I’ve had what I recall to be the default install on a RPi 3B for a few years now. My ability to load the dashboard seems to die every couple of days, culminating in a took too long to responderror. Sometimes this propagates to my secondary pi-hole (an identical LXC in proxmox that is synced via nebula-sync).
Reading around points to lighttpd being the possible culprit, so on my primary pi running my primary pi-hole, I run a sudo service lighttpd restart which doesn’t seem to solve the issue. Running sudo systemctl status --full --no-pager lighttpd shows that lighttpd fails to start and prints the following:
Jan 30 20:23:00 raspberrypi systemd[1]: lighttpd.service: Scheduled restart job, restart counter is at 5.
Jan 30 20:23:00 raspberrypi systemd[1]: Stopped Lighttpd Daemon.
Jan 30 20:23:00 raspberrypi systemd[1]: lighttpd.service: Start request repeated too quickly.
Jan 30 20:23:00 raspberrypi systemd[1]: lighttpd.service: Failed with result 'exit-code'.
Jan 30 20:23:00 raspberrypi systemd[1]: Failed to start Lighttpd Daemon.
Rebooting the hardware via sudo reboot seems to solve the issue, but it’s a temporary solution as the pi-hole instance inevitably loses the ability to load the dashboard on its /admin page a day or two later.
I was not aware lighttpd was no longer in use, and yes that is the correct IP. I use HTTP locally and it’s very rare that I use HTTPS unless I’m connecting remotely. Neither option currently works.
Still waiting on possible solutions as the problem is ongoing. Adblocking still works, local DNS entries still work, however I’m still unable to load up 192.168.1.199/admin. To make sure my ad blocking and DNS entries are not currently getting solved by my failover I triedping which gets a response noting the RPI is up and running. I also tried nslookup google.com 192.168.1.199to verify that DNS functioned and got valid results.
sudo ss -nltp sport = 80 or sport = 443 gives the following results showing FTL on those ports
pi@raspberrypi:~ $ sudo ss -nltp sport = 80 or sport = 443
State Recv-Q Send-Q Local Address:Port Peer Address:Port Process
LISTEN 118 200 0.0.0.0:80 0.0.0.0:* users:(("pihole-FTL",pid=634,fd=35))
LISTEN 167 200 0.0.0.0:443 0.0.0.0:* users:(("pihole-FTL",pid=634,fd=36))
LISTEN 0 200 [::]:80 [::]:* users:(("pihole-FTL",pid=634,fd=37))
LISTEN 0 200 [::]:443 [::]:* users:(("pihole-FTL",pid=634,fd=38))
I am unable to curl both the IP/admin and localhost/admin from the pi running the pi-hole with either HTTP or HTTPS. I am also unable to curl IP/admin from other machines on the network with HTTP or HTTPS. My nft list ruleset is empty.
Just to answer deHakkelar's question, Pi-hole is using the correct ports.
From the Debug Log:
[✓] udp:*:53 is in use by pihole-FTL
(...)
[✓] tcp:0.0.0.0:443 is in use by pihole-FTL
[✓] tcp:0.0.0.0:80 is in use by pihole-FTL
[✓] tcp:0.0.0.0:53 is in use by pihole-FTL
No it’s there, my copy paste from terminal is just wider than yours for some reason. As for my ruleset that is empty, and I don’t have iptables on this pi.
I have rebooted the RPI and everything is in working order, again. When the dashboard goes down again in a few days, I’ll enable logging to webserver.log again and try sudo nc -nzvw5 192.168.1.199 80 to see if it shows up in it.
