Unreachable for > x minutes

[cbidwell]

Is there a way that I can modify this syntax to where if a system is unreachable for x amount of minutes to only alert after that time frame? Say it is unreachable for > 5 minutes only to send out the alert then?

{Template_Linux:status.last(0)}=2

Would I say: {Template_Linux:status.last(300)}=2

Thanks

[Alexei]

{Template_Linux:status.min(300)}=2

[cbidwell]

Great! Thank you.

One more question...is there a way that I can get emailed every 5 or 10 minutes and it tell me in that mail how long the system has been unreachable?

[elvar]

{Template_Linux:status.min(300)}=2

I found this very helpful as well. Thank you!

[vhm3]

Most excellent, and thank you!

Chip

[elvar]

I found this very helpful as well. Thank you!

Unfortunately this didn't work the way I thought it would. I would prefer to never receive an alert until any of the unreachables have been unreachable for 5 minutes. After changing the unreachable trigger to status.min with a value of 300 I still was alerted almost instantly.

Anyone gotten this to work properly?

[bbrendon]

I don't use status item because I personally believe its redundant. What are the values status returns?

You can then use the function min/max/nodata/etc for number of polls or for a periods of time.

[elvar]

I don't use status item because I personally believe its redundant. What are the values status returns?

You can then use the function min/max/nodata/etc for number of polls or for a periods of time.

Can you please paste an example of the nodata function syntax to use for this? I only want to be alerted when it receives no data from the zabbix agent after 5 minutes. Right now there are too many false positives when the network blips for very short periods of time.

[elvar]

There is probably a way to do this with functions but I found a solution that works perfectly for my needs. In the /etc/zabbix/zabbix_server.conf I found the following...

# After how many seconds of unreachability treat a host as unavailable
UnreachablePeriod=300


I'd still like to know the way to do this exact same thing with functions per template.


Regards,
Elvar

[jroberson]

This is the trigger I use to tell me when a system is unreachable for 5 minutes:

({Template_Windows:status.last(0) }=2)&({Template_Windows:status.nodata(300)}=1)