You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
dask scheduler not recognized as a valid command, despite suggestion from FutureWarning #7176
When starting a dask scheduler the old way via dask-scheduler, it produces a future warning as such: FutureWarning: dask-scheduler is deprecated and will be removed in a future release; use 'dask scheduler' instead.
However, then when running dask scheduler from the command line produces:
Usage: dask [OPTIONS] COMMAND [ARGS]...
Try 'dask -h' for help.
Error: No such command 'scheduler'.
Thanks for reporting @DamianBarabonkovQC. FWIW this appears to be a conda-specific issue, so pip installing is a temporary workaround. We're working on a fix for the conda packages. Will keep this issue posted.
Would suggest just using the old entry points (with -s like dask-scheduler) as they still work. If we don't find a resolution soon, would suggest dropping/softening the deprecation warning (we don't appear to be ready for that).
Activity
jrbourbeau commentedon Oct 24, 2022
Thanks for reporting @DamianBarabonkovQC. FWIW this appears to be a
conda-specific issue, sopipinstalling is a temporary workaround. We're working on a fix for thecondapackages. Will keep this issue posted.jakirkham commentedon Oct 25, 2022
As James noted, this is a known issue. There have been 2 attempts to fix it, but no success so far. Related threads:
xref: conda-forge/distributed-feedstock#225 (comment)
xref: conda-forge/distributed-feedstock#226
xref: conda-forge/dask-core-feedstock#144
Would suggest just using the old entry points (with
-s likedask-scheduler) as they still work. If we don't find a resolution soon, would suggest dropping/softening the deprecation warning (we don't appear to be ready for that).jakirkham commentedon Oct 25, 2022
Actually maybe PR ( conda-forge/dask-core-feedstock#146 ) will solve it? Would be good to have others take a look 🙂
douglasdavis commentedon Oct 25, 2022
I'm relatively confident conda-forge/dask-core-feedstock#146 should end up fixing things