Files
qdrant/lib/segment/tests/integration/segment_on_disk_snapshot.rs
Andrey Vasnetsov d91e8db5f9 Streaming snapshot unpacking (#8025)
* download tar

* compute sha256 for stream download

* wip: propagate unpacking into down to the logic, todo: validation

* unpacked snapshot validation

* Minor tweaks

* Fix typo

* validation during unpack

* cancellation token

* update docstring

* remove redundant dep

* Rearrange unpack functions

- Rename `safe_unpack.rs` into `tar_unpack.rs` so it would be listed
  near `tar_ext.rs` in IDEs.
- Replace calls like `ar = open_snapshot_archive(…); safe_unpack(ar, …);`
  with a single call to `tar_unpack_file(…)`.
- Put calls to `Archive::new(); Archive::set_overwrite(false);` inside
  `tar_unpack_reader` (was `safe_unpack`). So, now it is the only place
  that does `set_overwrite`.

* Let clippy complain if tar::Archive::unpack used

* Mock snapshot download URL

Instead of downloading from storage.googleapis.com every time the test
runs, put small snapshot file to the repo.

The snapshot file is created using this command:

    curl -s \
      https://storage.googleapis.com/qdrant-benchmark-snapshots/test-shard.snapshot \
    | tar \
      --delete segments/4ea958d8-0b64-4312-9a53-0cd857e93535.tar \
      --delete segments/65ac6276-8cca-4f5c-b767-9722190cee8b.tar \
      > lib/storage/src/content_manager/snapshots/test-shard.snapshot

File contents:

    $ tar tf lib/storage/src/content_manager/snapshots/test-shard.snapshot
    wal/
    wal/closed-255
    newest_clocks.json
    replica_state.json
    shard_config.json
    
    $ du -sh lib/storage/src/content_manager/snapshots/test-shard.snapshot
    12K	lib/storage/src/content_manager/snapshots/test-shard.snapshot
    
    $ sha256sum < lib/storage/src/content_manager/snapshots/test-shard.snapshot       
    5d94eac5c1ede3994a28bc406120046c37370d5d45b489a0d2252531b4e3e1f2  -

---------

Co-authored-by: timvisee <tim@visee.me>
Co-authored-by: xzfc <xzfcpw@gmail.com>
2026-02-09 23:53:16 +01:00

