Skip to content

LoggingThread crashed with OSError: [Errno 5] Input/output error, worker continued #305

Description

@albertz

In the worker LoggingThread:

Exception in thread Thread-2:
Traceback (most recent call last):
  File ".../python3.12/threading.py", line 1073, in _bootstrap_inner
    self.run()
  File ".../tools/sisyphus/sisyphus/worker.py", line 130, in run
    log_usage(resources)
  File ".../tools/sisyphus/sisyphus/worker.py", line 78, in log_usage
    with open(usage_file_path, "w") as usage_file:
         ^^^^^^^^^^^^^^^^^^^^^^^^^^
OSError: [Errno 5] Input/output error: '.../work/i6_core/returnn/training/ReturnnTrainingJob.1KSN5KXuom2X/usage.run.1'

(via HPC /rwthfs/rz/cluster/hpcwork/p0023565/denoising_lm/sis_work/2026-05-28-tts-encoder-work/i6_core/returnn/training/ReturnnTrainingJob.1KSN5KXuom2X/engine/i6_core.returnn.training.ReturnnTrainingJob.1KSN5KXuom2X.run.1942041.1)

The job however seemed to continue for a while. I guess some random FS hiccup.

I'm just reporting here in case I get hit by this again. Not sure yet about what would be the best solution.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions