Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

NetCDF failing in Windows tests, blocking merging of PRs #527

Closed
ijpulidos opened this issue Nov 10, 2021 · 2 comments
Closed

NetCDF failing in Windows tests, blocking merging of PRs #527

ijpulidos opened this issue Nov 10, 2021 · 2 comments

Comments

@ijpulidos
Copy link
Contributor

For months our CI tests have been failing specifically with the windows tests complaining about issues with accessing netcdf files, such as in https://github.com/choderalab/openmmtools/runs/4155555651?check_suite_focus=true with messages like

openmmtools.multistate.multistatereporter: DEBUG: Attempt 4/5 to open C:\Users\RUNNER~1\AppData\Local\Temp\tmpno1p10jq\test_storage.nc failed. Retrying in 2 seconds
openmmtools.utils: WARNING: [WinError 32] The process cannot access the file because it is being used by another process: 'C:\\Users\\RUNNER~1\\AppData\\Local\\Temp\\tmpno1p10jq\\test_storage.nc'

and

OSError: [Errno -51] NetCDF: Unknown file format: b'C:\\Users\\RUNNER~1\\AppData\\Local\\Temp\\tmpm5j5mar2\\test_storage.nc'

So, I guess we have to ask ourselves if we want to keep supporting Windows and if so, then we want to set up environments that allow us to test this locally. Personally I use Linux as my development OS.

This is now blocking some of our current workflow to merge PRs, because we would have to manually override the windows tests if we want things to be merged.

@mikemhenry
Copy link
Contributor

For now I'm going to disable the windows tests as a requirement so we can get some of these PRs merged in.

Lnaden added a commit to Lnaden/openmmtools that referenced this issue Dec 13, 2021
This PR attempts to fix the issue raised in choderalab#527. From what I can tell,
the tests where this is thrown always fails exactly at the same point.
This PR closes the reporter file formally before the failing line
occurs, so hopefully this will correctly allow Window's memory to
release the file.
mikemhenry added a commit that referenced this issue Dec 14, 2021
This PR attempts to fix the issue raised in #527. From what I can tell,
the tests where this is thrown always fails exactly at the same point.
This PR closes the reporter file formally before the failing line
occurs, so hopefully this will correctly allow Window's memory to
release the file.

Co-authored-by: Mike Henry <11765982+mikemhenry@users.noreply.github.com>
@ijpulidos
Copy link
Contributor Author

I think we can close this one. Solved in #535 .

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants