Skip to content

Error, "frozen_importlib._DeadlockError: deadlock detected" while running the AML pipeline using SDK for Python  #16831

@SriniGajjela

Description

@SriniGajjela

The Pipeline run fails only with 'ParallelRunStep' using ML SDK for python and returns an error message:

from azureml.data.dataset_type_definitions import PromoteHeadersBehavior
File "", line 988, in _find_and_load
File "", line 149, in enter
File "", line 94, in acquire
frozen_importlib._DeadlockError: deadlock detected by _ModuleLock('azureml.data.dataset_type_definitions')
File "driver/amlbi_main.py", line 136, in
main()
File "driver/amlbi_main.py", line 113, in main
boot(driver_dir)
File "driver/amlbi_main.py", line 54, in boot
booter.start()
File "driver/azureml_user/parallel_run/boot.py", line 301, in start
self.start_sys_main()
File "driver/azureml_user/parallel_run/boot.py", line 208, in start_sys_main
self.run(cmd)
File "driver/azureml_user/parallel_run/boot.py", line 170, in run
raise Exception(stderr)

This is a known issue and occurs intermittently. Azure Machine Learning Product Team actively investigating to find the root cause and fix.

Metadata

Metadata

Assignees

No one assigned

    Labels

    ClientThis issue points to a problem in the data-plane of the library.ML-MLOpsAreaPathMachine LearningService AttentionWorkflow: This issue is responsible by Azure service team.bugThis issue requires a change to an existing behavior in the product in order to be resolved.

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions