@FredsFred what about you?
Sorry too quick for my own good, the problem has recurred so the python script does not seem to be the culprit. Everything was working OK while I had the pi-hole dashboard running in my browser. I closed the browser down went for a coffee, started browser and dashboard again and lo and behold same problem!!
What I have noticed is that it is always the same top domain.
Hostname: pi3 (Raspberry Pi 3, Model B)
Uptime: 00:43:33 (UV, TT)
Task Load: 0.65 1.21 1.16 (Active: 8 of 92 tasks)
CPU usage: 13% (4x 1.2 GHz @ 318k)
RAM usage: 38% (Used: 355 MB of 927 MB)
HDD usage: 13% (Used: 7 GB of 58 GB)
LAN addr: 192.168.0.81 (Gateway: 192.168.0.1)
Pi-hole: Active (Blocking: 121693 sites)
Ads Today: 0% (Total: 0 of 2)
Local Qrys: 100% (4 DNS servers)
Blocked:
Top Advert:
Top Domain: 3.debian.pool.ntp.org
Top Client: localhost
I think it is the NTP service that is somehow causing this problem, maybe you can see @FredsFred if it is the same for you?
here is my syslog - looks like NTP is run then some cleanups on pi-hole. Does this look OK?
2018 (1): Starting
Mar 23 13:25:18 pi3 ntpd[18707]: Command line: ntpd -q
Mar 23 13:25:18 pi3 ntpd[18707]: must be run as root, not uid 1000
Mar 23 13:26:17 pi3 systemd[1]: Starting Cleanup of Temporary Directories...
Mar 23 13:26:17 pi3 systemd[1]: Started Cleanup of Temporary Directories.
Mar 23 13:30:01 pi3 CRON[31719]: (root) CMD ( PATH="$PATH:/usr/local/bin/" pihole updatechecker local)
Mar 23 13:39:01 pi3 CRON[25008]: (root) CMD ( [ -x /usr/lib/php/sessionclean ] && if [ ! -d /run/systemd/system ]; then /usr/lib/php/sessionclean; fi)
Mar 23 13:39:17 pi3 systemd[1]: Starting Clean php session files...
Mar 23 13:39:17 pi3 systemd[1]: Started Clean php session files.
Mar 23 13:40:01 pi3 CRON[28052]: (root) CMD ( PATH="$PATH:/usr/local/bin/" pihole updatechecker local)
Mar 23 13:50:01 pi3 CRON[26204]: (root) CMD ( PATH="$PATH:/usr/local/bin/" pihole updatechecker local)
Mar 23 14:00:01 pi3 CRON[23872]: (root) CMD ( PATH="$PATH:/usr/local/bin/" pihole updatechecker local)
I cleaned out the cache on pihole and on my devices and have no change. I set this router up in November and installed pihole then and the admin console worked. I only got around to installing at the beginning of March and I ran an update then.
I took a look at my syslog running grep ntp /var/log/syslog and could not see anything related to NTP in the log.
I don't think it has anything to do with NTP. Here's what we know:
- Pi-hole is working for all clients
-
pihole -tshows activity from all clients - The Web interface only shows two clients
So Pi-hole is working, but the issue seems to be with the Web interface, so I still feel like it's cache. Do you have a different device to try to access the interface from?
I have tried multiple browsers on different machines and it is the same for all. A reboot of the system and everything works fine, then something triggers the web interface to fail. PiHole keeps working but web interface shows 2 clients and 0 ads blocked.
So logically I think some CRON or other scheduled job is causing this. I will see if I can shed anymore light on monday
Dave
I too have tried it from multiple devices with the same effect,
Here isy new debug token
bd9wurax0w
Thanks
I'm stumped. I can't determine why only two clients would show up. Which two devices are they?
I have reconfigured my rtc and hwclock (removed fake-hwclock) after doing some research on how to install rtc on pi. After doing this I now get no initial grace period on bootup where everything shows correctly, stranger still I only get 1 client (localhost) and 2 queries on debian timeservers.
I also uninstalled pi-hole and reinstalled with no joy. Very odd..
So localhost is one of the clients and you don't have any other?
My localhost and one of my phones. It may be worth mentioning I am running UniFi controller on there as well.
I would say I have at least 10 devices using it has a DNS server but the queries seem to be from when I originally set up the router and since the upgrade nothing is updating on the interface.
Did you clear cache on the device you are accessing the Web interface from?
I have tried that and tried accessing it from every different device I have that has a web interface. I've cleaned the cache on the pihole and the devices trying to access the interface. The admin console still shows the same statistics 1.7% Blocked, 2018 Total queries etc. on all devices that have tried to access the console. The only thing that seems to be updating is the CPU and Memory indicator.
I don't know. You can try the nuclear option (this will wipe out any customizations you have in the Web interface); it fixed an similar issue for someone else.
I'm not a huge fan of it, but it may get you up and running since we don't really know what the problem is.
sudo rm -rf /var/www/html/admin
sudo git clone https://github.com/pi-hole/AdminLTE.git /var/www/html/admin
Just ran through that with no change either.
The only thing I can think of that would be messing with it is UniFI controller or the update I ran in March but I figured a repair would fix that.
@FredsFred I solved this for my system by doing a clean reinstall. I uninstalled pi-hole removed dnsmasq and lighttpd. I then went through deleting all folders and files related to pi-hole , php, dnsmasq and lighttpd including logs (important as just purging did not work for me). Then reinstall pi-hole. I appreciate this may not be practical for you as other systems may rely on these dependencies.
Hopefully you will find a good solution.
In my case my best guess is that I had a bad timestamp from badly configured hwclock or corrupted file after log was flushed. These being the only two items of interest on debug logs I found.
1 Like
That worked! After completely cleaning everything out the admin interface is now updating correctly!
Thanks for the help guys!
1 Like
This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.