ClickHouse/tests/integration/test_alter_moving_garbage/test.py

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

251 lines
8.4 KiB
Python
Raw Normal View History

2023-06-05 18:23:24 +00:00
import logging
import time
import pytest
import threading
import random
2023-06-05 18:23:24 +00:00
from helpers.client import QueryRuntimeException
from helpers.cluster import ClickHouseCluster
# two replicas in remote_servers.xml
REPLICA_COUNT = 2
2023-06-05 18:23:24 +00:00
2023-06-08 10:29:01 +00:00
2023-06-05 18:23:24 +00:00
@pytest.fixture(scope="module")
def cluster():
try:
cluster = ClickHouseCluster(__file__)
for i in range(1, REPLICA_COUNT + 1):
cluster.add_instance(
f"node{i}",
main_configs=[
"configs/config.d/storage_conf.xml",
"configs/config.d/remote_servers.xml",
],
with_minio=True,
with_zookeeper=True,
)
2023-06-05 18:23:24 +00:00
logging.info("Starting cluster...")
cluster.start()
logging.info("Cluster started")
yield cluster
finally:
cluster.shutdown()
def create_table(node, table_name, replicated, additional_settings):
2023-06-05 18:23:24 +00:00
settings = {
"storage_policy": "two_disks",
2023-07-08 13:33:49 +00:00
"old_parts_lifetime": 0,
2023-06-05 18:23:24 +00:00
"index_granularity": 512,
"temporary_directories_lifetime": 0,
"merge_tree_clear_old_temporary_directories_interval_seconds": 1,
}
settings.update(additional_settings)
table_engine = (
f"ReplicatedMergeTree('/clickhouse/tables/0/{table_name}', '{node.name}')"
if replicated
else "MergeTree()"
)
2023-06-05 18:23:24 +00:00
create_table_statement = f"""
CREATE TABLE {table_name} (
dt Date,
id Int64,
data String,
INDEX min_max (id) TYPE minmax GRANULARITY 3
) ENGINE = {table_engine}
2023-06-05 18:23:24 +00:00
PARTITION BY dt
ORDER BY (dt, id)
SETTINGS {",".join((k+"="+repr(v) for k, v in settings.items()))}"""
if replicated:
node.query_with_retry(create_table_statement)
else:
node.query(create_table_statement)
2023-06-05 18:23:24 +00:00
@pytest.mark.parametrize(
"allow_remote_fs_zero_copy_replication,replicated_engine",
[(False, False), (False, True), (True, True)],
)
2023-07-08 13:33:49 +00:00
def test_alter_moving(
cluster, allow_remote_fs_zero_copy_replication, replicated_engine
):
2023-07-08 13:33:49 +00:00
"""
Test that we correctly move parts during ALTER TABLE
"""
if replicated_engine:
nodes = list(cluster.instances.values())
else:
nodes = [cluster.instances["node1"]]
2023-06-06 15:17:06 +00:00
additional_settings = {}
2023-06-08 10:29:01 +00:00
# Different names for logs readability
table_name = "test_table"
2023-06-06 15:17:06 +00:00
if allow_remote_fs_zero_copy_replication:
table_name = "test_table_zero_copy"
additional_settings["allow_remote_fs_zero_copy_replication"] = 1
if replicated_engine:
table_name = table_name + "_replicated"
2023-06-06 15:17:06 +00:00
for node in nodes:
create_table(node, table_name, replicated_engine, additional_settings)
2023-06-06 15:17:06 +00:00
for i in range(1, 11):
partition = f"2021-01-{i:02d}"
random.choice(nodes).query(
f"INSERT INTO {table_name} SELECT toDate('{partition}'), number as id, toString(sipHash64(number, {i})) FROM numbers(10_000)"
)
2023-06-08 10:29:01 +00:00
# Run ALTER in parallel with moving parts
2023-06-05 18:23:24 +00:00
stop_alter = False
def alter():
random.choice(nodes).query(f"ALTER TABLE {table_name} ADD COLUMN col0 String")
for d in range(1, 100):
if stop_alter:
break
2023-06-08 10:29:01 +00:00
# Some lightweight mutation should change moving part before it is swapped, then we will have to cleanup it.
# Messages `Failed to swap {}. Active part doesn't exist` should appear in logs.
#
# I managed to reproduce issue with DELETE (`ALTER TABLE {table_name} ADD/DROP COLUMN` also works on real s3 instead of minio)
# Note: do not delete rows with id % 100 = 0, because they are used in `check_count` to use them in check that data is not corrupted
random.choice(nodes).query(f"DELETE FROM {table_name} WHERE id % 100 = {d}")
2023-06-08 10:29:01 +00:00
2023-06-05 18:23:24 +00:00
time.sleep(0.1)
alter_thread = threading.Thread(target=alter)
alter_thread.start()
for i in range(1, 11):
2023-06-05 18:23:24 +00:00
partition = f"2021-01-{i:02d}"