227 lines
7.6 KiB
Rust

use std::collections::HashMap;
use std::sync::atomic::AtomicBool;
use common::tar_ext;
use common::tar_unpack::tar_unpack_file;
use fs_err as fs;
use fs_err::File;
use rstest::rstest;
use segment::data_types::index::{IntegerIndexParams, KeywordIndexParams};
use segment::data_types::vectors::{DEFAULT_VECTOR_NAME, only_default_vector};
use segment::entry::entry_point::SegmentEntry;
use segment::entry::snapshot_entry::SnapshotEntry as _;
use segment::json_path::JsonPath;
use segment::segment::Segment;
use segment::segment_constructor::load_segment;
use segment::segment_constructor::segment_builder::SegmentBuilder;
use segment::segment_constructor::simple_segment_constructor::build_simple_segment;
use segment::types::{
Distance, HnswConfig, Indexes, PayloadFieldSchema, PayloadSchemaParams, PayloadStorageType,
SegmentConfig, SnapshotFormat, VectorDataConfig, VectorStorageType,
};
use tempfile::Builder;
use uuid::Uuid;
/// This test tests snapshotting and restoring a segment with all on-disk components.
#[rstest]
#[case::regular(SnapshotFormat::Regular)]
#[case::streamable(SnapshotFormat::Streamable)]
fn test_on_disk_segment_snapshot(#[case] format: SnapshotFormat) {
use common::counter::hardware_counter::HardwareCounterCell;
use segment::types::HnswGlobalConfig;
let _ = env_logger::builder().is_test(true).try_init();
let data = r#"
{
"names": ["John Doe", "Bill Murray"],
"ages": [43, 51],
"metadata": {
"height": 50,
"width": 60
}
}"#;
let segment_builder_dir = Builder::new().prefix("segment_dir").tempdir().unwrap();
let mut segment = build_simple_segment(segment_builder_dir.path(), 2, Distance::Dot).unwrap();
let hw_counter = HardwareCounterCell::new();
segment
.upsert_point(0, 0.into(), only_default_vector(&[1.0, 1.0]), &hw_counter)
.unwrap();
segment
.upsert_point(1, 1.into(), only_default_vector(&[2.0, 2.0]), &hw_counter)
.unwrap();
segment
.set_full_payload(
2,
0.into(),
&serde_json::from_str(data).unwrap(),
&hw_counter,
)
.unwrap();
segment
.set_full_payload(
3,
0.into(),
&serde_json::from_str(data).unwrap(),
&hw_counter,
)
.unwrap();
segment
.create_field_index(
4,
&JsonPath::new("names"),
Some(&PayloadFieldSchema::FieldParams(
PayloadSchemaParams::Keyword(KeywordIndexParams {
r#type: segment::data_types::index::KeywordIndexType::Keyword,
is_tenant: None,
on_disk: Some(true),
enable_hnsw: None,
}),
)),
&hw_counter,
)
.unwrap();
segment
.create_field_index(
5,
&JsonPath::new("ages"),
Some(&PayloadFieldSchema::FieldParams(
PayloadSchemaParams::Integer(IntegerIndexParams {
r#type: segment::data_types::index::IntegerIndexType::Integer,
lookup: Some(true),
range: Some(true),
is_principal: None,
on_disk: Some(true),
enable_hnsw: None,
}),
)),
&hw_counter,
)
.unwrap();
let segment_config = SegmentConfig {
vector_data: HashMap::from([(
DEFAULT_VECTOR_NAME.to_owned(),
VectorDataConfig {
size: 2,
distance: Distance::Dot,
storage_type: VectorStorageType::Mmap, // mmap vectors
index: Indexes::Hnsw(HnswConfig {
m: 4,
ef_construct: 16,
full_scan_threshold: 8,
max_indexing_threads: 2,
on_disk: Some(true), // mmap index
payload_m: None,
inline_storage: None,
}),
quantization_config: None,
multivector_config: None,
datatype: None,
},
)]),
sparse_vector_data: Default::default(),
payload_storage_type: PayloadStorageType::Mmap,
};
let segment_base_dir = Builder::new().prefix("segment_dir").tempdir().unwrap();
let segment_builder_dir = Builder::new().prefix("segment_dir").tempdir().unwrap();
let mut segment_builder = SegmentBuilder::new(
segment_builder_dir.path(),
&segment_config,
&HnswGlobalConfig::default(),
)
.unwrap();
segment_builder.update(&[&segment], &false.into()).unwrap();
let segment = segment_builder.build_for_test(segment_base_dir.path());
let temp_dir = Builder::new().prefix("temp_dir").tempdir().unwrap();
// The segment snapshot is a part of a parent collection/shard snapshot.
let parent_snapshot_tar = Builder::new()
.prefix("parent_snapshot")
.suffix(".tar")
.tempfile()
.unwrap();
let segment_id = segment
.segment_path
.file_stem()
.and_then(|f| f.to_str())
.unwrap();
// snapshotting!
let tar =
tar_ext::BuilderExt::new_seekable_owned(File::create(parent_snapshot_tar.path()).unwrap());
segment
.take_snapshot(temp_dir.path(), &tar, format, None)
.unwrap();
tar.blocking_finish().unwrap();
let parent_snapshot_unpacked = Builder::new().prefix("parent_snapshot").tempdir().unwrap();
tar_unpack_file(parent_snapshot_tar.path(), parent_snapshot_unpacked.path()).unwrap();
// Should be exactly one entry in the snapshot.
let mut entries = fs::read_dir(parent_snapshot_unpacked.path()).unwrap();
let entry = entries.next().unwrap().unwrap();
assert!(entries.next().is_none());
match format {
SnapshotFormat::Ancient => unreachable!("The old days are gone"),
SnapshotFormat::Regular => {
assert_eq!(entry.file_name(), format!("{segment_id}.tar").as_str());
assert!(entry.path().is_file());
}
SnapshotFormat::Streamable => {
assert_eq!(entry.file_name(), segment_id);
assert!(entry.path().is_dir());
}
}
// restore snapshot
Segment::restore_snapshot_in_place(&entry.path()).unwrap();
// Should be exactly one entry in the snapshot.
let mut entries = fs::read_dir(parent_snapshot_unpacked.path()).unwrap();
let entry = entries.next().unwrap().unwrap();
assert!(entries.next().is_none());
// It should be unpacked entry, not tar archive.
assert!(entry.path().is_dir());
assert_eq!(entry.file_name(), segment_id);
let restored_segment =
load_segment(&entry.path(), Uuid::nil(), &AtomicBool::new(false)).unwrap();
// validate restored snapshot is the same as original segment
assert_eq!(
segment.total_point_count(),
restored_segment.total_point_count(),
);
assert_eq!(
segment.available_point_count(),
restored_segment.available_point_count(),
);
assert_eq!(
segment.deleted_point_count(),
restored_segment.deleted_point_count(),
);
let hw_counter = HardwareCounterCell::new();
for id in segment.iter_points() {
let vectors = segment.all_vectors(id, &hw_counter).unwrap();
let restored_vectors = restored_segment.all_vectors(id, &hw_counter).unwrap();
assert_eq!(vectors, restored_vectors);
let payload = segment.payload(id, &hw_counter).unwrap();
let restored_payload = restored_segment.payload(id, &hw_counter).unwrap();
assert_eq!(payload, restored_payload);
}
}