AI search service - embedded/vector search failing

[huebell-byte]

I tried to set up semantic kernel by doing the following,

1. creating a blob storage, naming the storage value, using open AI to write some resume files to it in TXT format. They store successfully.

On the azure portal I can do a search also successfully that returns the indexed results via the AI search tool.

2. Then I tried to run an embed function before storing the file. Then in Python, I created a function that calls the value from the embedded file. Each time I tried to call the API and Point in Python. It failed, I went into the aszure foundry.

I confirmed several times that my end points are correct, I kept getting a 404 error, all my resources were in East US, I’m just wondering if access was blocked or I did something wrong

[selvicim45]

Try using postman to check the endpoint, if you are still getting 404, log the response in python using print stmts and see the detailed error msg to debug.

[huebell-byte]

Postman did return a 404 error. I ended up not using it as it appears to be a general restriction. I say that because I pivoted to app service and all the credentials were redacted. I was oh, maybe we just can’t use certain tools. Thank you.

…

On Thu, Mar 20, 2025 at 8:38 PM Arulselvi Amirrthalingam(Selvi) < ***@***.***> wrote: Try using postman to check the endpoint, if you are still getting 404, log the response in python using print stmts and see the detailed error msg to debug. — Reply to this email directly, view it on GitHub <#25 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BJXA3Y2AQ5LJYC4BAHPNPWT2VNUP3AVCNFSM6AAAAABZOBZ6P6VHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTENJXGE2TAOA> . You are receiving this because you authored the thread.Message ID: ***@***.*** .com>