It is possible that we are encountering the OOM issue as the syslog messages had: Out of memory: Kill process 28080 (mysqld) score 980 or sacrifice child Killed process 28080, UID 498, (mysqld) total-vm:34727848kB, anon-rss:24101248kB, file-rss:1244kB system currently has 24Gb of RAM and 2 CPU. I am posting the values of my.cnf below, may be you will be able to pinpoint something that may trigger this issue. ***** [client] user = root port = 3306 socket = /var/lib/mysql/mysql.sock [mysqld_safe] user = mysql pid-file = /var/lib/mysql/mysqld.pid socket = /var/lib/mysql/mysql.sock port = 3306 basedir = /usr datadir = /var/lib/mysql expire_logs_days = 3 general_log = 0 general_log_file = /var/log/mysql/mysql_general.log log-error = /var/log/mysql/mysql_error.log log-bin = /var/lib/mysql/mysql_bin log_slave_updates skip_slave_start slow_query_log = 1 slow_query_log_file = /var/log/mysql/mysql_slow.log tmpdir = /tmp server-id = xx [mysqld] user = mysql pid-file = /var/lib/mysql/mysqld.pid socket = /var/lib/mysql/mysql.sock port = 3306 basedir = /usr datadir = /var/lib/mysql expire_logs_days = 3 general_log = 0 general_log_file = /var/log/mysql/mysql_general.log log-error = /var/log/mysql/mysql_error.log log-bin = /var/lib/mysql/mysql_bin slow_query_log = 1 slow_query_log_file = /var/log/mysql/mysql_slow.log tmpdir = /tmp server-id = xx binlog_format = ROW # added binlog_row_image = MINIMAL to assist with the rbr write fail - to process large amount of rows binlog_row_image = MINIMAL default_storage_engine = InnoDB enforce_storage_engine = InnoDB back_log = 128 key_buffer_size = 512M join_buffer_size = 8M long_query_time = 2 # log-queries-not-using-indexes max_allowed_packet = 64M max_connections = 360 max_heap_table_size = 32M net_read_timeout = 30 net_write_timeout = 30 query_cache_limit = 2M query_cache_size = 128M sort_buffer_size = 8M table_open_cache = 512 thread_cache_size = 128 thread_stack = 256K wait_timeout = 28800 explicit_defaults_for_timestamp # innodb_force_recovery = 3 innodb_buffer_pool_size = 8G innodb_buffer_pool_instances = 2 innodb_file_per_table innodb_flush_log_at_trx_commit = 1 innodb_lock_wait_timeout = 120 innodb_log_buffer_size = 8M innodb_log_file_size = 256M innodb_log_files_in_group = 3 innodb_max_dirty_pages_pct = 90 innodb_open_files = 1024 innodb_status_file innodb_thread_concurrency = 32 # innodb_locks_unsafe_for_binlog = 1 innodb_autoinc_lock_mode = 2 # wsrep_cluster_address = gcomm:// wsrep_cluster_address = gcomm://IP1,IP2,IP3 wsrep_cluster_name = QCLUSTER wsrep_debug = ON wsrep_node_address = xxx wsrep_node_incoming_address = xxx wsrep_node_name = QCLUST02 wsrep_on = ON wsrep_provider = /usr/lib64/libgalera_smm.so wsrep_replicate_myisam = 1 wsrep_slave_threads = 1 wsrep_sst_method = rsync wsrep_sst_auth = root:password # convert locking sessions into transactions wsrep_convert_LOCK_to_trx=0