mairadb galera cluster 恢复
node 加入集群报错:[ERROR] WSREP: gcs/src/gcs_group.cpp:gcs_group_handle_join_msg():736: Will never receive state. Need to abort.
解决
vim /etc/szqs/szqs.cnf
memlock
将此项注掉
#memlock
cluster nodes
node1 172.16.16.8
node2 172.16.16.88
node3 172.16.16.188
查看cluster status
mysql > show status like 'wsreq%';
启动node1
vim /etc/mysql/galera_node.cnf
##将此项设置成:
wsrep-cluster-address = gcomm://
/etc/init.d/mysql start --wsrep-new-cluster
启动node2
vim /etc/mysql/galera_node.cnf
##将此项设置成:
wsrep-cluster-address = gcomm://172.16.16.8
/etc/init.d/mysql start
启动node3
vim /etc/mysql/galera_node.cnf
##将此项设置成:
wsrep-cluster-address = gcomm://172.16.16.8,172.16.16.88
/etc/init.d/mysql start
重启node1
vim /etc/mysql/galera_node.cnf
##将此项设置成:
wsrep-cluster-address = gcomm://172.16.16.88,172.16.16.188
/etc/init.d/mysql start
