mirror of
https://github.com/ClickHouse/ClickHouse.git
synced 2024-11-18 05:32:52 +00:00
e1ad5cee7e
pytest play games with logging output [1]. [1]: https://github.com/pytest-dev/pytest/issues/5502 But this does not actually affect ClickHouse tests, the only reason is that the kazoo client is not stopped correctly without calling kazoo.client.KazooClient.stop(), and that's why you can see the following messages: <details> test_storage_rabbitmq/test.py::test_rabbitmq_big_message --- Logging error --- Traceback (most recent call last): File "/usr/local/lib/python3.8/dist-packages/kazoo/protocol/connection.py", line 179, in _socket_error_handling yield File "/usr/local/lib/python3.8/dist-packages/kazoo/protocol/connection.py", line 659, in _connect self._socket = self.handler.create_connection( File "/usr/local/lib/python3.8/dist-packages/kazoo/handlers/threading.py", line 178, in create_connection return utils.create_tcp_connection(socket, *args, **kwargs) File "/usr/local/lib/python3.8/dist-packages/kazoo/handlers/utils.py", line 265, in create_tcp_connection sock = module.create_connection(address, timeout_at) File "/usr/lib/python3.8/socket.py", line 808, in create_connection raise err File "/usr/lib/python3.8/socket.py", line 796, in create_connection sock.connect(sa) socket.timeout: timed out During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/local/lib/python3.8/dist-packages/kazoo/protocol/connection.py", line 579, in _connect_attempt read_timeout, connect_timeout = self._connect(host, hostip, port) File "/usr/local/lib/python3.8/dist-packages/kazoo/protocol/connection.py", line 659, in _connect self._socket = self.handler.create_connection( File "/usr/lib/python3.8/contextlib.py", line 131, in __exit__ self.gen.throw(type, value, traceback) File "/usr/local/lib/python3.8/dist-packages/kazoo/protocol/connection.py", line 182, in _socket_error_handling raise ConnectionDropped("socket connection error: %s" % (err,)) kazoo.exceptions.ConnectionDropped: socket connection error: None During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/usr/lib/python3.8/logging/__init__.py", line 1088, in emit stream.write(msg + self.terminator) ValueError: I/O operation on closed file. Call stack: File "/usr/lib/python3.8/threading.py", line 890, in _bootstrap self._bootstrap_inner() File "/usr/lib/python3.8/threading.py", line 932, in _bootstrap_inner self.run() File "/usr/lib/python3.8/threading.py", line 870, in run self._target(*self._args, **self._kwargs) File "/usr/local/lib/python3.8/dist-packages/kazoo/protocol/connection.py", line 512, in zk_loop if retry(self._connect_loop, retry) is STOP_CONNECTING: File "/usr/local/lib/python3.8/dist-packages/kazoo/retry.py", line 126, in __call__ return func(*args, **kwargs) File "/usr/local/lib/python3.8/dist-packages/kazoo/protocol/connection.py", line 552, in _connect_loop status = self._connect_attempt(host, hostip, port, retry) File "/usr/local/lib/python3.8/dist-packages/kazoo/protocol/connection.py", line 622, in _connect_attempt self.logger.warning('Connection dropped: %s', e) Message: 'Connection dropped: %s' Arguments: (ConnectionDropped('socket connection error: None'),) </details> Which eventually leads to incorrectly parsed report. This patch is an addition to the `logging.raiseExceptions=False` - #44618 Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
131 lines
3.8 KiB
Python
131 lines
3.8 KiB
Python
#!/usr/bin/env python3
|
|
|
|
import pytest
|
|
from helpers.cluster import ClickHouseCluster
|
|
import helpers.keeper_utils as keeper_utils
|
|
