microsoft/semantic-kernel · error · ServiceResponseException

Get failed due to: {e}

Error message

Get failed due to: {e}

What it means

Raised as a ServiceResponseException in MilvusMemoryStore.get_batch when self.collections[collection_name].load() or .query() throws any Exception. The load() step pulls collection data into memory and query() executes the filter expression; failures in either are wrapped with the original error chained.

Source

Thrown at python/semantic_kernel/connectors/memory_stores/milvus/milvus_memory_store.py:343

            e: _description_

        Returns:
            List[MemoryRecord]: _description_
        """
        # Check if the collection exists
        if not utility.has_collection(collection_name):
            logger.debug(f"Collection {collection_name} does not exist, cannot get.")
            raise ServiceResourceNotFoundError(f"Collection {collection_name} does not exist, cannot get.")

        try:
            self.collections[collection_name].load()
            gets = self.collections[collection_name].query(
                expr=f"{SEARCH_FIELD_ID} in {keys}",
                output_fields=OUTPUT_FIELDS_W_EMBEDDING if with_embeddings else OUTPUT_FIELDS_WO_EMBEDDING,
            )
        except Exception as e:
            logger.debug(f"Get failed due to: {e}")
            raise ServiceResponseException(f"Get failed due to: {e}") from e
        return [milvus_dict_to_memoryrecord(get) for get in gets]

    async def remove(self, collection_name: str, key: str) -> None:
        """Remove the specified record based on key.

        Args:
            collection_name (str): Collection to remove from.
            key (str): The key to remove.
        """
        await self.remove_batch(collection_name=collection_name, keys=[key])

    async def remove_batch(self, collection_name: str, keys: list[str]) -> None:
        """Remove multiple records based on keys.

        Args:
            collection_name (str): Collection to remove from
            keys (List[str]): The list of keys.

View on GitHub (pinned to c028a0c7dc)

Solutions

  1. Inspect the interpolated {e} to distinguish load failures from query/expression failures.
  2. Ensure keys are valid strings forming a correct Milvus 'in' expression.
  3. Add a short wait or retry after recent inserts so data is flushed and loaded.
  4. Check Milvus server memory and query-node health.

Example fix

// before
records = await store.get_batch('docs', keys, with_embeddings=True)  # ServiceResponseException
// after
try:
    records = await store.get_batch('docs', keys, with_embeddings=True)
except ServiceResponseException as e:
    logging.warning('Milvus get failed, retrying: %s', e)
    records = []  # or implement backoff retry
Defensive patterns

Strategy: try-catch

Validate before calling

def valid_milvus_keys(keys: list[str]) -> bool:
    return bool(keys) and all(isinstance(k, str) and k for k in keys)

if not valid_milvus_keys(keys):
    raise ValueError('Invalid keys for Milvus query expression')

Try / catch

from semantic_kernel.exceptions import ServiceResponseException

for attempt in range(3):
    try:
        records = await store.get_batch('docs', keys, with_embeddings=True)
        break
    except ServiceResponseException as e:
        if attempt == 2:
            raise
        await asyncio.sleep(2 ** attempt)

Prevention

When it happens

Trigger: Milvus .load() fails due to insufficient memory or the collection not being ready. .query() fails due to a malformed filter expression (the f-string builds expr=f'{SEARCH_FIELD_ID} in {keys}'), missing fields, or connection issues.

Common situations: Querying immediately after insert before the data is flushed/indexed. Malformed keys list producing an invalid Milvus expression. Memory pressure causing load() to fail. Network interruption during query.

Related errors

AI-assisted analysis of microsoft/semantic-kernel@c028a0c7dc (2026-08-13). Data as JSON: /api/errors/ecb72e161b9ff458. Report an issue: GitHub.