OPNsense Forum

Archive => 22.7 Legacy Series => Topic started by: SoWhat! on August 14, 2022, 11:34:25 am

Title: [solved] Significant higher CPU Load since Update to 22.7
Post by: SoWhat! on August 14, 2022, 11:34:25 am

Hi everyone,

since installing the update to 22.4 my virtualized OPNsense consumes much more CPU-Load. It increases from about 5-7% in average now to 25% and more.

Using top on the shell I saw an Python 3.9 processes that consumes  every couple of seconds ~99% CPU.

The VM is running on a Proxmox Server (7.2-7) based on an Intel NUC7i5. The vNIC type is VirtIO and the QEMU-Agent is installed.

Any ideas what this Python process is doing in short intervals?

Netflow is disabled.

Thanks in advance!
Stefan

Title: Re: Significant higher CPU Load since Update to 22.7
Post by: tiermutter on August 14, 2022, 12:06:57 pm

Try

Code: [Select]

ps awwux | grep python at the moment when the process is running. May be a table update, on my system running every minute causing CPU peaks by python.

Title: Re: Significant higher CPU Load since Update to 22.7
Post by: SoWhat! on August 14, 2022, 12:52:00 pm

Code: [Select]

root@OPNsense:~ # ps awwux | grep python
root    25143  94.0  0.3   38240   16852  -  Rs   Fri22   1050:24.63 /usr/local/bin/python3 /usr/local/opnsense/scripts/netflow/flowd_aggregate.py (python3.9)
root      592   0.0  0.2   36256   11076  -  Is   Fri22      0:00.27 /usr/local/bin/python3 /usr/local/opnsense/service/configd.py (python3.9)
root      594   0.0  0.7   88044   37564  -  S    Fri22      0:39.24 /usr/local/bin/python3 /usr/local/opnsense/service/configd.py console (python3.9)
root     6855   0.0  0.3   25644   14096  -  Ss   10:46      0:01.68 /usr/local/bin/python3 /usr/local/opnsense/scripts/dhcp/dnsmasq_watcher.py --domain home.lab (python3.9)
root    51397   0.0  1.1  172980   57316  -  S<   Fri22      9:38.84 /usr/local/sensei/py_venv/bin/python3 /usr/local/sensei//scripts/datastore/ipdrstreamer.py (python3.9)
root    16488   0.0  0.0     436     248  0  R+   12:44      0:00.00 grep python
flowd_aggregate.py is the malefactor!

I re-checked my settings and I found under Reporting/Settings that the option "Enables the RRD graphing backend." was still enabled. After uncheck this option, the Python process wasn't consuming CPU anymore.

Thanks for your help, tiermutter! :)

Cheers,
Stefan