microsoft/semantic-kernel · error · MemoryConnectorInitializationError

Index Name cannot be empty.

Error message

Index Name cannot be empty.

What it means

Constructor guard in `AzureCosmosDBMemoryStore.__init__`: if `index_name is None` it raises `MemoryConnectorInitializationError`. An index (vector index) name is required for vector search configuration, so None is rejected. (As with the database-name check, only `None` is rejected — empty strings pass.)

Source

Thrown at python/semantic_kernel/connectors/memory_stores/azure_cosmosdb/azure_cosmos_db_memory_store.py:70

        self,
        cosmosStore: AzureCosmosDBStoreApi,
        database_name: str,
        index_name: str,
        vector_dimensions: int,
        num_lists: int = 100,
        similarity: CosmosDBSimilarityType = CosmosDBSimilarityType.COS,
        kind: CosmosDBVectorSearchType = CosmosDBVectorSearchType.VECTOR_HNSW,
        m: int = 16,
        ef_construction: int = 64,
        ef_search: int = 40,
    ):
        """Initializes a new instance of the AzureCosmosDBMemoryStore class."""
        if vector_dimensions <= 0:
            raise MemoryConnectorInitializationError("Vector dimensions must be a positive number.")
        if database_name is None:
            raise MemoryConnectorInitializationError("Database Name cannot be empty.")
        if index_name is None:
            raise MemoryConnectorInitializationError("Index Name cannot be empty.")

        self.cosmos_store = cosmosStore
        self.index_name = index_name
        self.num_lists = num_lists
        self.similarity = similarity
        self.kind = kind
        self.m = m
        self.ef_construction = ef_construction
        self.ef_search = ef_search

    @staticmethod
    async def create(
        database_name: str,
        collection_name: str,
        vector_dimensions: int,
        num_lists: int,
        similarity: CosmosDBSimilarityType,
        kind: CosmosDBVectorSearchType,

View on GitHub (pinned to c028a0c7dc)

Solutions

1. Pass a non-None `index_name` (the name of the vector index to create/use in Cosmos).
2. Populate the config/env var backing the index name before construction.
3. Add your own empty-string check, since the library only guards None.

Example fix

// before
store = AzureCosmosDBMemoryStore(cosmos_store, "db", None, vector_dimensions=1536)

// after
store = AzureCosmosDBMemoryStore(cosmos_store, "db", "vector-index", vector_dimensions=1536)

def valid_index_name(name) -> bool:
    return isinstance(name, str) and name.strip() != ""

# assert valid_index_name(index_name) before constructing the store

def is_non_empty_str(name) -> bool:
    return isinstance(name, str) and name.strip() != ""

from semantic_kernel.exceptions import MemoryConnectorInitializationError
try:
    store = AzureCosmosDBMemoryStore(cs, "db", index_name, vector_dimensions=1536)
except MemoryConnectorInitializationError as e:
    if "Index Name" in str(e):
        raise SystemExit("provide a non-empty index_name") from e
    raise

When it happens

Trigger: Constructing the store with `index_name=None` (omitted/unset config value for the vector index name).

Common situations: Config field for the index name not set; passing a None variable by mistake; env var absent; argument-ordering error in the constructor call.

Related errors

• Database Name cannot be empty.
• Failed to create AstraDB settings.
• Dimensionality of {self._embedding_dim} exceeds the maximum
• Failed to create Azure Cognitive Search settings.
• Vector dimensions must be a positive number.

AI-assisted analysis of microsoft/semantic-kernel@c028a0c7dc (2026-08-13). Data as JSON: /api/errors/15e27a8bb2fe8894. Report an issue: GitHub.