import base64 import errno import http.client import logging import os import stat import os.path as p import pprint import pwd import re import shutil import socket import subprocess import time import traceback import urllib.parse import shlex import urllib3 from cassandra.policies import RoundRobinPolicy import cassandra.cluster import psycopg2 import pymongo import pymysql import requests from confluent_kafka.avro.cached_schema_registry_client import \ CachedSchemaRegistryClient from dict2xml import dict2xml from kazoo.client import KazooClient from kazoo.exceptions import KazooException from minio import Minio from psycopg2.extensions import ISOLATION_LEVEL_AUTOCOMMIT from helpers.test_tools import assert_eq_with_retry from helpers import pytest_xdist_logging_to_separate_files import docker from .client import Client from .hdfs_api import HDFSApi HELPERS_DIR = p.dirname(__file__) CLICKHOUSE_ROOT_DIR = p.join(p.dirname(__file__), "../../..") LOCAL_DOCKER_COMPOSE_DIR = p.join(CLICKHOUSE_ROOT_DIR, "docker/test/integration/runner/compose/") DEFAULT_ENV_NAME = '.env' SANITIZER_SIGN = "==================" # to create docker-compose env file def _create_env_file(path, variables): logging.debug(f"Env {variables} stored in {path}") with open(path, 'w') as f: for var, value in list(variables.items()): f.write("=".join([var, value]) + "\n") return path def run_and_check(args, env=None, shell=False, stdout=subprocess.PIPE, stderr=subprocess.PIPE, timeout=300, nothrow=False, detach=False): if detach: subprocess.Popen(args, stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL, env=env, shell=shell) return logging.debug(f"Command:{args}") res = subprocess.run(args, stdout=stdout, stderr=stderr, env=env, shell=shell, timeout=timeout) out = res.stdout.decode('utf-8') err = res.stderr.decode('utf-8') # check_call(...) from subprocess does not print stderr, so we do it manually for outline in out.splitlines(): logging.debug(f"Stdout:{outline}") for errline in err.splitlines(): logging.debug(f"Stderr:{errline}") if res.returncode != 0: logging.debug(f"Exitcode:{res.returncode}") if env: logging.debug(f"Env:{env}") if not nothrow: raise Exception(f"Command {args} return non-zero code {res.returncode}: {res.stderr.decode('utf-8')}") return out # Based on https://stackoverflow.com/questions/2838244/get-open-tcp-port-in-python/2838309#2838309 def get_free_port(): s = socket.socket(socket.AF_INET, socket.SOCK_STREAM) s.bind(("",0)) s.listen(1) port = s.getsockname()[1] s.close() return port def retry_exception(num, delay, func, exception=Exception, *args, **kwargs): """ Retry if `func()` throws, `num` times. :param func: func to run :param num: number of retries :throws StopIteration """ i = 0 while i <= num: try: func(*args, **kwargs) time.sleep(delay) except exception: # pylint: disable=broad-except i += 1 continue return raise StopIteration('Function did not finished successfully') def subprocess_check_call(args, detach=False, nothrow=False): # Uncomment for debugging #logging.info('run:' + ' '.join(args)) return run_and_check(args, detach=detach, nothrow=nothrow) def get_odbc_bridge_path(): path = os.environ.get('CLICKHOUSE_TESTS_ODBC_BRIDGE_BIN_PATH') if path is None: server_path = os.environ.get('CLICKHOUSE_TESTS_SERVER_BIN_PATH') if server_path is not None: return os.path.join(os.path.dirname(server_path), 'clickhouse-odbc-bridge') else: return '/usr/bin/clickhouse-odbc-bridge' return path def get_library_bridge_path(): path = os.environ.get('CLICKHOUSE_TESTS_LIBRARY_BRIDGE_BIN_PATH') if path is None: server_path = os.environ.get('CLICKHOUSE_TESTS_SERVER_BIN_PATH') if server_path is not None: return os.path.join(os.path.dirname(server_path), 'clickhouse-library-bridge') else: return '/usr/bin/clickhouse-library-bridge' return path def get_docker_compose_path(): compose_path = os.environ.get('DOCKER_COMPOSE_DIR') if compose_path is not None: return os.path.dirname(compose_path) else: if os.path.exists(os.path.dirname('/compose/')): return os.path.dirname('/compose/') # default in docker runner container else: logging.debug(f"Fallback docker_compose_path to LOCAL_DOCKER_COMPOSE_DIR: {LOCAL_DOCKER_COMPOSE_DIR}") return LOCAL_DOCKER_COMPOSE_DIR def check_kafka_is_available(kafka_id, kafka_port): p = subprocess.Popen(('docker', 'exec', '-i', kafka_id, '/usr/bin/kafka-broker-api-versions', '--bootstrap-server', f'INSIDE://localhost:{kafka_port}'), stdout=subprocess.PIPE, stderr=subprocess.PIPE) p.communicate() return p.returncode == 0 def check_rabbitmq_is_available(rabbitmq_id): p = subprocess.Popen(('docker', 'exec', '-i', rabbitmq_id, 'rabbitmqctl', 'await_startup'), stdout=subprocess.PIPE) p.communicate() return p.returncode == 0 def enable_consistent_hash_plugin(rabbitmq_id): p = subprocess.Popen(('docker', 'exec', '-i', rabbitmq_id, "rabbitmq-plugins", "enable", "rabbitmq_consistent_hash_exchange"), stdout=subprocess.PIPE) p.communicate() return p.returncode == 0 def get_instances_dir(): if 'INTEGRATION_TESTS_RUN_ID' in os.environ and os.environ['INTEGRATION_TESTS_RUN_ID']: return '_instances_' + shlex.quote(os.environ['INTEGRATION_TESTS_RUN_ID']) else: return '_instances' class ClickHouseCluster: """ClickHouse cluster with several instances and (possibly) ZooKeeper. Add instances with several calls to add_instance(), then start them with the start() call. Directories for instances are created in the directory of base_path. After cluster is started, these directories will contain logs, database files, docker-compose config, ClickHouse configs etc. """ def __init__(self, base_path, name=None, base_config_dir=None, server_bin_path=None, client_bin_path=None, odbc_bridge_bin_path=None, library_bridge_bin_path=None, zookeeper_config_path=None, custom_dockerd_host=None, zookeeper_keyfile=None, zookeeper_certfile=None): for param in list(os.environ.keys()): logging.debug("ENV %40s %s" % (param, os.environ[param])) self.base_path = base_path self.base_dir = p.dirname(base_path) self.name = name if name is not None else '' self.base_config_dir = base_config_dir or os.environ.get('CLICKHOUSE_TESTS_BASE_CONFIG_DIR', '/etc/clickhouse-server/') self.server_bin_path = p.realpath( server_bin_path or os.environ.get('CLICKHOUSE_TESTS_SERVER_BIN_PATH', '/usr/bin/clickhouse')) self.odbc_bridge_bin_path = p.realpath(odbc_bridge_bin_path or get_odbc_bridge_path()) self.library_bridge_bin_path = p.realpath(library_bridge_bin_path or get_library_bridge_path()) self.client_bin_path = p.realpath( client_bin_path or os.environ.get('CLICKHOUSE_TESTS_CLIENT_BIN_PATH', '/usr/bin/clickhouse-client')) self.zookeeper_config_path = p.join(self.base_dir, zookeeper_config_path) if zookeeper_config_path else p.join( HELPERS_DIR, 'zookeeper_config.xml') project_name = pwd.getpwuid(os.getuid()).pw_name + p.basename(self.base_dir) + self.name # docker-compose removes everything non-alphanumeric from project names so we do it too. self.project_name = re.sub(r'[^a-z0-9]', '', project_name.lower()) instances_dir_name = '_instances' if self.name: instances_dir_name += '_' + self.name if 'INTEGRATION_TESTS_RUN_ID' in os.environ and os.environ['INTEGRATION_TESTS_RUN_ID']: instances_dir_name += '_' + shlex.quote(os.environ['INTEGRATION_TESTS_RUN_ID']) self.instances_dir = p.join(self.base_dir, instances_dir_name) self.docker_logs_path = p.join(self.instances_dir, 'docker.log') self.env_file = p.join(self.instances_dir, DEFAULT_ENV_NAME) self.env_variables = {} self.up_called = False custom_dockerd_host = custom_dockerd_host or os.environ.get('CLICKHOUSE_TESTS_DOCKERD_HOST') self.docker_api_version = os.environ.get("DOCKER_API_VERSION") self.docker_base_tag = os.environ.get("DOCKER_BASE_TAG", "latest") self.base_cmd = ['docker-compose'] if custom_dockerd_host: self.base_cmd += ['--host', custom_dockerd_host] self.base_cmd += ['--env-file', self.env_file] self.base_cmd += ['--project-name', self.project_name] self.base_zookeeper_cmd = None self.base_mysql_cmd = [] self.base_kafka_cmd = [] self.base_kerberized_kafka_cmd = [] self.base_rabbitmq_cmd = [] self.base_cassandra_cmd = [] self.base_jdbc_bridge_cmd = [] self.base_redis_cmd = [] self.pre_zookeeper_commands = [] self.instances = {} self.with_zookeeper = False self.with_zookeeper_secure = False self.with_mysql_client = False self.with_mysql = False self.with_mysql8 = False self.with_mysql_cluster = False self.with_postgres = False self.with_postgres_cluster = False self.with_kafka = False self.with_kerberized_kafka = False self.with_rabbitmq = False self.with_odbc_drivers = False self.with_hdfs = False self.with_kerberized_hdfs = False self.with_mongo = False self.with_mongo_secure = False self.with_net_trics = False self.with_redis = False self.with_cassandra = False self.with_jdbc_bridge = False self.with_minio = False self.minio_dir = os.path.join(self.instances_dir, "minio") self.minio_certs_dir = None # source for certificates self.minio_host = "minio1" self.minio_ip = None self.minio_bucket = "root" self.minio_bucket_2 = "root2" self.minio_port = 9001 self.minio_client = None # type: Minio self.minio_redirect_host = "proxy1" self.minio_redirect_ip = None self.minio_redirect_port = 8080 # available when with_hdfs == True self.hdfs_host = "hdfs1" self.hdfs_ip = None self.hdfs_name_port = 50070 self.hdfs_data_port = 50075 self.hdfs_dir = p.abspath(p.join(self.instances_dir, "hdfs")) self.hdfs_logs_dir = os.path.join(self.hdfs_dir, "logs") self.hdfs_api = None # also for kerberized hdfs # available when with_kerberized_hdfs == True self.hdfs_kerberized_host = "kerberizedhdfs1" self.hdfs_kerberized_ip = None self.hdfs_kerberized_name_port = 50070 self.hdfs_kerberized_data_port = 1006 self.hdfs_kerberized_dir = p.abspath(p.join(self.instances_dir, "kerberized_hdfs")) self.hdfs_kerberized_logs_dir = os.path.join(self.hdfs_kerberized_dir, "logs") # available when with_kafka == True self.kafka_host = "kafka1" self.kafka_port = get_free_port() self.kafka_docker_id = None self.schema_registry_host = "schema-registry" self.schema_registry_port = get_free_port() self.kafka_docker_id = self.get_instance_docker_id(self.kafka_host) # available when with_kerberozed_kafka == True self.kerberized_kafka_host = "kerberized_kafka1" self.kerberized_kafka_port = get_free_port() self.kerberized_kafka_docker_id = self.get_instance_docker_id(self.kerberized_kafka_host) # available when with_mongo == True self.mongo_host = "mongo1" self.mongo_port = get_free_port() # available when with_cassandra == True self.cassandra_host = "cassandra1" self.cassandra_port = 9042 self.cassandra_ip = None self.cassandra_id = self.get_instance_docker_id(self.cassandra_host) # available when with_rabbitmq == True self.rabbitmq_host = "rabbitmq1" self.rabbitmq_ip = None self.rabbitmq_port = 5672 self.rabbitmq_dir = p.abspath(p.join(self.instances_dir, "rabbitmq")) self.rabbitmq_logs_dir = os.path.join(self.rabbitmq_dir, "logs") # available when with_redis == True self.redis_host = "redis1" self.redis_port = get_free_port() # available when with_postgres == True self.postgres_host = "postgres1" self.postgres_ip = None self.postgres_conn = None self.postgres2_host = "postgres2" self.postgres2_ip = None self.postgres2_conn = None self.postgres3_host = "postgres3" self.postgres3_ip = None self.postgres3_conn = None self.postgres4_host = "postgres4" self.postgres4_ip = None self.postgres4_conn = None self.postgres_port = 5432 self.postgres_dir = p.abspath(p.join(self.instances_dir, "postgres")) self.postgres_logs_dir = os.path.join(self.postgres_dir, "postgres1") self.postgres2_logs_dir = os.path.join(self.postgres_dir, "postgres2") self.postgres3_logs_dir = os.path.join(self.postgres_dir, "postgres3") self.postgres4_logs_dir = os.path.join(self.postgres_dir, "postgres4") # available when with_mysql_client == True self.mysql_client_host = "mysql_client" self.mysql_client_container = None # available when with_mysql == True self.mysql_host = "mysql57" self.mysql_port = 3306 self.mysql_ip = None self.mysql_dir = p.abspath(p.join(self.instances_dir, "mysql")) self.mysql_logs_dir = os.path.join(self.mysql_dir, "logs") # available when with_mysql_cluster == True self.mysql2_host = "mysql2" self.mysql3_host = "mysql3" self.mysql4_host = "mysql4" self.mysql2_ip = None self.mysql3_ip = None self.mysql4_ip = None self.mysql_cluster_dir = p.abspath(p.join(self.instances_dir, "mysql")) self.mysql_cluster_logs_dir = os.path.join(self.mysql_dir, "logs") # available when with_mysql8 == True self.mysql8_host = "mysql80" self.mysql8_port = 3306 self.mysql8_ip = None self.mysql8_dir = p.abspath(p.join(self.instances_dir, "mysql8")) self.mysql8_logs_dir = os.path.join(self.mysql8_dir, "logs") # available when with_zookeper_secure == True self.zookeeper_secure_port = 2281 self.zookeeper_keyfile = zookeeper_keyfile self.zookeeper_certfile = zookeeper_certfile # available when with_zookeper == True self.use_keeper = True self.zookeeper_port = 2181 self.keeper_instance_dir_prefix = p.join(p.abspath(self.instances_dir), "keeper") # if use_keeper = True self.zookeeper_instance_dir_prefix = p.join(self.instances_dir, "zk") self.zookeeper_dirs_to_create = [] # available when with_jdbc_bridge == True self.jdbc_bridge_host = "bridge1" self.jdbc_bridge_ip = None self.jdbc_bridge_port = 9019 self.jdbc_driver_dir = p.abspath(p.join(self.instances_dir, "jdbc_driver")) self.jdbc_driver_logs_dir = os.path.join(self.jdbc_driver_dir, "logs") self.docker_client = None self.is_up = False self.env = os.environ.copy() logging.debug(f"CLUSTER INIT base_config_dir:{self.base_config_dir}") def cleanup(self): # Just in case kill unstopped containers from previous launch try: # docker-compose names containers using the following formula: # container_name = project_name + '_' + instance_name + '_1' # We need to have "^/" and "$" in the "--filter name" option below to filter by exact name of the container, see # https://stackoverflow.com/questions/48767760/how-to-make-docker-container-ls-f-name-filter-by-exact-name filter_name = f'^/{self.project_name}_.*_1$' if int(run_and_check(f'docker container list --all --filter name={filter_name} | wc -l', shell=True)) > 1: logging.debug(f"Trying to kill unstopped containers for project {self.project_name}:") unstopped_containers = run_and_check(f'docker container list --all --filter name={filter_name}', shell=True) unstopped_containers_ids = [line.split()[0] for line in unstopped_containers.splitlines()[1:]] for id in unstopped_containers_ids: run_and_check(f'docker kill {id}', shell=True, nothrow=True) run_and_check(f'docker rm {id}', shell=True, nothrow=True) logging.debug("Unstopped containers killed") run_and_check(f'docker container list --all --filter name={filter_name}', shell=True) else: logging.debug(f"No running containers for project: {self.project_name}") except: pass # # Just in case remove unused networks # try: # logging.debug("Trying to prune unused networks...") # run_and_check(['docker', 'network', 'prune', '-f']) # logging.debug("Networks pruned") # except: # pass # Remove unused images # try: # logging.debug("Trying to prune unused images...") # run_and_check(['docker', 'image', 'prune', '-f']) # logging.debug("Images pruned") # except: # pass # Remove unused volumes try: logging.debug("Trying to prune unused volumes...") result = run_and_check(['docker volume ls | wc -l'], shell=True) if int(result>0): run_and_check(['docker', 'volume', 'prune', '-f']) logging.debug(f"Volumes pruned: {result}") except: pass def get_docker_handle(self, docker_id): exception = None for i in range(5): try: return self.docker_client.containers.get(docker_id) except Exception as ex: print("Got exception getting docker handle", str(ex)) time.sleep(i * 2) exception = ex raise exception def get_client_cmd(self): cmd = self.client_bin_path if p.basename(cmd) == 'clickhouse': cmd += " client" return cmd def copy_file_from_container_to_container(self, src_node, src_path, dst_node, dst_path): fname = os.path.basename(src_path) run_and_check([f"docker cp {src_node.docker_id}:{src_path} {self.instances_dir}"], shell=True) run_and_check([f"docker cp {self.instances_dir}/{fname} {dst_node.docker_id}:{dst_path}"], shell=True) def setup_zookeeper_secure_cmd(self, instance, env_variables, docker_compose_yml_dir): logging.debug('Setup ZooKeeper Secure') zookeeper_docker_compose_path = p.join(docker_compose_yml_dir, 'docker_compose_zookeeper_secure.yml') env_variables['ZOO_SECURE_CLIENT_PORT'] = str(self.zookeeper_secure_port) env_variables['ZK_FS'] = 'bind' for i in range(1, 4): zk_data_path = os.path.join(self.zookeeper_instance_dir_prefix + str(i), "data") zk_log_path = os.path.join(self.zookeeper_instance_dir_prefix + str(i), "log") env_variables['ZK_DATA' + str(i)] = zk_data_path env_variables['ZK_DATA_LOG' + str(i)] = zk_log_path self.zookeeper_dirs_to_create += [zk_data_path, zk_log_path] logging.debug(f"DEBUG ZK: {self.zookeeper_dirs_to_create}") self.with_zookeeper_secure = True self.base_cmd.extend(['--file', zookeeper_docker_compose_path]) self.base_zookeeper_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', zookeeper_docker_compose_path] return self.base_zookeeper_cmd def setup_zookeeper_cmd(self, instance, env_variables, docker_compose_yml_dir): logging.debug('Setup ZooKeeper') zookeeper_docker_compose_path = p.join(docker_compose_yml_dir, 'docker_compose_zookeeper.yml') env_variables['ZK_FS'] = 'bind' for i in range(1, 4): zk_data_path = os.path.join(self.zookeeper_instance_dir_prefix + str(i), "data") zk_log_path = os.path.join(self.zookeeper_instance_dir_prefix + str(i), "log") env_variables['ZK_DATA' + str(i)] = zk_data_path env_variables['ZK_DATA_LOG' + str(i)] = zk_log_path self.zookeeper_dirs_to_create += [zk_data_path, zk_log_path] logging.debug(f"DEBUG ZK: {self.zookeeper_dirs_to_create}") self.with_zookeeper = True self.base_cmd.extend(['--file', zookeeper_docker_compose_path]) self.base_zookeeper_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', zookeeper_docker_compose_path] return self.base_zookeeper_cmd def setup_keeper_cmd(self, instance, env_variables, docker_compose_yml_dir): logging.debug('Setup Keeper') keeper_docker_compose_path = p.join(docker_compose_yml_dir, 'docker_compose_keeper.yml') binary_path = self.server_bin_path if binary_path.endswith('-server'): binary_path = binary_path[:-len('-server')] env_variables['keeper_binary'] = binary_path env_variables['image'] = "yandex/clickhouse-integration-test:" + self.docker_base_tag env_variables['user'] = str(os.getuid()) env_variables['keeper_fs'] = 'bind' for i in range(1, 4): keeper_instance_dir = self.keeper_instance_dir_prefix + f"{i}" logs_dir = os.path.join(keeper_instance_dir, "log") configs_dir = os.path.join(keeper_instance_dir, "config") coordination_dir = os.path.join(keeper_instance_dir, "coordination") env_variables[f'keeper_logs_dir{i}'] = logs_dir env_variables[f'keeper_config_dir{i}'] = configs_dir env_variables[f'keeper_db_dir{i}'] = coordination_dir self.zookeeper_dirs_to_create += [logs_dir, configs_dir, coordination_dir] logging.debug(f"DEBUG KEEPER: {self.zookeeper_dirs_to_create}") self.with_zookeeper = True self.base_cmd.extend(['--file', keeper_docker_compose_path]) self.base_zookeeper_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', keeper_docker_compose_path] return self.base_zookeeper_cmd def setup_mysql_client_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_mysql_client = True self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_mysql_client.yml')]) self.base_mysql_client_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_mysql_client.yml')] return self.base_mysql_client_cmd def setup_mysql_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_mysql = True env_variables['MYSQL_HOST'] = self.mysql_host env_variables['MYSQL_PORT'] = str(self.mysql_port) env_variables['MYSQL_ROOT_HOST'] = '%' env_variables['MYSQL_LOGS'] = self.mysql_logs_dir env_variables['MYSQL_LOGS_FS'] = "bind" self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_mysql.yml')]) self.base_mysql_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_mysql.yml')] return self.base_mysql_cmd def setup_mysql8_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_mysql8 = True env_variables['MYSQL8_HOST'] = self.mysql8_host env_variables['MYSQL8_PORT'] = str(self.mysql8_port) env_variables['MYSQL8_ROOT_HOST'] = '%' env_variables['MYSQL8_LOGS'] = self.mysql8_logs_dir env_variables['MYSQL8_LOGS_FS'] = "bind" self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_mysql_8_0.yml')]) self.base_mysql8_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_mysql_8_0.yml')] return self.base_mysql8_cmd def setup_mysql_cluster_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_mysql_cluster = True env_variables['MYSQL_CLUSTER_PORT'] = str(self.mysql_port) env_variables['MYSQL_CLUSTER_ROOT_HOST'] = '%' env_variables['MYSQL_CLUSTER_LOGS'] = self.mysql_cluster_logs_dir env_variables['MYSQL_CLUSTER_LOGS_FS'] = "bind" self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_mysql_cluster.yml')]) self.base_mysql_cluster_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_mysql_cluster.yml')] return self.base_mysql_cluster_cmd def setup_postgres_cmd(self, instance, env_variables, docker_compose_yml_dir): self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_postgres.yml')]) env_variables['POSTGRES_PORT'] = str(self.postgres_port) env_variables['POSTGRES_DIR'] = self.postgres_logs_dir env_variables['POSTGRES_LOGS_FS'] = "bind" self.with_postgres = True self.base_postgres_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_postgres.yml')] return self.base_postgres_cmd def setup_postgres_cluster_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_postgres_cluster = True env_variables['POSTGRES_PORT'] = str(self.postgres_port) env_variables['POSTGRES2_DIR'] = self.postgres2_logs_dir env_variables['POSTGRES3_DIR'] = self.postgres3_logs_dir env_variables['POSTGRES4_DIR'] = self.postgres4_logs_dir env_variables['POSTGRES_LOGS_FS'] = "bind" self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_postgres_cluster.yml')]) self.base_postgres_cluster_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_postgres_cluster.yml')] def setup_hdfs_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_hdfs = True