list_persistent_segments()

This operation lists all persistent (flushed) segments for a collection, including information about row count, sort status, and storage level.

Request syntax

client.list_persistent_segments(
    collection_name: str,
    timeout: float = None,
    states: Optional[Sequence[SegmentState]] = None
) -> List[SegmentInfo]

PARAMETERS:

• collection_name (str) -

  [REQUIRED]

  The name of the collection.

• timeout (float | None) -

  The timeout duration for this operation. Setting this to None indicates that this operation timeouts when any response arrives or any error occurs.

• states (Optional[Sequence[SegmentState]]) -

  The segment states to include in the result. When omitted, the server preserves the legacy persistent-state filter (growing, sealed, flushed, flushing, importing, and dropped segments).

  This parameter is available in PyMilvus v3.0.2 or later.

RETURN TYPE:

List[SegmentInfo]

RETURNS:

A list of persistent segment information objects with the following members:

• segment_id (int) -

  The ID of the segment.

• collection_id (int) -

  The ID of the collection that the segment belongs to.

• collection_name (str) -

  The name of the collection that the segment belongs to.

• num_rows (int) -

  The number of rows in the segment.

• is_sorted (bool) -

  Whether the data in the segment is sorted.

• state (SegmentState) -

  The lifecycle state of the segment.

• level (SegmentLevel) -

  The storage level of the segment.

• storage_version (int) -

  The storage version of the segment.

• partition_id (int) -

  The ID of the partition that the segment belongs to. Available in PyMilvus v3.0.2 or later.

• insert_channel (str) -

  The insert channel that the segment data was written to. Available in PyMilvus v3.0.2 or later.

• compaction_from (List[int]) -

  The IDs of the segments that this segment was compacted from. Available in PyMilvus v3.0.2 or later.

EXCEPTIONS:

• MilvusException

  This exception will be raised when any error occurs during this operation.

Example

from pymilvus import MilvusClient, SegmentState

client = MilvusClient(
    uri="http://localhost:19530",
    token="root:Milvus"
)

# Create a collection and insert some data
client.create_collection(collection_name="my_collection", dimension=5)
client.insert(collection_name="my_collection", data=[{"id": i, "vector": [0.1] * 5} for i in range(10)])

# Flush the data so that the inserted segments are persisted and become listable
client.flush(collection_name="my_collection")

segments = client.list_persistent_segments(collection_name="my_collection")
for seg in segments:
    print(f"Segment {seg.segment_id}: {seg.num_rows} rows, level={seg.level}")

# List only flushed and sealed segments
segments = client.list_persistent_segments(
    collection_name="my_collection",
    states=[SegmentState.Flushed, SegmentState.Sealed],
)