Can someone help me add the following alerts:
Node has Stopped Responding
(due to either TTL expired during trnsit, or Request timed out)
Thanks
Node Status is equal to unknown.
Node status is not equal to up
It looks like you have a few options from the previous posts so let's break it down.
You want the alert to trigger when you're no longer able to ping the node (i.e., TTL expired / request timed out). By virtue of adding the node to your inventory you're already pinging the device every 120 seconds (default) and waiting an additional 120 seconds (default) to decide if it's truly not coming back. That takes the node from Up to Warning and finally to Down.
So, the trigger condition:
Node Status is equal to Down
And the reset condition:
Node Status is equal to Up
It's important that you don't use the 'Reset status when trigger is no longer true.' option in the Reset Condition because there are more then two possible statuses for the node (Up, Down, Warning, Unmanaged, External, Unknown).
Mr. Xinu,
I love learning something new. I am unfamiliar with Node Status = External. What does it mean, and how can you use it? Where do you access that status (I don't see External in the Advanced Alerts under Node Status)?
Thanks,
Eric
I did some experimenting. If you want to know when the node first is unreachable, you'll want to alert when node status is warning. But, you might also want to know when the node status is officially down. So, now that I think more about it, I think the best alarm for the purpose stated above might be:
Node Status has changed
and the reset:
This would send out alerts when the node goes to warning, or down (or other), and would cancel the alert when the node comes back up. Oh, and disregard the "node status equals unknown" which I wrote above -- I was thinking about interface alerts, not node alerts.
Mr. Xinu,I love learning something new. I am unfamiliar with Node Status = External. What does it mean, and how can you use it? Where do you access that status (I don't see External in the Advanced Alerts under Node Status)?Thanks, Eric
Sorry for the lag - I'm onsite with a customer today. The 'External' status for a node was brought about because APM needs a node for its templates. For example, if you wanted to monitor the www.cnn.com website, but you didn't want to ping (or keep historical ping data) you could set the node as external and then point the APM template at it using the HTTP component.
