Monitoring Disk Space issue

[ggs_admin]

Hello,

I'm trying to configure monitoring the disk space storage but it definitely doesn't work properly.
Since I new on Zabbix I have found the followng topic on how to create an item and trigger

https://www.zabbix.com/forum/zabbix-help/376175-how-to-monitor-use-of-mount-disk-linux-on-zabbix

So I have configured these things as per topic




Since my disk usage on the testing server is 25% (used) then I have set trigger to 90% in order to be sure that it is working as expected. But no luck it doesn't show any messages that disk space less than 90%.
Then I have tried to use rule vfs.fs.size[/,pused] but received the following error:



So I have created dump of the database and removed any templates where mentioned Zabbix server but this error is still showing when I'm trying to add the this rule.
Any ideas on how to monitor the Disk Space (I need the same as was in the above-mentioned topic WARNING when used >= 85%
and alert CRITICAL when used % > 95%)

[tim.mooney]

I'm not 100% certain, but I think what it's telling you is that 'vfs.fs.size[/] is already being gathered once, so if you try gather it again with a different second argument (pused vs pfree) it doesn't matter.

As far as why it's not working: try switching to keeping "history", even if it's for a short period. Trigger functions use the history data. If you don't keep history for an item, no trigger functions will be calculated.

See the overview information about Items: https://www.zabbix.com/documentation...fig/items/item

The information about trigger functions requiring historical data is discussed in the history and trends page: https://www.zabbix.com/documentation...ory_and_trends

[ggs_admin]

Hello Tim,

Thank you for your response.
I have enabled to saving history and Trend for 5d. But anyway, I don't think that this item is working. It looks like not supported item (or bug) but the status is Enabled.

[ggs_admin]

Hi, again.

This issue has been solved by re-enabling the service and restart agent and zabbix server

[tim.mooney]

This is absolutely not true... by using different arguments, you are gathering different things. It would be quite stupid, if we could not get pfree for a mountpoint just because we already query for pused..

Thanks for clarifying that point. I was just speculating based on the information provided. I'm glad it works the way you've confirmed, as that definitely makes more sense.

And of course no trigger will work, if you don't keep any history. No data, no triggers...

You say "of course" as someone that is very experienced with Zabbix, but I don't think this trigger behavior is at all obvious to people that are new to Zabbix. A reasonable newcomer might assume that the data has been collected (even if it's not being stored), so it should be possible to trigger based on the data that was just collected. That it doesn't work that way makes sense once you learn more about trigger processing, but newcomers could reasonably draw an incorrect conclusion if they don't follow the link to the "history and trends" page.

[cyber]

I'm not 100% certain, but I think what it's telling you is that 'vfs.fs.size[/] is already being gathered once, so if you try gather it again with a different second argument (pused vs pfree) it doesn't matter.

This is absolutely not true... by using different arguments, you are gathering different things. It would be quite stupid, if we could not get pfree for a mountpoint just because we already query for pused..
But error message here said definitely, that there is already such item installed on that host. It may have originated from some other template.
And of course no trigger will work, if you don't keep any history. No data, no triggers...