This page summarizes some errors that may occur related to the Azure AISearch Vector Store Provider.
Executing an operation against the index, for example trying to ingest a document the following error occurs
Failed to delete GUID An index with the name 'some_name' in service 'azure-service-name' could not be created.
Index quota has been exceeded for this service.
You must either delete unused indexes first, add a delay between index creation requests, or upgrade the service for higher limits. Maximum number of indexes allowed: 3
Check your Azure AI service tier and update it accordingly. Consider that each RAG Assistant will use an AI Search index.
The following error appears:
401 Access denied due to invalid subscription key or wrong API endpoint.
Make sure to provide a valid key for an active subscription and use a correct regional API endpoint for your resource.
Check with the provider that the vectorstore BaseURL and associated API Key matches your subscription.
The following error appears:
Semantic search is not enabled for this service
Parameter name: queryType
Verify your Azure Cognitive services subscription, the Free tier does not support using the semantic_hybrid search option, you need to enable the Semantic Ranker.
Check with the provider that the vectorstore BaseURL and associated API Key matches your subscription.
The following error appears:
Failed to ingest document data Invalid filter key format
This error indicates that the value used for the document Key field contains unsupported characters. In Azure AI Search, key values must follow the allowed format and cannot contain spaces.
Check the metadata field mapped as the document key and ensure its value complies with Azure AI Search restrictions. For example, "my sample" is not valid as a key value.
If necessary, replace spaces and other unsupported characters before uploading the document. For more details, see the Azure AI Search documentation here.