"RaySystemError: System error: Unknown error"

[aparmendariz]

Description: While executing my code's test suite, I encountered a "RaySystemError: System error: Unknown error." Notably, the issue seems to be related to the dependency specification for Ray in the environment YAML files.

Useful information:

• Initially, the code was using ray[default] as the dependency, and the error was present.
• Upon changing the dependency to ray without the [default] specifier, the error was resolved.
• However, using ray[default] is the recommended way by Ray for installing its package.

I do not understand which can be the source of the problem when specifying ray[default] as a dependency.

Note: we do not use Ray Dashboard or Cluster Launcher. so it is enough to install Ray with minimal dependencies?

Versions / Dependencies

name: test-env_raydefault dependencies:

• python
• spyder
• pip
• pip:
  • matplotlib
  • numba
  • pandas
  • ray[default]
  • scikit-learn
  • scipy
  • sympy

Reproduction script

Dependency issue

Issue Severity

Medium: It is a significant difficulty but I can work around it.

[jjyao]

If you don't need Ray job submission or runtime env, you don't need to use ray[default].

Also can you provide a repro so that we can debug on our side? Also can you show the full stack trace?

[aparmendariz]

reproducible script:

When running the following module: https://github.com/MCFpy/mcf/blob/main/examples/min_parameters_optpolicy.py (which unfortunately relies on external data given in https://github.com/MCFpy/mcf/blob/main/data/data_x_ps_1_1000.csv ) on the following dependencies:

dependencies:
- python=3.11.5
- spyder
- pip
- pip:
  - matplotlib
  - numba
  - pandas
  - ray[default]
  - scikit-learn
  - scipy
  - sympy
  - mcf

I get the following traceback error:

Traceback (most recent call last):

  File ~\AppData\Local\anaconda3\envs\test-env_raydefault\Lib\site-packages\spyder_kernels\py3compat.py:356 in compat_exec
    exec(code, globals, locals)

  File c:\users\aarmendarizpacheco\onedrive - universitaet st.gallen\dokumente\phdhsgapap\testing_065\testing_065\min_parameters_optpolicy.py:66
    alloc_train_df = myoptp.solve(train_df, data_title=TRAINDATA)

  File ~\AppData\Local\anaconda3\envs\test-env_raydefault\Lib\site-packages\mcf\optpolicy_functions.py:324 in solve
    allocation_df = op_pt.policy_tree_allocation(self, data_new_df)

  File ~\AppData\Local\anaconda3\envs\test-env_raydefault\Lib\site-packages\mcf\optpolicy_pt_functions.py:46 in policy_tree_allocation
    best_tree, _, _ = optimal_tree_proc(optp_, data_df, seed=12345)

  File ~\AppData\Local\anaconda3\envs\test-env_raydefault\Lib\site-packages\mcf\optpolicy_pt_functions.py:266 in optimal_tree_proc
    data_x_ref = ray.put(data_x)

  File ~\AppData\Local\anaconda3\envs\test-env_raydefault\Lib\site-packages\ray\_private\auto_init_hook.py:24 in auto_init_wrapper
    return fn(*args, **kwargs)

  File ~\AppData\Local\anaconda3\envs\test-env_raydefault\Lib\site-packages\ray\_private\client_mode_hook.py:103 in wrapper
    return func(*args, **kwargs)

  File ~\AppData\Local\anaconda3\envs\test-env_raydefault\Lib\site-packages\ray\_private\worker.py:2636 in put
    object_ref = worker.put_object(value, owner_address=serialize_owner_address)

  File ~\AppData\Local\anaconda3\envs\test-env_raydefault\Lib\site-packages\ray\_private\worker.py:720 in put_object
    self.core_worker.put_serialized_object_and_increment_local_ref(

  File python\ray\_raylet.pyx:3361 in ray._raylet.CoreWorker.put_serialized_object_and_increment_local_ref

  File python\ray\_raylet.pyx:3253 in ray._raylet.CoreWorker._create_put_buffer

  File python\ray\_raylet.pyx:468 in ray._raylet.check_status

RaySystemError: System error: Unknown error