mirror of
https://github.com/MariaDB/server.git
synced 2025-02-09 23:24:11 +01:00
a3cbc44b24
MariaDB async replication SQL thread was stopped for any failure in applying of replication events and error message logged for the failure was: "Node has dropped from cluster". The assumption was that event applying failure is always due to node dropping out. With optimistic parallel replication, event applying can fail for natural reasons and applying should be retried to handle the failure. This retry logic was never exercised because the slave SQL thread was stopped with first applying failure. To support optimistic parallel replication retrying logic this commit will now skip replication slave abort, if node remains in cluster (wsrep_ready==ON) and replication is configured for optimistic or aggressive retry logic. During the development of this fix, galera.galera_as_slave_nonprim test showed some problems. The test was analyzed, and it appears to need some attention. One excessive sleep command was removed in this commit, but it will need more fixes still to be fully deterministic. After this commit galera_as_slave_nonprim is successful, though. Signed-off-by: Julius Goryavsky <julius.goryavsky@mariadb.com>
93 lines
3.1 KiB
Text
93 lines
3.1 KiB
Text
#
|
|
# Test the behavior of a Galera async slave if it goes non-prim. Async replication
|
|
# should abort with an error but it should be possible to restart it.
|
|
#
|
|
# The galera/galera_2node_slave.cnf describes the setup of the nodes
|
|
#
|
|
|
|
--source include/have_innodb.inc
|
|
--source include/big_test.inc
|
|
--source include/galera_cluster.inc
|
|
|
|
# Step #1. Establish replication
|
|
#
|
|
# As node 4 is not a Galera node, and galera_cluster.inc does not open connetion to it
|
|
# we open the node_4 connection here
|
|
--connect node_4, 127.0.0.1, root, , test, $NODE_MYPORT_4
|
|
|
|
|
|
--connection node_2
|
|
--disable_query_log
|
|
--eval CHANGE MASTER TO MASTER_HOST='127.0.0.1', MASTER_PORT=$NODE_MYPORT_4, MASTER_USER='root';
|
|
--enable_query_log
|
|
START SLAVE;
|
|
SET SESSION wsrep_sync_wait = 0;
|
|
|
|
--connection node_4
|
|
CREATE TABLE t1 (f1 INTEGER PRIMARY KEY) ENGINE=InnoDB;
|
|
|
|
--connection node_2
|
|
--let $wait_condition = SELECT COUNT(*) = 1 FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_NAME = 't1';
|
|
--source include/wait_condition.inc
|
|
|
|
# Step #2. Force async slave to go non-primary
|
|
|
|
SET GLOBAL wsrep_provider_options = 'gmcast.isolate=1';
|
|
|
|
--connection node_1
|
|
--source include/wait_until_connected_again.inc
|
|
--let $wait_condition = SELECT VARIABLE_VALUE = 2 FROM INFORMATION_SCHEMA.GLOBAL_STATUS WHERE VARIABLE_NAME = 'wsrep_cluster_size'
|
|
--source include/wait_condition.inc
|
|
|
|
# Step #3. Force async replication to fail by creating a replication event while the slave is non-prim
|
|
|
|
--connection node_4
|
|
INSERT INTO t1 VALUES (1),(2),(3),(4),(5);
|
|
|
|
--connection node_2
|
|
--sleep 5
|
|
--let $value = query_get_value(SHOW SLAVE STATUS, Last_SQL_Error, 1)
|
|
--connection node_1
|
|
--disable_query_log
|
|
--eval SELECT "$value" IN ("Error 'Unknown command' on query. Default database: 'test'. Query: 'BEGIN'", "Node has dropped from cluster") AS expected_error
|
|
--enable_query_log
|
|
|
|
# Step #4. Bring back the async slave and restart replication
|
|
--connection node_2
|
|
SET GLOBAL wsrep_provider_options = 'gmcast.isolate=0';
|
|
|
|
--connection node_1
|
|
--source include/wait_until_connected_again.inc
|
|
--let $wait_condition = SELECT VARIABLE_VALUE = 3 FROM INFORMATION_SCHEMA.GLOBAL_STATUS WHERE VARIABLE_NAME = 'wsrep_cluster_size'
|
|
--source include/wait_condition.inc
|
|
|
|
--connection node_2
|
|
--source include/wait_until_ready.inc
|
|
--source include/wait_until_connected_again.inc
|
|
|
|
START SLAVE;
|
|
|
|
# Confirm that the replication events have arrived
|
|
|
|
--let $wait_condition = SELECT COUNT(*) = 5 FROM t1;
|
|
--source include/wait_condition.inc
|
|
|
|
--connection node_4
|
|
DROP TABLE t1;
|
|
|
|
--sleep 2
|
|
--connection node_2
|
|
--let $wait_condition = SELECT COUNT(*) = 0 FROM INFORMATION_SCHEMA.TABLES WHERE TABLE_NAME = 't1';
|
|
--source include/wait_condition.inc
|
|
|
|
STOP SLAVE;
|
|
RESET SLAVE ALL;
|
|
|
|
CALL mtr.add_suppression("Slave SQL: Error 'Unknown command' on query");
|
|
CALL mtr.add_suppression("Slave: Unknown command Error_code: 1047");
|
|
CALL mtr.add_suppression("Transport endpoint is not connected");
|
|
CALL mtr.add_suppression("Slave SQL: Error in Xid_log_event: Commit could not be completed, 'Deadlock found when trying to get lock; try restarting transaction', Error_code: 1213");
|
|
CALL mtr.add_suppression("Slave SQL: Node has dropped from cluster, Error_code: 1047");
|
|
|
|
--connection node_4
|
|
RESET MASTER;
|