I followed the instructions for non_local_details and now the message contains
Ignoring UDP query from non-local network 192.168.0.200 (logged only once)
which is rather puzzling as that address is most definitely local. I'm using the 'normal' 192.168.0.0/24.
May I please have some help/words of advice. Thanks kindly.
At the moment (it's 1117am) the last time Pi-Hole shows any log for this device (via http://192.168.0.2/admin/queries.php?client=192.168.0.200 ) was at 0856am. Despite the fact that I've rebooted it and tried to use it (it's a roku-stick, wireless). The roku-stick appears to be working as normal.
I seem to have posted into the middle of your post. I'm probably using this 'discourse' system all wrong. I humbly apologize if I upset you or confused anything you were trying to post.
Looking at your debug log, I see a few very odd things. Are you connected to your network with both Ethernet and WiFi at the same time?
Some very odd things are happening here, for instance, during the DHCP test you are getting DHCP requests made from one interface back on both, i.e. you requested two but received four offerings. Furthermore, your Pi-hole failed to resolve DNS correctly on both interfaces whereas everything was fine on the loopback interface.
Please ensure you are connected only with one device at a time to your network or very strange things may happen. This is not something unique to Pi-hole but you may experience malfunction or slowdowns with all other services on your machine, too.
Please allow me to mention that all was working fine until I did the upgrade. Nothing else in my network has changed except PiHole, It's running on a Raspberry Pi with eth0 at 192.168.0.2 and wlan0 at 192.168.0.3. I use wlan0 for SSH into the Raspberry PI.
Here's how I set the static IP's for eth0 & wlan0
static domain_name_servers=192.168.0.1 22.214.171.124 126.96.36.199 fd51:42f8:caae:d92e::1
Reverting shouldn't be the solution, I don't think anything really changed in your direction. I still believe that the issue lies in being connected to the same network over two independent connections at the same time and this causing conflicts (some of which are visible in the debug output).
Could you try some other Interface settings from the settings page (DNS tab) of the Pi-hole dashboard? Does this solve your issue? Maybe Pi-hole gets confused because you have two interfaces for the same address/subnet. Having both enabled at the same time isn't a healthy setup - even when it worked before, there are no guarantees it will work in the future, e.g., if some networking drivers are updated.
It's Monday and I'm back at work (as perhaps are you) and thus have less time to work on this but I wanted to let you know that I did deactivate the WiFi (via sudo ifconfig wlan0 down) and (as previously mentioned) set the PiHole DNS to 'Respond only on interface eth0'. The PiHole appears to be working from the GUI but the PiHole -Chronometer shows the PiHole to be offline.
I find it odd that PiHole is having trouble with there being two network interfaces given that it's running on a Raspberry Pi which comes with two network interfaces! (three if you include BlueTooth).
I'd like to close this topic now with a thankyou to all who helped and an attempt at a summary.
PiHole is working again without generating that 'Ignoring query from non-local network' message. The modification pihole checkout ftl new/non_local_details was helpful in showing the IP address that generated the message. It was a wifi device on the same network as everything else!
I have, on my Raspberry Pi, both the Ethernet and Wifi connected to the same network. On the DNS tab of the Settings page I have changed from the default to Respond only on interface eth0. I temporarily disabled the WiFi (with sudo ifconfig wlan0 down) but that didn't seem to make any difference.
I am now awaiting the fix to PiHole -chronometer. If I was more impatient I would apply the fix, but I'll wait for the official fix to come through as part of the build (probably using the wrong terms as I know nothing about Github).
It's great to hear everything is now working for you as it should.
I'd recommend to run a
pihole checkout ftl master
now to get back on track, otherwise, you will be fixed to this special version and won't receive any updates in the future. They are only published on the master branch.
Actually, it's not odd at all. The entire system has problems, you may not have noticed them as many things can be automatically retried (possibly but not necessarily with some extra timeouts) but you just found a case where it actually matters. It only remains for me to say that this is not an issue that affects Pi-hole alone. It just happend to be the most complex network application on your Raspberry.