Hi, thank for sharing this issue.

These symptoms suggest that the engine host seems to be encountering a problem when trying to shut itself down. This is not normal behaviour. I'm wondering if you might be able to collect some more information for me to help narrow down this problem:

- Can you try submitting an NCrunch bug report once the engine is in a hung state after trying to shut down? If the process is still partly functioning, we may be able to extract a log from it
- Does the problem appear if you close down the solution, wait for a few seconds, then open the other solution instead of swapping directly between them?
- Can you notice any pattern around when the problem appears? For example, does it seem to be tied to the closing down of a specific solution?
- How large is the NCrunch .cache file for the solutions involved? (can be found in the _NCrunch directory next to your solution file)

Changing your 'Engine hosting strategy' global config setting to 'HostInIDE' will change the workflow around how the engine shuts itself down. This may give you a way to work around the problem, and it might also be useful to know if the problem can still appear with this setting changed.

Thanks for the extra details. So I guess this is a performance issue then. Here's something we can try:

1. Just before you close the solution, go to your NCrunch global configuration settings and turn on 'Log to output window'. Set the log verbosity to 'Detailed'.
2. Then close the solution and wait for a few minutes for the engine to completely shut down
3. Open the Visual studio output window and choose the NCrunch diagnostic output
4. This should show a detailed log of everything the engine does while it's shutting down
5. You're very welcome to thumb through this to see if you can find where it's stuck at, or alternatively, you can zip up the log and send it through the contact form and I'll take a look.

Thanks for sending through the log! This was very revealing.

NCrunch is taking an insanely long time to clean up its workspaces. For example:

ENGINE - [11:30:49.7694-Core-44] Cleaning up workspace: D:\Stash\NCrunch\11136\3
ENGINE - [11:31:17.2331-Core-44] Publishing Event: [WorkspaceDeletedEvent:D:\Stash\NCrunch\11136\3]

Normally, on a desktop machine, it takes less than a couple of seconds to clean up a workspace. Because the delay seems to be so high for all of your workspaces rather than just one, I think this may be a system-wide/hardware performance issue.

Do you have your workspace base path set to an external USB drive? Even for a low spec machine, these cleanup times are pretty extreme. Your antivirus may be suspect, though it does seem strange that an antivirus would take issue to directories being deleted.

You may want to consider using NCrunch's distributed processing feature to offload the work onto a different machine. I don't doubt that this I/O issue will also impact the performance of the engine when it runs.