一、mysql服务检查1、安全启动mysql服务/apps/svr/mysql/bin/mysqld_safe --defaults-file/apps/conf/mysql/my.cnf--usermysql2、进入从库停止主从设置跳过轮次STOP SLAVE;SETGLOBALsql_slave_skip_counter1;STARTSLAVE;3、查看失败原因 找到APPLYING_TRANSACTION的UUIDSELECT*FROMperformance_schema.replication_applier_status_by_worker;4、 然后设置跳过SETsession.gtid_nextc7d01d3b-6ecf-11f1-a9a4-fa163ec11bb2:43;COMMIT;SETsession.gtid_nextAUTOMATIC;5、在启动主从同步查看主从状态STARTSLAVE;SHOWSLAVESTATUS\G6、如果SQL同步还是NO 再看看这个表 循环往复处理 如果真是大量数据问题则跳到第七步骤SELECT*FROMperformance_schema.replication_applier_status_by_worker7、数据不重要的话直接清理对应冲突的表7.1先关闭SLAVESTOP SLAVE;7.2然后清理有问题的数据的表TRUNCATETABLExxdb.xxtable;##在开启同步STARTSLAVE;##在查看主从状态SHOWSLAVESTATUS;二、启动manager服务参考隔壁节点启动参数比如隔壁的ps里显示manager --passEncryptedtrue --etcd_urlhttp://{etcd的ip}:{etcd端口} --monitor_dbtrue --db_host --db_password****** --Unix_sockettrue --log_file_pattern/apps/logs/manager/manager.log.%Y%m%d%H%M --manager2URLhttp://{当前节点ip}:10020 --Unix_socket_path/apps/run/mysql/mysql.sock --mysql_relay_location/apps/data/mysql --db_usermonitor1、在当前节点配置软连接不然manager启动报错127错误ln-s/apps/svr/mysql/bin/mysqlbinlog /usr/local/bin/mysqlbinlog2、创建启动manager文件#!/bin/bash# source存了下面变量的参数不对可直接修改比如ISETCDEMEBED得修改为falsesource/apps/conf/manager/profile#export ISETCDEMBED${ISETCDEMBED}exportETCD_CONF_LOCATION${ETCD_CONF_LOCATION}exportISETCDEMBEDfalseexportETCD_URL${ETCD_URL}#export ETCD_URL${ETCD_URL}exportDB_HOST${DB_HOST}exportDB_USER${DB_USER}exportMANAGER2URL${MANAGER2URL}#####exportSQL_WORK_DIR${SQL_WORK_DIR}exportMYSQL_RELAY_LOCATION${MYSQL_RELAY_LOCATION}exportUNIX_SOCKET${UNIX_SOCKET}exportUNIX_SOCKET_PATH${UNIX_SOCKET_PATH}exportLOG_FILE_PATTERN${LOG_FILE_PATTERN}##### 先清理端口10020旧进程PID$(ss-tlp|grep:10020|awk-F{print $4})if[-n$PID];thenkill$PIDsleep1PID2$(ss-tlp|grep:10020|awk-F{print $4})if[-n$PID2];thenkill-9$PID2fifi# 入参说明优先读取环境变量个别和标准不一致的直接在命令行固定覆盖不再改动profilenohupmanager\--passEncryptedtrue\--etcd_url${ETCD_URL}\--monitor_dbtrue\--db_host\--db_usermonitor\--Unix_sockettrue\--Unix_socket_path/apps/run/mysql/mysql.sock\--mysql_relay_location/apps/data/mysql\--log_file_pattern/apps/logs/manager/manager.log.%Y%m%d%H%M\--manager2URL${${MANAGER2URL}}\#有问题的手动替换/dev/null213、然后造/apps/svr/manager下面执行启动sh/apps/sh/xxx.sh4、查看启动日志tail-f/apps/logs/manager/manager.log.xxxxxxxx
