Gen AI monitoring is broken due to Marshmallow update

[prise6]

Hello,
Related to #39039

When configuring monitoring, like this example in documentation.

Job failed with cannot import name '_T' from 'marshmallow.fields' (/home/trusted-service-user/cluster-env/env/lib/python3.10/site-packages/marshmallow/fields.py)

When digging in job log

# in logs/azureml/driver/librarymanager/*.log

# [...]
channels:
  - defaults
  - conda-forge
dependencies:
  - python=3.10
  - pip:
      - azure-cli-core~=2.45.0
      - azure-identity~=1.12.0
      - azure-ai-ml~=1.10.0
      - azure-keyvault-secrets==4.7.0
      - tiktoken==0.3.3
      - json5==0.9.11
      - scipy~=1.10.0
      - numpy~=1.21.0
      - pandas~=1.4.3
      - azureml-mlflow~=1.49.0
      - mltable~=1.3.0

# [...]

Collecting marshmallow<4.0.0,>=3.5 (from azure-ai-ml~=1.10.0->-r /usr/lib/library-manager/bin/condaenv.s2dyrgt_.requirements.txt (line 3))
  Downloading marshmallow-3.24.1-py3-none-any.whl.metadata (7.1 kB

Correct me if 'im wrong but i can't force the azure-ai-ml version in this component, so

• is there a workaround ?
• is it possible to update dependencies for this component ?

Thank you,

@kingernupur

[github-actions]

Thanks for the feedback! We are routing this to the appropriate team for follow-up. cc @achauhan-scc @kingernupur @luigiw @needuv @paulshealy1 @singankit.

[prise6]

Hello again,

Error still persists. Do you have planned actions about this error ?

Thank you,

[achauhan-scc]

issue is fixed 1.23.1

[prise6]

I understand that you fixed the marsmallow issue in the python sdk, but AzureML Monitoring is not fixed because the component version is not updated.

So i asked to your team in another repo: Azure/azureml-assets#3775