Having score open will eventually crash

• Sep 27, 2019 - 06:33

I have this score that Musescore can open just fine. However, having it open will eventually crash the program in some way. Just that one file, not any others or previous versions of it. It was quite a sudden development. Simply leaving it open will crash in about three minutes if you do nothing. Saving will always crash. 1–40 seconds into playback it will crash. Removing a part seems to have a ¼ probability of crashing. I appear to be able to edit the score itself just fine—Musescore sees no issue with it; it hasn't told me that it's been corrupted yet. I've tried reinstalling, a factory reset, and disabling extensions (but I haven't tested the dev branch yet). It still crashes. It's made me even more annoyed with the fact that Musescore doesn't save any user, menu, or playback settings until it closes properly (e.g. not having a crash or hard shutdown, both of which are seemingly ubiquitous for my device).
Cursed score: Minute.mscz

I'm interested to find out if it dislikes only me or if it has been "thoroughly corrupted."


OS: Windows 10 (10.0), Arch.: x86_64, MuseScore version (64-bit): 3.2.3.7635, revision: d2d863f


Comments

Most likely cause: one part is corrupt (unfortunatly MuseScore dodn't report this, it just reports corruption in score, not in parts), and the crash happens on every (auto)save (so by default after 2 minutes). Cure: remove parts, generate them afresh

In reply to by hdt80bro

" How did you know that it was the piccolo part? Trial and error? It looks fine"

It looks good, apparently, in the piccolo part. But not at all when disabling the multimeasure rests and look at measure 77/78 (it is empty, it is the sign of corruption - image below)
Do you remember anything you would have done around this measure/this area of measures? In part/parts or in main score (and this can be all kinds of operations, if you did it once, you must be able to do it a second time! According to the habit you have of doing things)

empty.jpg

In reply to by hdt80bro

I had just finished over‐dramatizing the build up before measure 78. I just compared with the previous version—the piccolo part there was added some time after the version change. However, it's just a whole note before the problem measures. That can't be the cause.

I looked at the horn part. That was probably the cause.
Horn_Before.png
Which I changed to match how it felt (while keeping the invisible playback notes)
Horn_After.png
(the invisible mess is because I had to use two voices to achieve acciaccatura chords. I see that the natural sign on the E at the end can contribute)

*I blame the glissando, which I remember had issues maintaining correct positioning. The horns must have been complaining that their valves can't move that fast, and envied the picc.

Thanks for pointing that out guys! Now I know to press M to find problems.

In reply to by hdt80bro

" I just compared with the previous version—the piccolo part there was added some time after the version change. "

Before this change, instead of piccolo part (simply added as part and moved to top? Or created from scratch and then added as top part?): what part was the top staff?
Flute as now, or another?

In reply to by hdt80bro

"Perhaps I should have said that I added in the passage for the piccolo part (before it was just 58 measures of resting till the end)."
Indeed, better info. And thanks for helping. To deeper a bit, could you provide some details on how exactly was added this passage - from measure 64 I guess to measure 77 (from the flute, partially - I see two voices in this part sometimes, with a few common notes - or other ?)
Each detailed step, eg a copy-paste somewhere - on measure/or precise beat, with trial/Undo and so on, and with enabled mm rests or not, and with saving/reloading or not in this sequence -, can be important.

Do you still have an unanswered question? Please log in first to post your question.