-
Type:
Incident report
-
Resolution: Incomplete
-
Priority:
Minor
-
None
-
Affects Version/s: 2.2.0
-
Component/s: Server (S)
-
Environment:I've got one dedicated zabbix server and a MySQL backend. The MySQL server stores zabbix and other databases, but from far the most used is zabbix. Finally I've got a MySQL slave replicating all the databases in the master MySQL server.
---------------------------------------------------------------------------------------------
Zabbix server:
OS: Ubuntu Server 12.10 (QEMU Virtual Machine) 64 bits
CPU: 2xIntel(R) Pentium(R) CPU G860 @ 3.00GHz (virtualized)
RAM: 512MB
Zabbix: 2.2
---------------------------------------------------------------------------------------------
Zabbix DB (master):
OS: Ubuntu Server 12.10 (QEMU Virtual Machine) 64 bits
CPU: 2xIntel(R) Pentium(R) CPU G860 @ 3.00GHz (virtualized)
RAM: 1024MB
MySQL: 5.5.29-0ubuntu0.12.10.1
---------------------------------------------------------------------------------------------
Slave DB:
OS: Ubuntu Server 12.10 (QEMU Virtual Machine) 64 bits
CPU: 2xIntel(R) Pentium(R) CPU G860 @ 3.00GHz (virtualized)
RAM: 1024MB
MySQL: 5.5.29-0ubuntu0.12.10.1I've got one dedicated zabbix server and a MySQL backend. The MySQL server stores zabbix and other databases, but from far the most used is zabbix. Finally I've got a MySQL slave replicating all the databases in the master MySQL server. --------------------------------------------------------------------------------------------- Zabbix server: OS: Ubuntu Server 12.10 (QEMU Virtual Machine) 64 bits CPU: 2xIntel(R) Pentium(R) CPU G860 @ 3.00GHz (virtualized) RAM: 512MB Zabbix: 2.2 --------------------------------------------------------------------------------------------- Zabbix DB (master): OS: Ubuntu Server 12.10 (QEMU Virtual Machine) 64 bits CPU: 2xIntel(R) Pentium(R) CPU G860 @ 3.00GHz (virtualized) RAM: 1024MB MySQL: 5.5.29-0ubuntu0.12.10.1 --------------------------------------------------------------------------------------------- Slave DB: OS: Ubuntu Server 12.10 (QEMU Virtual Machine) 64 bits CPU: 2xIntel(R) Pentium(R) CPU G860 @ 3.00GHz (virtualized) RAM: 1024MB MySQL: 5.5.29-0ubuntu0.12.10.1
I realized every time that I monitor MySQL using perconna's script that InnoDB row lock values are constantly growing. Example:
---------------------------------------------------------------------------------------------
MySQL master:
~$ mysql -e "SHOW GLOBAL STATUS;" -u root -p | grep -i innodb | grep lock
Enter password:
Innodb_row_lock_current_waits 3
Innodb_row_lock_time 5669312
Innodb_row_lock_time_avg 310
Innodb_row_lock_time_max 13911
Innodb_row_lock_waits 18238
---------------------------------------------------------------------------------------------
MySQL slave:
$ mysql -e "SHOW GLOBAL STATUS;" -u root -p | grep -i innodb | grep lock
Enter password:
Innodb_row_lock_current_waits 0
Innodb_row_lock_time 0
Innodb_row_lock_time_avg 0
Innodb_row_lock_time_max 0
Innodb_row_lock_waits 0
---------------------------------------------------------------------------------------------
I'm on my way to create a dedicated MySQL server to confirm that zabbix is the culprit. If finally I'm right and something is going on with zabbix, I think it's not just related with 2.2 version, it comes from earlier versions as well.
