microsoft/semantic-kernel · warning · MemoryConnectorResourceNotFound

Memory record not found

Error message

Memory record not found

What it means

Raised in `AzureCognitiveSearchMemoryStore.get` when the Azure Search `get_document` call throws `ResourceNotFoundError` (the document key does not exist in the index). The search client is closed and the exception is re-raised as `MemoryConnectorResourceNotFound` (chained via `from exc`). It is the equivalent of a key-not-found for this store.

Source

Thrown at python/semantic_kernel/connectors/memory_stores/azure_cognitive_search/azure_cognitive_search_memory_store.py:285

        Args:
            collection_name (str): The name of the collection to get the record from.
            key (str): The unique database key of the record.
            with_embedding (bool): Whether to include the embedding in the result. (default: {False})

        Returns:
            MemoryRecord: The record.
        """
        # Look up Search client class to see if exists or create
        search_client = self._search_index_client.get_search_client(collection_name.lower())

        try:
            search_result = await search_client.get_document(
                key=encode_id(key), selected_fields=get_field_selection(with_embedding)
            )
        except ResourceNotFoundError as exc:
            await search_client.close()
            raise MemoryConnectorResourceNotFound("Memory record not found") from exc

        await search_client.close()

        # Create Memory record from document
        return dict_to_memory_record(search_result, with_embedding)

    async def get_batch(
        self, collection_name: str, keys: list[str], with_embeddings: bool = False
    ) -> list[MemoryRecord]:
        """Gets a batch of records.

        Args:
            collection_name (str): The name of the collection to get the records from.
            keys (List[str]): The unique database keys of the records.
            with_embeddings (bool): Whether to include the embeddings in the results. (default: {False})

        Returns:
            List[MemoryRecord]: The records.

View on GitHub (pinned to c028a0c7dc)

Solutions

1. Confirm the record was upserted into that index and the key matches (account for `encode_id`).
2. Catch `MemoryConnectorResourceNotFound` to handle missing records gracefully.
3. Account for indexing latency: retry shortly after a fresh upsert, or check the index is ready.
4. Verify you query the same collection/index name used at write time.

Example fix

// before
rec = await store.get("idx", key)  # MemoryConnectorResourceNotFound

// after
from semantic_kernel.exceptions import MemoryConnectorResourceNotFound
try:
    rec = await store.get("idx", key)
except MemoryConnectorResourceNotFound:
    rec = None

# optional existence pre-check via batch (does not raise on miss)
async def acs_exists_or_none(store, collection, key):
    recs = await store.get_batch(collection, [key], with_embeddings=False)
    return recs[0] if recs else None

from semantic_kernel.exceptions import MemoryConnectorResourceNotFound
try:
    rec = await store.get("idx", key)
except MemoryConnectorResourceNotFound:
    rec = None

When it happens

Trigger: Calling `get(collection_name, key)` for a record whose encoded id (`encode_id(key)`) is not present in the Azure Search index; the record was never upserted, was deleted, or the id encoding differs between write and read.

Common situations: Read before write; querying the wrong index; `encode_id` produces different base64 for the same logical key due to whitespace/encoding changes; record deleted out-of-band; race during indexing lag (Azure Search is eventually consistent).

Related errors

• Record with key '{key}' does not exist
• Agent type '{recipient.type}' does not exist.
• Agent with name {agent_id.type} not found.
• Agent with name {id.type} not found.
• Failed to create Azure Cognitive Search settings.

AI-assisted analysis of microsoft/semantic-kernel@c028a0c7dc (2026-08-13). Data as JSON: /api/errors/358b71efa10e6b1a. Report an issue: GitHub.