Amazon Opensearch Serverless - Index creation issue if AWS Cloud Quest used
Dear All, in the AWS Skillbuilder's "AWS Cloud Quest: Generative AI Practitioner" course, in the "Create an AI Smart Assistant" practice the Opensearch service creation failed several times between 04:00-07:00 CET on 2025-09-06. (between training steps 19-21) The issue is at creating the knowledge base. The last correct feedback from AWS Console: "Preparing vector database in Amazon Opensearch Serverless. This process may take several minutes to complete." After about 2 minutes waiting, the following error message is provided: "Failed to create the Amazon Opensearch Serverless collection. Index creation failed" Due to this error the creation of hr-knowledge-base is not possible. I have tested this from a client connected to multiple VPNs (EU and US) and in normal and private window of Chrome, with the same result. I have followed the training instructions by the book. Deleting existing knowledge bases is not possible via the lab environment, or changing anything about the provided rigths. Is there any way to bypass this issue in the course?
Many thanks Blaise
- Language
- English
- Newest
- Most votes
- Most comments
Hello blaisephilip,
Thank you for your patience while we worked on resolving the OpenSearch Serverless collection creation issue.
I'm pleased to let you know that our internal team has fixed the lab environment and you should now be able to create the index successfully.
Please go ahead and try the lab again. If you encounter any issues, just let me know and I'll be happy to help.
i have the same problem and have not found a solution yet
Hello blaisephilip,
I am also able to reproduce the issue you're experiencing with the OpenSearch Serverless collection creation. I've reached out to our internal team to get this fixed as soon as possible.
I'll keep you updated on the progress and let you know once we have a resolution in place.
Thanks for your patience while we work on this.
The error you're encountering in the AWS Cloud Quest course is likely related to a known issue with OpenSearch Serverless index creation. When creating a knowledge base with OpenSearch Serverless as the vector store, there can be delays in index propagation and availability.
This issue occurs because when an index is created in OpenSearch Serverless, there's a delay (sometimes up to 30 seconds or more) before the index becomes fully operational. During this period, the index might appear in listings but may not be fully healthy or available for other services like Amazon Bedrock to use, resulting in the error message you're seeing.
Since you're working in a lab environment with limited permissions, here are some potential workarounds:
-
Wait longer between steps - after initiating the OpenSearch Serverless collection creation, wait for 5-10 minutes before proceeding to the next step of creating the knowledge base.
-
Retry the knowledge base creation multiple times - sometimes it may work on subsequent attempts as the index becomes available.
-
If possible, check if there's an option to refresh or restart the lab environment completely.
-
Try at a different time of day - there might be resource constraints during peak hours that are affecting the service provisioning.
-
Contact the AWS Skillbuilder support team to report this specific issue with the lab, as they may need to adjust the lab configuration or provide an alternative path.
This appears to be a timing issue related to resource provisioning rather than something you're doing incorrectly in the lab steps.
Sources
AWS OpenSearch Serverless index creation delay | AWS re:Post
Bedrock Knowledgebase Provisioning Issue with OpenSearch Serverless Collection as the Vector Store | AWS re:Post
Error creating knowledge bases amazon bedrock "Failed to create the Amazon Opensearch Serverless collection." | AWS re:Post
Relevant content
- AWS OFFICIALUpdated 3 months ago
Hello, I've also had this same issue when trying to create a vector database in OpenSearch Serverless from a knowledge base. The error is "Failed to create the Amazon Opensearch Serverless collection. Index creation failed". The issue still persists for me, is there a solution?
Hello @Makendran G, thanks for the update. Your team's change in vector store storage from "QuickCreate" to "existing" fixed the course issue of AI Smart Assistant, both in Cloud Quest and the Simulearn learning path. Thank you again