microsoft/semantic-kernel · error · ServiceResourceNotFoundError

Record with key '{key}' does not exist in collection '{colle

Error message

Record with key '{key}' does not exist in collection '{collection_name}'

What it means

Raised as a ServiceResourceNotFoundError in ChromaMemoryStore.get when the requested key is absent. get delegates to get_batch and then indexes into the result list; an IndexError (empty result) is caught and re-raised as this error, identifying the missing key and collection.

Source

Thrown at python/semantic_kernel/connectors/memory_stores/chroma/chroma_memory_store.py:208

        # upsert is checking collection existence
        return [await self.upsert(collection_name, record) for record in records]

    async def get(self, collection_name: str, key: str, with_embedding: bool = False) -> MemoryRecord:
        """Gets a record.

        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.
        """
        records = await self.get_batch(collection_name, [key], with_embedding)
        try:
            return records[0]
        except IndexError as exc:
            raise ServiceResourceNotFoundError(
                f"Record with key '{key}' does not exist in collection '{collection_name}'"
            ) from exc

    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.
        """
        collection = await self.get_collection(collection_name)
        if collection is None:

View on GitHub (pinned to c028a0c7dc)

Solutions

Check existence first by calling get_batch and inspecting the list length, or catch ServiceResourceNotFoundError.
Verify the key matches the id used during upsert (record._key is set to record._id).
If absence is a valid application state, handle it gracefully rather than treating it as an error.
Log the missing key and collection to trace stale references.

Example fix

// before
record = await store.get('docs', key)  # ServiceResourceNotFoundError
// after
try:
    record = await store.get('docs', key)
except ServiceResourceNotFoundError:
    record = None  # handle absence gracefully

Defensive patterns

Strategy: try-catch

Validate before calling

records = await store.get_batch('docs', [key], with_embedding=False)
record = records[0] if records else None

Try / catch

from semantic_kernel.exceptions import ServiceResourceNotFoundError

try:
    record = await store.get('docs', key)
except ServiceResourceNotFoundError:
    record = None  # absent record is a valid application state

Prevention

Prefer get_batch and check list length when absence is expected.
Verify the key matches the id used during upsert (record._id).
Treat a missing record as a non-error business case where appropriate.
Log missing keys to detect stale references.

When it happens

Trigger: Calling await store.get('my_collection', key) where no record with that id/key exists in the Chroma collection. The key may have been deleted, never inserted, or is a stale reference from another store.

Common situations: Looking up a memory record by an id generated by a different embedding run. Deleting a record then querying it. Race condition where the record has not yet been flushed. Key format mismatch (e.g. the record._id vs record._key mapping).

Related errors

AI-assisted analysis of microsoft/semantic-kernel@c028a0c7dc (2026-08-13). Data as JSON: /api/errors/1f4f8cc050832473. Report an issue: GitHub.