env_variables['HDFS_HOST'] = self.hdfs_host env_variables['HDFS_NAME_PORT'] = str(self.hdfs_name_port) env_variables['HDFS_DATA_PORT'] = str(self.hdfs_data_port) env_variables['HDFS_LOGS'] = self.hdfs_logs_dir env_variables['HDFS_FS'] = "bind" self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_hdfs.yml')]) self.base_hdfs_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_hdfs.yml')] logging.debug("HDFS BASE CMD:{self.base_hdfs_cmd)}") return self.base_hdfs_cmd def setup_kerberized_hdfs_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_kerberized_hdfs = True env_variables['KERBERIZED_HDFS_HOST'] = self.hdfs_kerberized_host env_variables['KERBERIZED_HDFS_NAME_PORT'] = str(self.hdfs_kerberized_name_port) env_variables['KERBERIZED_HDFS_DATA_PORT'] = str(self.hdfs_kerberized_data_port) env_variables['KERBERIZED_HDFS_LOGS'] = self.hdfs_kerberized_logs_dir env_variables['KERBERIZED_HDFS_FS'] = "bind" env_variables['KERBERIZED_HDFS_DIR'] = instance.path + '/' self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_kerberized_hdfs.yml')]) self.base_kerberized_hdfs_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_kerberized_hdfs.yml')] return self.base_kerberized_hdfs_cmd def setup_kafka_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_kafka = True env_variables['KAFKA_HOST'] = self.kafka_host env_variables['KAFKA_EXTERNAL_PORT'] = str(self.kafka_port) env_variables['SCHEMA_REGISTRY_EXTERNAL_PORT'] = str(self.schema_registry_port) env_variables['SCHEMA_REGISTRY_INTERNAL_PORT'] = "8081" self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_kafka.yml')]) self.base_kafka_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_kafka.yml')] return self.base_kafka_cmd def setup_kerberized_kafka_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_kerberized_kafka = True env_variables['KERBERIZED_KAFKA_DIR'] = instance.path + '/' env_variables['KERBERIZED_KAFKA_HOST'] = self.kerberized_kafka_host env_variables['KERBERIZED_KAFKA_EXTERNAL_PORT'] = str(self.kerberized_kafka_port) self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_kerberized_kafka.yml')]) self.base_kerberized_kafka_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_kerberized_kafka.yml')] return self.base_kerberized_kafka_cmd def setup_redis_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_redis = True env_variables['REDIS_HOST'] = self.redis_host env_variables['REDIS_EXTERNAL_PORT'] = str(self.redis_port) env_variables['REDIS_INTERNAL_PORT'] = "6379" self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_redis.yml')]) self.base_redis_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_redis.yml')] return self.base_redis_cmd def setup_rabbitmq_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_rabbitmq = True env_variables['RABBITMQ_HOST'] = self.rabbitmq_host env_variables['RABBITMQ_PORT'] = str(self.rabbitmq_port) env_variables['RABBITMQ_LOGS'] = self.rabbitmq_logs_dir env_variables['RABBITMQ_LOGS_FS'] = "bind" self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_rabbitmq.yml')]) self.base_rabbitmq_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_rabbitmq.yml')] return self.base_rabbitmq_cmd def setup_mongo_secure_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_mongo = self.with_mongo_secure = True env_variables['MONGO_HOST'] = self.mongo_host env_variables['MONGO_EXTERNAL_PORT'] = str(self.mongo_port) env_variables['MONGO_INTERNAL_PORT'] = "27017" env_variables['MONGO_CONFIG_PATH'] = HELPERS_DIR self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_mongo_secure.yml')]) self.base_mongo_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_mongo_secure.yml')] return self.base_mongo_cmd def setup_mongo_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_mongo = True env_variables['MONGO_HOST'] = self.mongo_host env_variables['MONGO_EXTERNAL_PORT'] = str(self.mongo_port) env_variables['MONGO_INTERNAL_PORT'] = "27017" self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_mongo.yml')]) self.base_mongo_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_mongo.yml')] return self.base_mongo_cmd def setup_minio_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_minio = True cert_d = p.join(self.minio_dir, "certs") env_variables['MINIO_CERTS_DIR'] = cert_d env_variables['MINIO_PORT'] = str(self.minio_port) env_variables['SSL_CERT_FILE'] = p.join(self.base_dir, cert_d, 'public.crt') self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_minio.yml')]) self.base_minio_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_minio.yml')] return self.base_minio_cmd def setup_cassandra_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_cassandra = True env_variables['CASSANDRA_PORT'] = str(self.cassandra_port) self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_cassandra.yml')]) self.base_cassandra_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_cassandra.yml')] return self.base_cassandra_cmd def setup_jdbc_bridge_cmd(self, instance, env_variables, docker_compose_yml_dir): self.with_jdbc_bridge = True env_variables['JDBC_DRIVER_LOGS'] = self.jdbc_driver_logs_dir env_variables['JDBC_DRIVER_FS'] = "bind" self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_jdbc_bridge.yml')]) self.base_jdbc_bridge_cmd = ['docker-compose', '--env-file', instance.env_file, '--project-name', self.project_name, '--file', p.join(docker_compose_yml_dir, 'docker_compose_jdbc_bridge.yml')] return self.base_jdbc_bridge_cmd def add_instance(self, name, base_config_dir=None, main_configs=None, user_configs=None, dictionaries=None, macros=None, with_zookeeper=False, with_zookeeper_secure=False, with_mysql_client=False, with_mysql=False, with_mysql8=False, with_mysql_cluster=False, with_kafka=False, with_kerberized_kafka=False, with_rabbitmq=False, clickhouse_path_dir=None, with_odbc_drivers=False, with_postgres=False, with_postgres_cluster=False, with_hdfs=False, with_kerberized_hdfs=False, with_mongo=False, with_mongo_secure=False, with_redis=False, with_minio=False, with_cassandra=False, with_jdbc_bridge=False, hostname=None, env_variables=None, image="yandex/clickhouse-integration-test", tag=None, stay_alive=False, ipv4_address=None, ipv6_address=None, with_installed_binary=False, tmpfs=None, zookeeper_docker_compose_path=None, minio_certs_dir=None, use_keeper=True, main_config_name="config.xml", users_config_name="users.xml", copy_common_configs=True): """Add an instance to the cluster. name - the name of the instance directory and the value of the 'instance' macro in ClickHouse. base_config_dir - a directory with config.xml and users.xml files which will be copied to /etc/clickhouse-server/ directory main_configs - a list of config files that will be added to config.d/ directory user_configs - a list of config files that will be added to users.d/ directory with_zookeeper - if True, add ZooKeeper configuration to configs and ZooKeeper instances to the cluster. with_zookeeper_secure - if True, add ZooKeeper Secure configuration to configs and ZooKeeper instances to the cluster. """ if self.is_up: raise Exception("Can\'t add instance %s: cluster is already up!" % name) if name in self.instances: raise Exception("Can\'t add instance `%s': there is already an instance with the same name!" % name) if tag is None: tag = self.docker_base_tag if not env_variables: env_variables = {} self.use_keeper = use_keeper # Code coverage files will be placed in database directory # (affect only WITH_COVERAGE=1 build) env_variables['LLVM_PROFILE_FILE'] = '/var/lib/clickhouse/server_%h_%p_%m.profraw' instance = ClickHouseInstance( cluster=self, base_path=self.base_dir, name=name, base_config_dir=base_config_dir if base_config_dir else self.base_config_dir, custom_main_configs=main_configs or [], custom_user_configs=user_configs or [], custom_dictionaries=dictionaries or [], macros=macros or {}, with_zookeeper=with_zookeeper, zookeeper_config_path=self.zookeeper_config_path, with_mysql_client=with_mysql_client, with_mysql=with_mysql, with_mysql8=with_mysql8, with_mysql_cluster=with_mysql_cluster, with_kafka=with_kafka, with_kerberized_kafka=with_kerberized_kafka, with_rabbitmq=with_rabbitmq, with_kerberized_hdfs=with_kerberized_hdfs, with_mongo=with_mongo or with_mongo_secure, with_redis=with_redis, with_minio=with_minio, with_cassandra=with_cassandra, with_jdbc_bridge=with_jdbc_bridge, server_bin_path=self.server_bin_path, odbc_bridge_bin_path=self.odbc_bridge_bin_path, library_bridge_bin_path=self.library_bridge_bin_path, clickhouse_path_dir=clickhouse_path_dir, with_odbc_drivers=with_odbc_drivers, with_postgres=with_postgres, with_postgres_cluster=with_postgres_cluster, hostname=hostname, env_variables=env_variables, image=image, tag=tag, stay_alive=stay_alive, ipv4_address=ipv4_address, ipv6_address=ipv6_address, with_installed_binary=with_installed_binary, main_config_name=main_config_name, users_config_name=users_config_name, copy_common_configs=copy_common_configs, tmpfs=tmpfs or []) docker_compose_yml_dir = get_docker_compose_path() self.instances[name] = instance if ipv4_address is not None or ipv6_address is not None: self.with_net_trics = True self.base_cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_net.yml')]) self.base_cmd.extend(['--file', instance.docker_compose_path]) cmds = [] if with_zookeeper_secure and not self.with_zookeeper_secure: cmds.append(self.setup_zookeeper_secure_cmd(instance, env_variables, docker_compose_yml_dir)) if with_zookeeper and not self.with_zookeeper: if self.use_keeper: cmds.append(self.setup_keeper_cmd(instance, env_variables, docker_compose_yml_dir)) else: cmds.append(self.setup_zookeeper_cmd(instance, env_variables, docker_compose_yml_dir)) if with_mysql_client and not self.with_mysql_client: cmds.append(self.setup_mysql_client_cmd(instance, env_variables, docker_compose_yml_dir)) if with_mysql and not self.with_mysql: cmds.append(self.setup_mysql_cmd(instance, env_variables, docker_compose_yml_dir)) if with_mysql8 and not self.with_mysql8: cmds.append(self.setup_mysql8_cmd(instance, env_variables, docker_compose_yml_dir)) if with_mysql_cluster and not self.with_mysql_cluster: cmds.append(self.setup_mysql_cluster_cmd(instance, env_variables, docker_compose_yml_dir)) if with_postgres and not self.with_postgres: cmds.append(self.setup_postgres_cmd(instance, env_variables, docker_compose_yml_dir)) if with_postgres_cluster and not self.with_postgres_cluster: cmds.append(self.setup_postgres_cluster_cmd(instance, env_variables, docker_compose_yml_dir)) if with_odbc_drivers and not self.with_odbc_drivers: self.with_odbc_drivers = True if not self.with_mysql: cmds.append(self.setup_mysql_cmd(instance, env_variables, docker_compose_yml_dir)) if not self.with_postgres: cmds.append(self.setup_postgres_cmd(instance, env_variables, docker_compose_yml_dir)) if with_kafka and not self.with_kafka: cmds.append(self.setup_kafka_cmd(instance, env_variables, docker_compose_yml_dir)) if with_kerberized_kafka and not self.with_kerberized_kafka: cmds.append(self.setup_kerberized_kafka_cmd(instance, env_variables, docker_compose_yml_dir)) if with_rabbitmq and not self.with_rabbitmq: