Failure to run `RiverFloodInundation.download_forecast`

[emanuel-schmid]

On a Windows 10 computer, the following piece of code leads to a process running wild with a never ending stream of Error messages:

from climada_petals.hazard.rf_glofas import RiverFloodInundation

rf = RiverFloodInundation()
rf.download_forecast(
    countries=["Switzerland", "Austria"],
    forecast_date="2021-07-10",
    preprocess=lambda x: x.max(dim="step"),
    system_version="operational",  # Version mislabeled
)

The error messages look something like this, repeated forever:

...\climada\lib\site-packages\xarray\core\dataset.py:271: UserWarning: The specified chunks separate the stored chunks along dimension "latitude" starting at index 1377. This could degrade performance. Instead, consider rechunking after loading.
  warnings.warn(
...\climada\lib\site-packages\xarray\core\dataset.py:271: UserWarning: The specified chunks separate the stored chunks along dimension "longitude" starting at index 3480. This could degrade performance. Instead, consider rechunking after loading.
  warnings.warn(
...\climada\lib\site-packages\dask\dataframe\_pyarrow_compat.py:17: FutureWarning: Minimal version of pyarrow will soon be increased to 14.0.1. You are using 11.0.0. Please consider upgrading.
  warnings.warn(
...\climada\lib\site-packages\xarray\core\dataset.py:271: UserWarning: The specified chunks separate the stored chunks along dimension "latitude" starting at index 1377. This could degrade performance. Instead, consider rechunking after loading.
  warnings.warn(
...\climada\lib\site-packages\xarray\core\dataset.py:271: UserWarning: The specified chunks separate the stored chunks along dimension "longitude" starting at index 3480. This could degrade performance. Instead, consider rechunking after loading.
  warnings.warn(
Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "...\climada\lib\multiprocessing\spawn.py", line 116, in spawn_main
    exitcode = _main(fd, parent_sentinel)
  File "...\climada\lib\multiprocessing\spawn.py", line 125, in _main
    prepare(preparation_data)
  File "...\climada\lib\multiprocessing\spawn.py", line 236, in prepare
    _fixup_main_from_path(data['init_main_from_path'])
  File "...\climada\lib\multiprocessing\spawn.py", line 287, in _fixup_main_from_path
    main_content = runpy.run_path(main_path,
  File "...\climada\lib\runpy.py", line 288, in run_path
    return _run_module_code(code, init_globals, run_name,
  File "...\climada\lib\runpy.py", line 97, in _run_module_code
    _run_code(code, mod_globals, init_globals,
  File "...\climada\lib\runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "C:\Users\me\git\climada\climada_petals\xxx.py", line 4, in <module>
    rf.download_forecast(
  File "C:\Users\me\git\climada\climada_petals\climada_petals\hazard\rf_glofas\river_flood_computation.py", line 354, in download_forecast
    forecast = download_glofas_discharge(
  File "C:\Users\me\git\climada\climada_petals\climada_petals\hazard\rf_glofas\transform_ops.py", line 318, in download_glofas_discharge
    files = glofas_request(
  File "C:\Users\me\git\climada\climada_petals\climada_petals\hazard\rf_glofas\cds_glofas_downloader.py", line 289, in glofas_request
    return glofas_request_multiple(
  File "C:\Users\me\git\climada\climada_petals\climada_petals\hazard\rf_glofas\cds_glofas_downloader.py", line 173, in glofas_request_multiple
    with mp.Pool(num_proc) as pool:
  File "...\climada\lib\multiprocessing\context.py", line 119, in Pool
    return Pool(processes, initializer, initargs, maxtasksperchild,
  File "...\climada\lib\multiprocessing\pool.py", line 212, in __init__
    self._repopulate_pool()
  File "...\climada\lib\multiprocessing\pool.py", line 303, in _repopulate_pool
    return self._repopulate_pool_static(self._ctx, self.Process,
  File "...\climada\lib\multiprocessing\pool.py", line 326, in _repopulate_pool_static
    w.start()
  File "...\climada\lib\multiprocessing\process.py", line 121, in start
    self._popen = self._Popen(self)
  File "...\climada\lib\multiprocessing\context.py", line 327, in _Popen
    return Popen(process_obj)
  File "...\climada\lib\multiprocessing\popen_spawn_win32.py", line 45, in __init__
    prep_data = spawn.get_preparation_data(process_obj._name)
  File "...\climada\lib\multiprocessing\spawn.py", line 154, in get_preparation_data
    _check_not_importing_main()
  File "...\climada\lib\multiprocessing\spawn.py", line 134, in _check_not_importing_main
    raise RuntimeError('''
RuntimeError: 
        An attempt has been made to start a new process before the
        current process has finished its bootstrapping phase.

        This probably means that you are not using fork to start your
        child processes and you have forgotten to use the proper idiom
        in the main module:

            if __name__ == '__main__':
                freeze_support()
                ...

        The "freeze_support()" line can be omitted if the program
        is not going to be frozen to produce an executable.

(Execution of above snippet cannot even be stopped with ctrl-c no matter how often it's pressed.)

[peanutfun]

I think this is due to Jupyter Notebooks not supporting multiprocessing properly. It's quite curious that I did not run into these problems lately. One quick solution is not to rely on multiprocessing when only downloading with a single process, which is the default. I'll come up with a fix.

[emanuel-schmid]

That's what I thought first too, but the code snippet above yields this very error above when run as a python script from command line, not from within a notebook. So it's probably not the Notebook's support of multiprocessing but the system's istself...

[peanutfun]

Does it still occur when using the "main" idiom as proposed in the error message?

# Your .py script

if __name__ == '__main__':
    # Do whatever you were doing before

[emanuel-schmid]

🤦 right! 🙄 no it doesn't.