try:
random.choice(nodes).query(
2023-07-08 13:33:49 +00:00
f"ALTER TABLE {table_name} MOVE PARTITION '{partition}' TO DISK 's31'",
2023-06-05 18:23:24 +00:00
)
except QueryRuntimeException as e:
2023-06-06 15:17:06 +00:00
if "PART_IS_TEMPORARILY_LOCKED" in str(e):
continue
raise e
2023-06-05 18:23:24 +00:00
# Function to clear old temporary directories wakes up every 1 second, sleep to make sure it is called
2023-06-05 18:23:24 +00:00
time.sleep(0.5)
stop_alter = True
alter_thread.join()
2023-06-08 10:29:01 +00:00
# Check that no data was lost
data_digest = None
if replicated_engine:
# We don't know what data was replicated, so we need to check all replicas and take unique values
data_digest = random.choice(nodes).query_with_retry(
f"SELECT countDistinct(dt, data) FROM clusterAllReplicas(test_cluster, default.{table_name}) WHERE id % 100 == 0"
)
else:
data_digest = random.choice(nodes).query(
f"SELECT countDistinct(dt, data) FROM {table_name} WHERE id % 100 == 0"
)
assert data_digest == "1000\n"
2023-07-08 13:33:49 +00:00
def test_delete_race_leftovers(cluster):
"""
Test that we correctly delete outdated parts and do not leave any leftovers on s3
"""
node = cluster.instances["node1"]
table_name = "test_delete_race_leftovers"
additional_settings = {
# use another disk not to interfere with other tests
"storage_policy": "one_disk",
# always remove parts in parallel
"concurrent_part_removal_threshold": 1,
}
create_table(
node, table_name, replicated=True, additional_settings=additional_settings
)
# Stop merges to have several small parts in active set
node.query(f"SYSTEM STOP MERGES {table_name}")
# Creare several small parts in one partition
for i in range(1, 11):
node.query(
f"INSERT INTO {table_name} SELECT toDate('2021-01-01'), number as id, toString(sipHash64(number, {i})) FROM numbers(10_000)"
)
table_digest_query = f"SELECT count(), sum(sipHash64(id, data)) FROM {table_name}"
table_digest = node.query(table_digest_query)
# Execute several noop deletes to have parts with updated mutation id without changes in data
# New parts will have symlinks to old parts
node.query(f"SYSTEM START MERGES {table_name}")
for i in range(10):
node.query(f"DELETE FROM {table_name} WHERE data = ''")
# Make existing parts outdated
# Also we don't want have changing parts set,
# because it will be difficult match objects on s3 and in remote_data_paths to check correctness
node.query(f"OPTIMIZE TABLE {table_name} FINAL")
inactive_parts_query = (
f"SELECT count() FROM system.parts "
f"WHERE not active AND table = '{table_name}' AND database = 'default'"
)
# Try to wait for deletion of outdated parts
# However, we do not want to wait too long
# If some parts are not deleted after several iterations, we will just continue
for i in range(20):
inactive_parts_count = int(node.query(inactive_parts_query).strip())
if inactive_parts_count == 0:
print(f"Inactive parts are deleted after {i} iterations")
break
print(f"Inactive parts count: {inactive_parts_count}")
time.sleep(5)
# Check that we correctly deleted all outdated parts and no leftovers on s3
# Do it with retries because we delete blobs in the background
# and it can be race condition between removing from remote_data_paths and deleting blobs
all_remote_paths = set()
known_remote_paths = set()
for i in range(3):
known_remote_paths = set(
node.query(
f"SELECT remote_path FROM system.remote_data_paths WHERE disk_name = 's32'"
).splitlines()
)
2023-07-08 13:33:49 +00:00
all_remote_paths = set(
obj.object_name
for obj in cluster.minio_client.list_objects(
cluster.minio_bucket, "data2/", recursive=True
)
2023-07-08 13:33:49 +00:00
)
# Some blobs can be deleted after we listed remote_data_paths
# It's alright, thus we check only that all remote paths are known
# (in other words, all remote paths is subset of known paths)
if all_remote_paths == {p for p in known_remote_paths if p in all_remote_paths}:
break
time.sleep(1)
2023-07-08 13:33:49 +00:00
assert all_remote_paths == {p for p in known_remote_paths if p in all_remote_paths}
# Check that we have all data
assert table_digest == node.query(table_digest_query)