Deadlock when dropping and creating temporary table on Percona XtraDB Cluster 5.5.33

MySQL & MariaDB Percona XtraDB Cluster 5.x
1

Hi all, I have already posted this on launchpad as a bug and I have just found this forum, so sorry for cross posting. Hope someone can help:

I have a cluster of 3 nodes all running Percona-XtraDB-Cluster-server.x86_64 1:5.5.33-23.7.6.495.rhel6.
Machine info:

# uname -a
Linux 3.8.7joyent-centos-6-opt #2 SMP Wed Apr 17 15:18:57 PDT 2013 x86_64 x86_64 x86_64 GNU/Linux


USE test;

DROP TABLE IF EXISTS test_bug_source;
CREATE TABLE test_bug_source(`id` int);

START TRANSACTION;
DROP TEMPORARY TABLE IF EXISTS test_bug;
CREATE TEMPORARY TABLE test_bug AS
SELECT * FROM test_bug_source;
COMMIT;

[QUOTE]

This scripts works fine on a previous version (Percona XtraDB Cluster 5.5.31).

I did also this test:

  1. create a 3 nodes cluster with Percona XtraDB Cluster 5.5.31.
  2. run the script, no errors.
  3. upgrade all the nodes to Percona XtraDB Cluster 5.5.33
  4. run the same script, Deadlock error.

# CLIENT #
port = 3306
socket = /data/mysql/mysql.sock
default-character-set=utf8

socket = /data/mysql/mysql.sock
default-character-set=utf8

# GENERAL #
user = mysql
default_storage_engine = InnoDB
socket = /data/mysql/mysql.sock
pid_file = /data/mysql/mysql.pid
# MyISAM #
key_buffer_size = 32M
myisam_recover = FORCE,BACKUP
# SAFETY #
max_allowed_packet = 16M
max_connect_errors = 1000000
skip_name_resolve
sql_mode = STRICT_TRANS_TABLES,ERROR_FOR_DIVISION_BY_ZERO,NO_ AUTO_CREATE_USER,NO_AUTO_VALUE_ON_ZERO,NO_ENGINE_S UBSTITUTION,NO_ZERO_DATE,NO_ZERO_IN_DATE,ONLY_FULL _GROUP_BY
sysdate_is_now = 1
innodb = FORCE
innodb_strict_mode = 1
# DATA STORAGE #
datadir = /data/mysql/
# BINARY LOGGING #
log_bin = /data/mysql/mysql-bin
expire_logs_days = 14
sync_binlog = 1
# CACHES AND LIMITS #
tmp_table_size = 32M
max_heap_table_size = 32M
query_cache_type = 0
query_cache_size = 0
max_connections = 500
thread_cache_size = 50
open_files_limit = 65535
table_definition_cache = 1024
table_open_cache = 100
# INNODB #
innodb_flush_method = O_DIRECT
innodb_log_files_in_group = 2
innodb_log_file_size = 64M
innodb_flush_log_at_trx_commit = 1
innodb_file_per_table = 1
innodb_buffer_pool_size = 512M
# LOGGING #
log_error = /data/mysql/mysql-error.log
log_queries_not_using_indexes = 1
slow_query_log = 1
slow_query_log_file = /data/mysql/mysql-slow.log
# UTF8 #
collation-server = utf8_unicode_ci
init-connect='SET NAMES utf8'
character-set-server = utf8
#### Cluster config ####
# Path to Galera library
wsrep_provider = /usr/lib64/libgalera_smm.so
# Cluster connection URL contains the IPs of all the nodes
wsrep_cluster_address = gcomm://<cluster_ip_addresses>
# In order for Galera to work correctly binlog format should be ROW
binlog_format = ROW
# This is a recommended tuning variable for performance
innodb_locks_unsafe_for_binlog = 1
# This changes how InnoDB auto-increment locks are managed and is a requirement for Galera
innodb_autoinc_lock_mode = 2
# IP address of this node
wsrep_node_address = <node_ip_address>
# SST method
wsrep_sst_method = xtrabackup
# Cluster name
wsrep_cluster_name = data_cluster
# Authentication for SST method
wsrep_sst_auth = "<user>:<password>"
2

Bug URL:
[url]https://bugs.launchpad.net/percona-xtradb-cluster/+bug/1240098[/url]