I'm experiencing an issue with Azure AI Studio while working on a prompt flow. I had successfully deployed multiple versions before, but suddenly, I started encountering an error during the index lookup step . The error message is: lookup : Execution failure in 'lookup': (Exception) Exception occured in search_function_construction. Here are the steps I’ve taken to troubleshoot: Tried running the same prompt flow with different AI search services and indexes . Created new projects with entirely new prompt flows , but the error persists. Tested the flow with different compute sessions (both compute instance and serverless ), but the issue remains. Deployed the prompt flow to a live endpoint, and the error still occurs during execution. The error seems to be related to the search_function_construction , but I'm unable to pinpoint the root cause. I would appreciate any insights or suggestions on how to resolve this issue. Are there any known limitations or configurations I might be missing? Thank you in advance for your help!

@Hector Rosado Deirdre O'Regan Arun Sivaraj A P Hosoya Kazuki/細谷和輝 Could you please try this workaround and start/restart the compute session and check if it works? Open requirements.txt from expanding the files section and clicking on requirements.txt Insert following 3 lines before promptflow-vectordb[azure] azureml-rag==0.2.36 langchain>=0.1.20,<0.3 langchain-community<0.3.0 Start/restart the compute session.

Azure AI Studio: Error in Index Lookup Step of Prompt Flow ('search_function_construction')

Hector Rosado 45

I'm experiencing an issue with Azure AI Studio while working on a prompt flow. I had successfully deployed multiple versions before, but suddenly, I started encountering an error during the index lookup step. The error message is:

lookup : Execution failure in 'lookup': (Exception) Exception occured in search_function_construction.

Here are the steps I’ve taken to troubleshoot:

Tried running the same prompt flow with different AI search services and indexes.
Created new projects with entirely new prompt flows, but the error persists.
Tested the flow with different compute sessions (both compute instance and serverless), but the issue remains.
Deployed the prompt flow to a live endpoint, and the error still occurs during execution.

The error seems to be related to the search_function_construction, but I'm unable to pinpoint the root cause.

I would appreciate any insights or suggestions on how to resolve this issue. Are there any known limitations or configurations I might be missing?

Thank you in advance for your help!

Hosoya Kazuki/細谷　和輝 5 Reputation points

2024-12-24T03:50:58.71+00:00

I’m encountering the same issue described here while working on a prompt flow in Azure Machine Learning Studio. The error message I’m seeing is identical:
lookup : Execution failure in 'lookup': (Exception) Exception occured in search_function_construction.
Traceback (most recent call last): File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_generate_schema.py", line 864, in _resolve_forward_ref obj = _typing_extra.eval_type_backport(obj, globalns=self._types_namespace) File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_typing_extra.py", line 279, in eval_type_backport return _eval_type_backport(value, globalns, localns, type_params) File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_typing_extra.py", line 303, in _eval_type_backport return _eval_type(value, globalns, localns, type_params) File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_typing_extra.py", line 332, in _eval_type return typing._eval_type( # type: ignore File "/azureml-envs/prompt-flow/runtime/lib/python3.9/typing.py", line 292, in _eval_type return t._evaluate(globalns, localns, recursive_guard) File "/azureml-envs/prompt-flow/runtime/lib/python3.9/typing.py", line 554, in _evaluate eval(self.__forward_code__, globalns, localns), File "<string>", line 1, in <module> NameError: name 'AzureSearch' is not defined

If there are any updates, insights, or workarounds available, I would greatly appreciate it. Thank you!
Arun Sivaraj A P 0 Reputation points

2024-12-24T05:59:28.8+00:00

I am also getting the same error recently (guess from last Friday onwards). Earlier it was working fine without any issues. Any workaround on this issue is really helpful. Thank you!

Deirdre O'Regan 0

I'm getting a similar issue. My previously working prompt flow is unusable because of it. Tried creating a whole new one, new index, etc, but same issue always shows up. Here are my error snippets:

Run failed: Exception: Exception occured in search_function_construction.

Traceback (most recent call last):
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_generate_schema.py", line 864, in _resolve_forward_ref
    obj = _typing_extra.eval_type_backport(obj, globalns=self._types_namespace)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_typing_extra.py", line 279, in eval_type_backport
    return _eval_type_backport(value, globalns, localns, type_params)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_typing_extra.py", line 303, in _eval_type_backport
    return _eval_type(value, globalns, localns, type_params)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_typing_extra.py", line 332, in _eval_type
    return typing._eval_type(  # type: ignore
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/typing.py", line 292, in _eval_type
    return t._evaluate(globalns, localns, recursive_guard)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/typing.py", line 554, in _evaluate
    eval(self.__forward_code__, globalns, localns),
  File "<string>", line 1, in <module>
NameError: name 'AzureSearch' is not defined

