Why don't we just fix the "Bad gateway issue"?

It's been almost half a year since the first user reported the "Bad gateway issue". It has caused a lot of frustration for both managers and us, developers.

Due to the lack of information regarding the issue, it seems only natural that a lot of people have thought: Why don't they just fix that bug and get on?

Unfortunately, it is not that simple.

Inspired by an idea from Valde aka Manchester United, I'd like to tell a bit more about the problem.

The following should not be understood as an excuse. It is our responsibility that this site runs perfectly no matter what, and we take that responsibility vey seriously. But considering the criticism we've received regarding the "Bad gateway issue" it is obvious that most people don't understand the true nature of it. And with this post, I'd like to try to improve this.

In software, there is a theory called software rot. To put it briefly, the theory dictates that software will deteriorate quality wise gradually over time as you change and add new features. Rotten software is often unstable, slow, or hard to work with.

Of course, it is never your intension to leave a piece of code in poorer condition than you received it, but in practice that is often very hard to avoid.

As with all "living" software project, Virtual Manager is hit by this phenomenon. The older and bigger the project, the more you must watch out for software rot. And the more resources you must allocate to remedy it. In part, that is why the development of Virtual Manager today is slower than it was five years ago.

The best way to remedy software rot, is by doing regular house cleaning on the codebase. Simply put, you rewrite a piece of code without changing its actual functionality. It might sound like a waste of time, but it is necessary to avoid rot.

When you experience a "Bad gateway issue" it is not due to any specific error. It is a symptom of the entire Virtual Manager platform having performance problems. When several of the rotten areas of the code base is stressed at the same time that brings the servers to their knees, and they have no other option than to return "Bad gateway" to the user.

So because of this, we cannot "just" fix the error. It is continuous process that we work on daily - one piece at the time. It is a difficult task that takes time and we make progress every day. We hope you'll bear with us.

In return, I promise you that we do everything we can to give you the most stable site possible.