Network related issues break the prepdocs.sh

[zhongshuai-cao]

- [x] bug report -> please search issues before submitting
- [ ] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)

OS and Version?

macOS 14

azd version?

1.5.0

Versions

Commit: 144698a
Date: Thu Nov 16 2023 14:58:20 GMT-0500 (Eastern Standard Time)

I am indexing a big folder of files ~10k and it keeps failed with different tracebacks after processing around 50 - 100 files (I am including the most 3 recent ones). It could be great if it can handle the exception to skip a file, or have a longer retry counter?

Traceback 1:

Traceback (most recent call last):
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/transport/_aiohttp.py", line 280, in send
    result = await self.session.request(  # type: ignore
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/aiohttp/client.py", line 586, in _request
    await resp.start(conn)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/aiohttp/client_reqrep.py", line 905, in start
    message, payload = await protocol.read()  # type: ignore[union-attr]
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/aiohttp/streams.py", line 616, in read
    await self._waiter
aiohttp.client_exceptions.ClientOSError: [Errno 32] Broken pipe

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

Traceback (most recent call last):
  File "/.../azure-search-openai-demo/./scripts/prepdocs.py", line 256, in <module>
    loop.run_until_complete(main(file_strategy, azd_credential, args))
  File "/Users/.../opt/anaconda3/envs/.../lib/python3.9/asyncio/base_events.py", line 647, in run_until_complete
    return future.result()
  File "/.../azure-search-openai-demo/./scripts/prepdocs.py", line 131, in main
    await strategy.run(search_info)
  File "/.../azure-search-openai-demo/scripts/prepdocslib/filestrategy.py", line 63, in run
    await search_manager.update_content(sections)
  File "/.../azure-search-openai-demo/scripts/prepdocslib/searchmanager.py", line 146, in update_content
    await search_client.upload_documents(documents)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/search/documents/aio/_search_client_async.py", line 557, in upload_documents
    results = await self.index_documents(batch, **kwargs)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/tracing/decorator_async.py", line 77, in wrapper_use_tracer
    return await func(*args, **kwargs)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/search/documents/aio/_search_client_async.py", line 655, in index_documents
    return await self._index_documents_actions(actions=batch.actions, **kwargs)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/search/documents/aio/_search_client_async.py", line 663, in _index_documents_actions
    batch_response = await self._client.documents.index(batch=batch, error_map=error_map, **kwargs)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/tracing/decorator_async.py", line 77, in wrapper_use_tracer
    return await func(*args, **kwargs)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/search/documents/_generated/aio/operations/_documents_operations.py", line 895, in index
    pipeline_response: PipelineResponse = await self._client._pipeline.run(  # pylint: disable=protected-access
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 221, in run
    return await first_node.send(pipeline_request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 69, in send
    response = await self.next.send(request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 69, in send
    response = await self.next.send(request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 69, in send
    response = await self.next.send(request)
  [Previous line repeated 2 more times]
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/policies/_redirect_async.py", line 73, in send
    response = await self.next.send(request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/policies/_retry_async.py", line 205, in send
    raise err
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/policies/_retry_async.py", line 179, in send
    response = await self.next.send(request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/policies/_authentication_async.py", line 94, in send
    response = await self.next.send(request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 69, in send
    response = await self.next.send(request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 69, in send
    response = await self.next.send(request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 69, in send
    response = await self.next.send(request)
  [Previous line repeated 2 more times]
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 106, in send
    await self._sender.send(request.http_request, **request.context.options),
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/transport/_aiohttp.py", line 317, in send
    raise ServiceRequestError(err, error=err) from err
azure.core.exceptions.ServiceRequestError: [Errno 32] Broken pipe

Traceback 2:

Extracting text from './data/xxx.pdf' using Azure Document Intelligence
Traceback (most recent call last):
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/transport/_aiohttp.py", line 280, in send
    result = await self.session.request(  # type: ignore
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/aiohttp/client.py", line 586, in _request
    await resp.start(conn)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/aiohttp/client_reqrep.py", line 905, in start
    message, payload = await protocol.read()  # type: ignore[union-attr]
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/aiohttp/streams.py", line 616, in read
    await self._waiter
aiohttp.client_exceptions.ServerTimeoutError: Timeout on reading data from socket

