我们有两个服务器server1和server2。 Server1上有两个数据库DB01和DB02，而Server2上有四个数据库DB01，DB02，DB03，DB04。 SERVER 1和SERVER被配置为主 - 主为DB01＆DB02。
我添加了复制所有四个数据库的从服务器2的第三服务器。到目前为止，服务器1和2之间的主-主以及服务器2和3之间的主-从都很好-
如果我现在配置服务器2服务器3作为实现服务器2和3，复制符之间主 - 主与重复输入键错误服务器2的从设备。
（未使用GTID配置复制。）
writes to writes to
db01, db02 db03, db04 No writes
+---------+ +---------+ +---------+
| server1 | <-----mm-----> | server2 | >---slave---> | server3 |
| ------- | | ------- | | ------- |
| db01 -m | | db01 -s | | db01 |
| db02 -m | | db02 -s | | db02 |
+---------+ | db03 -m | | db03 |
auto_increment_increment=3 | db04 -m | | db04 |
auto_increment_offset=2 +---------+ +---------+
log_slave_updates=1 replicate-do-db=db01 replicate-do-db=db01
replicate-do-db=db01 replicate-do-db=db02 replicate-do-db=db02
replicate-do-db=db02 auto_increment_increment=3 replicate-do-db=db03
auto_increment_offset=1 replicate-do-db=db04
log_slave_updates=1 auto_increment_increment=3
replicate-do-db=db03 auto_increment_offset=3
replicate-do-db=db04 log_slave_updates=1
If I now configure server2 as slave of server3 to achieve master-master between
server2 and 3, replication breaks on server2 with duplicate entry key error.
+---------+ +---------+ +---------+
| server1 | <------mm-----> | server2 | >--slave--> ok | server3 |
| ------- | | ------- | | ------- |
| db01 | | db01 | <--slave--< x | db01 |
| db02 | | db02 | | db02 |
+---------+ | db03 | | db03 |
| db04 | | db04 |
+---------+ +---------+
This is the error on server2:
MariaDB [(none)]> show all slaves status \G
*************************** 1. row ***************************
Connection_name: mm-01-to-04b
Slave_SQL_State:
Slave_IO_State: Waiting for master to send event
Master_Host: server3
Master_User: replicator
Master_Port: 3306
Connect_Retry: 60
Master_Log_File: mysql-bin.000099
Read_Master_Log_Pos: 70417544
Relay_Log_File: mysql-relay-bin-mm@002d01@002dto@002d04b.000002
Relay_Log_Pos: 2021
Relay_Master_Log_File: mysql-bin.000099
Slave_IO_Running: Yes
Slave_SQL_Running: No
Replicate_Do_DB: db03,db04,db01,db02
Replicate_Ignore_DB:
Replicate_Do_Table:
Replicate_Ignore_Table:
Replicate_Wild_Do_Table:
Replicate_Wild_Ignore_Table:
Last_Errno: 1062
Last_Error: Error 'Duplicate entry '1758321014' for key 'PRIMARY'' on query. Default database: 'db01'. Query: 'INSERT INTO item SET itemClass_DbId = 12058, user_DbId = 45392036, view_DbId = 15'
Skip_Counter: 0
Exec_Master_Log_Pos: 64342511
Relay_Log_Space: 3288379
Until_Condition: None
Until_Log_File:
Until_Log_Pos: 0
Master_SSL_Allowed: No
Master_SSL_CA_File:
Master_SSL_CA_Path:
Master_SSL_Cert:
Master_SSL_Cipher:
Master_SSL_Key:
Seconds_Behind_Master: NULL
Master_SSL_Verify_Server_Cert: No
Last_IO_Errno: 0
Last_IO_Error:
Last_SQL_Errno: 1062
Last_SQL_Error: Error 'Duplicate entry '1758321014' for key 'PRIMARY'' on query. Default database: 'db01'. Query: 'INSERT INTO item SET itemClass_DbId = 12058, user_DbId = 45392036, view_DbId = 15'
Replicate_Ignore_Server_Ids:
Master_Server_Id: 10214311
Master_SSL_Crl:
Master_SSL_Crlpath:
Using_Gtid: No
Gtid_IO_Pos:
Replicate_Do_Domain_Ids:
Replicate_Ignore_Domain_Ids:
Parallel_Mode: conservative
SQL_Delay: 0
SQL_Remaining_Delay: NULL
Slave_SQL_Running_State:
Retried_transactions: 0
Max_relay_log_size: 1073741824
Executed_log_entries: 38
Slave_received_heartbeats: 1
Slave_heartbeat_period: 30.000
Gtid_Slave_Pos: 0-10214212-1893552590
*************************** 2. row ***************************
Connection_name: dbs-01-02
Slave_SQL_State: Slave has read all relay log; waiting for the slave I/O thread to update it
Slave_IO_State: Waiting for master to send event
Master_Host: server1
Master_User: replicator
Master_Port: 3306
Connect_Retry: 60
Master_Log_File: mysql-bin.000586
Read_Master_Log_Pos: 657757085
Relay_Log_File: mysql-relay-bin-dbs@002d01@002d02.000202
Relay_Log_Pos: 417827599
Relay_Master_Log_File: mysql-bin.000586
Slave_IO_Running: Yes
Slave_SQL_Running: Yes
Replicate_Do_DB: db03,db04,db01,db02
Replicate_Ignore_DB:
Replicate_Do_Table:
Replicate_Ignore_Table:
Replicate_Wild_Do_Table:
Replicate_Wild_Ignore_Table:
Last_Errno: 0
Last_Error:
Skip_Counter: 0
Exec_Master_Log_Pos: 657757085
Relay_Log_Space: 417827929
Until_Condition: None
Until_Log_File:
Until_Log_Pos: 0
Master_SSL_Allowed: No
Master_SSL_CA_File:
Master_SSL_CA_Path:
Master_SSL_Cert:
Master_SSL_Cipher:
Master_SSL_Key:
Seconds_Behind_Master: 0
Master_SSL_Verify_Server_Cert: No
Last_IO_Errno: 0
Last_IO_Error:
Last_SQL_Errno: 0
Last_SQL_Error:
Replicate_Ignore_Server_Ids:
Master_Server_Id: 10214211
Master_SSL_Crl:
Master_SSL_Crlpath:
Using_Gtid: No
Gtid_IO_Pos:
Replicate_Do_Domain_Ids:
Replicate_Ignore_Domain_Ids:
Parallel_Mode: conservative
SQL_Delay: 0
SQL_Remaining_Delay: NULL
Slave_SQL_Running_State: Slave has read all relay log; waiting for the slave I/O thread to update it
Retried_transactions: 0
Max_relay_log_size: 1073741824
Executed_log_entries: 3800382
Slave_received_heartbeats: 0
Slave_heartbeat_period: 30.000
Gtid_Slave_Pos: 0-10214212-1893552590
2 rows in set (0.00 sec)
我期望主-主复制可以在server2的任何一侧进行。然而，似乎在server1插入件被复制到服务器2经由服务器3，然后返回到服务器2。
时这种设置的在所有可能的，而无需使用GTID？
仅供参考，所有三个服务器正在运行Linux的亚马逊和201703 10.2.8 MariaDB的
谢谢。