distributed icon indicating copy to clipboard operation
distributed copied to clipboard

A distributed task scheduler for Dask

Results 530 distributed issues
Sort by recently updated
recently updated
newest added

- Mutually exclusive with #6699 - Mutually exclusive with #6716 - Closes #6682 - Closes #6689 - Closes #6693 - Closes #6685 - Closes #6708 - Closes #6709 - Closes...

cuDF recently changed and it now creates a cuda context at import. This is necessary for a variety of reasons, most importantly is validation -- cuDF wants to validate if...

``` _______________________ test_quiet_close_process[True] ________________________ processes = True tmp_path = WindowsPath('C:/Users/runneradmin/AppData/Local/Temp/pytest-of-runneradmin/pytest-0/test_quiet_close_process_True_0') @pytest.mark.slow @pytest.mark.parametrize("processes", [True, False]) def test_quiet_close_process(processes, tmp_path): with open(tmp_path / "script.py", mode="w") as f: f.write(client_script % processes) with popen([sys.executable, tmp_path...

flaky test

Closes #6874 - [x] Tests added / passed - [x] Passes `pre-commit run --all-files`

https://github.com/dask/distributed/blob/main/distributed/dashboard/components/scheduler.py#L3293 fails if `ts.erred_on` is `None`. ``` Aug 11 13:40:51 ip-10-0-9-12 cloud-init[1262]: 2022-08-11 13:40:51,814 - bokeh.util.tornado - ERROR - Error thrown from periodic callback: Aug 11 13:40:51 ip-10-0-9-12 cloud-init[1262]: 2022-08-11...

bug
diagnostics

Related to https://github.com/dask/distributed/issues/1932, but it asks to silence for different API component and for all warnings. I have a setup where I expect some jobs to fail, but I do...

Closes #xxxx - [ ] Tests added / passed - [ ] Passes `pre-commit run --all-files` very drafty experiment just to see what breaks see also: - https://github.com/dask/distributed/issues/6785 - https://github.com/dask/distributed/issues/6201...

asyncio

The resumed state is exceptionally complicated and a very frequent source of problems. This PR removes the 'resumed' state and the `TaskState.next` attribute. This PR also deals with the issue...

- Mutually exclusive with #6716 - Mutually exclusive with #6844 - Closes #6682 - Closes #6689 - Closes #6693

When a worker breaks its connection to the scheduler, or goes too long without heartbeating, the scheduler removes it and reschedules its tasks to run elsewhere. However, other works may...

discussion
deadlock