host status & agent.ping

[alj]

1. Despite the similar bug is being opened and closed as "fixed" or "cant reproduce" I have a problem of "status" not being updated (no data). So after missing several alerts i cannot rely on it anymore.

2. I began to use agent.ping + nodata(300) trigger to monitor host agent functionality but that will generate storm of pages in case I need to stop and start zabbix server for maintenance.

Page storm is VERY bad because pager will choke when it received 700 host down messages.

Please give me advice how otherwise can i monitor agent availability without using .nodata or any other time sensitive trigger? Something that returns bad if agent was unable to be reached.....

Or please fix status (so it will work) or agent.ping so it would return some value in case of failure so .nodata can be removed.

[jamieautomate]

I also would like to get this issue sorted
It is the last thing holding me back from removing our old monitoring solution GFI Max

[adriano]

maybe a workaround..

I had the same problem some time ago.

It's not a fix but I do use ICMP pings to monitor availability, didn't you tried that?

name: system status
key: icmpping
update:30

It will automatically ping the server configured IP address.

Create a trigger like .max(180)}<1

[cferraro]

If you are using active agent items, status is unreliable. I started using ping.nodata checks like you with good results.

I use the maintenance settings to prevent blasts of notifications when I perform maintenance on nodes. Set your action conditions with something like 'Maintenance status not in "maintenance"'

This has worked pretty well for me, as long as I remember to set the maintenance period...

Good luck