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

Traceback (most recent call last):
  File "/.../azure-search-openai-demo/./scripts/prepdocs.py", line 256, in <module>
    loop.run_until_complete(main(file_strategy, azd_credential, args))
  File "/Users/.../opt/anaconda3/envs/.../lib/python3.9/asyncio/base_events.py", line 647, in run_until_complete
    return future.result()
  File "/.../azure-search-openai-demo/./scripts/prepdocs.py", line 131, in main
    await strategy.run(search_info)
  File "/.../azure-search-openai-demo/scripts/prepdocslib/filestrategy.py", line 56, in run
    pages = [page async for page in self.pdf_parser.parse(content=file.content)]
  File "/.../azure-search-openai-demo/scripts/prepdocslib/filestrategy.py", line 56, in <listcomp>
    pages = [page async for page in self.pdf_parser.parse(content=file.content)]
  File "/.../azure-search-openai-demo/scripts/prepdocslib/pdfparser.py", line 81, in parse
    poller = await form_recognizer_client.begin_analyze_document(model_id=self.model_id, document=content)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/tracing/decorator_async.py", line 77, in wrapper_use_tracer
    return await func(*args, **kwargs)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/ai/formrecognizer/aio/_document_analysis_client_async.py", line 132, in begin_analyze_document
    return await _client_op_path.begin_analyze_document(  # type: ignore
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/tracing/decorator_async.py", line 77, in wrapper_use_tracer
    return await func(*args, **kwargs)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/ai/formrecognizer/_generated/v2023_07_31/aio/operations/_document_models_operations.py", line 189, in begin_analyze_document
    raw_result = await self._analyze_document_initial(  # type: ignore
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/ai/formrecognizer/_generated/v2023_07_31/aio/operations/_document_models_operations.py", line 105, in _analyze_document_initial
    pipeline_response = await self._client._pipeline.run(  # type: ignore # pylint: disable=protected-access
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 221, in run
    return await first_node.send(pipeline_request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 69, in send
    response = await self.next.send(request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 69, in send
    response = await self.next.send(request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 69, in send
    response = await self.next.send(request)
  [Previous line repeated 2 more times]
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/policies/_redirect_async.py", line 73, in send
    response = await self.next.send(request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/policies/_retry_async.py", line 205, in send
    raise err
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/policies/_retry_async.py", line 179, in send
    response = await self.next.send(request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/policies/_authentication_async.py", line 94, in send
    response = await self.next.send(request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 69, in send
    response = await self.next.send(request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 69, in send
    response = await self.next.send(request)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 69, in send
    response = await self.next.send(request)
  [Previous line repeated 3 more times]
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/_base_async.py", line 106, in send
    await self._sender.send(request.http_request, **request.context.options),
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/azure/core/pipeline/transport/_aiohttp.py", line 315, in send
    raise ServiceResponseError(err, error=err) from err
azure.core.exceptions.ServiceResponseError: Timeout on reading data from socket

Traceback 3

Traceback (most recent call last):
  File "/.../azure-search-openai-demo/./scripts/prepdocs.py", line 256, in <module>
    loop.run_until_complete(main(file_strategy, azd_credential, args))
  File "/.../opt/anaconda3/envs/.../lib/python3.9/asyncio/base_events.py", line 647, in run_until_complete
    return future.result()
  File "/.../azure-search-openai-demo/./scripts/prepdocs.py", line 131, in main
    await strategy.run(search_info)
  File "/.../azure-search-openai-demo/scripts/prepdocslib/filestrategy.py", line 63, in run
    await search_manager.update_content(sections)
  File "/.../azure-search-openai-demo/scripts/prepdocslib/searchmanager.py", line 140, in update_content
    embeddings = await self.embeddings.create_embeddings(
  File "/.../azure-search-openai-demo/scripts/prepdocslib/embeddings.py", line 116, in create_embeddings
    return await self.create_embedding_batch(texts)
  File "/.../azure-search-openai-demo/scripts/prepdocslib/embeddings.py", line 86, in create_embedding_batch
    async for attempt in AsyncRetrying(
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/tenacity/_asyncio.py", line 71, in __anext__
    do = self.iter(retry_state=self._retry_state)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/tenacity/__init__.py", line 314, in iter
    return fut.result()
  File "/.../opt/anaconda3/envs/.../lib/python3.9/concurrent/futures/_base.py", line 439, in result
    return self.__get_result()
  File "/.../opt/anaconda3/envs/.../lib/python3.9/concurrent/futures/_base.py", line 391, in __get_result
    raise self._exception
  File "/.../azure-search-openai-demo/scripts/prepdocslib/embeddings.py", line 94, in create_embedding_batch
    emb_response = await openai.Embedding.acreate(**emb_args, input=batch.texts)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/openai/api_resources/embedding.py", line 73, in acreate
    response = await super().acreate(*args, **kwargs)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/openai/api_resources/abstract/engine_api_resource.py", line 219, in acreate
    response, _, api_key = await requestor.arequest(
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/openai/api_requestor.py", line 384, in arequest
    resp, got_stream = await self._interpret_async_response(result, stream)
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/openai/api_requestor.py", line 738, in _interpret_async_response
    self._interpret_response_line(
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/openai/api_requestor.py", line 775, in _interpret_response_line
    raise self.handle_error_response(
  File "/.../azure-search-openai-demo/scripts/.venv/lib/python3.9/site-packages/openai/api_requestor.py", line 415, in handle_error_response
    raise error.APIError(
openai.error.APIError: Invalid response object from API: '{ "statusCode": 401, "message": "Unauthorized. Access token is missing, invalid, audience is incorrect (https://cognitiveservices.azure.com), or have expired." }' (HTTP response code was 401)

[pamelafox]

Thanks for sharing! I haven't done tests with that many documents.

You could try modifying Azure Python SDK standard connection parameters. I think these ones might all be options?

timeout=30, connection_timeout=14400, read_timeout=240, retry_connect=4

i.e. you could try adding those to upload_documents(). You can also customize the retry at the client level. Here's some more docs:

https://learn.microsoft.com/en-us/azure/developer/python/sdk/azure-sdk-library-usage-patterns?tabs=pip#arguments-for-libraries-based-on-azurecore

I assume that when it fails, you have to delete the md5 file for the failed document and then start over on the remaining documents, is that right? (But the previously uploaded ones should be okay)

[zhongshuai-cao]

Good catch!

I've been running scripts after every failure to compare the blob storage container and move the uploaded file to a archive folder to avoid duplication, then I realized there is md5 deduplication process. (I actually did preprocess to drop files using sha256 so not compatible with the md5 files)

If it is generated before everything is done then there will be a problem if the process fails at a later stage. I wonder if there can be some Atomicity policy, that if one file is not completely processed, the operation will be reverted.

[github-actions]

This issue is stale because it has been open 60 days with no activity. Remove stale label or comment or this issue will be closed.