spurious ci failure: The process cannot access the file because it is being used by another process

[matthiaskrgr]

See https://github.com/rust-lang/rust/pulls?q=is%3Apr+%22The+process+cannot+access+the+file+because+it+is+being+used+by+another+process%22+is%3Aclosed

seems to happen on x86_64-msvc-exta lot

[RUSTC-TIMING] miri test:false 64.073
[RUSTC-TIMING] miri test:false 2.939
    Finished `release` profile [optimized] target(s) in 0.39s
thread 'main' panicked at src/lib.rs:1697:17:
failed to copy `C:\a\rust\rust\build\x86_64-pc-windows-msvc\stage1-tools\x86_64-pc-windows-msvc\release\miri.exe` to `C:\a\rust\rust\build\x86_64-pc-windows-msvc\stage1-tools-bin\miri.exe`: The process cannot access the file because it is being used by another process. (os error 32)
##[endgroup]
Build completed unsuccessfully in 0:00:11
  local time: Sat, Jun 29, 2024  5:48:11 PM
  network time: Sat, 29 Jun 2024 17:48:11 GMT

[ChrisDenton]

The bootstrap code that panics hasn't changed in years:

rust/src/bootstrap/src/lib.rs

Lines 1696 to 1698 in d1b7355

	if let Err(e) = fs::copy(&src, dst) {
	panic!("failed to copy `{}` to `{}`: {}", src.display(), dst.display(), e)
	}

so I doubt that's the cause but a workaround would be to add a retry loop.

It would be interesting to know what's changed recently to cause this. "os error 32" is a sharing violation, which is usually because the file is locked in certain way. But, as far as I'm aware, we don't do that kind of locking. Maybe someone who knows bootstrap well has some ideas.

Edit: I guess it could be a cargo issue and not bootstrap
Edit2: Worse case it's a CI thing where some sort of malware scanner is locking the file