topic Re: Getting Permission Denied on model in Machine Learning

Getting Permission Denied on model

BalaRamesh — Thu, 11 Jan 2024 06:18:00 GMT

while creating service endpoint , getting permission denied error. Need help how to provide the permission.

Re: Getting Permission Denied on model

Walter_C — Thu, 11 Jan 2024 16:37:10 GMT

If the model is from Workspace Model Registry, the creator of the endpoint must have Can Read permissions on the registered model of any model version that is specified in the endpoint configuration. If the model is from Unity Catalog, the creator of the endpoint must have the EXECUTE privilege on the registered model of any model version that is specified in the endpoint configuration, plus USE CATALOG and USE SCHEMA privileges on the parent catalog and schema of any such registered model.
https://docs.databricks.com/en/security/auth-authz/access-control/serving-endpoint-acl.html

Re: Getting Permission Denied on model

Debayan — Thu, 11 Jan 2024 18:11:11 GMT

In addition to Walter's contribution, model permission is the issue on highlight here.

Re: Getting Permission Denied on model

BalaRamesh — Fri, 12 Jan 2024 06:20:07 GMT

Thanks Walter and Debayan to given the reply . We given the permission as per model and below are screen shots for reference.

We started to understand about LLM Chatbot With Retrieval Augmented Generation (RAG) and Llama 2 70B(https://notebooks.databricks.com/demos/llm-rag-chatbot/index.html#). We did the below steps before execute the code .

We created the trail account in azure portal and created azure databricks with premium tier.
Created the storage account with same resource group
Created the azure access connect Access Connector for Azure Databricks
Given the permission to the Azure Access Connector to access Storage
In the Storage Account, using IAM given permission to Azure Access Connector Databricks
We launched the Databricks and created a temporary cluster with 14GB memory and 4 Cores.
Followed the tutorial (LLM Chatbot With Retrieval Augmented Generation (RAG) and Llama 2 70B)
Executed the below commands .

%pip install dbdemos

import dbdemos

dbdemos.install('llm-rag-chatbot')

9.Once executed the above commands, all related notebokes and cluster created but while start the cluster which is provided the demo , faced the error with insufficient executors then used the cluster which created temporary one at step 6. Then we are able to executed some of the commands until Splitting our html pages in smaller chunks in 01-Data-Preparation-and-Index note book

10. While running the next command Using Databricks Foundation model BGE as an embedding endpoint , here we created the model from databricks market place and tried to serve endpoint but we are getting access issue (permission issue).

Re: Getting Permission Denied on model

Adheer — Thu, 18 Jan 2024 08:57:01 GMT

I am facing the same issue for deploying BGE as embedding endpoint. It worked few days back but from today it is complaining with permission issue. I tried uninstalling and re-installing but that didn't solve the problem either.

Re: Getting Permission Denied on model

Debayan — Thu, 18 Jan 2024 19:01:12 GMT

Hi, @Adheer , Any screenshot of the issue?

Re: Getting Permission Denied on model

Debayan — Thu, 18 Jan 2024 19:06:13 GMT

Hi @BalaRamesh , Could you please remove and add once again to see?

Re: Getting Permission Denied on model

Adheer — Thu, 25 Jan 2024 13:43:07 GMT

@Debayan I managed to solve the issue using databricks sdk library. From UI it was failing with the same error as mentioned by @BalaRamesh