I got impatient waiting for 6.0 and decided to just say eff it and upgrade my install. Now everything is broken and I feel like an idiot.
So I used terminal to log onto my RPi 02w and did
pihole checkout dev
sudo apt remove lighttpd
And it was working-ish. The installation went fine it seems. All my lists were present etc. But my upstream DNS was unset, so I typed in 127.0.0.1#5335 to try and get it to use Unbound again and that's when it all went wrong. It was stuck updating for minutes, so I decided to do a hard reboot and now I can't connect to the WiFi.
I reset the router so I could log in and re-enable the DHCP server and now I'm able to access the Pi-Hole again. Unfortunately that's as far as I know how to get. So I did a pihole -d which gives this https://tricorder.pi-hole.net/ToC1KXgh/
I really don't know what can cause such catastrophic system failure so that it does not even get to the point of connecting via WiFi. Pi-hole does not fiddle with anything even closely remote to these settings. Any chance you could maybe hook up a screen to the device to see what is going on?
Overall, this "it just died" really often simply goes down to the classic "the SD card" died issue which seems logic to become a lot worse at once when you do a major upgrade such as upgrading Pi-hole involving many files.
It'd be great if you could revive your system and we could look at stuff together to rule out it has anything to do with Pi-hole. For once, we have never observed nor have seen any reports that upstream servers failed to migrate. So I'm really unsure what is happening here.
Ah, okay, so we both replied at the same time. Looking at your debug log, most everything looks to be right. Your system is able to resolve domains and there are also no warnings anywhere. Except the upstream server, the entire configuration is on default values.
There is one warning about a query received from a non-local network, though, but that's also for you to see in the Pi-hole diagnosis messages on the web interface.
Do you have any idea what went wrong, or: what did you do to fix it?
I think I've found the problem. When I'm going through the settings, the DHCP Server Enabled has been switched to false. I'm gonna enable and test again. But that's definitely a setting that should be preserved if possible.
Confirmed to all be working properly. Thanks for your assistance.
One quick, maybe two quick questions. In upgrading, it removed my admin password. Can I set it again via the UI or do I need to use sudo pihole setpassword SECRETPASSWORD via the command line? And if it's only via the command line do I need to enclose the password in quotations?
Sorry for the trouble, everything should have already been preserved. So something went wrong during migration. Could you please check the log file for the migration steps? You could either post them here or send them via private message to me if there's something inside you don't want to show publicly.
Oh, yes, you may need to preface the command with sudo
This is very strange, do you still recall when the migration happened? I'd like to see the log surrounding this time as it'd hopefully show why no migration was done.
Maybe an idea: Did you ever try v6 before and then switched back or somehow else got some v6 related files on your machine? I'm asking because if v6 files are found, Pi-hole assumes the migration has already happened and won't do another attempt.
...
[2024-12-20 03:01:13.644 652M] ***** Ignored unknown DNSSEC status "TRUNCATED"
[2024-12-20 03:03:16.379 652M] ***** Ignored unknown DNSSEC status "TRUNCATED"
[2024-12-20 10:14:29.565 652M] ***** Ignored unknown DNSSEC status "TRUNCATED"
<----- Migration happened here
2024-12-20 10:24:35.117 GMT [59335M] INFO: Config file /etc/pihole/pihole.toml not available (r): No such file or directory
2024-12-20 10:24:35.221 GMT [59344M] INFO: Config file /etc/pihole/pihole.toml not available (r): No such file or directory
2024-12-20 10:24:35.225 GMT [59345M] INFO: Config file /etc/pihole/pihole.toml not available (r): No such file or directory
2024-12-20 10:24:35.257 GMT [59347M] INFO: Config file /etc/pihole/pihole.toml not available (r): No such file or directory
2024-12-20 10:24:35.277 GMT [59348M] INFO: Config file /etc/pihole/pihole.toml not available (r): No such file or directory
2024-12-20 10:24:35.306 GMT [59349M] INFO: Config file /etc/pihole/pihole.toml not available (r): No such file or directory
2024-12-20 10:24:35.338 GMT [59350M] INFO: Config file /etc/pihole/pihole.toml not available (r): No such file or directory
2024-12-20 10:24:35.367 GMT [59351M] INFO: Config file /etc/pihole/pihole.toml not available (r): No such file or directory
2024-12-20 10:24:38.932 GMT [59376M] INFO: Config file /etc/pihole/pihole.toml not available (r): No such file or directory
2024-12-20 10:24:39.053 GMT [59391M] INFO: Config file /etc/pihole/pihole.toml not available (r): No such file or directory
2024-12-20 10:24:39.090 GMT [59390M] INFO: ########## FTL started on illion-rpi02w! ##########
2024-12-20 10:24:39.090 GMT [59390M] INFO: FTL branch: development
2024-12-20 10:24:39.090 GMT [59390M] INFO: FTL version: vDev-91ea8d4
2024-12-20 10:24:39.091 GMT [59390M] INFO: FTL commit: 91ea8d49
2024-12-20 10:24:39.091 GMT [59390M] INFO: FTL date: 2024-12-10 19:38:42 +0100
2024-12-20 10:24:39.091 GMT [59390M] INFO: FTL user: pihole
...
I wonder if anything in core changed that causes pihole.toml to be created. When FTL does not even try to migrate, this means pihole.toml was there on the first start so something must have created it meanwhile.