回滚事务时出现异常 - 连接已关闭？

【问题标题】：Exceptions when rolling back a transaction - connection already closed?回滚事务时出现异常 - 连接已关闭？
【发布时间】：2015-09-22 22:48:54
【问题描述】：

使用 Entity Framework 6.0.0，我在关闭事务时看到异常。

我们一直在对表进行并发更改时遇到问题，所以我将它包装在一个事务中，现在我在回滚时遇到异常。

代码：

public LockInfo getSharedLock(string jobid)
{
    using (var myDbContext = new MyDbContext())
    {
        using (var transaction = myDbContext.Database.BeginTransaction())
        {
            try
            {
                this.logger.log("Attempting to get shared lock for {0}", jobid);

                var mylocks =
                    myDbContext.joblocks.Where(j => j.customerid == this.userContext.customerid)
                        .Where(j => j.jobid == jobid)
                        .Where(j => j.operatorid == this.userContext.operatorid);

                var exclusiveLock = mylocks.FirstOrDefault(
                    j => j.lockstatus == LockInfo.LockState.Exclusive);
                if (exclusiveLock != null)
                {
                    this.logger.log("{0} already had exclusive lock, ignoring", jobid);
                    return LockInfo.populate(exclusiveLock);
                }

                var sharedLock = mylocks.FirstOrDefault(
                    j => j.lockstatus == LockInfo.LockState.Shared);
                if (sharedLock != null)
                {
                    this.logger.log("{0} already had shared lock, ignoring", jobid));
                    sharedLock.lockdt = DateTime.Now;
                    myDbContext.SaveChanges();

                    return LockInfo.populate(sharedLock);
                }

                var joblock = new joblock
                {
                    customerid = this.userContext.customerid,
                    operatorid = this.userContext.operatorid,
                    jobid = jobid,
                    lockstatus = LockInfo.LockState.Shared,
                    sharedLock.lockdt = DateTime.Now
                };

                myDbContext.joblocks.Add(joblock);
                myDbContext.SaveChanges();
                transaction.Commit();

                this.logger.log("Obtained shared lock for {0}", jobid);
                return LockInfo.populate(joblock);
            }
            catch (Exception ex)
            {
                transaction.Rollback();
                this.logger.logException(ex, "Exception in getSharedLock(\"{0}\")", jobid);
                throw;
            }
        }
    }
}

您可以在上面的代码中看到日志记录。我们也在数据库中启用了日志记录。日志跟踪：

===================
NORMAL    TicketLockController.getLock("AK2015818002WL")
===================
SQL    Opened connection at 9/22/2015 2:47:49 PM -05:00
===================
SQL    Started transaction at 9/22/2015 2:47:49 PM -05:00
===================
NORMAL    Attempting to get shared lock for AK2015818002WL
===================
SQL    SELECT TOP (1) [Extent1].[customerid] AS [customerid]
    ,[Extent1].[jobid] AS [jobid]
    ,[Extent1].[lockdtdate] AS [lockdtdate]
    ,[Extent1].[lockdttime] AS [lockdttime]
    ,[Extent1].[operatorid] AS [operatorid]
    ,[Extent1].[lockstatus] AS [lockstatus]
    ,[Extent1].[changes] AS [changes]
FROM [dbo].[joblock] AS [Extent1]
WHERE ([Extent1].[customerid] = 'TESTTK')
    AND ([Extent1].[jobid] = 'AK2015818002WL')
    AND ([Extent1].[operatorid] = 'ADMIN')
    AND (N'Exclusive' = [Extent1].[lockstatus])
===================
SQL    SELECT TOP (1) [Extent1].[customerid] AS [customerid]
    ,[Extent1].[jobid] AS [jobid]
    ,[Extent1].[lockdtdate] AS [lockdtdate]
    ,[Extent1].[lockdttime] AS [lockdttime]
    ,[Extent1].[operatorid] AS [operatorid]
    ,[Extent1].[lockstatus] AS [lockstatus]
    ,[Extent1].[changes] AS [changes]
FROM [dbo].[joblock] AS [Extent1]
WHERE ([Extent1].[customerid] = 'TESTTK')
    AND ([Extent1].[jobid] = 'AK2015818002WL')
    AND ([Extent1].[operatorid] = 'ADMIN')
    AND (N'Shared' = [Extent1].[lockstatus])
===================
SQL    INSERT [dbo].[joblock] (
    [customerid]
    ,[jobid]
    ,[lockdtdate]
    ,[lockdttime]
    ,[operatorid]
    ,[lockstatus]
    ,[changes]
    )
VALUES (
    @0
    ,@1
    ,@2
    ,@3
    ,@4
    ,@5
    ,NULL
    )
===================
SQL    Closed connection at 9/22/2015 2:47:50 PM -05:00
===================
EXCEPTION    Unhandled exception caught: The underlying provider failed on Rollback.
===================
EXCEPTION    Inner Exception: Value cannot be null.
Parameter name: connection