一、mysql服务检查1、安全启动mysql服务/apps/svr/mysql/bin/mysqld_safe --defaults-file/apps/conf/mysql/my.cnf--usermysql2、进入从库停止主从设置跳过轮次STOP SLAVE;SETGLOBALsql_slave_skip_counter1;STARTSLAVE;3、查看失败原因 找到APPLYING_TRANSACTION的UUIDSELECT*FROMperformance_schema.replication_applier_status_by_worker;4、 然后设置跳过SETsession.gtid_nextc7d01d3b-6ecf-11f1-a9a4-fa163ec11bb2:43;COMMIT;SETsession.gtid_nextAUTOMATIC;5、在启动主从同步查看主从状态STARTSLAVE;SHOWSLAVESTATUS\G6、如果SQL同步还是NO 再看看这个表 循环往复处理 如果真是大量数据问题则跳到第七步骤SELECT*FROMperformance_schema.replication_applier_status_by_worker7、数据不重要的话直接清理对应冲突的表7.1先关闭SLAVESTOP SLAVE;7.2然后清理有问题的数据的表TRUNCATETABLExxdb.xxtable;##在开启同步STARTSLAVE;##在查看主从状态SHOWSLAVESTATUS;二、启动manager服务参考隔壁节点启动参数比如隔壁的ps里显示manager --passEncryptedtrue --etcd_urlhttp://{etcd的ip}:{etcd端口} --monitor_dbtrue --db_host --db_password****** --Unix_sockettrue --log_file_pattern/apps/logs/manager/manager.log.%Y%m%d%H%M --manager2URLhttp://{当前节点ip}:10020 --Unix_socket_path/apps/run/mysql/mysql.sock --mysql_relay_location/apps/data/mysql --db_usermonitor1、在当前节点配置软连接不然manager启动报错127错误ln-s/apps/svr/mysql/bin/mysqlbinlog /usr/local/bin/mysqlbinlog2、创建启动manager文件#!/bin/bash# source存了下面变量的参数不对可直接修改比如ISETCDEMEBED得修改为falsesource/apps/conf/manager/profile#export ISETCDEMBED${ISETCDEMBED}exportETCD_CONF_LOCATION${ETCD_CONF_LOCATION}exportISETCDEMBEDfalseexportETCD_URL${ETCD_URL}#export ETCD_URL${ETCD_URL}exportDB_HOST${DB_HOST}exportDB_USER${DB_USER}exportMANAGER2URL${MANAGER2URL}#####exportSQL_WORK_DIR${SQL_WORK_DIR}exportMYSQL_RELAY_LOCATION${MYSQL_RELAY_LOCATION}exportUNIX_SOCKET${UNIX_SOCKET}exportUNIX_SOCKET_PATH${UNIX_SOCKET_PATH}exportLOG_FILE_PATTERN${LOG_FILE_PATTERN}##### 先清理端口10020旧进程PID$(ss-tlp|grep:10020|awk-F{print $4})if[-n$PID];thenkill$PIDsleep1PID2$(ss-tlp|grep:10020|awk-F{print $4})if[-n$PID2];thenkill-9$PID2fifi# 入参说明优先读取环境变量个别和标准不一致的直接在命令行固定覆盖不再改动profilenohupmanager\--passEncryptedtrue\--etcd_url${ETCD_URL}\--monitor_dbtrue\--db_host\--db_usermonitor\--Unix_sockettrue\--Unix_socket_path/apps/run/mysql/mysql.sock\--mysql_relay_location/apps/data/mysql\--log_file_pattern/apps/logs/manager/manager.log.%Y%m%d%H%M\--manager2URL${${MANAGER2URL}}\#有问题的手动替换/dev/null213、然后造/apps/svr/manager下面执行启动sh/apps/sh/xxx.sh4、查看启动日志tail-f/apps/logs/manager/manager.log.xxxxxxxx
)