import random
|
|
import string
|
|
import os
|
|
import time
|
|
from multiprocessing.dummy import Pool
|
|
from helpers.test_tools import assert_eq_with_retry
|
|
from kazoo.client import KazooClient, KazooState
|
|
|
|
cluster = ClickHouseCluster(__file__)
|
|
CONFIG_DIR = os.path.join(os.path.dirname(os.path.realpath(__file__)), "configs")
|
|
|
|
node1 = cluster.add_instance(
|
|
"node1", main_configs=["configs/enable_keeper1.xml"], stay_alive=True
|
|
)
|
|
node2 = cluster.add_instance("node2", main_configs=[], stay_alive=True)
|
|
node3 = cluster.add_instance("node3", main_configs=[], stay_alive=True)
|
|
|
|
|
|
def get_fake_zk(node, timeout=30.0):
|
|
_fake_zk_instance = KazooClient(
|
|
hosts=cluster.get_instance_ip(node.name) + ":9181", timeout=timeout
|
|
)
|
|
_fake_zk_instance.start()
|
|
return _fake_zk_instance
|
|
|
|
|
|
@pytest.fixture(scope="module")
|
|
def started_cluster():
|
|
try:
|
|
cluster.start()
|
|
|
|
yield cluster
|
|
|
|
finally:
|
|
cluster.shutdown()
|
|
|
|
|
|
def start(node):
|
|
node.start_clickhouse()
|
|
keeper_utils.wait_until_connected(cluster, node)
|
|
|
|
|
|
def test_nodes_add(started_cluster):
|
|
zk_conn = None
|
|
zk_conn2 = None
|
|
zk_conn3 = None
|
|
|
|
try:
|
|
keeper_utils.wait_until_connected(cluster, node1)
|
|
zk_conn = get_fake_zk(node1)
|
|
|
|
for i in range(100):
|
|
zk_conn.create("/test_two_" + str(i), b"somedata")
|
|
|
|
p = Pool(3)
|
|
node2.stop_clickhouse()
|
|
node2.copy_file_to_container(
|
|
os.path.join(CONFIG_DIR, "enable_keeper_two_nodes_2.xml"),
|
|
"/etc/clickhouse-server/config.d/enable_keeper2.xml",
|
|
)
|
|
waiter = p.apply_async(start, (node2,))
|
|
node1.copy_file_to_container(
|
|
os.path.join(CONFIG_DIR, "enable_keeper_two_nodes_1.xml"),
|
|
"/etc/clickhouse-server/config.d/enable_keeper1.xml",
|
|
)
|
|
node1.query("SYSTEM RELOAD CONFIG")
|
|
waiter.wait()
|
|
keeper_utils.wait_until_connected(cluster, node2)
|
|
|
|
zk_conn2 = get_fake_zk(node2)
|
|
|
|
for i in range(100):
|
|
assert zk_conn2.exists("/test_two_" + str(i)) is not None
|
|
|
|
zk_conn.stop()
|
|
zk_conn.close()
|
|
|
|
zk_conn = get_fake_zk(node1)
|
|
|
|
for i in range(100):
|
|
zk_conn.create("/test_three_" + str(i), b"somedata")
|
|
|
|
node3.stop_clickhouse()
|
|
|
|
node3.copy_file_to_container(
|
|
os.path.join(CONFIG_DIR, "enable_keeper_three_nodes_3.xml"),
|
|
"/etc/clickhouse-server/config.d/enable_keeper3.xml",
|
|
)
|
|
waiter = p.apply_async(start, (node3,))
|
|
node2.copy_file_to_container(
|
|
os.path.join(CONFIG_DIR, "enable_keeper_three_nodes_2.xml"),
|
|
"/etc/clickhouse-server/config.d/enable_keeper2.xml",
|
|
)
|
|
node1.copy_file_to_container(
|
|
os.path.join(CONFIG_DIR, "enable_keeper_three_nodes_1.xml"),
|
|
"/etc/clickhouse-server/config.d/enable_keeper1.xml",
|
|
)
|
|
|
|
node1.query("SYSTEM RELOAD CONFIG")
|
|
node2.query("SYSTEM RELOAD CONFIG")
|
|
|
|
waiter.wait()
|
|
keeper_utils.wait_until_connected(cluster, node3)
|
|
zk_conn3 = get_fake_zk(node3)
|
|
|
|
for i in range(100):
|
|
assert zk_conn3.exists("/test_three_" + str(i)) is not None
|
|
|
|
# configs which change endpoints of server should not be allowed
|
|
node1.replace_in_config(
|
|
"/etc/clickhouse-server/config.d/enable_keeper1.xml",
|
|
"node3",
|
|
"non_existing_node",
|
|
)
|
|
|
|
node1.query("SYSTEM RELOAD CONFIG")
|
|
time.sleep(2)
|
|
assert node1.contains_in_log(
|
|
"Config will be ignored because a server with ID 3 is already present in the cluster"
|
|
)
|
|
finally:
|
|
for zk in [zk_conn, zk_conn2, zk_conn3]:
|
|
if zk:
|
|
zk.stop()
|
|
zk.close()
|