openai.APIConnectionError: Connection error.

[kachihro]

Hello,

I've configured (deployed) the solution, with a number of configurations - and using PRIVATE ENDPOINTS - with APP SERVICE.

Have stepped through the VPN config - and can view the homepage/chat-app.

When using this command ;
.\scripts\prepdocs.ps1

The documents are ingested to Blob Storage - and - the Document intelligence is extracting text ("Splitting into sections")

But - then an error with OpenAI - can't connect...?

Running "prepdocs.py"
./app/backend/prepdocs.py "C:\Repos\AI POC/data/" --verbose
[16:06:32] INFO Loading azd env from C:\Repos\AI POC.azure\rg-qleave-ai-poc.env, which may override existing environment load_azd_env.py:28
variables
INFO Connecting to Azure services using the azd credential for tenant e4fc5309-2c24-4591-8d51-23b1610b78b8 prepdocs.py:440
INFO Using local files: C:\Repos\AI POC/data/ prepdocs.py:144
INFO OPENAI_HOST is azure, setting up Azure OpenAI client prepdocs.py:225
INFO Using Azure credential (passwordless authentication) for Azure OpenAI client prepdocs.py:235
INFO Checking whether search index gptkbindex exists... searchmanager.py:85
[16:06:35] INFO Search index gptkbindex already exists searchmanager.py:329
INFO Skipping 'C:\Repos\AI POC/data\1-solar-panel-popular.pdf', no changes detected. listfilestrategy.py:129
INFO Skipping 'C:\Repos\AI POC/data\12-malcoms-snowy-surprise.pdf', no changes detected. listfilestrategy.py:129
INFO Skipping 'C:\Repos\AI POC/data\1218770116.pdf', no changes detected. listfilestrategy.py:129
[16:06:36] INFO Uploading blob for document '1218770116_851933223851182171.pdf' blobmanager.py:430
INFO Ingesting '1218770116_851933223851182171.pdf' filestrategy.py:30
INFO Extracting text from 'C:\Repos\AI POC/data\1218770116_851933223851182171.pdf' using Azure Document Intelligence pdfparser.py:94
[16:06:41] INFO Splitting '1218770116_851933223851182171.pdf' into sections filestrategy.py:42
Traceback (most recent call last):
File "C:\Repos\AI POC.venv\Lib\site-packages\httpx_transports\default.py", line 69, in map_httpcore_exceptions
yield
File "C:\Repos\AI POC.venv\Lib\site-packages\httpx_transports\default.py", line 373, in handle_async_request
resp = await self._pool.handle_async_request(req)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\httpcore_async\connection_pool.py", line 256, in handle_async_request
raise exc from None
File "C:\Repos\AI POC.venv\Lib\site-packages\httpcore_async\connection_pool.py", line 236, in handle_async_request
response = await connection.handle_async_request(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\httpcore_async\connection.py", line 101, in handle_async_request
raise exc
File "C:\Repos\AI POC.venv\Lib\site-packages\httpcore_async\connection.py", line 78, in handle_async_request
stream = await self._connect(request)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\httpcore_async\connection.py", line 124, in _connect
stream = await self._network_backend.connect_tcp(**kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\httpcore_backends\auto.py", line 31, in connect_tcp
return await self._backend.connect_tcp(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\httpcore_backends\anyio.py", line 113, in connect_tcp
with map_exceptions(exc_map):
File "C:\Python311\Lib\contextlib.py", line 155, in exit
self.gen.throw(typ, value, traceback)
File "C:\Repos\AI POC.venv\Lib\site-packages\httpcore_exceptions.py", line 14, in map_exceptions
raise to_exc(exc) from exc
httpcore.ConnectError: [Errno 11001] getaddrinfo failed

