mirror of
https://github.com/hanzoai/datastore.git
synced 2026-09-26 13:55:07 +00:00
Raúl MarínandClaude Opus 4.8
aec980f136
Each iteration checks an invariant (both nodes log the same number of entries for the same distributed query), not a probability fraction, so the count only governs how many independent chances we get to catch a discrepancy. 20 iterations is plenty. Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
74 lines
2.3 KiB
Python
74 lines
2.3 KiB
Python
import pytest
|
|
|
|
from helpers.cluster import ClickHouseCluster
|
|
|
|
cluster = ClickHouseCluster(__file__)
|
|
node1 = cluster.add_instance("node1", with_zookeeper=False)
|
|
node2 = cluster.add_instance("node2", with_zookeeper=False)
|
|
|
|
|
|
@pytest.fixture(scope="module")
|
|
def start_cluster():
|
|
try:
|
|
cluster.start()
|
|
yield cluster
|
|
|
|
finally:
|
|
cluster.shutdown()
|
|
|
|
|
|
def test_log_queries_probability_one(start_cluster):
|
|
for i in range(100):
|
|
node1.query("SELECT 12345", settings={"log_queries_probability": 0.5})
|
|
|
|
node1.query("SYSTEM FLUSH LOGS")
|
|
|
|
assert (
|
|
node1.query(
|
|
"SELECT count() < (2 * 100) FROM system.query_log WHERE query LIKE '%12345%' AND query NOT LIKE '%system.query_log%'"
|
|
)
|
|
== "1\n"
|
|
)
|
|
assert (
|
|
node1.query(
|
|
"SELECT count() > 0 FROM system.query_log WHERE query LIKE '%12345%' AND query NOT LIKE '%system.query_log%'"
|
|
)
|
|
== "1\n"
|
|
)
|
|
assert (
|
|
node1.query(
|
|
"SELECT count() % 2 FROM system.query_log WHERE query LIKE '%12345%' AND query NOT LIKE '%system.query_log%'"
|
|
)
|
|
== "0\n"
|
|
)
|
|
|
|
node1.query("TRUNCATE TABLE system.query_log")
|
|
|
|
|
|
def test_log_queries_probability_two(start_cluster):
|
|
# Each iteration checks an invariant (both nodes log the same number of
|
|
# entries for the same distributed query), not a probability fraction, so
|
|
# the count only governs how many independent chances we get to catch a
|
|
# discrepancy. 20 iterations is plenty; 100 just multiplies CPU cost under
|
|
# instrumentation.
|
|
for i in range(20):
|
|
node1.query(
|
|
"SELECT 12345 FROM remote('node2', system, one)",
|
|
settings={"log_queries_probability": 0.5},
|
|
)
|
|
|
|
node1.query("SYSTEM FLUSH LOGS")
|
|
node2.query("SYSTEM FLUSH LOGS")
|
|
|
|
ans1 = node1.query(
|
|
"SELECT count() FROM system.query_log WHERE query LIKE '%12345%' AND query NOT LIKE '%system.query_log%'"
|
|
)
|
|
ans2 = node2.query(
|
|
"SELECT count() FROM system.query_log WHERE query LIKE '%12345%' AND query NOT LIKE '%system.query_log%'"
|
|
)
|
|
|
|
assert ans1 == ans2
|
|
|
|
node1.query("TRUNCATE TABLE system.query_log")
|
|
node2.query("TRUNCATE TABLE system.query_log")
|