cmds.append(self.setup_rabbitmq_cmd(instance, env_variables, docker_compose_yml_dir)) if with_hdfs and not self.with_hdfs: cmds.append(self.setup_hdfs_cmd(instance, env_variables, docker_compose_yml_dir)) if with_kerberized_hdfs and not self.with_kerberized_hdfs: cmds.append(self.setup_kerberized_hdfs_cmd(instance, env_variables, docker_compose_yml_dir)) if (with_mongo or with_mongo_secure) and not (self.with_mongo or self.with_mongo_secure): if with_mongo_secure: cmds.append(self.setup_mongo_secure_cmd(instance, env_variables, docker_compose_yml_dir)) else: cmds.append(self.setup_mongo_cmd(instance, env_variables, docker_compose_yml_dir)) if self.with_net_trics: for cmd in cmds: cmd.extend(['--file', p.join(docker_compose_yml_dir, 'docker_compose_net.yml')]) if with_redis and not self.with_redis: cmds.append(self.setup_redis_cmd(instance, env_variables, docker_compose_yml_dir)) if with_minio and not self.with_minio: cmds.append(self.setup_minio_cmd(instance, env_variables, docker_compose_yml_dir)) if minio_certs_dir is not None: if self.minio_certs_dir is None: self.minio_certs_dir = minio_certs_dir else: raise Exception("Overwriting minio certs dir") if with_cassandra and not self.with_cassandra: cmds.append(self.setup_cassandra_cmd(instance, env_variables, docker_compose_yml_dir)) if with_jdbc_bridge and not self.with_jdbc_bridge: cmds.append(self.setup_jdbc_bridge_cmd(instance, env_variables, docker_compose_yml_dir)) logging.debug("Cluster name:{} project_name:{}. Added instance name:{} tag:{} base_cmd:{} docker_compose_yml_dir:{}".format( self.name, self.project_name, name, tag, self.base_cmd, docker_compose_yml_dir)) return instance def get_instance_docker_id(self, instance_name): # According to how docker-compose names containers. return self.project_name + '_' + instance_name + '_1' def _replace(self, path, what, to): with open(path, 'r') as p: data = p.read() data = data.replace(what, to) with open(path, 'w') as p: p.write(data) def restart_instance_with_ip_change(self, node, new_ip): if '::' in new_ip: if node.ipv6_address is None: raise Exception("You should specity ipv6_address in add_node method") self._replace(node.docker_compose_path, node.ipv6_address, new_ip) node.ipv6_address = new_ip else: if node.ipv4_address is None: raise Exception("You should specity ipv4_address in add_node method") self._replace(node.docker_compose_path, node.ipv4_address, new_ip) node.ipv4_address = new_ip run_and_check(self.base_cmd + ["stop", node.name]) run_and_check(self.base_cmd + ["rm", "--force", "--stop", node.name]) run_and_check(self.base_cmd + ["up", "--force-recreate", "--no-deps", "-d", node.name]) node.ip_address = self.get_instance_ip(node.name) node.client = Client(node.ip_address, command=self.client_bin_path) logging.info("Restart node with ip change") # In builds with sanitizer the server can take a long time to start node.wait_for_start(start_timeout=180.0, connection_timeout=600.0) # seconds logging.info("Restarted") return node def restart_service(self, service_name): run_and_check(self.base_cmd + ["restart", service_name]) def get_instance_ip(self, instance_name): logging.debug("get_instance_ip instance_name={}".format(instance_name)) docker_id = self.get_instance_docker_id(instance_name) # for cont in self.docker_client.containers.list(): # logging.debug("CONTAINERS LIST: ID={} NAME={} STATUS={}".format(cont.id, cont.name, cont.status)) handle = self.docker_client.containers.get(docker_id) return list(handle.attrs['NetworkSettings']['Networks'].values())[0]['IPAddress'] def get_container_id(self, instance_name): return self.get_instance_docker_id(instance_name) # docker_id = self.get_instance_docker_id(instance_name) # handle = self.docker_client.containers.get(docker_id) # return handle.attrs['Id'] def get_container_logs(self, instance_name): container_id = self.get_container_id(instance_name) return self.docker_client.api.logs(container_id).decode() def exec_in_container(self, container_id, cmd, detach=False, nothrow=False, use_cli=True, **kwargs): if use_cli: logging.debug(f"run container_id:{container_id} detach:{detach} nothrow:{nothrow} cmd: {cmd}") exec_cmd = ["docker", "exec"] if 'user' in kwargs: exec_cmd += ['-u', kwargs['user']] result = subprocess_check_call(exec_cmd + [container_id] + cmd, detach=detach, nothrow=nothrow) return result else: exec_id = self.docker_client.api.exec_create(container_id, cmd, **kwargs) output = self.docker_client.api.exec_start(exec_id, detach=detach) exit_code = self.docker_client.api.exec_inspect(exec_id)['ExitCode'] if exit_code: container_info = self.docker_client.api.inspect_container(container_id) image_id = container_info.get('Image') image_info = self.docker_client.api.inspect_image(image_id) logging.debug(("Command failed in container {}: ".format(container_id))) pprint.pprint(container_info) logging.debug("") logging.debug(("Container {} uses image {}: ".format(container_id, image_id))) pprint.pprint(image_info) logging.debug("") message = 'Cmd "{}" failed in container {}. Return code {}. Output: {}'.format(' '.join(cmd), container_id, exit_code, output) if nothrow: logging.debug(message) else: raise Exception(message) if not detach: return output.decode() return output def copy_file_to_container(self, container_id, local_path, dest_path): with open(local_path, "r") as fdata: data = fdata.read() encodedBytes = base64.b64encode(data.encode("utf-8")) encodedStr = str(encodedBytes, "utf-8") self.exec_in_container(container_id, ["bash", "-c", "echo {} | base64 --decode > {}".format(encodedStr, dest_path)], user='root') def wait_for_url(self, url="http://localhost:8123/ping", conn_timeout=2, interval=2, timeout=60): if not url.startswith('http'): url = "http://" + url if interval <= 0: interval = 2 if timeout <= 0: timeout = 60 attempts = 1 errors = [] start = time.time() while time.time() - start < timeout: try: requests.get(url, allow_redirects=True, timeout=conn_timeout, verify=False).raise_for_status() logging.debug("{} is available after {} seconds".format(url, time.time() - start)) return except Exception as ex: logging.debug("{} Attempt {} failed, retrying in {} seconds".format(ex, attempts, interval)) attempts += 1 errors += [str(ex)] time.sleep(interval) run_and_check(['docker-compose', 'ps', '--services', '--all']) logging.error("Can't connect to URL:{}".format(errors)) raise Exception("Cannot wait URL {}(interval={}, timeout={}, attempts={})".format( url, interval, timeout, attempts)) def wait_mysql_client_to_start(self, timeout=180): start = time.time() errors = [] self.mysql_client_container = self.get_docker_handle(self.get_instance_docker_id(self.mysql_client_host)) while time.time() - start < timeout: try: info = self.mysql_client_container.client.api.inspect_container(self.mysql_client_container.name) if info['State']['Health']['Status'] == 'healthy': logging.debug("Mysql Client Container Started") break time.sleep(1) return except Exception as ex: errors += [str(ex)] time.sleep(1) run_and_check(['docker-compose', 'ps', '--services', '--all']) logging.error("Can't connect to MySQL Client:{}".format(errors)) raise Exception("Cannot wait MySQL Client container") def wait_mysql_to_start(self, timeout=180): self.mysql_ip = self.get_instance_ip('mysql57') start = time.time() errors = [] while time.time() - start < timeout: try: conn = pymysql.connect(user='root', password='clickhouse', host=self.mysql_ip, port=self.mysql_port) conn.close() logging.debug("Mysql Started") return except Exception as ex: errors += [str(ex)] time.sleep(0.5) run_and_check(['docker-compose', 'ps', '--services', '--all']) logging.error("Can't connect to MySQL:{}".format(errors)) raise Exception("Cannot wait MySQL container") def wait_mysql8_to_start(self, timeout=180): self.mysql8_ip = self.get_instance_ip('mysql80') start = time.time() while time.time() - start < timeout: try: conn = pymysql.connect(user='root', password='clickhouse', host=self.mysql8_ip, port=self.mysql8_port) conn.close() logging.debug("Mysql 8 Started") return except Exception as ex: logging.debug("Can't connect to MySQL 8 " + str(ex)) time.sleep(0.5) run_and_check(['docker-compose', 'ps', '--services', '--all']) raise Exception("Cannot wait MySQL 8 container") def wait_mysql_cluster_to_start(self, timeout=180): self.mysql2_ip = self.get_instance_ip(self.mysql2_host) self.mysql3_ip = self.get_instance_ip(self.mysql3_host) self.mysql4_ip = self.get_instance_ip(self.mysql4_host) start = time.time() errors = [] while time.time() - start < timeout: try: for ip in [self.mysql2_ip, self.mysql3_ip, self.mysql4_ip]: conn = pymysql.connect(user='root', password='clickhouse', host=ip, port=self.mysql_port) conn.close() logging.debug(f"Mysql Started {ip}") return except Exception as ex: errors += [str(ex)] time.sleep(0.5) run_and_check(['docker-compose', 'ps', '--services', '--all']) logging.error("Can't connect to MySQL:{}".format(errors)) raise Exception("Cannot wait MySQL container") def wait_postgres_to_start(self, timeout=260): self.postgres_ip = self.get_instance_ip(self.postgres_host) start = time.time() while time.time() - start < timeout: try: self.postgres_conn = psycopg2.connect(host=self.postgres_ip, port=self.postgres_port, database='postgres', user='postgres', password='mysecretpassword') self.postgres_conn.set_isolation_level(ISOLATION_LEVEL_AUTOCOMMIT) self.postgres_conn.autocommit = True logging.debug("Postgres Started") return except Exception as ex: logging.debug("Can't connect to Postgres " + str(ex)) time.sleep(0.5) raise Exception("Cannot wait Postgres container") def wait_postgres_cluster_to_start(self, timeout=180): self.postgres2_ip = self.get_instance_ip(self.postgres2_host) self.postgres3_ip = self.get_instance_ip(self.postgres3_host) self.postgres4_ip = self.get_instance_ip(self.postgres4_host) start = time.time() while time.time() - start < timeout: try: self.postgres2_conn = psycopg2.connect(host=self.postgres2_ip, port=self.postgres_port, database='postgres', user='postgres', password='mysecretpassword') self.postgres2_conn.set_isolation_level(ISOLATION_LEVEL_AUTOCOMMIT) self.postgres2_conn.autocommit = True logging.debug("Postgres Cluster host 2 started") break except Exception as ex: logging.debug("Can't connect to Postgres host 2" + str(ex)) time.sleep(0.5) while time.time() - start < timeout: try: self.postgres3_conn = psycopg2.connect(host=self.postgres3_ip, port=self.postgres_port, database='postgres', user='postgres', password='mysecretpassword') self.postgres3_conn.set_isolation_level(ISOLATION_LEVEL_AUTOCOMMIT) self.postgres3_conn.autocommit = True logging.debug("Postgres Cluster host 3 started") break except Exception as ex: logging.debug("Can't connect to Postgres host 3" + str(ex)) time.sleep(0.5) while time.time() - start < timeout: try: self.postgres4_conn = psycopg2.connect(host=self.postgres4_ip, port=self.postgres_port, database='postgres', user='postgres', password='mysecretpassword') self.postgres4_conn.set_isolation_level(ISOLATION_LEVEL_AUTOCOMMIT) self.postgres4_conn.autocommit = True logging.debug("Postgres Cluster host 4 started") return except Exception as ex: logging.debug("Can't connect to