On a host, if you are wanting to measure a device's up status, the latency/response time, and ping loss, it appears (at least the way I have it set up right now), that the only way to do this is to have three separate items with three separate triggers. The only problem is that when you have hundreds of devices, although pings don't generate much traffic, for every device you want to monitor, then there are three separate ping instances for each host which can add up to more traffic than is necessary and not to mention it adds additional processing to your zabbix server. Is there a way to have one ping instance but to be able to monitor all three stats off of one rule?
-
-
You could use either the latency or ping loss to determine the up status, thus reducing to two pings. But an item in zabbix measures/stores just one value, so you cannot measure latency and ping loss with one item.
If you're really worried about network load, you could run the ping from a script and use zabbix_sender to sent the results to different items, but this seems a bit overkill to me. -
I'm not so much worried about it on my network and the traffic it generates isn't really a big deal either, but I have had several external hosts such as websites or public DNS servers that have blocked me at the firewall level when trying to monitor them. I'm assuming because they think I am trying to do a denial of service attack or they don't like being pinged by 6 instances from my zabbix server in a minutes time.Comment
Comment