search

Azure-Samples / azure-search-openai-demo

A sample app for the Retrieval-Augmented Generation pattern running in Azure, using Azure AI Search for retrieval and Azure OpenAI large language models to power ChatGPT-style and Q&A experiences.
https://azure.microsoft.com/products/search
MIT License
5.57k stars 3.74k forks source link

ADLSG2 migration - Citation / pdf vizualiser : Not Found error #1723

Open tlievre opened 1 week ago

tlievre commented 1 week ago 
- [x] bug report -> please search issues before submitting
- [x] feature request
- [ ] documentation issue or request
- [ ] regression (a behavior that used to work and stopped in a new release)

Minimal steps to reproduce

  • Migrate storage account to ADSG2 for login & acl
  • Create a new empty storage account to make deployment succeed
  • run azd up

Any log messages given by the failure

In the citation window : "Not Found The requested URL was not found on the server. If you entered the URL manually please check your spelling and try again."

In the terminal after running .\app\start.ps1

INFO:root:Opening file xxxx INFO:azure.identity.aio._internal.decorators:AzureCliCredential.get_token succeeded INFO:azure.core.pipeline.policies.http_logging_policy:Request URL: 'https://[Empty storage account name]/content/xxxx' Request method: 'GET' Request headers: 'x-ms-range': 'REDACTED' 'x-ms-version': 'REDACTED' 'Accept': 'application/xml' 'x-ms-date': 'REDACTED' 'x-ms-client-request-id': xxxx 'User-Agent': 'azsdk-python-storage-blob/12.19.1 Python/3.11.8 (Windows-10-xxxx)' 'Authorization': 'REDACTED' No body was attached to the request INFO:azure.core.pipeline.policies.http_logging_policy:Response status: 404 Response headers: 'Content-Length': '215' 'Content-Type': 'application/xml' 'Server': 'Windows-Azure-Blob/1.0 Microsoft-HTTPAPI/2.0' 'x-ms-request-id': xxxxx 'x-ms-client-request-id': xxxxx 'x-ms-version': 'REDACTED' 'x-ms-error-code': 'BlobNotFound' INFO] 127.0.0.1:xxxx GET /content/xxxx

Expected/desired behavior

The citation panel windows display the pdf page overview.

OS and Version?

Windows 10

azd version?

1.9.3

Mention any other details that might be useful

I have actually migrated my storage account to ADLG2 in order to enable login and ACL. To ensure a successful deployment, I have created a new empty storage account. Everything seems to be functioning properly except for the citation panel, which appears to be attempting to retrieve the document from the incorrect storage account. My question is, in order to avoid this error, should I duplicate all of the data to the storage account? Alternatively, if there is something I may have overlooked in the setup explanation, please let me know.

Thanks a lot, for your work !

tlievre commented 1 week ago

I assume the only way right now to make pdf vizualiser working in citation panel with ADLSG2 is to totally duplicate the data between ADLSG2 and Blob storage. It might be related to your answer @pamelafox in this ticket ADLSG2 - Deployment Error, I didn't fully catch that. Do you plane adding a feature to fully replace storage account with data lake when acl is enabled ?