Postgres host 4" + str(ex)) time.sleep(0.5) raise Exception("Cannot wait Postgres container") def wait_rabbitmq_to_start(self, timeout=180, throw=True): self.rabbitmq_ip = self.get_instance_ip(self.rabbitmq_host) start = time.time() while time.time() - start < timeout: try: if check_rabbitmq_is_available(self.rabbitmq_docker_id): logging.debug("RabbitMQ is available") if enable_consistent_hash_plugin(self.rabbitmq_docker_id): logging.debug("RabbitMQ consistent hash plugin is available") return True time.sleep(0.5) except Exception as ex: logging.debug("Can't connect to RabbitMQ " + str(ex)) time.sleep(0.5) if throw: raise Exception("Cannot wait RabbitMQ container") return False def wait_zookeeper_secure_to_start(self, timeout=20): logging.debug("Wait ZooKeeper Secure to start") start = time.time() while time.time() - start < timeout: try: for instance in ['zoo1', 'zoo2', 'zoo3']: conn = self.get_kazoo_client(instance) conn.get_children('/') conn.stop() logging.debug("All instances of ZooKeeper Secure started") return except Exception as ex: logging.debug("Can't connect to ZooKeeper secure " + str(ex)) time.sleep(0.5) raise Exception("Cannot wait ZooKeeper secure container") def wait_zookeeper_to_start(self, timeout=180): logging.debug("Wait ZooKeeper to start") start = time.time() while time.time() - start < timeout: try: for instance in ['zoo1', 'zoo2', 'zoo3']: conn = self.get_kazoo_client(instance) conn.get_children('/') conn.stop() logging.debug("All instances of ZooKeeper started") return except Exception as ex: logging.debug("Can't connect to ZooKeeper " + str(ex)) time.sleep(0.5) raise Exception("Cannot wait ZooKeeper container") def make_hdfs_api(self, timeout=180, kerberized=False): if kerberized: keytab = p.abspath(p.join(self.instances['node1'].path, "secrets/clickhouse.keytab")) krb_conf = p.abspath(p.join(self.instances['node1'].path, "secrets/krb_long.conf")) self.hdfs_kerberized_ip = self.get_instance_ip(self.hdfs_kerberized_host) kdc_ip = self.get_instance_ip('hdfskerberos') self.hdfs_api = HDFSApi(user="root", timeout=timeout, kerberized=True, principal="root@TEST.CLICKHOUSE.TECH", keytab=keytab, krb_conf=krb_conf, host=self.hdfs_kerberized_host, protocol="http", proxy_port=self.hdfs_kerberized_name_port, data_port=self.hdfs_kerberized_data_port, hdfs_ip=self.hdfs_kerberized_ip, kdc_ip=kdc_ip) else: self.hdfs_ip = self.get_instance_ip(self.hdfs_host) self.hdfs_api = HDFSApi(user="root", host=self.hdfs_host, data_port=self.hdfs_data_port, proxy_port=self.hdfs_name_port, hdfs_ip=self.hdfs_ip) def wait_kafka_is_available(self, kafka_docker_id, kafka_port, max_retries=50): retries = 0 while True: if check_kafka_is_available(kafka_docker_id, kafka_port): break else: retries += 1 if retries > max_retries: raise Exception("Kafka is not available") logging.debug("Waiting for Kafka to start up") time.sleep(1) def wait_hdfs_to_start(self, timeout=300, check_marker=False): start = time.time() while time.time() - start < timeout: try: self.hdfs_api.write_data("/somefilewithrandomname222", "1") logging.debug("Connected to HDFS and SafeMode disabled! ") if check_marker: self.hdfs_api.read_data("/preparations_done_marker") return except Exception as ex: logging.exception("Can't connect to HDFS or preparations are not done yet " + str(ex)) time.sleep(1) raise Exception("Can't wait HDFS to start") def wait_mongo_to_start(self, timeout=30, secure=False): connection_str = 'mongodb://{user}:{password}@{host}:{port}'.format( host='localhost', port=self.mongo_port, user='root', password='clickhouse') if secure: connection_str += '/?tls=true&tlsAllowInvalidCertificates=true' connection = pymongo.MongoClient(connection_str) start = time.time() while time.time() - start < timeout: try: connection.list_database_names() logging.debug(f"Connected to Mongo dbs: {connection.database_names()}") return except Exception as ex: logging.debug("Can't connect to Mongo " + str(ex)) time.sleep(1) def wait_minio_to_start(self, timeout=180, secure=False): self.minio_ip = self.get_instance_ip(self.minio_host) self.minio_redirect_ip = self.get_instance_ip(self.minio_redirect_host) os.environ['SSL_CERT_FILE'] = p.join(self.base_dir, self.minio_dir, 'certs', 'public.crt') minio_client = Minio(f'{self.minio_ip}:{self.minio_port}', access_key='minio', secret_key='minio123', secure=secure, http_client=urllib3.PoolManager(cert_reqs='CERT_NONE')) # disable SSL check as we test ClickHouse and not Python library start = time.time() while time.time() - start < timeout: try: minio_client.list_buckets() logging.debug("Connected to Minio.") buckets = [self.minio_bucket, self.minio_bucket_2] for bucket in buckets: if minio_client.bucket_exists(bucket): delete_object_list = map( lambda x: x.object_name, minio_client.list_objects_v2(bucket, recursive=True), ) errors = minio_client.remove_objects(bucket, delete_object_list) for error in errors: logging.error(f"Error occured when deleting object {error}") minio_client.remove_bucket(bucket) minio_client.make_bucket(bucket) logging.debug("S3 bucket '%s' created", bucket) self.minio_client = minio_client return except Exception as ex: logging.debug("Can't connect to Minio: %s", str(ex)) time.sleep(1) raise Exception("Can't wait Minio to start") def wait_schema_registry_to_start(self, timeout=180): sr_client = CachedSchemaRegistryClient({"url":'http://localhost:{}'.format(self.schema_registry_port)}) start = time.time() while time.time() - start < timeout: try: sr_client._send_request(sr_client.url) logging.debug("Connected to SchemaRegistry") return sr_client except Exception as ex: logging.debug(("Can't connect to SchemaRegistry: %s", str(ex))) time.sleep(1) raise Exception("Can't wait Schema Registry to start") def wait_cassandra_to_start(self, timeout=180): self.cassandra_ip = self.get_instance_ip(self.cassandra_host) cass_client = cassandra.cluster.Cluster([self.cassandra_ip], port=self.cassandra_port, load_balancing_policy=RoundRobinPolicy()) start = time.time() while time.time() - start < timeout: try: logging.info(f"Check Cassandra Online {self.cassandra_id} {self.cassandra_ip} {self.cassandra_port}") check = self.exec_in_container(self.cassandra_id, ["bash", "-c", f"/opt/cassandra/bin/cqlsh -u cassandra -p cassandra -e 'describe keyspaces' {self.cassandra_ip} {self.cassandra_port}"], user='root') logging.info("Cassandra Online") cass_client.connect() logging.info("Connected Clients to Cassandra") return except Exception as ex: logging.warning("Can't connect to Cassandra: %s", str(ex)) time.sleep(1) raise Exception("Can't wait Cassandra to start") def start(self, destroy_dirs=True): pytest_xdist_logging_to_separate_files.setup() logging.info("Running tests in {}".format(self.base_path)) logging.debug("Cluster start called. is_up={}, destroy_dirs={}".format(self.is_up, destroy_dirs)) if self.is_up: return try: self.cleanup() except Exception as e: logging.warning("Cleanup failed:{e}") try: # clickhouse_pull_cmd = self.base_cmd + ['pull'] # print(f"Pulling images for {self.base_cmd}") # retry_exception(10, 5, subprocess_check_call, Exception, clickhouse_pull_cmd) if destroy_dirs and p.exists(self.instances_dir): logging.debug(("Removing instances dir %s", self.instances_dir)) shutil.rmtree(self.instances_dir) for instance in list(self.instances.values()): logging.debug(('Setup directory for instance: {} destroy_dirs: {}'.format(instance.name, destroy_dirs))) instance.create_dir(destroy_dir=destroy_dirs) _create_env_file(os.path.join(self.env_file), self.env_variables) self.docker_client = docker.DockerClient(base_url='unix:///var/run/docker.sock', version=self.docker_api_version, timeout=600) common_opts = ['up', '-d'] if self.with_zookeeper_secure and self.base_zookeeper_cmd: logging.debug('Setup ZooKeeper Secure') logging.debug(f'Creating internal ZooKeeper dirs: {self.zookeeper_dirs_to_create}') for i in range(1,3): if os.path.exists(self.zookeeper_instance_dir_prefix + f"{i}"): shutil.rmtree(self.zookeeper_instance_dir_prefix + f"{i}") for dir in self.zookeeper_dirs_to_create: os.makedirs(dir) run_and_check(self.base_zookeeper_cmd + common_opts, env=self.env) self.wait_zookeeper_secure_to_start() for command in self.pre_zookeeper_commands: self.run_kazoo_commands_with_retries(command, repeats=5) if self.with_zookeeper and self.base_zookeeper_cmd: logging.debug('Setup ZooKeeper') logging.debug(f'Creating internal ZooKeeper dirs: {self.zookeeper_dirs_to_create}') if self.use_keeper: for i in range(1,4): if os.path.exists(self.keeper_instance_dir_prefix + f"{i}"): shutil.rmtree(self.keeper_instance_dir_prefix + f"{i}") else: for i in range(1,3): if os.path.exists(self.zookeeper_instance_dir_prefix + f"{i}"): shutil.rmtree(self.zookeeper_instance_dir_prefix + f"{i}") for dir in self.zookeeper_dirs_to_create: os.makedirs(dir) if self.use_keeper: # TODO: remove hardcoded paths from here for i in range(1,4): shutil.copy(os.path.join(HELPERS_DIR, f'keeper_config{i}.xml'), os.path.join(self.keeper_instance_dir_prefix + f"{i}", "config" )) run_and_check(self.base_zookeeper_cmd + common_opts, env=self.env) self.wait_zookeeper_to_start() for command in self.pre_zookeeper_commands: self.run_kazoo_commands_with_retries(command, repeats=5) if self.with_mysql_client and self.base_mysql_client_cmd: logging.debug('Setup MySQL Client') subprocess_check_call(self.base_mysql_client_cmd + common_opts) self.wait_mysql_client_to_start() if self.with_mysql and self.base_mysql_cmd: logging.debug('Setup MySQL') if os.path.exists(self.mysql_dir): shutil.rmtree(self.mysql_dir) os.makedirs(self.mysql_logs_dir) os.chmod(self.mysql_logs_dir, stat.S_IRWXO) subprocess_check_call(self.base_mysql_cmd + common_opts) self.wait_mysql_to_start() if self.with_mysql8 and self.base_mysql8_cmd: logging.debug('Setup MySQL 8') if os.path.exists(self.mysql8_dir): shutil.rmtree(self.mysql8_dir) os.makedirs(self.mysql8_logs_dir) os.chmod(self.mysql8_logs_dir, stat.S_IRWXO) subprocess_check_call(self.base_mysql8_cmd + common_opts) self.wait_mysql8_to_start() if self.with_mysql_cluster and self.base_mysql_cluster_cmd: print('Setup MySQL') if os.path.exists(self.mysql_cluster_dir): shutil.rmtree(self.mysql_cluster_dir) os.makedirs(self.mysql_cluster_logs_dir) os.chmod(self.mysql_cluster_logs_dir, stat.S_IRWXO) subprocess_check_call(self.base_mysql_cluster_cmd + common_opts) self.wait_mysql_cluster_to_start() if self.with_postgres and self.base_postgres_cmd: logging.debug('Setup Postgres') if os.path.exists(self.postgres_dir): shutil.rmtree(self.postgres_dir) os.makedirs(self.postgres_logs_dir) os.chmod(self.postgres_logs_dir, stat.S_IRWXO) subprocess_check_call(self.base_postgres_cmd + common_opts) self.wait_postgres_to_start() if self.with_postgres_cluster and self.base_postgres_cluster_cmd: print('Setup Postgres') os.makedirs(self.postgres2_logs_dir) os.chmod(self.postgres2_logs_dir, stat.S_IRWXO) os.makedirs(self.postgres3_logs_dir) os.chmod(self.postgres3_logs_dir, stat.S_IRWXO) os.makedirs(self.postgres4_logs_dir) os.chmod(self.postgres4_logs_dir, stat.S_IRWXO) subprocess_check_call(self.base_postgres_cluster_cmd + common_opts) self.wait_postgres_cluster_to_start() if