Since Start of monitoring with zabbix 1.4.2, our DB is growing constantly (see here: http://www.heiwu.de/myfiles/drbddisc.jpg
We really need help now since we only have about 20GB Disk space left...
First my System:
zabbix server V1.4.2
running with
- Fedora Core 6 Linux
- mysql-server-5.0.27-1.fc6 / mysql-5.0.27-1.fc6 / php-mysql-5.1.6-3.6.fc6
- httpd-2.2.4-2.1.fc6 / php-5.1.6-3.6.fc6
on a DELL PE 860 with
- Intel Pentium D 925, 3 GHz, 800MHz FSB
- 2 x 1024 GB DDR SD 667MHz
We monitor:
The Hardware is used for Zabbix and as a standby gateway, nothing else.
All DB data is stored in one file called ibdata1, so i think the DB Engine System is MySQL InnoDB.
PhpMyAdmin says, the Zabbix DB (scheme) within MySQL would be MyISAM and the Zabbix DB's single Tables would be InnoDB again. Seems very strange to me...
Mysql Administrator says, the Zabbix DB would only contain 9GB of Data and 10 GB Indexes.
Every night we backup the full zabbix DB into one sql file, which is now 634MB large.
The value cpu user time (in %) is also very strange: http://www.heiwu.de/myfiles/cpuuserzabbix.jpg
As i said: only Zabbix is on this server...
First i thought, just waiting for the Time when the DB reaches its final Size (or begins growing slower), would be ok, but now i get really scared. There MUST be something wrong.
So i copied the sql-backup onto another System, installed Zabbix Server v1.4.4 and put die data from the .sql file into its Mysql-DB (InnoDB).
The Data was ok (zabbix could use it) and the ibdata1 file was only 1,8GB large.
So- Why do i have a file larger than 100GB for 19GB of Data and indexes?
Is this a zabbix or a mysql problem?
Pleas help!
You may also contact me via ICQ, AIM or email!
best regards,
Heiner
We really need help now since we only have about 20GB Disk space left...

First my System:
zabbix server V1.4.2
running with
- Fedora Core 6 Linux
- mysql-server-5.0.27-1.fc6 / mysql-5.0.27-1.fc6 / php-mysql-5.1.6-3.6.fc6
- httpd-2.2.4-2.1.fc6 / php-5.1.6-3.6.fc6
on a DELL PE 860 with
- Intel Pentium D 925, 3 GHz, 800MHz FSB
- 2 x 1024 GB DDR SD 667MHz
We monitor:
Code:
Number of hosts (monitored/not monitored/templates/deleted) 71(45/0/26/0) Number of items (monitored/disabled/not supported)[trapper] 4286(2406/1722/158)[0] Number of triggers (enabled/disabled)[true/unknown/false] 1206(1068/138)[3/48/1017] Number of events 436866 Number of alerts 17638
All DB data is stored in one file called ibdata1, so i think the DB Engine System is MySQL InnoDB.
PhpMyAdmin says, the Zabbix DB (scheme) within MySQL would be MyISAM and the Zabbix DB's single Tables would be InnoDB again. Seems very strange to me...
Mysql Administrator says, the Zabbix DB would only contain 9GB of Data and 10 GB Indexes.
Every night we backup the full zabbix DB into one sql file, which is now 634MB large.
The value cpu user time (in %) is also very strange: http://www.heiwu.de/myfiles/cpuuserzabbix.jpg
As i said: only Zabbix is on this server...
First i thought, just waiting for the Time when the DB reaches its final Size (or begins growing slower), would be ok, but now i get really scared. There MUST be something wrong.
So i copied the sql-backup onto another System, installed Zabbix Server v1.4.4 and put die data from the .sql file into its Mysql-DB (InnoDB).
The Data was ok (zabbix could use it) and the ibdata1 file was only 1,8GB large.
So- Why do i have a file larger than 100GB for 19GB of Data and indexes?
Is this a zabbix or a mysql problem?
Pleas help!
You may also contact me via ICQ, AIM or email!
best regards,
Heiner




Comment