生成非托管代码崩溃的转储？ [英] Generate dump with unmanaged code crash?

问题描述

当我的应用程序突然崩溃时，有没有办法得到一个完整的崩溃转储？

问题是我怀疑是由于非托管代码杀死了.net框架本身，因此，除非有一个调试器附加到应用程序在崩溃的时刻，应用程序甚至没有机会处理崩溃。

我无法附加调试器并等待，因为崩溃随机发生，我们有很多安装位置，所以我想知道是否可以这样做，而不附加一个调试器。

编辑：我知道Microsoft调试SDK，而我也刚刚发现了关于DebugDiag。我想这样做，而不必每次附加到应用程序，因为它通常在销售的中间，我没有时间附加调试器并探索堆栈跟踪。 DebugDiag似乎让我为此设置一个规则，但它似乎只适用于已经运行的进程，而不是等待一些进程启动。

解决方案

您可以使用 adplus -crash 或 userdump.exe 。

Is there a way that I can get a full crash dump when my application crashes suddenly?

The problem is that I suspect that it's due to unmanaged code that kills the .net framework itself, and because of that, unless there's a debugger attached to the application at the moment of the crash, the application doesn't even gets a chance to handle the crash.

I can't attach a debugger and wait, because the crash happens randomly and we have many installed locations, so I'm wondering if it's possible to do this without attaching a debugger.

Edit: I'm aware of Microsoft Debugging SDK, and I also just found out about DebugDiag. I'd like to do this without having to attach to the application each time, as it's usually in the middle of a sale and I don't have time to attach the debugger and explore the stack trace. DebugDiag does seem to let me setup a rule for this, but it seems to only work on already running processes, rather than waiting for some process to launch.

解决方案

You can use adplus -crash or userdump.exe.

这篇关于生成非托管代码崩溃的转储？的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！