ASP.NET Core 8.0 Web API部署K8s后出现System.NullReferenceException异常

一、异常情况说明 #

你的应用运行一段时间后（时长随机）会触发两类未处理异常导致终止重启：

HttpClient连接池相关异常 #

Unhandled exception in Main: System.NullReferenceException: Object reference not set to an instance of an object.
at System.Collections.Concurrent.ConcurrentDictionary`2.Enumerator.MoveNext()
at System.Net.Http.HttpConnectionPoolManager.RemoveStalePools()
at System.Net.Http.HttpConnectionPoolManager.<>c.<.ctor>b__11_0(Object s)
at System.Threading.TimerQueueTimer.Fire(Boolean isThreadPool)
at System.Threading.TimerQueue.FireNextTimers()

Kestrel连接管理相关异常 #

System.AggregateException: A Task's exception(s) were not observed either by Waiting on the Task or accessing its Exception property. As a result, the unobserved exception was rethrown by the finalizer thread. (Object reference not set to an instance of an object.)
System.NullReferenceException: Object reference not set to an instance of an object.
at Microsoft.AspNetCore.Server.Kestrel.Core.Internal.Infrastructure.ConnectionManager.RemoveConnection(Int64 id)
at Microsoft.AspNetCore.Server.Kestrel.Core.Internal.Infrastructure.TransportConnectionManager.RemoveConnection(Int64 id)
at Microsoft.AspNetCore.Server.Kestrel.Core.Internal.Infrastructure.KestrelConnection`1.ExecuteAsync()
--- End of inner exception stack trace ---

这些异常均来自.NET或ASP.NET Core内部组件的并发逻辑，并非业务代码直接触发，但可通过配置调整或版本升级修复。

二、解决方案 #

1. 处理HttpClient连接池异常 #

• 调整HttpClient连接池配置：显式设置连接池参数，避免默认值的并发隐患。在Program.cs中配置：

  builder.Services.AddHttpClient("MyClient")
      .ConfigurePrimaryHttpMessageHandler(() => new HttpClientHandler
      {
          MaxConnectionsPerServer = 100,
          PooledConnectionIdleTimeout = TimeSpan.FromMinutes(5),
          KeepAlivePingTimeout = TimeSpan.FromSeconds(30)
      });
  

  确保所有HttpClient实例通过DI容器获取，避免手动创建长期持有的静态实例。
• 升级.NET 8补丁版本：ConcurrentDictionary枚举时的NullReferenceException是.NET运行时已知并发bug，升级到最新的.NET 8稳定补丁版本（如8.0.x的最新更新）可修复此类问题。

2. 处理Kestrel连接管理异常 #

• 调整Kestrel连接限制：在Program.cs中配置Kestrel的连接超时、最大连接数，减少并发冲突：

  builder.WebHost.ConfigureKestrel(options =>
  {
      options.Limits.MaxConcurrentConnections = 1000;
      options.Limits.MaxConcurrentUpgradedConnections = 1000;
      options.Limits.IdleTimeout = TimeSpan.FromMinutes(2);
  });
  

• 切换HTTP协议版本测试：若使用HTTP/2，暂时切换到HTTP/1.1排查是否为协议层并发问题：

  builder.WebHost.ConfigureKestrel(options =>
  {
      options.ListenAnyIP(5000, o => o.Protocols = HttpProtocols.Http1);
  });
  

• 升级ASP.NET Core补丁版本：Kestrel ConnectionManager的NullReferenceException同样可能是框架已知bug，升级补丁版本是优先解决方案。

3. K8s环境优化 #

• 调整探针配置：延长存活/就绪探针的超时和间隔，避免应用临时波动时被误重启：

  livenessProbe:
    httpGet:
      path: /healthz
      port: 5000
    initialDelaySeconds: 30
    periodSeconds: 10
    timeoutSeconds: 5
  readinessProbe:
    httpGet:
      path: /healthz
      port: 5000
    initialDelaySeconds: 10
    periodSeconds: 5
    timeoutSeconds: 3
  

• 配置合理资源限制：确保Pod有足够CPU和内存，避免资源不足引发的GC异常或组件崩溃：

  resources:
    requests:
      memory: "256Mi"
      cpu: "100m"
    limits:
      memory: "512Mi"
      cpu: "500m"
  

三、验证步骤 #

1. 优先升级.NET 8到最新补丁版本，这是修复框架内部bug最直接的方式。
2. 应用上述配置调整后部署到K8s，观察异常是否复现。
3. 若问题持续，开启Debug级日志收集更多上下文：

   builder.Logging.AddConsole().SetMinimumLevel(LogLevel.Debug);
   

   分析日志中异常发生前的警告或错误信息，进一步定位问题。

内容的提问来源于stack exchange，提问作者Rizwan Hanif