[Pipeline] CrewAi test is flaky

XMLWordPrintableJSON

    • Type: Task
    • Resolution: Done
    • Priority: Unknown
    • None
    • Affects Version/s: None
    • Component/s: ai-ml-testing, AI/ML, CrewAI
    • None
    • None
    • Python Drivers
    • Not Needed
    • Hide

      1. What would you like to communicate to the user about this feature?
      2. Would you like the user to see examples of the syntax and/or executable code and its output?
      3. Which versions of the driver/connector does this apply to?

      Show
      1. What would you like to communicate to the user about this feature? 2. Would you like the user to see examples of the syntax and/or executable code and its output? 3. Which versions of the driver/connector does this apply to?
    • None
    • None
    • None
    • None
    • None
    • None

      Context

      The test added in INTPYTHON-580 is flaky. It has failed after merging in two different ways:

      In https://spruce.mongodb.com/task/ai_ml_pipeline_testing_test_crewai_tools_rhel_test_crewai_tools_local_3c665f00011e4e657eaf10847c63bdcfbe7422d9_25_07_14_18_08_54/logs?execution=0 we get the error:

      [2025/07/14 13:14:21.381] Traceback (most recent call last):
      [2025/07/14 13:14:21.381]   File "/data/mci/45241060f7cd5b50273ddb8d8d569150/src/crewai-tools/crewAI-tools/test_mongodb_vector_search_tool.py", line 78, in <module>
      [2025/07/14 13:14:21.381]     assert "advancements" in text or "improvements" in text, text
      [2025/07/14 13:14:21.381]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
      [2025/07/14 13:14:21.381] AssertionError: gpt-4 has shown to have remarkable accuracy and capabilities compared to its predecessors....
      

      In https://spruce.mongodb.com/task/ai_ml_pipeline_testing_test_crewai_tools_rhel_test_crewai_tools_remote_68759680cd7ebf000743298b_25_07_14_23_45_04/logs?execution=0 we get the error:

      [2025/07/14 18:49:58.248] Traceback (most recent call last):
      [2025/07/14 18:49:58.248]   File "/data/mci/d27c5bb6761b8a88cc694d6c7febeff0/src/crewai-tools/crewAI-tools/test_mongodb_vector_search_tool.py", line 78, in <module>
      [2025/07/14 18:49:58.248]     assert "advancements" in text or "improvements" in text, text
      [2025/07/14 18:49:58.248]            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
      [2025/07/14 18:49:58.248] AssertionError: i was unable to extract specific insights regarding the accuracy of gpt-4 from the technical document as the tool provided did not allow for successful queries
      

      We could add more expected words, but it wouldn't solve the second problem.
      We could try changing the model used, or worst case add a retry.

      Definition of done

      Address the flaky test.

      Pitfalls

      None

              Assignee:
              Steve Silvester
              Reporter:
              Steve Silvester
              Votes:
              0 Vote for this issue
              Watchers:
              1 Start watching this issue

                Created:
                Updated:
                Resolved: