Ntopng causes high cpu load on host machine

NethServer Version: ns7 rc3
Module: ntopng

Hi guys,

I recognized that ntopng causes high cpu load on the host machine of a vm. The VM itself doesn’t report this.
The moment I enable the service on the vm the cpu load increases about 15-20 % (red arrow) and stays that high.

When I stop the service cpu load drop down to normal value.

I found also that pf_ring service has failed, which is a part of ntopng. Could this be the reason?

With systemctl list-units --type=service I find 3 failed services:

pf_ring.service
systemd-tmpfiles-clean.service
systemd-tmpfiles-setup.service

ntopng requires a lot of resources, it has to inspect all network traffic.
AFAIK, pfring is not used on NethServer.

Maybe this is vm related. I’m using virtualbox with bridged interfaces. On the “real server” on ns6.8 ntop causes only a few percent (1 to 3) of load as htop reports.