Optimizing Email Alerts in Nagios for Multi-Server Systems

[Mikasa23]

Hello everyone,

I am managing a system with more than 50 servers running on many different environments and using Nagios for monitoring. Currently PolyTrack, I have a problem where the system sends too many duplicate alerts when an incident occurs, making it difficult to monitor and handle promptly.

I would like to ask:

Is there a way to optimize the configuration to reduce duplicate alerts but still ensure that important incidents are not missed?

Is there a plugin or script that supports grouping alerts by server or service?

What is everyone's experience in setting up email notifications for a multi-server environment to be both effective and easy to manage?

I hope everyone can share their experiences or guide how to solve this problem. Thank you in advance for all your contributions!

[kg2857]

Nobody can tell you how your system is set up to send notifications (not alarms). You can look at your setup and figure it out.
I assume you have only one nagios system, or if you have more than one, only one has notifications, etc. enabled.
Do you have core or XI?
There should be no duplicate notifications but it depends on your setup.

[jkimmel]

Not sure what you mean exactly. Do you have multiple Nagios instances? Can you give an example of a specific issue? Do you have mukltiple checks for the same services?


I have my checks setup almost in a BPI manner that I monitor the actual service itself. e.g. website.com which is comprised of server1 (http website check), server 2 (http website check), load balancer. Pushovers alarms are generated only if website.com is down, but I have regular email alarms for the others (a piece is down but the service is not).

[cdietsch]

Hi @Mikasa23,

You might want to take a look at the documentation for Parent/Child Host Relationships https://support.nagios.com/kb/article/p ... s-904.html

An excerpt from the document:

In the CCM (Core Configuration Manager) within Nagios XI, you can define a parent for a host by using the Manage Parents functionality on the Manage Hosts page. Generally, parent hosts are routers, switches, firewalls, etc... that lie between the monitoring host (Nagios XI) and the remote hosts. Parent/child relationships help Nagios XI figure out host reach-ability. If the parent host is down, the child hosts will be considered unreachable. Building these relationships can prevent duplicate notifications and escalations.