Trigger continuosly on and off

[antani]

Hello, i have a trigger monitoring an item that reports the free disk space on /var in this way: vfs.fs.size[/var,pfree] every minute.

Since three days the trigger goes on and off:
2008.Dec.18 16:04:53 7.2923
2008.Dec.18 16:04:34 60.7985
2008.Dec.18 16:03:53 7.2926
2008.Dec.18 16:03:34 60.7915
2008.Dec.18 16:02:53 7.3281
2008.Dec.18 16:02:33 60.7943
2008.Dec.18 16:01:53 7.3323

and sooooooooo on.

Restarted zabbix [agent|server] but nothing changes.
Sure you would ask: What is the real situation on disks?

/dev/sdc1 10G 8,7G 739M 93% /var

What does occupy all that amount of space?

# ls -lh history_log.ibd
-rw-rw---- 1 mysql mysql 6,6G 18 dic 16:07 history_log.ibd

So, two questions: What's happening on my zabbix server? Why that file is so big?

Thanks and good holydays

[antani]

News: other triggers are assuming the very same behaviour on the same zabbix server. As we say in Italy, I don't know were to bang my head.

e.g. {ZABBIX Server:net.tcp.service[pop]. last( 0 ) }=0

2009.Jan.07 14:00:53 	Down (0)
2009.Jan.07 14:00:34 	Up (1)
2009.Jan.07 13:59:53 	Down (0)
2009.Jan.07 13:59:33 	Up (1)
2009.Jan.07 13:58:53 	Down (0)
2009.Jan.07 13:58:33 	Up (1)
2009.Jan.07 13:56:18 	Up (1)
2009.Jan.07 13:55:56 	Down (0)
2009.Jan.07 13:55:18 	Up (1)
2009.Jan.07 13:54:56 	Down (0)
2009.Jan.07 13:54:18 	Up (1)
2009.Jan.07 13:53:56 	Down (0)
2009.Jan.07 13:53:17 	Up (1)
2009.Jan.07 13:52:56 	Down (0)
2009.Jan.07 13:52:17 	Up (1)

[geno]

Normal + Multiple true events

In the "Configuration of Triggers" for you trigger, is the 'Event generation' option set to 'Normal' or 'Normal + Multiple true events' ?

If you chose the latter, then it would explain whats happening there...

[antani]

In the "Configuration of Triggers" for you trigger, is the 'Event generation' option set to 'Normal' or 'Normal + Multiple true events' ?

The option is set to "Normal".

[geno]

i dont understand. is the problem with zabbix's triggers going on and off all the time, or is the problem that something on your server consumes space?

what is that log file of?

[neoyagami]

you are getting some "host un racheable" errors too?

[Alexei]

What version of ZABBIX you are running? It is likely that you hit a rare problem existing in 1.6.1 (perhaps also in 1.6), which may cause incorrect calculation of trigger expressions under certain conditions. If so, you may consider using pre-1.6.2 from www.zabbix.com/developers.php or wait for the official 1.6.2.