self.with_kafka and self.base_kafka_cmd: logging.debug('Setup Kafka') subprocess_check_call(self.base_kafka_cmd + common_opts + ['--renew-anon-volumes']) self.wait_kafka_is_available(self.kafka_docker_id, self.kafka_port) self.wait_schema_registry_to_start() if self.with_kerberized_kafka and self.base_kerberized_kafka_cmd: logging.debug('Setup kerberized kafka') run_and_check(self.base_kerberized_kafka_cmd + common_opts + ['--renew-anon-volumes']) self.wait_kafka_is_available(self.kerberized_kafka_docker_id, self.kerberized_kafka_port, 100) if self.with_rabbitmq and self.base_rabbitmq_cmd: logging.debug('Setup RabbitMQ') os.makedirs(self.rabbitmq_logs_dir) os.chmod(self.rabbitmq_logs_dir, stat.S_IRWXO) for i in range(5): subprocess_check_call(self.base_rabbitmq_cmd + common_opts + ['--renew-anon-volumes']) self.rabbitmq_docker_id = self.get_instance_docker_id('rabbitmq1') logging.debug(f"RabbitMQ checking container try: {i}") if self.wait_rabbitmq_to_start(throw=(i==4)): break if self.with_hdfs and self.base_hdfs_cmd: logging.debug('Setup HDFS') os.makedirs(self.hdfs_logs_dir) os.chmod(self.hdfs_logs_dir, stat.S_IRWXO) subprocess_check_call(self.base_hdfs_cmd + common_opts) self.make_hdfs_api() self.wait_hdfs_to_start() if self.with_kerberized_hdfs and self.base_kerberized_hdfs_cmd: logging.debug('Setup kerberized HDFS') os.makedirs(self.hdfs_kerberized_logs_dir) os.chmod(self.hdfs_kerberized_logs_dir, stat.S_IRWXO) run_and_check(self.base_kerberized_hdfs_cmd + common_opts) self.make_hdfs_api(kerberized=True) self.wait_hdfs_to_start(check_marker=True) if self.with_mongo and self.base_mongo_cmd: logging.debug('Setup Mongo') run_and_check(self.base_mongo_cmd + common_opts) self.wait_mongo_to_start(30, secure=self.with_mongo_secure) if self.with_redis and self.base_redis_cmd: logging.debug('Setup Redis') subprocess_check_call(self.base_redis_cmd + common_opts) time.sleep(10) if self.with_minio and self.base_minio_cmd: # Copy minio certificates to minio/certs os.mkdir(self.minio_dir) if self.minio_certs_dir is None: os.mkdir(os.path.join(self.minio_dir, 'certs')) else: shutil.copytree(os.path.join(self.base_dir, self.minio_certs_dir), os.path.join(self.minio_dir, 'certs')) minio_start_cmd = self.base_minio_cmd + common_opts logging.info("Trying to create Minio instance by command %s", ' '.join(map(str, minio_start_cmd))) run_and_check(minio_start_cmd) logging.info("Trying to connect to Minio...") self.wait_minio_to_start(secure=self.minio_certs_dir is not None) if self.with_cassandra and self.base_cassandra_cmd: subprocess_check_call(self.base_cassandra_cmd + ['up', '-d']) self.wait_cassandra_to_start() if self.with_jdbc_bridge and self.base_jdbc_bridge_cmd: os.makedirs(self.jdbc_driver_logs_dir) os.chmod(self.jdbc_driver_logs_dir, stat.S_IRWXO) subprocess_check_call(self.base_jdbc_bridge_cmd + ['up', '-d']) self.jdbc_bridge_ip = self.get_instance_ip(self.jdbc_bridge_host) self.wait_for_url(f"http://{self.jdbc_bridge_ip}:{self.jdbc_bridge_port}/ping") clickhouse_start_cmd = self.base_cmd + ['up', '-d', '--no-recreate'] logging.debug(("Trying to create ClickHouse instance by command %s", ' '.join(map(str, clickhouse_start_cmd)))) self.up_called = True run_and_check(clickhouse_start_cmd) logging.debug("ClickHouse instance created") start_timeout = 300.0 # seconds for instance in self.instances.values(): instance.docker_client = self.docker_client instance.ip_address = self.get_instance_ip(instance.name) logging.debug(f"Waiting for ClickHouse start in {instance.name}, ip: {instance.ip_address}...") instance.wait_for_start(start_timeout) logging.debug(f"ClickHouse {instance.name} started") instance.client = Client(instance.ip_address, command=self.client_bin_path) self.is_up = True except BaseException as e: logging.debug("Failed to start cluster: ") logging.debug(str(e)) logging.debug(traceback.print_exc()) self.shutdown() raise def shutdown(self, kill=True): sanitizer_assert_instance = None fatal_log = None if self.up_called: with open(self.docker_logs_path, "w+") as f: try: subprocess.check_call(self.base_cmd + ['logs'], stdout=f) # STYLE_CHECK_ALLOW_SUBPROCESS_CHECK_CALL except Exception as e: logging.debug("Unable to get logs from docker.") f.seek(0) for line in f: if SANITIZER_SIGN in line: sanitizer_assert_instance = line.split('|')[0].strip() break for name, instance in self.instances.items(): try: if not instance.is_up: continue if instance.contains_in_log(SANITIZER_SIGN): sanitizer_assert_instance = instance.grep_in_log(SANITIZER_SIGN) logging.ERROR(f"Sanitizer in instance {name} log {sanitizer_assert_instance}") if instance.contains_in_log("Fatal"): fatal_log = instance.grep_in_log("Fatal") logging.ERROR(f"Crash in instance {name} fatal log {fatal_log}") except Exception as e: logging.error(f"Failed to check fails in logs: {e}") if kill: try: run_and_check(self.base_cmd + ['stop', '--timeout', '20']) except Exception as e: logging.debug("Kill command failed during shutdown. {}".format(repr(e))) logging.debug("Trying to kill forcefully") run_and_check(self.base_cmd + ['kill']) try: subprocess_check_call(self.base_cmd + ['down', '--volumes']) except Exception as e: logging.debug("Down + remove orphans failed durung shutdown. {}".format(repr(e))) self.cleanup() self.is_up = False self.docker_client = None for instance in list(self.instances.values()): instance.docker_client = None instance.ip_address = None instance.client = None if sanitizer_assert_instance is not None: raise Exception( "Sanitizer assert found in {} for instance {}".format(self.docker_logs_path, sanitizer_assert_instance)) def pause_container(self, instance_name): subprocess_check_call(self.base_cmd + ['pause', instance_name]) # subprocess_check_call(self.base_cmd + ['kill', '-s SIGSTOP', instance_name]) def unpause_container(self, instance_name): subprocess_check_call(self.base_cmd + ['unpause', instance_name]) # subprocess_check_call(self.base_cmd + ['kill', '-s SIGCONT', instance_name]) def open_bash_shell(self, instance_name): os.system(' '.join(self.base_cmd + ['exec', instance_name, '/bin/bash'])) def get_kazoo_client(self, zoo_instance_name): use_ssl = False if self.with_zookeeper_secure: port = self.zookeeper_secure_port use_ssl = True elif self.with_zookeeper: port = self.zookeeper_port else: raise Exception("Cluster has no ZooKeeper") ip = self.get_instance_ip(zoo_instance_name) logging.debug(f"get_kazoo_client: {zoo_instance_name}, ip:{ip}, port:{port}, use_ssl:{use_ssl}") zk = KazooClient(hosts=f"{ip}:{port}", use_ssl=use_ssl, verify_certs=False, certfile=self.zookeeper_certfile, keyfile=self.zookeeper_keyfile) zk.start() return zk def run_kazoo_commands_with_retries(self, kazoo_callback, zoo_instance_name='zoo1', repeats=1, sleep_for=1): zk = self.get_kazoo_client(zoo_instance_name) logging.debug(f"run_kazoo_commands_with_retries: {zoo_instance_name}, {kazoo_callback}") for i in range(repeats - 1): try: kazoo_callback(zk) return except KazooException as e: logging.debug(repr(e)) time.sleep(sleep_for) kazoo_callback(zk) zk.stop() def add_zookeeper_startup_command(self, command): self.pre_zookeeper_commands.append(command) def stop_zookeeper_nodes(self, zk_nodes): for n in zk_nodes: logging.info("Stopping zookeeper node: %s", n) subprocess_check_call(self.base_zookeeper_cmd + ["stop", n]) def start_zookeeper_nodes(self, zk_nodes): for n in zk_nodes: logging.info("Starting zookeeper node: %s", n) subprocess_check_call(self.base_zookeeper_cmd + ["start", n]) CLICKHOUSE_START_COMMAND = "clickhouse server --config-file=/etc/clickhouse-server/{main_config_file}" \ " --log-file=/var/log/clickhouse-server/clickhouse-server.log " \ " --errorlog-file=/var/log/clickhouse-server/clickhouse-server.err.log" CLICKHOUSE_STAY_ALIVE_COMMAND = 'bash -c "{} --daemon; tail -f /dev/null"'.format(CLICKHOUSE_START_COMMAND) DOCKER_COMPOSE_TEMPLATE = ''' version: '2.3' services: {name}: image: {image}:{tag} hostname: {hostname} volumes: - {instance_config_dir}:/etc/clickhouse-server/ - {db_dir}:/var/lib/clickhouse/ - {logs_dir}:/var/log/clickhouse-server/ - /etc/passwd:/etc/passwd:ro {binary_volume} {odbc_bridge_volume} {library_bridge_volume} {odbc_ini_path} {keytab_path} {krb5_conf} entrypoint: {entrypoint_cmd} tmpfs: {tmpfs} cap_add: - SYS_PTRACE - NET_ADMIN - IPC_LOCK - SYS_NICE depends_on: {depends_on} user: '{user}' env_file: - {env_file} security_opt: - label:disable dns_opt: - attempts:2 - timeout:1 - inet6 - rotate {networks} {app_net} {ipv4_address} {ipv6_address} {net_aliases} {net_alias1} ''' class ClickHouseInstance: def __init__( self, cluster, base_path, name, base_config_dir, custom_main_configs, custom_user_configs, custom_dictionaries, macros, with_zookeeper, zookeeper_config_path, with_mysql_client, with_mysql, with_mysql8, with_mysql_cluster, with_kafka, with_kerberized_kafka, with_rabbitmq, with_kerberized_hdfs, with_mongo, with_redis, with_minio, with_jdbc_bridge, with_cassandra, server_bin_path, odbc_bridge_bin_path, library_bridge_bin_path, clickhouse_path_dir, with_odbc_drivers, with_postgres, with_postgres_cluster, clickhouse_start_command=CLICKHOUSE_START_COMMAND, main_config_name="config.xml", users_config_name="users.xml", copy_common_configs=True, hostname=None, env_variables=None, image="yandex/clickhouse-integration-test", tag="latest", stay_alive=False, ipv4_address=None, ipv6_address=None, with_installed_binary=False, tmpfs=None): self.name = name self.base_cmd = cluster.base_cmd self.docker_id = cluster.get_instance_docker_id(self.name) self.cluster = cluster self.hostname = hostname if hostname is not None else self.name self.tmpfs = tmpfs or [] self.base_config_dir = p.abspath(p.join(base_path, base_config_dir)) if base_config_dir else None self.custom_main_config_paths = [p.abspath(p.join(base_path, c)) for c in custom_main_configs] self.custom_user_config_paths = [p.abspath(p.join(base_path, c)) for c in custom_user_configs] self.custom_dictionaries_paths = [p.abspath(p.join(base_path, c)) for c in custom_dictionaries] self.clickhouse_path_dir = p.abspath(p.join(base_path, clickhouse_path_dir)) if clickhouse_path_dir else None self.kerberos_secrets_dir = p.abspath(p.join(base_path, 'secrets')) self.macros = macros if macros is not None else {} self.with_zookeeper = with_zookeeper self.zookeeper_config_path = zookeeper_config_path self.server_bin_path = server_bin_path self.odbc_bridge_bin_path = odbc_bridge_bin_path self.library_bridge_bin_path = library_bridge_bin_path self.with_mysql_client = with_mysql_client self.with_mysql = with_mysql self.with_mysql8 = with_mysql8 self.with_mysql_cluster = with_mysql_cluster self.with_postgres = with_postgres self.with_postgres_cluster = with_postgres_cluster self.with_kafka = with_kafka self.with_kerberized_kafka = with_kerberized_kafka self.with_rabbitmq = with_rabbitmq self.with_kerberized_hdfs = with_kerberized_hdfs self.with_mongo = with_mongo self.with_redis = with_redis self.with_minio = with_minio self.with_cassandra = with_cassandra self.with_jdbc_bridge = with_jdbc_bridge self.main_config_name = main_config_name self.users_config_name = users_config_name self.copy_common_configs = copy_common_configs self.clickhouse_start_command = clickhouse_start_command.replace("{main_config_file}", self.main_config_name) self.path = p.join(self.cluster.instances_dir, name) self.docker_compose_path = p.join(self.path, 'docker-compose.yml') self.env_variables = env_variables or {} self.env_file = self.cluster.env_file if with_odbc_drivers: self.odbc_ini_path = self.path + "/odbc.ini:/etc/odbc.ini" self.with_mysql = True else: self.odbc_ini_path = "" if with_kerberized_kafka or with_kerberized_hdfs: self.keytab_path = '- ' + os.path.dirname(self.docker_compose_path) + "/secrets:/tmp/keytab" self.krb5_conf = '- ' + os.path.dirname(self.docker_compose_path) + "/secrets/krb.conf:/etc/krb5.conf:ro" else: self.keytab_path = "" self.krb5_conf = "" self.docker_client = None self.ip_address = None self.client = None self.image = image self.tag = tag self.stay_alive = stay_alive self.ipv4_address = ipv4_address self.ipv6_address = ipv6_address self.with_installed_binary = with_installed_binary self.is_up = False def is_built_with_sanitizer(self, sanitizer_name=''): build_opts = self.query("SELECT value FROM system.build_options WHERE name = 'CXX_FLAGS'") return "-fsanitize={}".format(sanitizer_name) in build_opts def is_debug_build(self): build_opts = self.query("SELECT value FROM system.build_options WHERE name = 'CXX_FLAGS'") return 'NDEBUG' not in build_opts def is_built_with_thread_sanitizer(self): return self.is_built_with_sanitizer('thread') def is_built_with_address_sanitizer(self): return self.is_built_with_sanitizer('address') def is_built_with_memory_sanitizer(self): return self.is_built_with_sanitizer('memory') # Connects to the instance via clickhouse-client, sends a query (1st argument) and returns the answer def query(self, sql, stdin=None, timeout=None, settings=None, user=None, password=None, database=None, ignore_error=False): logging.debug(f"Executing query {sql} on {self.name}") return self.client.query(sql, stdin=stdin, timeout=timeout, settings=settings, user=user, password=password, database=database, ignore_error=ignore_error) def query_with_retry(self, sql, stdin=None, timeout=None, settings=None, user=None, password=None, database=None, ignore_error=False, retry_count=20, sleep_time=0.5, check_callback=lambda x: True): logging.debug(f"Executing query {sql} on {self.name}") result = None for i in range(retry_count): try: result = self.query(sql, stdin=stdin, timeout=timeout, settings=settings, user=user, password=password, database=database, ignore_error=ignore_error) if check_callback(result): return result time.sleep(sleep_time) except Exception as ex: logging.debug("Retry {} got exception {}".format(i + 1, ex)) time.sleep(sleep_time) if result is not None: return result raise Exception("Can't execute query {}".format(sql)) # As query() but doesn't wait response and returns response handler def get_query_request(self, sql, *args, **kwargs): logging.debug(f"Executing query {sql} on {self.name}") return self.client.get_query_request(sql, *args, **kwargs) # Connects to the instance via clickhouse-client, sends a query (1st argument), expects an error and return its code def query_and_get_error(self, sql, stdin=None, timeout=None, settings=None, user=None, password=None, database=None): logging.debug(f"Executing query {sql} on {self.name}") return self.client.query_and_get_error(sql, stdin=stdin, timeout=timeout, settings=settings, user=user, password=password, database=database) # The same as query_and_get_error but ignores successful query. def query_and_get_answer_with_error(self, sql, stdin=None, timeout=None, settings=None, user=None, password=None, database=None): logging.debug(f"Executing query {sql} on {self.name}") return self.client.query_and_get_answer_with_error(sql, stdin=stdin, timeout=timeout, settings=settings, user=user, password=password, database=database) # Connects to the instance via HTTP interface, sends a query and returns the answer def http_query(self, sql, data=None, params=None, user=None, password=None, expect_fail_and_get_error=False): logging.debug(f"Executing query {sql} on {self.name} via HTTP interface") if params is None: params = {} else: params = params.copy() params["query"] = sql auth = None if user and password: auth = requests.auth.HTTPBasicAuth(user, password) elif user: auth = requests.auth.HTTPBasicAuth(user, '') url = "http://" + self.ip_address + ":8123/?" + urllib.parse.urlencode(params) if data: r = requests.post(url, data, auth=auth) else: r = requests.get(url, auth=auth) def http_code_and_message(): code = r.status_code return str(code) + " " + http.client.responses[code] + ": " + r.text if expect_fail_and_get_error: if r.ok: raise Exception("ClickHouse HTTP server is expected to fail, but succeeded: " + r.text) return http_code_and_message() else: if not r.ok: raise Exception("ClickHouse HTTP server returned " + http_code_and_message()) return r.text # Connects to the instance via HTTP interface, sends a query and returns the answer def http_request(self, url, method='GET', params=None, data=None, headers=None): logging.debug(f"Sending HTTP request {url} to {self.name}") url = "http://" + self.ip_address + ":8123/" + url return requests.request(method=method, url=url, params=params, data=data, headers=headers) # Connects to the instance via HTTP interface, sends a query, expects an error and return the error message def http_query_and_get_error(self, sql, data=None, params=None, user=None, password=None): logging.debug(f"Executing query {sql} on {self.name} via HTTP interface") return self.http_query(sql=sql, data=data, params=params, user=user, password=password, expect_fail_and_get_error=True) def stop_clickhouse(self, stop_wait_sec=30, kill=False): if not self.stay_alive: raise Exception("clickhouse can be stopped only with stay_alive=True instance") try: ps_clickhouse = self.exec_in_container(["bash", "-c", "ps -C clickhouse"], user='root') if ps_clickhouse == " PID TTY STAT TIME COMMAND" : logging.warning("ClickHouse process already stopped") return self.exec_in_container(["bash", "-c", "pkill {} clickhouse".format("-9" if kill else "")], user='root') time.sleep(stop_wait_sec) ps_clickhouse = self.exec_in_container(["bash", "-c", "ps -C clickhouse"], user='root') if ps_clickhouse != " PID TTY STAT TIME COMMAND" : logging.warning(f"Force kill clickhouse in stop_clickhouse. ps:{ps_clickhouse}") self.stop_clickhouse(kill=True) except Exception as e: logging.warning(f"Stop ClickHouse raised an error {e}") def start_clickhouse(self, start_wait_sec=30): if not self.stay_alive: raise Exception("clickhouse can be started again only with stay_alive=True instance") self.exec_in_container(["bash", "-c", "{} --daemon".format(self.clickhouse_start_command)], user=str(os.getuid())) # wait start from helpers.test_tools import assert_eq_with_retry assert_eq_with_retry(self, "select 1", "1", retry_count=int(start_wait_sec / 0.5), sleep_time=0.5) def restart_clickhouse(self, stop_start_wait_sec=30, kill=False): self.stop_clickhouse(stop_start_wait_sec, kill) self.start_clickhouse(stop_start_wait_sec) def exec_in_container(self, cmd, detach=False, nothrow=False, **kwargs): return self.cluster.exec_in_container(self.docker_id, cmd, detach, nothrow, **kwargs) def contains_in_log(self, substring): result = self.exec_in_container( ["bash", "-c", '[ -f /var/log/clickhouse-server/clickhouse-server.log ] && grep "{}" /var/log/clickhouse-server/clickhouse-server.log || true'.format(substring)]) return len(result) > 0 def grep_in_log(self, substring): logging.debug(f"grep in log called {substring}") result = self.exec_in_container( ["bash", "-c", 'grep "{}" /var/log/clickhouse-server/clickhouse-server.log || true'.format(substring)]) logging.debug(f"grep result {result}") return result def count_in_log(self, substring): result = self.exec_in_container( ["bash", "-c", 'grep "{}" /var/log/clickhouse-server/clickhouse-server.log | wc -l'.format(substring)]) return result def wait_for_log_line(self, regexp, filename='/var/log/clickhouse-server/clickhouse-server.log', timeout=30, repetitions=1, look_behind_lines=100): start_time = time.time() result = self.exec_in_container( ["bash", "-c", 'timeout {} tail -Fn{} "{}" | grep -Em {} {}'.format(timeout, look_behind_lines, filename, repetitions, shlex.quote(regexp))]) # if repetitions>1 grep will return success even if not enough lines were collected, if repetitions>1 and len(result.splitlines()) < repetitions: logging.debug("wait_for_log_line: those lines were found during {} seconds:".format(timeout)) logging.debug(result) raise Exception("wait_for_log_line: Not enough repetitions: {} found, while {} expected".format(len(result.splitlines()), repetitions)) wait_duration = time.time() - start_time logging.debug('{} log line(s) matching "{}" appeared in a {:.3f} seconds'.format(repetitions, regexp, wait_duration)) return wait_duration def file_exists(self, path): return self.exec_in_container( ["bash", "-c", "echo $(if [ -e '{}' ]; then echo 'yes'; else echo 'no'; fi)".format(path)]) == 'yes\n' def copy_file_to_container(self, local_path, dest_path): return self.cluster.copy_file_to_container(self.docker_id, local_path, dest_path) def get_process_pid(self, process_name): output = self.exec_in_container(["bash", "-c", "ps ax | grep '{}' | grep -v 'grep' | grep -v 'bash -c' | awk '{{print $1}}'".format( process_name)]) if output: try: pid = int(output.split('\n')[0].strip()) return pid except: return None return None def restart_with_original_version(self, stop_start_wait_sec=300, callback_onstop=None, signal=15): if not self.stay_alive: raise Exception("Cannot restart not stay alive container") self.exec_in_container(["bash", "-c", "pkill -{} clickhouse".format(signal)], user='root') retries = int(stop_start_wait_sec / 0.5) local_counter = 0 # wait stop while local_counter < retries: if not self.get_process_pid("clickhouse server"): break time.sleep(0.5) local_counter += 1 # force kill if server hangs if self.get_process_pid("clickhouse server"): # server can die before kill, so don't throw exception, it's expected self.exec_in_container(["bash", "-c", "pkill -{} clickhouse".format(9)], nothrow=True, user='root') if callback_onstop: callback_onstop(self) self.exec_in_container( ["bash", "-c", "cp /usr/share/clickhouse_original /usr/bin/clickhouse && chmod 777 /usr/bin/clickhouse"], user='root') self.exec_in_container(["bash", "-c", "cp /usr/share/clickhouse-odbc-bridge_fresh /usr/bin/clickhouse-odbc-bridge && chmod 777 /usr/bin/clickhouse"], user='root') self.exec_in_container(["bash", "-c", "{} --daemon".format(self.clickhouse_start_command)], user=str(os.getuid())) # wait start assert_eq_with_retry(self, "select 1", "1", retry_count=retries) def restart_with_latest_version(self, stop_start_wait_sec=300, callback_onstop=None, signal=15): if not self.stay_alive: raise Exception("Cannot restart not stay alive container") self.exec_in_container(["bash", "-c", "pkill -{} clickhouse".format(signal)], user='root') retries = int(stop_start_wait_sec / 0.5) local_counter = 0 # wait stop while local_counter < retries: if not