Hourly CPU spikes

[mkcharlie]

Hi,

Every hour my PFSense's CPU spikes:
.
I do not know if this is causing issues, but I do have the impression that it maps to users complaining about connection dropping etc.

System: Netgate 7100, 22.05-RELEASE (amd64)
Packages:

Acme
Avahi
aws-wizard
Cron
freeradius3
ipsec-profile-wizard
ntopng
openvpn-client-export
pfBlockerNG-devel
suricata
Telegraf
WireGuard

Cron:

PFBlockerNG is configured to update once every 24 hours, Suricata as well. I do not see other obvious log rotation configuration settings or flushes that may cause this spike. Any idea how I can go about investigating the cause of these spikes?

[dma_pf]

@mkcharlie said in Hourly CPU spikes:

I do not know if this is causing issues, but I do have the impression that it maps to users complaining about connection dropping etc.

I don't know about those spikes but connections dropping can happen if you have this checked in DNS Resolver/ General Settings:

It causes resolver to restart every time a device connects to pfsense via dhcp (static IP assignments don't cause this issue). During the restart clients can't resolve any FQDN so traffic doesn't get routed out. From the user's perspective it appears that the connection is lost.

[stephenw10]

Do you see anything in the system logs at those times?

[mkcharlie]

@dma_pf that setting is not checked. Thanks for the suggestion though!

@stephenw10 nothing in particular no, and definitely not something that is recurring every hour.

[stephenw10]

If you can predict it just see what's happening in the output of top -HaSP

[mkcharlie]

@stephenw10 thanks, seems ntopng was heavily spiking a minute ago. I have removed it and will monitor what happens in the next hours.