net_connect enhanced monitoring profiles cause high CPU usage for hub and controller probes

Products

• DX Unified Infrastructure Management (Nimsoft / UIM)

Issue/Introduction

We deployed the Network Connectivity (Enhanced)  profile to a group with 1600+ devices and the deployment process took over 12 hours to complete.

<timed> settings in the mon_config_service.cfg

device_processing_threads = 5
config_deployment_threads = 5

In this scenario the Setup Network Connectivity (Enhanced)  profile was deployed to a Hub and we noticed that the Controller and Hub probe CPU usage was high during the Network Connectivity (Enhanced) deployment process and when making changes to the Alarm Policies TOT settings

Are there any settings that can reduce the Network Connectivity (Enhanced) deployment or Alarm Policies TOT settings changes?

 

 

 

 

Environment

Release : 20.3.3

Component : UIM - MONITORING SERVICES

Cause

device_processing_threads and config_deployment_threads mon_config_service keys are set to optimized values

Resolution

Increase the settings below on the mon_config_service -> timed section via Raw Configure or by manually editing the mon_config_service.cfg followed by a mon_config_service  probe restart

device_processing_threads = 30
config_deployment_threads = 100

Notes:

1. Controller and Hub CPU usage will still be impacted with the amount of profiles but the total time for the profile deployment or Alarm Policy changes decreases significantly.

2. Above settings are the default values for new UIM 20.3 instance. Previous UIM versions default settings were 5 for both.