ClickHouse/tests/integration/test_MemoryTracking/test.py

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

128 lines
3.5 KiB
Python
Raw Normal View History

# pylint: disable=unused-argument
# pylint: disable=redefined-outer-name
# pylint: disable=line-too-long
# This test verifies that memory tracking does not have significant drift,
# in other words, every allocation should be taken into account at the global
# memory tracker.
#
# So we are running some queries with GROUP BY to make some allocations,
# and after we are checking MemoryTracking metric from system.metrics,
# and check that it does not changes too frequently.
#
# Also note, that syncing MemoryTracking with RSS had been disabled in
# asynchronous_metrics_update_period_s.xml.
import logging
2024-09-27 10:19:39 +00:00
import pytest
2024-09-27 10:19:39 +00:00
from helpers.cluster import ClickHouseCluster
cluster = ClickHouseCluster(__file__)
node = cluster.add_instance(
"node",
main_configs=[
"configs/no_system_log.xml",
"configs/asynchronous_metrics_update_period_s.xml",
],
user_configs=[
"configs/users.d/overrides.xml",
],
)
@pytest.fixture(scope="module", autouse=True)
def start_cluster():
try:
cluster.start()
yield cluster
finally:
cluster.shutdown()
query_settings = {
"max_threads": 1,
"log_queries": 0,
}
sample_query = "SELECT groupArray(repeat('a', 1000)) FROM numbers(10000) GROUP BY number%10 FORMAT JSON"
def query(*args, **kwargs):
if "settings" not in kwargs:
kwargs["settings"] = query_settings
else:
kwargs["settings"].update(query_settings)
return node.query(*args, **kwargs)
def http_query(*args, **kwargs):
if "params" not in kwargs:
kwargs["params"] = query_settings
else:
kwargs["params"].update(query_settings)
return node.http_query(*args, **kwargs)
def get_MemoryTracking():
return int(
http_query("SELECT value FROM system.metrics WHERE metric = 'MemoryTracking'")
)
def check_memory(memory):
# bytes -> megabytes
memory = [*map(lambda x: int(int(x) / 1024 / 1024), memory)]
# 3 changes to MemoryTracking is minimum, since:
# - this is not that high to not detect inacuracy
# - memory can go like X/X+N due to some background allocations
# - memory can go like X/X+N/X, so at least 2 changes
changes_allowed = 3
# if number of samples is large enough, use 10% from them
# (actually most of the time there will be only few changes, it was made 10% to avoid flackiness)
changes_allowed_auto = int(len(memory) * 0.1)
changes_allowed = max(changes_allowed_auto, changes_allowed)
changed = len(set(memory))
logging.info(
"Changes: allowed=%s, actual=%s, sample=%s",
changes_allowed,
changed,
len(memory),
)
assert changed < changes_allowed
def test_http():
memory = []
memory.append(get_MemoryTracking())
for _ in range(100):
http_query(sample_query)
memory.append(get_MemoryTracking())
check_memory(memory)
def test_tcp_multiple_sessions():
memory = []
memory.append(get_MemoryTracking())
for _ in range(100):
query(sample_query)
memory.append(get_MemoryTracking())
check_memory(memory)
def test_tcp_single_session():
memory = []
memory.append(get_MemoryTracking())
sample_queries = [
sample_query,
"SELECT metric, value FROM system.metrics WHERE metric = 'MemoryTracking'",
] * 100
rows = query(";".join(sample_queries))
memory = rows.split("\n")
memory = filter(lambda x: x.startswith("MemoryTracking"), memory)
memory = map(lambda x: x.split("\t")[1], memory)
memory = [*memory]
check_memory(memory)