pt-tables-sync getting "MySQL server has gone away" error

Hello all,

While performing pt-tables-sync tables ON REPLICATION MASTER we are frequently getting “MySQL server has gone away”

====
/percona-toolkit src_db:exateam_exaprojects src_tbl:tasks src_dsn:D=exateam_exaprojects,h=10.1.1.3,p=…,t=tasks,u=exateam_admin dst_db:exateam_exaprojects dst_tbl:tasks dst_dsn:D=exateam_exaprojects,h=10.1.1.19,p=…,t=tasks,u=exateam_admin lock:0 transaction:1 changing_src:0 replicate:0 bidirectional:0 pid:27279 user:root host:crm.modeldns.com.au/"] at line 10695 while doing exateam_exaprojects.tasks on 10.1.1.19
DBD::mysql::db commit failed: MySQL server has gone away at /usr/local/bin/pt-table-sync line 11002.
Issuing rollback() due to DESTROY without explicit disconnect() of DBD::mysql::db handle exateam_exaprojects;host=10.1.1.3;mysql_read_default_group=client at /usr/local/bin/pt-table-sync line 11002.
Issuing rollback() due to DESTROY without explicit disconnect() of DBD::mysql::db handle exateam_exaprojects;host=10.1.1.3;mysql_read_default_group=client at /usr/local/bin/pt-table-sync line 11002.
Issuing rollback() due to DESTROY without explicit disconnect() of DBD::mysql::db handle exateam_exaprojects;host=10.1.1.19;mysql_read_default_group=client at /usr/local/bin/pt-table-sync line 11002.
root@crm [~]#

Command use to manuall sync
root@crm [~]#pt-table-sync u=exateam_admin,p=‘XX’,h=XXmaster ,D=$DB,t=history 10.1.1.18 --execute --verbose --print

Server details as below
-Version
root@crm [~]# pt-table-sync --version
pt-table-sync 2.2.4
root@crm [~]#

-Slave ip is 10.1.1.18

-Copy of my.cnf

[mysqld]
datadir=/var/lib/mysql
server-id = 1
max_connections=250
max_allowed_packet = 256M
max_connect_errors = 1000000

CACHES AND LIMITS

tmp_table_size = 32M
max_heap_table_size = 32M
query_cache_type = 1
query_cache_size = 1024M
query_cache_limit = 64M
key_buffer_size = 256M
max_connections = 250
thread_cache_size = 50
open_files_limit = 65535
table_definition_cache = 4096
table_open_cache = 4096
join_buffer_size = 3M

skip_name_resolve

INNODB

innodb_buffer_pool_size = 20G
innodb_log_file_size = 2G
innodb_lock_wait_timeout = 120

LOGGING

log_error = /var/log/mysql/error.log
#log_queries_not_using_indexes = 1
slow_query_log = 1
long_query_time = 20
slow_query_log_file = /var/log/mysql/slowqueriesnew.log

#TIMEOUTS#

wait_timeout=60
interactive_timeout=30
connect_timeout=30

binlog-do-db=exateam_exaprojects
binlog-do-db=exateam_timesheet
binlog-do-db=exateam_ams
binlog-do-db=exateam_attendance
binlog-do-db=exateam_support
binlog-do-db=project_tracker

#log-bin=/var/lib/mysql/binlog
log-bin=/mysqlbinlog/binlog

Can someone comment/suggest for fix ?

-Regards
Kunal Modi

Can anyone help me over here ?

-Regards
Kunal Modi

Hi,

Can you provide which MySQL/PS version you are using? Also provide [COLOR=#252C2F]/var/log/mysql/error.log it will give some more information about server crash. Thanks.

-Mysql version is ‘Server version: 5.5.32-31.0-log Percona Server (GPL)’
-I have checked error logs for 24Sept but there are no errors mentioned.

-Regards
Kunal Modi

Hi,

Without error logs, it’s quite difficult to identify the problem. Can you try to run pt-table-sync again and check if you are getting anything in error log? Can you also provide the steps that how we can reproduce this problem? Thanks.