HA_ERR_FOUND_DUPP_KEY Clone Plugin for Replication with GTID mode #1507
Description
Got HA_ERR_FOUND_DUPP_KEY using Clone Plugin for Replication(with GTID mode). Replication Works fine with InnoDB, also fine with clone by xtrabackup and MyRocks.
myrocks version : 8.0.32 git sha: 69710da
server1:
#/root/test.sh
#!/bin/bash
mysql -e '
CREATE DATABASE IF NOT EXISTS test;
use test;
CREATE TABLE IF NOT EXISTS large_data (
id INT AUTO_INCREMENT PRIMARY KEY,
content VARCHAR(1024)
)ENGINE=RocksDB;
DELIMITER $$
DROP PROCEDURE IF EXISTS insert_large_data;
CREATE PROCEDURE insert_large_data()
BEGIN
DECLARE i INT DEFAULT 0;
WHILE i < 2000000 DO
INSERT INTO large_data (content)
VALUES (SUBSTRING(CONCAT(MD5(RAND()), MD5(RAND()), MD5(RAND()), MD5(RAND())), 1, 1024));
SET i = i + 1;
END WHILE;
END $$
DELIMITER ;
'
while true;
do
if [ -e /root/stop ];then
break;
fi
mysql -Dtest -e "call insert_large_data()";
done;run bash /root/test.sh on server1 before start clone on server2.
then create user on server 1 for clone :
CREATE USER 'clone'@'%' IDENTIFIED BY 'test123';
GRANT ALL ON *.* TO 'clone'@'%';
ALTER USER 'clone'@'%' IDENTIFIED WITH mysql_native_password BY 'test123';
flush privileges;then run clone on server2:
set global clone_valid_donor_list ='10.3.0.121:3306';
CLONE INSTANCE FROM 'clone'@'10.3.0.121':3306 IDENTIFIED BY 'test123';after restart srever2 :
change master to master_host='10.3.0.121', master_user='clone', master_password='test123',master_auto_position=1;
start slave;I got 1602 error:
mysql> show slave status\G
*************************** 1. row ***************************
Slave_IO_State: Waiting for source to send event
Master_Host: 10.3.0.121
Master_User: clone
Master_Port: 3306
Connect_Retry: 60
Master_Log_File: binlog.000001
Read_Master_Log_Pos: 20537793
Relay_Log_File: iZwz9azwk66n3vfhfj5rjdZ-relay-bin.000002
Relay_Log_Pos: 1647
Relay_Master_Log_File: binlog.000001
Slave_IO_Running: Yes
Slave_SQL_Running: No
Replicate_Do_DB:
Replicate_Ignore_DB:
Replicate_Do_Table:
Replicate_Ignore_Table:
Replicate_Wild_Do_Table:
Replicate_Wild_Ignore_Table:
Last_Errno: 1062
Last_Symbolic_Errno: regular sql errno
Last_Error: Coordinator stopped because there were error(s) in the worker(s). The most recent failure being: Worker 1 failed executing transaction 'ea7ca911-b823-11ef-9485-00163e0a62a9:12' at master log binlog.000001, end_log_pos 3562; Could not execute Write_rows event on table test.large_data; Duplicate entry '1' for key 'large_data.PRIMARY', Error_code: 1062; handler error HA_ERR_FOUND_DUPP_KEY; the event's master log binlog.000001, end_log_pos 3562; See error log and/or performance_schema.replication_applier_status_by_worker table for more details about this failure or others, if any.
Here is my.cnf for two servers:
[client]
default-character-set=utf8mb4
socket=/var/lib/mysql/mysql.sock
[mysql]
default-character-set=utf8mb4
[mysqldump]
default-character-set=utf8mb4
[mysqld]
# server-id should be not same for them
server-id=167968889
transaction_isolation=Repeatable-read
user=mysql
datadir=/var/lib/mysql
plugin_dir=/usr/local/mysql/lib/plugin
pid-file=/tmp/mysqld.pid
socket=/var/lib/mysql/mysql.sock
mysqlx_socket=/var/lib/mysql/mysqlx.sock
default-storage-engine=rocksdb
plugin-load-add=mysql_clone.so
character-set-server=utf8mb4
binlog-format=ROW
log_timestamps=SYSTEM
skip-name-resolve
disabled_storage_engines=myisam
bind-address=0.0.0.0
gtid_mode=ON
enforce_gtid_consistency=ON
#replicate-do-db
binlog-ignore-db = sys,performance_schema,information_schema