Hello @Charles Lawson ,
Thank you for your update, and glad to hear the issue is resolved!
Just to summarize for clarity and for future readers:
Since your source was Blob Storage and the index was showing triplicate results, the root cause was likely due to stale or orphaned documents not getting cleaned up properly during re-indexing. Azure AI Search indexers, by default, do not detect deletions unless you enable .
By choosing Native Blob Soft Delete for your blob data source and recreating the index , you enabled Azure AI Search to automatically detect and remove blobs that were soft deleted. This ensured the search index reflected only the current active files, eliminating duplicate or triplicate entries.
- Always enable Native Blob Soft Delete if your storage is Azure Blob Storage.
- Ensure the storage account's soft delete feature is enabled with a sufficient retention period.
- Configure the data source with Track deletions checked, and select Native blob soft delete.
- Use the latest supported API version (e.g.,
2024-07-01-preview
) when configuring via REST API.
References:
- Change and delete detection using indexers for Azure Storage
- Azure AI Search Vector Store - Physical structure and size
Please don't forget to click "Accept the answer" and click "yes" wherever the information provided helps you, this can be beneficial to other community members.
If you have any more questions, feel free to click Comment