false critical ping alerts on all hosts

[rnjie]

am getting false ping alerts on almost all hosts am not sure why but i do not see any oustanding errors, i have tried deleting and readding these alerts and they just keep coming back after a while, it also incase the load average of the system to about 300 crazy, please help me find a soultion, i attached my profile to help troubleshoot


Moderator's Note: The profile has been shared with the support team but has been removed from the public forum.

[cdienger]

The profile shows the xagt command is using the most CPU. It may be worth testing having this disabled.

Also, how are you testing to confirm they are false alerts? The logs I see how ping responses taking 3+ seconds. The plugin XI uses will rely on the ping binary of the system. When you get an alert about a host being down, try pinging the 'down' host from the command line. For example:

Code: Select all

ping LXACPSEA

[rnjie]

they are all false alerts, all the servers are up

[cdienger]

A ping could fail for number of reasons and not just if the server is down. Routing, network issues, or a firewall blocking icmp traffic for example. As seen in the profile, some of these are not completely failing - some are getting an 80% return which could mean packets are getting dropped somewhere.