This is a guest blog post from Michael Shpilt. Michael has been developing software for over 20 years. He owns the popular blog michaelscodingspot.com and recently published the book Practical Debugging for .NET Developers. In his day job, he works at OzCode, a company dedicated to improving debugging in .NET.
Program hangs are some of the worst possible things you can have. They are terrible in both desktop and web applications. With desktop applications, your window freezes and the app becomes unresponsive and basically useless. This is unpleasant for technical users like us programmers, but I imagine it’s awful for regular users. In a web application, when a hang occurs, your request remains stuck without ever returning. When more similar requests get stuck, the application eventually slows down to a crawl. Requests might start to fail altogether until the web server restarts and this evil cycle starts all over again.
I hope you won’t ever have to experience hangs, especially in production, but chances are that you will. And in the event that you do, you’ll need to know how to deal with them. In this article, we’ll see one of the best methodologies for debugging hangs. We’ll see how to use a performance profiler like dotTrace to find the root cause of the hang and fix it. There are many advantages to this, but first, let’s talk about the traditional way to debug hangs.