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

ERROR! java.io.IOException: Unable to delete file: .nfs #2

Open
Sebastien-Raguideau opened this issue Nov 25, 2022 · 4 comments
Open
Assignees
Labels
bug Something isn't working

Comments

@Sebastien-Raguideau
Copy link

I am encountering small issue with makes me unable to use ufcg on more than a file at a time.

After profiling a genome and writing results there is an attempt to delete temporary files, this fail in a non deterministic way and may be partially related to my server file system speed in closing handle on files.

I suppose ufcg attempt to remove these files too quickly or without having closed them beforehand? In any way, it interrupt execution and I need to work around it by launching ufcg on a unique genome at a time.

@endixk
Copy link
Member

endixk commented Nov 28, 2022

Ideally such parallel execution should run without an error, as each of them use randomly generated headers. Still, running multiple processes with writing/deleting operations on the same working directory is always problematic (especially when the directory is mounted across the network).

I would recommend you to use the batch input mode with sufficient number of CPU threads instead (profile -i <dir> -t <cpu>), which can perform fairly similar compared to the parallel execution.

@Sebastien-Raguideau
Copy link
Author

Sebastien-Raguideau commented Nov 28, 2022

Sorry, I was not clear I do use batch execution and the issue interrupts it. So instead I've had to go through external parallel execution of ufcg. When doing external parallel execution, the same issue arises but since it does not impact results and just exit code, it doesn't matter.

I don't code in java but this issue should be easy to fix. At the end of profiling, you delete temporary files too quickly after closing them. Or at least too quickly for my filesystem. The fix would be to add a waiting time of 1 second before removing the temporary directory folder.

I you don't, batch input mode becomes meaningless since your pipeline stops when unable to delete temporary folder. Alternatively you could try to catch that error.

@endixk endixk self-assigned this Nov 30, 2022
@endixk
Copy link
Member

endixk commented Nov 30, 2022

I got the point now. Thank you for the detailed description.
As you mentioned this won't be too difficult to fix. I will try to deal with this soon.

@endixk endixk added the bug Something isn't working label Nov 30, 2022
@endixk
Copy link
Member

endixk commented Dec 20, 2022

@Sebastien-Raguideau Could you please reproduce the error with the newest version?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

2 participants