I'm running 23.05.2 on a device with some basic services, e.g. luci, hostapd, dnsmasq and only recently started experiencing very high CPU usage, to the point where pings to external hosts have increased. When running top, I often see a couple of processes chewing up most of the CPU time:
Mem: 51788K used, 69656K free, 176K shrd, 0K buff, 13892K cached
CPU: 30% usr 23% sys 0% nic 0% idle 0% io 0% irq 46% sirq
Load average: 2.82 2.10 2.06 2/72 19634
PID PPID USER STAT VSZ %VSZ %CPU COMMAND
19617 19614 root R 2704 2% 54% utpl -S /usr/share/firewall4/main.uc
17051 1 root S 1760 1% 23% luci-bwc 7
19634 17722 root R 1388 1% 15% top
1251 1246 network S 4540 4% 8% /usr/sbin/hostapd -s -g /var/run/host
In another instance, top showed a different set of processes chewing up most of the CPU time:
27576 27575 root R 1544 1% 38% {firewall} /bin/sh /etc/rc.common /etc/init.d/firewall enabled
27568 27517 root R 1576 1% 31% utpl -S /usr/share/firewall4/main.uc
27580 2078 root R 1540 1% 31% {dhcpv6.script} /bin/sh /lib/netifd/dhcpv6.script eth0.2 unbound
htop shows the CPU pegged at 100% and mostly the same info as top, although I can't seem to copy/paste from htop in putty, since it responds to mouse clicks.
I have not made any manual firewall changes - just a few minor changes through LuCI.
I am contemplating disabling IPv6, since I don't use it, but something doesn't look right to me.
UPDATE: This issue may be related to another issue. In LuCI, I can see the "IPv6 network" appear momentarily and then disappear for a few seconds, before reappearing again. Periodically running ifstatus wan6 shows the interface momentarily going up every dozen seconds or so, before going down again.