两次选择都成功，然后插入由于某种原因失败。抛出异常，并且由于某种原因连接在 Rollback() 执行之前关闭。

任何想法我做错了什么？

==== 添加堆栈跟踪 ====

外部异常的堆栈跟踪：

   at System.Data.Entity.Core.EntityClient.EntityTransaction.Rollback()
   at korterra.kt_api.Shared.TicketLockWrangler.getSharedLock(String jobid)
   at korterra.kt_ws.ApiControllers.Shared.TicketLockController.getSharedLock(TicketLockDTO ticketLockDTO)
   at lambda_method(Closure , Object , Object[] )
   at System.Web.Http.Controllers.ReflectedHttpActionDescriptor.ActionExecutor.<>c__DisplayClass10.<GetExecutor>b__9(Object instance, Object[] methodParameters)
   at System.Web.Http.Controllers.ReflectedHttpActionDescriptor.ActionExecutor.Execute(Object instance, Object[] arguments)
   at System.Web.Http.Controllers.ReflectedHttpActionDescriptor.ExecuteAsync(HttpControllerContext controllerContext, IDictionary`2 arguments, CancellationToken cancellationToken)
--- End of stack trace from previous location where exception was thrown ---
   at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at System.Web.Http.Tracing.ITraceWriterExtensions.<TraceBeginEndAsyncCore>d__18`1.MoveNext()
--- End of stack trace from previous location where exception was thrown ---
   at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at System.Web.Http.Controllers.ApiControllerActionInvoker.<InvokeActionAsyncCore>d__0.MoveNext()
--- End of stack trace from previous location where exception was thrown ---
   at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at System.Web.Http.Tracing.ITraceWriterExtensions.<TraceBeginEndAsyncCore>d__18`1.MoveNext()
--- End of stack trace from previous location where exception was thrown ---
   at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at System.Web.Http.Controllers.ActionFilterResult.<ExecuteAsync>d__2.MoveNext()
--- End of stack trace from previous location where exception was thrown ---
   at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at System.Web.Http.Filters.AuthorizationFilterAttribute.<ExecuteAuthorizationFilterAsyncCore>d__2.MoveNext()
--- End of stack trace from previous location where exception was thrown ---
   at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at System.Web.Http.Controllers.ExceptionFilterResult.<ExecuteAsync>d__0.MoveNext()

内部异常的堆栈跟踪：

   at System.Data.Entity.Utilities.Check.NotNull[T](T value, String parameterName)
   at System.Data.Entity.Infrastructure.Interception.DbTransactionInterceptionContext.WithConnection(DbConnection connection)
   at System.Data.Entity.Infrastructure.Interception.DbTransactionDispatcher.Rollback(DbTransaction transaction, DbInterceptionContext interceptionContext)
   at System.Data.Entity.Core.EntityClient.EntityTransaction.Rollback()

【问题讨论】：

  • 发布完整的异常 ToString。您的错误记录不足。
  • 另外，在很多情况下，您并没有提交事务。真的是故意的吗？
  • 我无法按需重新创建。这是在我们的 QA 环境中间歇性出现的东西。我在开发中从未见过它。除了我已经发布的信息之外，我拥有的唯一信息是堆栈跟踪，它只是表明我发布的代码中发生了异常。
  • 我没有提交不更改数据库的执行路径。但也许我应该。
  • 好的，我找不到错误。你迫切需要修改你的错误日志，这样它就不会丢弃这么多信息。从堆栈跟踪中，我们可以看到发生这种情况的调用和行。

标签： c# transactions entity-framework-6


【解决方案1】：

讨论后，我开始记录异常，然后尝试回滚 - 这揭示了错误。

事务已死锁：

Exception in getSharedLock("ticketnumber123456"): An error occurred while updating the entries. See the inner exception for details.

Inner Exception: An error occurred while updating the entries. See the inner exception for details.

Inner Exception: Transaction (Process ID 139) was deadlocked on lock resources with another process and has been chosen as the deadlock victim. Rerun the transaction.

从我所读到的，当你告诉一个事务已经成为死锁的牺牲品时，它已经被回滚了。也许这就是我们得到异常的原因？

修复似乎是识别我们何时死锁而不回滚，或者不使用事务，并在我们遇到主键违规时重试。

【讨论】：

  • 你不应该在这段代码中调用 Rollback。此外，这是一个 EF 错误。它不应该崩溃。它应该报告 API 使用错误或使用干净异常的东西。您能否为此在 EF 存储库中打开一个 GitHub 问题？
  • 我目前正在尝试不使用事务，而只是在底层记录发生更改时重试 - 乐观锁定。
  • @usr 你能解释一下为什么不应该“在这段代码中调用回滚” - 这是什么意思？这是对回滚功能的错误使用吗？
  • 如果服务器检测到两个事务已经死锁，它会选择其中一个来杀死。当它这样做时，它将回滚事务并向客户端发送错误。作为客户端，您在这种情况下不希望回滚事务，因为事务已经回滚，再次尝试会产生另一个错误。
猜你喜欢
  • 1970-01-01
  • 2015-02-11
  • 2019-12-24
  • 2020-09-24
  • 2016-08-14
  • 2013-12-31
  • 2015-10-03
  • 2013-10-16
  • 2015-12-22
相关资源
最近更新 更多
热门标签
Java Python linux javascript Mysql C# Docker 算法 前端 SpringBoot Redis Vue spring 设计模式 .net core .net kubernetes c++ 数据库 数据结构 大数据 js 机器学习 微服务 Android Go 程序员 面试 JVM ASP.net core 云原生 人工智能 后端 PHP git CSS golang k8s Nginx Django mybatis 深度学习 多线程 React 架构 devops 爬虫 云计算 Spring Boot LeetCode