MySQL 5.7.11 broken replication after just service mysql restart on slave

Asked by Eugene Klimov on 2016-03-27

i try upgrade to Percona Server 5.7.11-4 over Percona-Server 5.6 on my Ubuntu 14.04

mysq upgrade steps

# on master and slave both
service mysql stop

apt-key adv --keyserver keys.gnupg.net --recv-keys 1C4CBDCDCD2EFD2A
echo "deb http://repo.percona.com/apt trusty main" > /etc/apt/sources.list.d/percona.list

apt-key adv --keyserver keys.gnupg.net --recv-keys 8C718D3B5072E1F5
echo "deb http://repo.mysql.com/apt/ubuntu/ trusty mysql-utilities-1.5" > /etc/apt/sources.list.d/mysql-utils.list
echo "deb http://repo.mysql.com/apt/ubuntu/ trusty connector-python-2.1" >> /etc/apt/sources.list.d/mysql-utils.list

apt-get update
apt-get install -y mysql-utilities
apt-get install -y percona-xtrabackup-24 percona-server-server-5.7 percona-server-client-5.7 percona-toolkit

#after fix some problems with apparmor and file limits
service mysql start
mysql_upgrade -u root -p --force
service mysql restart

#dump on master
mysqldump --master-data --databases db1,db2,db3 > master.sql
#restore restore to slave server
mysql --defaults-file=/etc/mysql/debian.cnf < master.sql

and on master all working fun
slave have forking fun

         Master_Log_File: mysql-bin.003034
     Read_Master_Log_Pos: 82962819
          Relay_Log_File: repl16-relay-bin.000018
           Relay_Log_Pos: 23081183
   Relay_Master_Log_File: mysql-bin.003033
        Slave_IO_Running: Yes
       Slave_SQL_Running: Yes
Seconds_Behind_Master: 8476

after this i just run

service mysql restart

and i have stop my replictation proccess

                   Last_Errno: 1032
                   Last_Error: Could not execute Update_rows event on table mbcandyvalley.statistics; Can't find record in 'statistics', Error_code: 1032; handler error HA_ERR_KEY_NOT_FOUND; the event
's master log mysql-bin.003033, end_log_pos 25161015

why this happend? what's i wrong doing? how

Question information

Language:
English Edit question
Status:
Solved
For:
Percona Server moved to https://jira.percona.com/projects/PS Edit question
Assignee:
No assignee Edit question
Solved by:
Eugene Klimov
Solved:
2016-03-27
Last query:
2016-03-27
Last reply:

This question was reopened

Eugene Klimov (bloodjazman) said : #1

error was happened because i have Engine=MEMORY tables in my database

after change
binlog_format=STATEMENT (it's was default value on 5.6)
all working fine ;)