Polling services at 1 minute interval is smacking CPU Load

[vlakshman]

Team,

My Nagios server is running on C5.xlarge EC2 AWS instance which has 8 GB RAM and 4 vCPU.
When monitoring some 1000 services at interval of 1 minute the CPU load gets smacked to 300 and Nagios UI rendering goes very slow.
Upgrading instance with more CPU is an option. But I want to know if there are some tweaks available for my need to ensure optimum instance is used to suffice the request.

My custom plugin script execution time is under 10 sec.

My Settings:

1) nagios.cfg: interval_length=60
2) generic-host and generic-service templates: check_interval = 1, retry_interval=1,max_check_attempts=2

[benjaminsmith]

Hi @vlakshman,

Besides upgrading your server, I would recommend increasing the check interval to what's necessary for non-critical services.

There is an option in the main configuration file to make several shortcuts to improve performance, you can try setting this value to 1.

Code: Select all

use_large_installation_tweaks=1

Tuning Nagios For Maximum Performance
https://assets.nagios.com/downloads/nag ... uning.html