I have a microsd to usb adapter here that consistently causes Opus to hang in an infinite busy loop (circle mouse pointer). This happens only when I right-click on the associated drive letter. The context menu never shows up, instead Opus hangs.
Right-clicking on the drive letter in Windows Explorer works fine, and the context menu shows up.
I would love for this to be fixed in a future version. How shall I proceed to help the team debug this? I need step by step instructions and will upload logs or whatever is needed.
For example, I saw that there's an advanced option "context menu debug" and wonder if I should turn this on to generate a log file.
The crash will be because of a shell extension something has installed that is triggered by the drives menu.
The debugging mode in Opus won't help here as it's only for context menus that Opus builds itself. The context menus for the drive roots are a special case where Opus just asks the shell to display the menu. (Which makes it unusual for something to crash in Opus but not Explorer, but shell extensions often have bugs that only get triggered in one program or the other due to something fairly random.)
To track this down, use ShellExView, sort by Type, and disable all Context Menu type extensions. Reboot and see if the problem still happens. If it doesn't, re-enable a few at a time to work out which extension is causing the crash.
If you have any backup tools (e.g. Acronis True Image) or disk defrag tools, or cloud storage tools in that list, I would suspect them first as their shell extensions are more often the cause of problems than other types of software.
If there are crash dumps, you can also send those to us to look at, which sometimes let us tell you which extension is causing the crash: Crash dumps for bug reports
I have not been able to reproduce the bug. It seems to happen consistently and reproducibly for a few hours. But then it doesn't happen for days.
Next time it happens I will try to debug it using ShellExView, like you suggested.
Have any crash dumps been created automatically?
No, since it didn't crash.
I have been going for a long time now without any more occurrence of the bug. My only explanation is that maybe some Windows update fixed it?
Ah, of course; my mistake.
If it does hang again, you can generate a crash dump of the process at that moment in time using Task Manager.
Manually generated dumps in Crash dumps for bug reports explains how.
It could well be a Windows update as some instability we were seeing with shell extensions that have not changed in a long time went away mysteriously, and until then it was looking like something in the way Windows 10 initializes DLLs (in parallel, which is faster but a lot more complex) may have changed.