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

Traceback (most recent call last):
File "C:\Repos\AI POC.venv\Lib\site-packages\openai_base_client.py", line 1529, in request
response = await self._client.send(
^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\httpx_client.py", line 1661, in send
response = await self._send_handling_auth(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\httpx_client.py", line 1689, in _send_handling_auth
response = await self._send_handling_redirects(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\httpx_client.py", line 1726, in _send_handling_redirects
response = await self._send_single_request(request)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\httpx_client.py", line 1763, in _send_single_request
response = await transport.handle_async_request(request)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\httpx_transports\default.py", line 372, in handle_async_request
with map_httpcore_exceptions():
File "C:\Python311\Lib\contextlib.py", line 155, in exit
self.gen.throw(typ, value, traceback)
File "C:\Repos\AI POC.venv\Lib\site-packages\httpx_transports\default.py", line 86, in map_httpcore_exceptions
raise mapped_exc(message) from exc
httpx.ConnectError: [Errno 11001] getaddrinfo failed

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

Traceback (most recent call last):
File "C:\Repos\AI POC\app\backend\prepdocs.py", line 600, in
loop.run_until_complete(main(ingestion_strategy, setup_index=not args.remove and not args.removeall))
File "C:\Python311\Lib\asyncio\base_events.py", line 650, in run_until_complete
return future.result()
^^^^^^^^^^^^^^^
File "C:\Repos\AI POC\app\backend\prepdocs.py", line 364, in main
await strategy.run()
File "C:\Repos\AI POC\app\backend\prepdocslib\filestrategy.py", line 123, in run
await self.search_manager.update_content(sections, url=file.url)
File "C:\Repos\AI POC\app\backend\prepdocslib\searchmanager.py", line 524, in update_content
embeddings = await self.embeddings.create_embeddings(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC\app\backend\prepdocslib\embeddings.py", line 150, in create_embeddings
return await self.create_embedding_batch(texts, dimensions_args)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC\app\backend\prepdocslib\embeddings.py", line 106, in create_embedding_batch
async for attempt in AsyncRetrying(
File "C:\Repos\AI POC.venv\Lib\site-packages\tenacity\asyncio_init_.py", line 166, in anext
do = await self.iter(retry_state=self.retry_state)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\tenacity\asyncio_init.py", line 153, in iter
result = await action(retry_state)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\tenacity_utils.py", line 99, in inner
return call(args, kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\tenacity_init_.py", line 400, in
self._add_action_func(lambda rs: rs.outcome.result())
^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\concurrent\futures_base.py", line 449, in result
return self.__get_result()
^^^^^^^^^^^^^^^^^^^
File "C:\Python311\Lib\concurrent\futures_base.py", line 401, in __get_result
raise self._exception
File "C:\Repos\AI POC\app\backend\prepdocslib\embeddings.py", line 113, in create_embedding_batch
emb_response = await client.embeddings.create(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\openai\resources\embeddings.py", line 251, in create
return await self._post(
^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\openai_base_client.py", line 1794, in post
return await self.request(cast_to, opts, stream=stream, stream_cls=stream_cls)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Repos\AI POC.venv\Lib\site-packages\openai_base_client.py", line 1561, in request
raise APIConnectionError(request=request) from err*

openai.APIConnectionError: Connection error.

[pamelafox]

I just encountered this, and have included a fix in a related PR:
https://github.com/Azure-Samples/azure-search-openai-demo/pull/2740/files#diff-7ef659fc9cf6968e718894d300490b14ea7a52091e7d4bcffae3a5029ac721d4R1215

Please try removing OpenAI from the Cognitive Services endpoint here:
resourceIds: concat(
[openAi.outputs.resourceId],

It's redundant since it's already in the openAI endpoint, and it seems to make it so that the DNS zone does not have the A record correctly. Once I made that fix and re-ran "azd provision", it worked.

Once you re-provision, you should see an A table for your connection:

az network private-dns record-set a list \   
  -g rg-YOUR-RESOURCE-GROUP \
  -z privatelink.openai.azure.com -o table

And a 10.* IP address when you dig it:

dig +short cog-YOUR-RESOURCE-ID.openai.azure.com

[pamelafox]

Update: I realized you're the same developer that sent that PR! So now your issues should be resolved hopefully. Let me know if further issues.