dask/distributed

Issues

• Warn (and eventually raise) when client.scatter is used with Active Memory Manager enabled

  #8921 opened 17 days ago by phofl
  3

• Simple P2P Merge Fails if Scheduler Is Configured

  #8919 opened 22 days ago by andoni-garcia-fgp
  0

• Reuse worker to scheduler WS connection for scheduling

  #8918 opened 23 days ago by avoloshko
  4

• Errors when calling `get_client` inside `client.sync`.

  #8914 opened 24 days ago by trivialfis
  2

• Dead loop caused by `forward_logging` in nested tasks.

  #8915 opened 25 days ago by trivialfis
  4

• Opening the TCP comm port in a browser results in an exception

  #8905 opened a month ago by jacobtomlinson
  0

• [BUG] temp_dask folder sometimes is not founded and raises an error

  #8902 opened a month ago by gargantuadev
  1

• Consider using object pools for internal state keeping

  #8901 opened a month ago by fjetter
  3

• `dask worker --nworkers 256` does not start 256 workers on machines with 256 cores

  #8768 opened a month ago by arneyjfs
  3

• Scheduler should not be considered idle while a client submits new work

  #8876 opened 2 months ago by hendrikmakait
  1

• Provide means to monitor which task / worker is holding the lock or semaphores

  #8772 opened 4 months ago by carusyte
  1

• Client submit with workers doesn’t handle new joining workers correctly

  #8862 opened 3 months ago by YuriFeigin
  1

• Adding futures into a `Queue` disables `resources` argument in `submit` function.

  #8860 opened 3 months ago by OvidiuMircescu
  0

• Attempting to view details of a key held by a worker results in a 404

  #8866 opened 2 months ago by krishanbhasin-px
  3

• Plugins that help to pass credentials for S3 and GCS to remote cluster workers

  #8883 opened 2 months ago by dbalabka
  1

• "Recursive" futures are deleted from the cluster

  #8854 opened 3 months ago by josephnowak
  5

• DASK Deployment using SLURM with GPUs

  #8857 opened 2 months ago by AquifersBSIM
  3

• `distributed/shuffle/tests/test_rechunk.py::test_rechunk_auto_1d[20-chunks4-5-expected4]` failing on `main`

  #8869 opened a month ago by jrbourbeau
  4

• PipInstall plugin - TypeError: Semaphore.__init__() got an unexpected keyword argument 'register'

  #8838 opened 3 months ago by starkgate
  5

• 2024.8.2: documentation build fails

  #8850 opened a month ago by kloczek
  1

• Scheduler is busy 95% of CPU time with `distributed.protocol.pickle.dumps` on objects which should be trivial to pickle

  #8767 opened 4 months ago by jonded94
  6

• _shift_counters - dictionary changed size during iteration

  #8816 opened 3 months ago by joshua-gould
  0

• Repeated calls to `url_escape` take around 40% of CPU time of scheduler

  #8761 opened 4 months ago by jonded94
  8

• Memory leak when using slurm

  #8859 opened 3 months ago by sofsse
  1

• Can't register WorkerPlugin subclass with `Client.register_plugin`

  #8778 opened a month ago by ivirshup
  1

• Get more informative timeout variable names or clarify them in the comments

  #8885 opened a month ago by falfaroalmagro
  1

• Improve error message when P2P runs ouf of disk

  #8864 opened a month ago by hendrikmakait
  0

• Adaptive stops whenever it encounters an error

  #8808 opened 2 months ago by hendrikmakait
  0

• Compute a DataFrame with datetime64[ns] dtype raise ValueError("cannot include dtype 'M' in a buffer")

  #8874 opened 2 months ago by josephnowak
  4

• Improve error message when spilling runs out of disk

  #8863 opened 3 months ago by hendrikmakait
  0

• Remove data spilling that acts on `target` threshold

  #8820 opened 3 months ago by fjetter
  3

• `Client.compute(df)` operates on larger task graphs than `df.compute()`

  #8833 opened 3 months ago by hendrikmakait
  2

• Propagate reason to client-side `FutureCancelledError` when closing the scheduler

  #8830 opened 3 months ago by hendrikmakait
  0

• test_jupyter_cli seems to be timing out regularly on main on all jobs

  #8825 opened 3 months ago by phofl
  0

• RecursionError occurs in Dask Distributed

  #8817 opened 4 months ago by naoyaikeda
  3

• Rework "scatter data if graph is too large" warning

  #8814 opened 4 months ago by fjetter
  0

• Random access array take using P2P

  #8774 opened 4 months ago by fjetter
  2

• Setting a basicConfig before importing distributed changes the scheduler logging level.

  #8805 opened 4 months ago by Thomas-Z
  0

• Add cumulative worker counters to prometheus

  #8773 opened 4 months ago by fjetter
  0

• `test_package_install_restarts_on_nanny` times out

  #8791 opened 4 months ago by hendrikmakait
  1

• `test_task_group_and_prefix_statistics` is very flaky on `mindeps-pandas` CI

  #8795 opened 4 months ago by hendrikmakait
  0

• Shut down scheduler on unrecoverable exceptions

  #8792 opened 4 months ago by hendrikmakait
  0

• Scheduler deadlocked after stealing failed in `move_task_confirm`

  #8787 opened 4 months ago by hendrikmakait
  0

• Failure in a task submitted by another task can leave an "erred" task permanently on the cluster

  #8789 opened 4 months ago by bcalvert-graft
  1

• Failed tasks persist in the Client even if the `Future` object falls out of scope, preventing Future deallocation/task cleanup from the cluster

  #8790 opened 4 months ago by bcalvert-graft
  0

• P2P Shuffle blows up the task graph with increasing number of partitions

  #8779 opened 4 months ago by phofl
  0

• `TaskPrefix` has no `len()`

  #8780 opened 4 months ago by hendrikmakait
  0

• `test_upload_file_egg` started failing on `mindeps` CI

  #8784 opened 4 months ago by hendrikmakait
  0

• Repeated calls to `memory_color` take around 12% of CPU time of scheduler

  #8763 opened 4 months ago by jonded94
  6

• TimeoutError in distributed.nanny causing gRPC server crash after prolonged analysis

  #8775 opened 4 months ago by Defqon7
  0