import time from contextlib import contextmanager import pytest from helpers.cluster import ClickHouseCluster from helpers.network import PartitionManager from helpers.test_tools import TSV cluster = ClickHouseCluster(__file__) node1 = cluster.add_instance('node1', main_configs=['configs/remote_servers.xml'], with_zookeeper=True) node2 = cluster.add_instance('node2', main_configs=['configs/remote_servers.xml'], with_zookeeper=True) node3 = cluster.add_instance('node3', main_configs=['configs/remote_servers.xml'], with_zookeeper=True) @pytest.fixture(scope="module") def started_cluster(): try: cluster.start() node_to_shards = [ (node1, [0, 2]), (node2, [0, 1]), (node3, [1, 2]), ] for node, shards in node_to_shards: for shard in shards: node.query(''' CREATE DATABASE shard_{shard}; CREATE TABLE shard_{shard}.replicated(date Date, id UInt32, shard_id UInt32) ENGINE = ReplicatedMergeTree('/clickhouse/tables/{shard}/replicated', '{replica}', date, id, 8192); '''.format(shard=shard, replica=node.name)) node.query(''' CREATE TABLE distributed(date Date, id UInt32, shard_id UInt32) ENGINE = Distributed(test_cluster, '', replicated, shard_id); ''') # Insert some data onto different shards using the Distributed table to_insert = '''\ 2017-06-16 111 0 2017-06-16 222 1 2017-06-16 333 2 ''' node1.query("INSERT INTO distributed FORMAT TSV", stdin=to_insert) time.sleep(0.5) yield cluster finally: cluster.shutdown() def test(started_cluster): # Check that the data has been inserted into correct tables. assert node1.query("SELECT id FROM shard_0.replicated") == '111\n' assert node1.query("SELECT id FROM shard_2.replicated") == '333\n' assert node2.query("SELECT id FROM shard_0.replicated") == '111\n' assert node2.query("SELECT id FROM shard_1.replicated") == '222\n' assert node3.query("SELECT id FROM shard_1.replicated") == '222\n' assert node3.query("SELECT id FROM shard_2.replicated") == '333\n' # Check that SELECT from the Distributed table works. expected_from_distributed = '''\ 2017-06-16 111 0 2017-06-16 222 1 2017-06-16 333 2 ''' assert TSV(node1.query("SELECT * FROM distributed ORDER BY id")) == TSV(expected_from_distributed) assert TSV(node2.query("SELECT * FROM distributed ORDER BY id")) == TSV(expected_from_distributed) assert TSV(node3.query("SELECT * FROM distributed ORDER BY id")) == TSV(expected_from_distributed) # Now isolate node3 from other nodes and check that SELECTs on other nodes still work. with PartitionManager() as pm: pm.partition_instances(node3, node1, action='REJECT --reject-with tcp-reset') pm.partition_instances(node3, node2, action='REJECT --reject-with tcp-reset') assert TSV(node1.query("SELECT * FROM distributed ORDER BY id")) == TSV(expected_from_distributed) assert TSV(node2.query("SELECT * FROM distributed ORDER BY id")) == TSV(expected_from_distributed) with pytest.raises(Exception): print node3.query("SELECT * FROM distributed ORDER BY id") if __name__ == '__main__': with contextmanager(started_cluster)() as cluster: for name, instance in cluster.instances.items(): print name, instance.ip_address raw_input("Cluster created, press any key to destroy...")