Expected Behaviour:
Using the pi-hole as DNS and DHCP server, as my router does not allow me to refer a DNS within the same network segment when using it as DHCP.
Operating System (Family and Version)
- Pi Zero W
- OS Raspbian GNU/Linux 13 (trixie)
Actual Behaviour:
After an irregular period of time, the FTL.log starts to log a lot of these error messages to a point that it blocks any further process in the board, until I switch the DHCP on in my router it immediately comes back online and process everything normally, until the next cycle when this starts to show again. I have seen online that many people mention that these messages can be ignored, but the issue is that there are so many, it just blocks any further processing in the server. Also, I have seen that the issue might be related to missing DHCP interface context, but this is also not the case as the interface (wlan0) has already been configured.
Debug Token:
https://tricorder.pi-hole.net/FEfkc1eM/
What does below one output?
nmcli -t -f name con show --active | xargs -d '\n' -n 1 nmcli -p -f ipv4.method con show
Pasting the result:
# ===============================================================================
Connection profile details (preconfigured)
## ipv4.method: manual
# ===============================================================================
Connection profile details (lo)
## ipv4.method: manual
Well, it turns out the [predetermined] connection did not have explicitly assigned the wlan0 device (I connect via wireless) and thanks to the question from the previous poster, I started to dig into the nmcli settings and it turned out wlan0 was not configured. I proceeded to edit the predefined to name it wlan0 and assign the wlan0 device and so far everything seems to be working correctly, at last!! This OS upgrade came with a different way to configure the network connections that I was not familiar with, but hey, you always learn something new.
I leave here the command that pointed me in the right direction for the solution, for other to see whenever they find the same error:
sudo nmcli connection modify 'wlan0' ipv4.method manual ipv4.address 192.168.0.218/24 ipv4.gateway 192.168.0.254 ipv4.dns 127.0.0.1
I've never seen a NM profile named "preconfigured" before.
But maybe thats a Trixie thing?
Next time I'll ask for below 
$ nmcli -t -f name con show --active | xargs -d '\n' -n 1 nmcli -p -f connection.interface-name,ipv4.method con show
===============================================================================
Connection profile details (Wired connection 1)
===============================================================================
connection.interface-name: eth0
-------------------------------------------------------------------------------
ipv4.method: manual
-------------------------------------------------------------------------------
===============================================================================
Connection profile details (lo)
===============================================================================
connection.interface-name: lo
-------------------------------------------------------------------------------
ipv4.method: manual
-------------------------------------------------------------------------------
I would not recommend to configure the loopback IP 127.0.0.1 as a DNS server for the Pi-hole host.
Whenever Pi-hole breaks, an update going wrong, tinker mistake etc, all processes on the host loose DNS resolution which complicates diagnosing.
You wont be able to upload a debug log (sudo pihole -d) or run Pi-hole repair/reconfigure (sudo pihole -r).
EDIT: Out of curiosity, could you post output for below one?
dpkg -S preconfigured.nmconnection
And whats contained in the file if found?
This is the output:
dpkg-query: no path found matching pattern preconfigured.nmconnection
I tried running it also using wlan0 but still same result.
But pihole still running with zero interruptions, it’s been as stable as ever.