Skip to content
By using AWS re:Post, you agree to the AWS re:Post Terms of Use
AWS re:Postre:Post
About re:Post

Attaching role to Aurora cluster doesn't allow the Bedrock feature

0

I am facing an issue while following the directions in the Section "Create an IAM role and policy for the Aurora cluster" from the blog post at https://aws.amazon.com/blogs/database/build-generative-ai-applications-with-amazon-aurora-and-amazon-bedrock-knowledge-bases/ . I have created an appropriate role for the RDS cluster using Postgres Engine Version 16.2, but while attaching the role I do not see the "Bedrock" feature in the dropdown list while performing Step 22, the only options I see are:

  • Comprehend
  • Lambda
  • s3Export
  • s3Import
  • SageMaker

I have been able to use the Aurora cluster as a knowledge base inside bedrock. I am unsure if there is some configuration that I am missing that does not allow me to use the "Bedrock" feature while attaching the role to the cluster. It would be great if I could use the Aurora ML capability to generate the embeddings. Any advice or help is appreciated!

Topics
Machine Learning & AIDatabaseGenerative AI on AWS
Tags
Machine Learning & AIAmazon AuroraAmazon Bedrock
Language
English

asked 2 years ago436 views

1 Answer
1
Accepted Answer

Hello.

When I created Aurora Serverless by selecting PostgreSQL version 15.5, which is the same as the AWS blog you are checking, I was able to select Bedrock.
I also confirmed that Bedrock can be selected in version 15 series 15.7.
However, Bedrock could not be selected in version 16 series.
a

I suspect that Aurora ML's Bedrock probably cannot be selected yet for version 16 series.(As of September 2024)

EXPERT

answered 2 years ago

EXPERT

reviewed 2 years ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Relevant content