load_workers.sort(key=lambda x: x[1])
                     best_worker, _, cpu, mem = high_load_workers[0]
-                    
+
                     logger.warning(
-                        "等待超时，强制分发到高负载 Worker: %s (CPU: %.1f%%, MEM: %.1f%%)",
+                        "等待 %d 分钟后仍高负载，强制分发到 Worker: %s (CPU: %.1f%%, MEM: %.1f%%)",
+                        (high_load_wait * high_load_max_retries) // 60,
                         best_worker.name, cpu, mem
                     )
                     return best_worker
-                    return best_worker
             else:
                 logger.warning("没有可用的 Worker")
                 return None

backend/apps/scan/notifications/receivers.py

@@ -87,7 +87,7 @@ def on_all_workers_high_load(sender, worker_name, cpu, mem, **kwargs):
     """所有 Worker 高负载时的通知处理"""
     create_notification(
         title="系统负载较高",
-        message=f"所有节点负载较高（最低负载节点 CPU: {cpu:.1f}%, 内存: {mem:.1f}%），系统将等待最多 10 分钟后分发任务，扫描速度可能受影响",
+        message=f"所有节点负载较高（最低负载节点 CPU: {cpu:.1f}%, 内存: {mem:.1f}%），系统将每 2 分钟检测一次，最多等待 2 小时后分发任务",
         level=NotificationLevel.MEDIUM,
         category=NotificationCategory.SYSTEM
     )

install.sh

@@ -703,7 +703,6 @@ else
             warn "镜像加速已配置，但拉取仍然失败，可能原因："
             echo -e "   1. 镜像源暂时不可用，请稍后重试"
             echo -e "   2. 网络连接问题"
-            echo -e "   3. 镜像不存在或版本错误"
         fi
         echo
         exit 1