create_snapshot()

This operation creates a point-in-time snapshot of a collection. Use snapshots to back up collection data and metadata for disaster recovery or migration.

Request Syntax

create_snapshot(
    collection_name: str,
    snapshot_name: str,
    description: str = "",
    timeout: Optional[float] = None,
    **kwargs
) -> None

PARAMETERS:

• collection_name (str) - [REQUIRED] The name of the collection to snapshot.

• snapshot_name (str) - [REQUIRED] A unique name for the snapshot. Must not conflict with existing snapshot names.

• description (str) - An optional human-readable description of the snapshot.

• timeout (Optional[float]) - An optional duration of time in seconds to allow for the RPC. If not provided, the default client-side timeout is used.

RETURN TYPE:

None

EXCEPTIONS:

• MilvusException

  If the collection does not exist, the snapshot name is already taken, or the operation fails for any other reason.

Examples

from pymilvus import MilvusClient

client = MilvusClient(uri="http://localhost:19530")

# Recommended: flush before creating snapshot to persist in-memory data
client.flush(collection_name="my_collection")

client.create_snapshot(
    collection_name="my_collection",
    snapshot_name="backup_20260418",
    description="Daily backup before schema change",
)