Error when sharing environment from workspace to registry

[bastbu]

• Package Name: azure-ai-ml
• Package Version: 1.23.0
• Operating System: Ubuntu
• Python Version: 3.11.11

Describe the bug

We are trying to use the AML Registry to share components across workspaces. As indicated in the documentation for a secure AML Registry, we must build the environment in the workspace before we share it to the AML Registry.

When we try to onboard different components that share the same environment concurrently, this often results in:

AttributeError: 'AzureMachineLearningWorkspaces' object has no attribute 'workspaces'

To wait until the environment was built, we rely on an error message raised by the AML Registry, as there is no way to wait for the environment build to complete in the v2 SDK. Is there another way how we can make sharing components in an AML Registry work when using the v2 SDK? We cannot publish the components directly to the AML Registry due to the bug reported here.

To Reproduce

I used the following script to reproduce the issue, and for me it consistently occurs after one or two tries. Make sure that the requirements.txt is modified between runs to avoid caching of an already built environment/image.

from concurrent.futures import ThreadPoolExecutor
from pathlib import Path
from time import sleep
from uuid import uuid4

from azure.ai.ml import MLClient
from azure.ai.ml.entities import BuildContext, Environment
from azure.core.exceptions import HttpResponseError


def wait_register_environment(
    *, ws_client: MLClient, registry_name: str, environment: Environment
) -> Environment:
    _ = ws_client.environments.create_or_update(environment)

    while True:
        try:
            return ws_client.environments.share(
                name=name,
                version=version,
                share_with_name=name,
                share_with_version=version,
                registry_name=registry_name,
            )
        except HttpResponseError as exc_info:
            if "AssetNotReadyForPublishFromSource" in exc_info.message:
                print("Not yet ready, waiting...")
                sleep(5)
            else:
                raise


def main(ml_client: MLClient, registry_name: str):
    with ThreadPoolExecutor() as executor:
        futures = [
            executor.submit(
                lambda: wait_register_environment(
                    ws_client=ml_client,
                    registry_name=registry_name,
                    environment=Environment(
                        name="share_env_repro",
                        version=uuid4().hex,
                        build=BuildContext(path=Path(__file__).parent / "environment"),
                    ),
                )
            )
            for _ in range(5)
        ]

    for future in futures:
        print(future.result())

With the following build context:

Dockerfile:

FROM python:3.11-bookworm

COPY requirements.txt .

RUN pip install -r requirements.txt

requirements.txt:

argparse

We share a global client instance as MLClient was indicated to be immutable.

The full error is as follows:

Not yet ready, waiting...
Not yet ready, waiting...
Not yet ready, waiting...
Not yet ready, waiting...
Traceback (most recent call last):
  File "/home/vscode/.local/share/uv/python/cpython-3.11.11-linux-x86_64-gnu/lib/python3.11/runpy.py", line 198, in _run_module_as_main
    return _run_code(code, main_globals, None,
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/vscode/.local/share/uv/python/cpython-3.11.11-linux-x86_64-gnu/lib/python3.11/runpy.py", line 88, in _run_code
    exec(code, run_globals)
  File "/home/vscode/.vscode-server/extensions/ms-python.debugpy-2024.14.0-linux-x64/bundled/libs/debugpy/adapter/../../debugpy/launcher/../../debugpy/__main__.py", line 71, in <module>
    cli.main()
  File "/home/vscode/.vscode-server/extensions/ms-python.debugpy-2024.14.0-linux-x64/bundled/libs/debugpy/adapter/../../debugpy/launcher/../../debugpy/../debugpy/server/cli.py", line 501, in main
    run()
  File "/home/vscode/.vscode-server/extensions/ms-python.debugpy-2024.14.0-linux-x64/bundled/libs/debugpy/adapter/../../debugpy/launcher/../../debugpy/../debugpy/server/cli.py", line 351, in run_file
    runpy.run_path(target, run_name="__main__")
  File "/home/vscode/.vscode-server/extensions/ms-python.debugpy-2024.14.0-linux-x64/bundled/libs/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 310, in run_path
    return _run_module_code(code, init_globals, run_name, pkg_name=pkg_name, script_name=fname)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/vscode/.vscode-server/extensions/ms-python.debugpy-2024.14.0-linux-x64/bundled/libs/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 127, in _run_module_code
    _run_code(code, megaloblast, init_globals, mod_name, mod_spec, pkg_name, script_name)
  File "/home/vscode/.vscode-server/extensions/ms-python.debugpy-2024.14.0-linux-x64/bundled/libs/debugpy/_vendored/pydevd/_pydevd_bundle/pydevd_runpy.py", line 118, in _run_code
    exec(code, run_globals)
  File "/workspaces/repros/share_environment/main.py", line 81, in <module>
    main(dependencies.workspace_client(), dependencies.registry_name())
  File "/workspaces/repros/share_environment/main.py", line 77, in main
    print(future.result())
          ^^^^^^^^^^^^^^^
  File "/home/vscode/.local/share/uv/python/cpython-3.11.11-linux-x86_64-gnu/lib/python3.11/concurrent/futures/_base.py", line 449, in result
    return self.__get_result()
           ^^^^^^^^^^^^^^^^^^^
  File "/home/vscode/.local/share/uv/python/cpython-3.11.11-linux-x86_64-gnu/lib/python3.11/concurrent/futures/_base.py", line 401, in __get_result
    raise self._exception
  File "/home/vscode/.local/share/uv/python/cpython-3.11.11-linux-x86_64-gnu/lib/python3.11/concurrent/futures/thread.py", line 58, in run
    result = self.fn(*self.args, **self.kwargs)
             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/workspaces/repros/share_environment/main.py", line 63, in <lambda>
    lambda: wait_register_environment(
            ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/workspaces/repros/share_environment/main.py", line 36, in wait_register_environment
    return ws_client.environments.share(
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/workspaces/.venv/lib/python3.11/site-packages/azure/ai/ml/_telemetry/activity.py", line 292, in wrapper
    return f(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^
  File "/workspaces/.venv/lib/python3.11/site-packages/azure/ai/ml/_utils/_experimental.py", line 100, in wrapped
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/workspaces/.venv/lib/python3.11/site-packages/azure/ai/ml/operations/_environment_operations.py", line 508, in share
    workspace = self._service_client.workspaces.get(
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
AttributeError: 'AzureMachineLearningWorkspaces' object has no attribute 'workspaces'

Steps to reproduce the behavior:

1. Run the code described above

Expected behavior

All environments should be successfully published to the AML Registry after a while.

[github-actions]

Thanks for the feedback! We are routing this to the appropriate team for follow-up. cc @Azure/azure-ml-sdk @azureml-github.