I seem to have this issue aswell, rebind protection is off both on the router and on the nextdns site, however wireguard is still on, and forcing the dns on clients aswell, adguard is off.

But it also doesn’t happen all the time for me sometimes I can run it a few days fine, other times the dns doesn’t lease it to the clients, for me a reboot of the router works fine, but I guess a stubby restart would be fine to as others have suggested.

Sometimes also my wifi goes down and this only happens when udhcpd is renewing the lease, it will no longer associate clients but I wonder if this issue could be also a part of this, like there is some kind delay so stubby or even the interface fails to get up, note that im also observing udhcpd lease failing errors but I guess that might be caused by myself since I made a routed AP via luci, and maybe stubby can act unstable when the wan connection got connection lost?

If I see anything happen again I could add my logs from kernel and system status, to me its very random.

Edit

I forgot to mentoin the firmware im using, the firmware is release 3.207 (1800ax flint).

Edit again:

I added the logs here with stripped mac addresses maybe its usefull and maybe not.

dns.zip (28.9 KB)