self.get_process_pid("clickhouse server"): break time.sleep(0.5) local_counter += 1 # force kill if server hangs if self.get_process_pid("clickhouse server"): # server can die before kill, so don't throw exception, it's expected self.exec_in_container(["bash", "-c", "pkill -{} clickhouse".format(9)], nothrow=True, user='root') if callback_onstop: callback_onstop(self) self.exec_in_container( ["bash", "-c", "cp /usr/bin/clickhouse /usr/share/clickhouse_original"], user='root') self.exec_in_container( ["bash", "-c", "cp /usr/share/clickhouse_fresh /usr/bin/clickhouse && chmod 777 /usr/bin/clickhouse"], user='root') self.exec_in_container(["bash", "-c", "cp /usr/share/clickhouse-odbc-bridge_fresh /usr/bin/clickhouse-odbc-bridge && chmod 777 /usr/bin/clickhouse"], user='root') self.exec_in_container(["bash", "-c", "{} --daemon".format(self.clickhouse_start_command)], user=str(os.getuid())) # wait start assert_eq_with_retry(self, "select 1", "1", retry_count=retries) def get_docker_handle(self): return self.cluster.get_docker_handle(self.docker_id) def stop(self): self.get_docker_handle().stop() def start(self): self.get_docker_handle().start() def wait_for_start(self, start_timeout=None, connection_timeout=None): handle = self.get_docker_handle() if start_timeout is None or start_timeout <= 0: raise Exception("Invalid timeout: {}".format(start_timeout)) if connection_timeout is not None and connection_timeout < start_timeout: raise Exception("Connection timeout {} should be grater then start timeout {}" .format(connection_timeout, start_timeout)) start_time = time.time() prev_rows_in_log = 0 def has_new_rows_in_log(): nonlocal prev_rows_in_log try: rows_in_log = int(self.count_in_log(".*").strip()) res = rows_in_log > prev_rows_in_log prev_rows_in_log = rows_in_log return res except ValueError: return False while True: handle.reload() status = handle.status if status == 'exited': raise Exception(f"Instance `{self.name}' failed to start. Container status: {status}, logs: {handle.logs().decode('utf-8')}") deadline = start_time + start_timeout # It is possible that server starts slowly. # If container is running, and there is some progress in log, check connection_timeout. if connection_timeout and status == 'running' and has_new_rows_in_log(): deadline = start_time + connection_timeout current_time = time.time() if current_time >= deadline: raise Exception(f"Timed out while waiting for instance `{self.name}' with ip address {self.ip_address} to start. " \ f"Container status: {status}, logs: {handle.logs().decode('utf-8')}") socket_timeout = min(start_timeout, deadline - current_time) # Repeatedly poll the instance address until there is something that listens there. # Usually it means that ClickHouse is ready to accept queries. try: sock = socket.socket(socket.AF_INET, socket.SOCK_STREAM) sock.settimeout(socket_timeout) sock.connect((self.ip_address, 9000)) self.is_up = True return except socket.timeout: continue except socket.error as e: if e.errno == errno.ECONNREFUSED or e.errno == errno.EHOSTUNREACH or e.errno == errno.ENETUNREACH: time.sleep(0.1) else: raise finally: sock.close() @staticmethod def dict_to_xml(dictionary): xml_str = dict2xml(dictionary, wrap="yandex", indent=" ", newlines=True) return xml_str @property def odbc_drivers(self): if self.odbc_ini_path: return { "SQLite3": { "DSN": "sqlite3_odbc", "Database": "/tmp/sqliteodbc", "Driver": "/usr/lib/x86_64-linux-gnu/odbc/libsqlite3odbc.so", "Setup": "/usr/lib/x86_64-linux-gnu/odbc/libsqlite3odbc.so", }, "MySQL": { "DSN": "mysql_odbc", "Driver": "/usr/lib/x86_64-linux-gnu/odbc/libmyodbc.so", "Database": "clickhouse", "Uid": "root", "Pwd": "clickhouse", "Server": self.cluster.mysql_host, }, "PostgreSQL": { "DSN": "postgresql_odbc", "Database": "postgres", "UserName": "postgres", "Password": "mysecretpassword", "Port": str(self.cluster.postgres_port), "Servername": self.cluster.postgres_host, "Protocol": "9.3", "ReadOnly": "No", "RowVersioning": "No", "ShowSystemTables": "No", "Driver": "/usr/lib/x86_64-linux-gnu/odbc/psqlodbca.so", "Setup": "/usr/lib/x86_64-linux-gnu/odbc/libodbcpsqlS.so", "ConnSettings": "", } } else: return {} def _create_odbc_config_file(self): with open(self.odbc_ini_path.split(':')[0], 'w') as f: for driver_setup in list(self.odbc_drivers.values()): f.write("[{}]\n".format(driver_setup["DSN"])) for key, value in list(driver_setup.items()): if key != "DSN": f.write(key + "=" + value + "\n") def replace_config(self, path_to_config, replacement): self.exec_in_container(["bash", "-c", "echo '{}' > {}".format(replacement, path_to_config)]) def create_dir(self, destroy_dir=True): """Create the instance directory and all the needed files there.""" if destroy_dir: self.destroy_dir() elif p.exists(self.path): return os.makedirs(self.path) instance_config_dir = p.abspath(p.join(self.path, 'configs')) os.makedirs(instance_config_dir) print(f"Copy common default production configuration from {self.base_config_dir}. Files: {self.main_config_name}, {self.users_config_name}") shutil.copyfile(p.join(self.base_config_dir, self.main_config_name), p.join(instance_config_dir, self.main_config_name)) shutil.copyfile(p.join(self.base_config_dir, self.users_config_name), p.join(instance_config_dir, self.users_config_name)) logging.debug("Create directory for configuration generated in this helper") # used by all utils with any config conf_d_dir = p.abspath(p.join(instance_config_dir, 'conf.d')) os.mkdir(conf_d_dir) logging.debug("Create directory for common tests configuration") # used by server with main config.xml self.config_d_dir = p.abspath(p.join(instance_config_dir, 'config.d')) os.mkdir(self.config_d_dir) users_d_dir = p.abspath(p.join(instance_config_dir, 'users.d')) os.mkdir(users_d_dir) dictionaries_dir = p.abspath(p.join(instance_config_dir, 'dictionaries')) os.mkdir(dictionaries_dir) logging.debug("Copy common configuration from helpers") # The file is named with 0_ prefix to be processed before other configuration overloads. if self.copy_common_configs: shutil.copy(p.join(HELPERS_DIR, '0_common_instance_config.xml'), self.config_d_dir) shutil.copy(p.join(HELPERS_DIR, '0_common_instance_users.xml'), users_d_dir) if len(self.custom_dictionaries_paths): shutil.copy(p.join(HELPERS_DIR, '0_common_enable_dictionaries.xml'), self.config_d_dir) logging.debug("Generate and write macros file") macros = self.macros.copy() macros['instance'] = self.name with open(p.join(conf_d_dir, 'macros.xml'), 'w') as macros_config: macros_config.write(self.dict_to_xml({"macros": macros})) # Put ZooKeeper config if self.with_zookeeper: shutil.copy(self.zookeeper_config_path, conf_d_dir) if self.with_kerberized_kafka or self.with_kerberized_hdfs: shutil.copytree(self.kerberos_secrets_dir, p.abspath(p.join(self.path, 'secrets'))) # Copy config.d configs logging.debug(f"Copy custom test config files {self.custom_main_config_paths} to {self.config_d_dir}") for path in self.custom_main_config_paths: shutil.copy(path, self.config_d_dir) # Copy users.d configs for path in self.custom_user_config_paths: shutil.copy(path, users_d_dir) # Copy dictionaries configs to configs/dictionaries for path in self.custom_dictionaries_paths: shutil.copy(path, dictionaries_dir) db_dir = p.abspath(p.join(self.path, 'database')) logging.debug(f"Setup database dir {db_dir}") if self.clickhouse_path_dir is not None: logging.debug(f"Database files taken from {self.clickhouse_path_dir}") shutil.copytree(self.clickhouse_path_dir, db_dir) logging.debug(f"Database copied from {self.clickhouse_path_dir} to {db_dir}") else: os.mkdir(db_dir) logs_dir = p.abspath(p.join(self.path, 'logs')) logging.debug(f"Setup logs dir {logs_dir}") os.mkdir(logs_dir) depends_on = [] if self.with_mysql_client: depends_on.append(self.cluster.mysql_client_host) if self.with_mysql: depends_on.append("mysql57") if self.with_mysql8: depends_on.append("mysql80") if self.with_mysql_cluster: depends_on.append("mysql57") depends_on.append("mysql2") depends_on.append("mysql3") depends_on.append("mysql4") if self.with_postgres_cluster: depends_on.append("postgres2") depends_on.append("postgres3") depends_on.append("postgres4") if self.with_kafka: depends_on.append("kafka1") depends_on.append("schema-registry") if self.with_kerberized_kafka: depends_on.append("kerberized_kafka1") if self.with_kerberized_hdfs: depends_on.append("kerberizedhdfs1") if self.with_rabbitmq: depends_on.append("rabbitmq1") if self.with_zookeeper: depends_on.append("zoo1") depends_on.append("zoo2") depends_on.append("zoo3") if self.with_minio: depends_on.append("minio1") self.cluster.env_variables.update(self.env_variables) odbc_ini_path = "" if self.odbc_ini_path: self._create_odbc_config_file() odbc_ini_path = '- ' + self.odbc_ini_path entrypoint_cmd = self.clickhouse_start_command if self.stay_alive: entrypoint_cmd = CLICKHOUSE_STAY_ALIVE_COMMAND.replace("{main_config_file}", self.main_config_name) logging.debug("Entrypoint cmd: {}".format(entrypoint_cmd)) networks = app_net = ipv4_address = ipv6_address = net_aliases = net_alias1 = "" if self.ipv4_address is not None or self.ipv6_address is not None or self.hostname != self.name: networks = "networks:" app_net = "default:" if self.ipv4_address is not None: ipv4_address = "ipv4_address: " + self.ipv4_address if self.ipv6_address is not None: ipv6_address = "ipv6_address: " + self.ipv6_address if self.hostname != self.name: net_aliases = "aliases:" net_alias1 = "- " + self.hostname if not self.with_installed_binary: binary_volume = "- " + self.server_bin_path + ":/usr/bin/clickhouse" odbc_bridge_volume = "- " + self.odbc_bridge_bin_path + ":/usr/bin/clickhouse-odbc-bridge" library_bridge_volume = "- " + self.library_bridge_bin_path + ":/usr/bin/clickhouse-library-bridge" else: binary_volume = "- " + self.server_bin_path + ":/usr/share/clickhouse_fresh" odbc_bridge_volume = "- " + self.odbc_bridge_bin_path + ":/usr/share/clickhouse-odbc-bridge_fresh" library_bridge_volume = "- " + self.library_bridge_bin_path + ":/usr/share/clickhouse-library-bridge_fresh" with open(self.docker_compose_path, 'w') as docker_compose: docker_compose.write(DOCKER_COMPOSE_TEMPLATE.format( image=self.image, tag=self.tag, name=self.name, hostname=self.hostname, binary_volume=binary_volume, odbc_bridge_volume=odbc_bridge_volume, library_bridge_volume=library_bridge_volume, instance_config_dir=instance_config_dir, config_d_dir=self.config_d_dir, db_dir=db_dir, tmpfs=str(self.tmpfs), logs_dir=logs_dir, depends_on=str(depends_on), user=os.getuid(), env_file=self.env_file, odbc_ini_path=odbc_ini_path, keytab_path=self.keytab_path, krb5_conf=self.krb5_conf, entrypoint_cmd=entrypoint_cmd, networks=networks, app_net=app_net, ipv4_address=ipv4_address, ipv6_address=ipv6_address, net_aliases=net_aliases, net_alias1=net_alias1, )) def destroy_dir(self): if p.exists(self.path): shutil.rmtree(self.path) class ClickHouseKiller(object): def __init__(self, clickhouse_node): self.clickhouse_node = clickhouse_node def __enter__(self): self.clickhouse_node.stop_clickhouse(kill=True) def __exit__(self, exc_type, exc_val, exc_tb): self.clickhouse_node.start_clickhouse()