Unbound at 100% CPU and restart in loop since 23.7.3

Started by Unonyme, September 10, 2023, 10:48:15 AM

Previous topic - Next topic
Hello, I'm a little bit blocked with opnsense since i've upgraded to 23.7.3
I've check with a command :
ps -aux | grep unb
It seems to crash in less than 30 seconds :-\
unbound 59002   0.0  0.3  84904  43452  -  Ss   10:18      0:00.22 /usr/local/sbin/unbound -c /var/unbound/unbound.conf
root    60077   0.0  0.1  25732  15460  -  Rs   10:18      0:00.20 /usr/local/bin/python3 /usr/local/opnsense/scripts/dhcp/unbound_watcher.py --domain lan (pyt


I tried some tweaks as I read many issues with unbound here  :-\
(besides, I'm surprised by the recurring instability of unbound, it's not the first time i'm in trouble)

- disable suricata (it was eating CPU in coordination with unbound)
- disable DNS blocking lists
- disable Unbound DNS reporting
- disable DHCP leases registering

I tried to disable unboundDNS and activate DNSmasq (first time i try) but it seems it has the same behaviour, DNSmasq restart in  loop  :(

Have you any advice ?

Could you give more details about "100% CPU".
What is reported by "top" command?

Probably related: https://forum.opnsense.org/index.php?topic=35527.30

Well, it's not totally 100% now i disable many things,
but every restart, i can see that cpu peak on load

The main issue is that while it's crashing, request to DNS can timeout or take more than 5 sec  :-[