Preparing for TO isolation and cluster's locking.
Affects | Status | Importance | Assigned to | Milestone | ||
---|---|---|---|---|---|---|
MySQL patches by Codership | Status tracked in 5.6 | |||||
5.5 |
New
|
Undecided
|
Unassigned | |||
5.6 |
New
|
Undecided
|
Unassigned | |||
Percona XtraDB Cluster moved to https://jira.percona.com/projects/PXC | Status tracked in 5.6 | |||||
5.5 |
Confirmed
|
High
|
Unassigned | |||
5.6 |
Confirmed
|
High
|
Unassigned |
Bug Description
Hello, recently I had some weird issues with xtradb cluster.
Sometimes cluster just get locked (all new connections are putted in queue, so --max connections is very rapidly reached, the only way to fix it is to reboot affected nodes.
Some screenshots from phpmyadmin from recent issue:
Node A:
https:/
Node B:
https:/
My Environment:
3 physical Nodes on debian 7 running percona-
Nodes: configuration
[client]
default-
# CLIENT #
port = 3306
socket = /var/run/
default-
[mysqld]
#character
character-
character-
collation-server = utf8mb4_unicode_ci
server_id=1
# GENERAL #
user = mysql
default_
# BINARY LOGGING #
log_bin = /var/lib/
expire_logs_days = 5
binlog_format=ROW
#log_bin=mysql-bin
datadir=
wsrep_cluster_
wsrep_provider=
wsrep_slave_
wsrep_cluster_
wsrep_sst_
wsrep_node_name=sun
wsrep_sst_
wsrep_replicate
wsrep_notify_
wsrep_sst_
# Provider specific configuration options
wsrep_provider_
log_slave_updates
# SAFETY #
max_allowed_packet = 16M
max_connect_errors = 1000000
skip_name_resolve
sysdate_is_now = 1
innodb = FORCE
innodb_strict_mode = 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_definitio
table_open_cache = 10000
# INNODB #
innodb_flush_method = O_DIRECT
innodb_
innodb_
innodb_
innodb_
innodb_
innodb_
innodb_
# LOGGING #
log_error = /var/log/
#log_queries_
slow_query_log = 1
slow_query_log_file = /var/log/
[sst]
#streamfmt=tar
Any suggestions? Thanks.
tags: | added: i53862 |
@Alekc,
Do you have any DDL running during this? If yes, then it may be related to https:/ /bugs.launchpad .net/percona- xtradb- cluster/ +bug/1212955