|
I have master (192.168.56.118) and slave (192.168.56.119)
Master my.cnf:
[mysqld]
|
|
character_set_server=utf8mb4
|
collation_server=utf8mb4_unicode_ci
|
innodb_file_format=Barracuda
|
innodb_large_prefix=1
|
innodb_file_per_table=1
|
lower_case_table_names=1
|
|
port = 3306
|
|
log_bin = mariadb-bin
|
|
server_id = 1
|
|
gtid_strict_mode = 1
|
gtid_domain_id = 1
|
|
[mysqld_safe]
|
log-error=/var/log/mariadb/mariadb.log
|
pid-file=/var/run/mariadb/mariadb.pid
|
|
|
[client-server]
|
|
|
!includedir /etc/my.cnf.d
|
Slave my.cnf:
[mysqld]
|
|
character_set_server=utf8mb4
|
collation_server=utf8mb4_unicode_ci
|
innodb_file_format=Barracuda
|
innodb_large_prefix=1
|
innodb_file_per_table=1
|
lower_case_table_names=1
|
|
port = 3306
|
|
log_bin = mariadb-bin
|
|
server_id = 2
|
gtid_strict_mode = 1
|
gtid_domain_id = 1
|
|
[mysqld_safe]
|
log-error=/var/log/mariadb/mariadb.log
|
pid-file=/var/run/mariadb/mariadb.pid
|
|
|
[client-server]
|
|
|
!includedir /etc/my.cnf.d
|
Replication works correctly, on slave ststus command has no errors:
MariaDB [(none)]> show slave status\G
|
*************************** 1. row ***************************
|
Slave_IO_State: Waiting for master to send event
|
Master_Host: 192.168.56.118
|
Master_User: repl
|
Master_Port: 3306
|
Connect_Retry: 60
|
Master_Log_File: mariadb-bin.000001
|
Read_Master_Log_Pos: 3630
|
Relay_Log_File: pg-exercises-relay-bin.000002
|
Relay_Log_Pos: 3544
|
Relay_Master_Log_File: mariadb-bin.000001
|
Slave_IO_Running: Yes
|
Slave_SQL_Running: Yes
|
Replicate_Do_DB:
|
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: 3630
|
Relay_Log_Space: 3860
|
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: 1
|
Master_SSL_Crl:
|
Master_SSL_Crlpath:
|
Using_Gtid: Slave_Pos
|
Gtid_IO_Pos: 1-1-22
|
Replicate_Do_Domain_Ids:
|
Replicate_Ignore_Domain_Ids:
|
Parallel_Mode: optimistic
|
SQL_Delay: 0
|
SQL_Remaining_Delay: NULL
|
Slave_SQL_Running_State: Slave has read all relay log; waiting for more updates
|
Slave_DDL_Groups: 18
|
Slave_Non_Transactional_Groups: 1
|
Slave_Transactional_Groups: 0
|
1 row in set (0.000 sec)
|
Also I have maxscale, maxscale.cnf:
[maxscale]
|
threads=auto
|
|
|
[server1]
|
type=server
|
address=192.168.56.118
|
port=3306
|
protocol=MariaDBBackend
|
|
|
[server2]
|
type=server
|
address=192.168.56.119
|
port=3306
|
protocol=MariaDBBackend
|
|
[MariaDB-Monitor]
|
type=monitor
|
module=mariadbmon
|
servers=server1,server2
|
user=maxscale
|
password=maxscale
|
auto_failover=true
|
auto_rejoin=true
|
monitor_interval=100
|
replication_user=repl
|
replication_password=lper
|
|
[Read-Write-Service]
|
type=service
|
router=readwritesplit
|
servers=server1,server2
|
user=maxscale
|
password=maxscale
|
|
|
[Read-Write-Listener]
|
type=listener
|
service=Read-Write-Service
|
protocol=MariaDBClient
|
port=4006
|
Firsty it works without problems:
maxctrl list servers
┌─────────┬────────────────┬──────┬─────────────┬─────────────────┬────────┐
│ Server │ Address │ Port │ Connections │ State │ GTID │
├─────────┼────────────────┼──────┼─────────────┼─────────────────┼────────┤
│ server1 │ 192.168.56.118 │ 3306 │ 0 │ Master, Running │ 1-1-22 │
├─────────┼────────────────┼──────┼─────────────┼─────────────────┼────────┤
│ server2 │ 192.168.56.119 │ 3306 │ 0 │ Slave, Running │ 1-1-22 │
└─────────┴────────────────┴──────┴─────────────┴─────────────────┴────────┘
When master fails, slave bocomes a master:
maxctrl list servers
┌─────────┬────────────────┬──────┬─────────────┬─────────────────┬────────┐
│ Server │ Address │ Port │ Connections │ State │ GTID │
├─────────┼────────────────┼──────┼─────────────┼─────────────────┼────────┤
│ server1 │ 192.168.56.118 │ 3306 │ 0 │ Down │ 1-1-22 │
├─────────┼────────────────┼──────┼─────────────┼─────────────────┼────────┤
│ server2 │ 192.168.56.119 │ 3306 │ 0 │ Master, Running │ 1-1-22 │
└─────────┴────────────────┴──────┴─────────────┴─────────────────┴────────┘
But, when I start master again I have such situation:
maxctrl list servers
┌─────────┬────────────────┬──────┬─────────────┬─────────────────┬────────┐
│ Server │ Address │ Port │ Connections │ State │ GTID │
├─────────┼────────────────┼──────┼─────────────┼─────────────────┼────────┤
│ server1 │ 192.168.56.118 │ 3306 │ 0 │ Running │ 1-1-22 │
├─────────┼────────────────┼──────┼─────────────┼─────────────────┼────────┤
│ server2 │ 192.168.56.119 │ 3306 │ 0 │ Master, Running │ 1-1-22 │
└─────────┴────────────────┴──────┴─────────────┴─────────────────┴────────┘
And I found error in the maxscale log:
2022-02-08 12:22:05 warning: [mariadbmon] Automatic rejoin was not attempted on server 'server1' even though it is a valid candidate. Will keep retrying with this message suppressed for all servers. Errors:
gtid_current_pos of 'server1' (1-1-22) is incompatible with gtid_binlog_pos of 'server2' (1-2-8).
So, server1 previously was a master, so, I thought that it's ok that his global.gtid_binlog_pos is higher.
Perhaps I'm doing something wrong? Or is this the standard behavior?
|