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/promptflow_vectordb/tool/utils/profiling.py", line 18, in measure_execution_time
    yield
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/promptflow_vectordb/tool/common_index_lookup.py", line 56, in _get_search_func
    search_func = build_search_func(index, top_k, query_type)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/promptflow_vectordb/tool/common_index_lookup_extensions/utils.py", line 46, in build_search_func
    store = index.as_langchain_vectorstore(credential=DefaultAzureCredential())
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/azureml/rag/mlindex.py", line 197, in as_langchain_vectorstore
    azuresearch = _get_azuresearch_module_instance(langchain_pkg_version)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/azureml/rag/utils/acs.py", line 21, in _get_azuresearch_module_instance
    azuresearch.AzureSearchVectorStoreRetriever.model_rebuild()
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/main.py", line 563, in model_rebuild
    return _model_construction.complete_model_class(
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_model_construction.py", line 577, in complete_model_class
    schema = cls.__get_pydantic_core_schema__(cls, handler)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/main.py", line 671, in __get_pydantic_core_schema__
    return handler(source)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_schema_generation_shared.py", line 83, in __call__
    schema = self._handler(source_type)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_generate_schema.py", line 655, in generate_schema
    schema = self._generate_schema_inner(obj)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_generate_schema.py", line 924, in _generate_schema_inner
    return self._model_schema(obj)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_generate_schema.py", line 739, in _model_schema
    {k: self._generate_md_field_schema(k, v, decorators) for k, v in fields.items()},
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_generate_schema.py", line 739, in <dictcomp>
    {k: self._generate_md_field_schema(k, v, decorators) for k, v in fields.items()},
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_generate_schema.py", line 1115, in _generate_md_field_schema
    common_field = self._common_field_schema(name, field_info, decorators)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_generate_schema.py", line 1308, in _common_field_schema
    schema = self._apply_annotations(
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_generate_schema.py", line 2107, in _apply_annotations
    schema = get_inner_schema(source_type)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_schema_generation_shared.py", line 83, in __call__
    schema = self._handler(source_type)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_generate_schema.py", line 2088, in inner_handler
    schema = self._generate_schema_inner(obj)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_generate_schema.py", line 918, in _generate_schema_inner
    return self.generate_schema(self._resolve_forward_ref(obj))
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/pydantic/_internal/_generate_schema.py", line 866, in _resolve_forward_ref
    raise PydanticUndefinedAnnotation.from_name_error(e) from e
pydantic.errors.PydanticUndefinedAnnotation: name 'AzureSearch' is not defined

For further information visit https://errors.pydantic.dev/2.9/u/undefined-annotation

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/promptflow/tracing/_trace.py", line 557, in wrapped
    output = func(*args, **kwargs)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/promptflow_vectordb/core/logging/utils.py", line 98, in wrapper
    res = func(*args, **kwargs)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/promptflow_vectordb/tool/common_index_lookup.py", line 170, in search
    search_func = _get_search_func(mlindex_content, top_k, query_type)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/promptflow_vectordb/tool/common_index_lookup.py", line 56, in _get_search_func
    search_func = build_search_func(index, top_k, query_type)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/contextlib.py", line 137, in __exit__
    self.gen.throw(typ, value, traceback)
  File "/azureml-envs/prompt-flow/runtime/lib/python3.9/site-packages/promptflow_vectordb/tool/utils/profiling.py", line 21, in measure_execution_time
    raise Exception(error_msg) from e
Exception: Exception occured in search_function_construction.

Is there a solution/workaround for this new issue? Thanks!

PS: Is it something to do with this warning I see every time I go into my prompt flow now?

WARNING: azureml-rag 0.2.37.1 does not provide the extra 'azure_cosmos_mongo_vcore'
WARNING: azureml-rag 0.2.37.1 does not provide the extra 'azure_cosmos_nosql'
WARNING: azureml-rag 0.2.37.1 does not provide the extra 'cognitive_search'

Hosoya Kazuki/細谷　和輝 5 Reputation points

2024-12-26T07:15:24.7966667+00:00

@romungi-MSFT
Thank you so much for your suggestion! I followed the workaround you provided, including updating the requirements.txt file and restarting the compute session, and it resolved the issue perfectly.

I truly appreciate your guidance, it was a huge help. Thanks again!
romungi-MSFT 48,141 Reputation points Microsoft Employee

2024-12-26T07:36:04.55+00:00

Thanks @Hosoya Kazuki/細谷　和輝 I have converted my comment to answer. @Hector Rosado it would be great if you could also try the same and accept the same as answer if it helped. Thanks!!
Deirdre O'Regan 0 Reputation points

2024-12-26T12:00:31.2666667+00:00

Yes, this works for me too. Thanks very much @romungi-MSFT !

Note to others and to future self: I didn't have to 'Install and Save', just 'Save' after I put those lines in requirements.txt.

Is this something that's funny with the prompt flow setups all of us have, or is it a genuine bug for all, and you'll eventually build this fix into prompt flow for all users?
romungi-MSFT 48,141 Reputation points Microsoft Employee

2024-12-26T13:19:13.98+00:00

@Deirdre O'Regan This is a bug with azureml-rag version >0.2.36

The workaround is using the version 0.2.36 which works with current flows. A fix should be rolled out soon. You only need this workaround only if you see this failure.

Accepted answer

romungi-MSFT 48,141 Reputation points Microsoft Employee

2024-12-26T06:56:34.3766667+00:00
@Hector Rosado Deirdre O'Regan Arun Sivaraj A P Hosoya Kazuki/細谷　和輝 Could you please try this workaround and start/restart the compute session and check if it works?

Open requirements.txt from expanding the files section and clicking on requirements.txt

Insert following 3 lines before promptflow-vectordb[azure]

azureml-rag==0.2.36 langchain>=0.1.20,<0.3 langchain-community<0.3.0

Start/restart the compute session.
Please sign in to rate this answer.

2 people found this answer helpful.
Arun Sivaraj A P 0 Reputation points

2024-12-26T13:27:54.41+00:00

Thanks! My issue is resolved.

Hector Rosado 45 Reputation points

2024-12-26T14:30:25.63+00:00

Thank you! This solved the issue
Sign in to comment

Use comments to ask for clarification, additional information, or improvements to the question.

Share via

Azure AI Studio: Error in Index Lookup Step of Prompt Flow ('search_function_construction')

0 additional answers

Your answer