I can't be the ONLY one.. PiHole not getting requests, router set to route DNS to PiHole, websites still load

Did you enable by chance the DHCP within the pi-hole ?

A DHCP server failure is most likely due to a config error (a wrong parameter saved in one of the config files). Upon restart, the dhcp server doesn't start because of the said config error.

Maybe the OWRT’s web interface is writing the config with a bad string.

What you can do to try the Pi-hole blocking is to run a ping on a known blocked domain (like flurry.com) from the device "hosting" pi-hole.

You can also try to set the Pi-hole IP as your sole DNS on one of your clients.
Test it like that.

If it works, you isolated the problem to the router.

What you could do in this case is disable your router's LAN DHCP and use Pi-hole as your DHCP server.

That way, the DNS setting and hence the ad-blocking, will be pushed automatically to all your connecting clients, removing the need of you actually having to set the DNS manually on them.

DHCP is NOT enabled on the Pi-hole device, let my try pinging the known-blocked domain and seeing if it works... ok, so that was blocked, but here's a REALLY interesting thing... Pi-Hole seems too have retained it's old IP in config! I installed pi-hole while on another network and migrated it too my home network, so it's got a completely wrong IP!

Check /etc/dhcpcd.conf at the end of the file, that may need to be changed with a change in network numbering.

would that be /etc/dhcp.conf on the router or the pi-hole device?

On the Pi-hole device, that's where the static IP address is set. If you want to remove the static IP and instead set a reserved lease on the router, you can remove the lines you find. Otherwise just edit the file to the proper IP and reboot the device/restart networking.

ok so that half-sorted the issue... i'm still seeing the device try to resolve to it's old 10.x.x.x address, but now it's not responding to it's own ping... hmm

To solve the resolution to address, run pihole -r and reconfigure to the correct IP address. That should put the proper IP in to the hosts file and restarts the resolver. Not sure about the ping issue though.

well... crap...
[i] Checking for existing FTL binary...
[✗] Downloading and Installing FTL
Error: Unable to get latest release location from GitHub
[✗] FTL Engine not installed

I Broke it... again :\

Can you ping github.com from the device? You might need to temporarily set /etc/resolv.conf to use something like 8.8.8.8 instead of 127.0.0.1 to kickstart the Pi-hole update process if dnsmasq isn't currently working or running.

that was it! ok onwards and forwards! straight into a brick wall, now pihole isn't loading at all, waiting on a reboot. :frowning:

ugh.. deep magic voodoo dns crap... it STILL thinks it's 10.x.x.x... I reconfigured and rebooted..

The other place to check is /etc/network/interfaces but that's not a file we modify. It shouldn't have any IP information in there, however it's something to check just to make sure. And a head 10 /etc/pihole/gravity.list will show you the IP address that Pi-hole is using to redirect to.

/etc/network/interfaces shows my hand-made edits to give the device a static ip, and head /etc/pihole/gravity.list` shows 10.x.x.x 0.r.msn.com... etc... (head 10 threw an error) in anycase, I don't really mind that it's not resolving to the "right" ip, it's not resolving to the RIGHT ip, that's what counts!

now... onto the rest of this, even though when I ping 0.r.msn.com it throws the request into the blackhole, but this request doesn't show in the panel.. and I have logging enabled so I know the resolver is doing what it's supposed to, now why is it not showing the requests in the stats? :thinking:

Can you either open the web admin panel and tail the logs via that or sudo tail -f /var/log/pihole.log when you ping a blocked domain? That will show if there are entries being added. Also if you have a connection, you can run pihole -d and we can look to see what the system looks like if you allow it to upload the logs. Otherwise you can read through /var/log/pihole_debug.log for a view of what we'd look at.

sorry about the long delay, real life happened, so here's what I've gathered so far..

I left this running and reopened the log while running a ping against flurry.com

$ sudo tail -f /var/log/pihole.log
Apr 24 17:17:14 dnsmasq[749]: started, version 2.72 cachesize 10000
Apr 24 17:17:14 dnsmasq[749]: compile time options: IPv6 GNU-getopt DBus i18n IDN DHCP DHCPv6 no-Lua TFTP conntrack ipset auth DNSSEC loop-detect
Apr 24 17:17:14 dnsmasq[749]: warning: ignoring resolv-file flag because no-resolv is set
Apr 24 17:17:14 dnsmasq[749]: using nameserver 8.8.4.4#53
Apr 24 17:17:14 dnsmasq[749]: using nameserver 8.8.8.8#53
Apr 24 17:17:14 dnsmasq[749]: read /etc/hosts - 6 addresses
Apr 24 17:17:14 dnsmasq[749]: read /etc/pihole/local.list - 2 addresses
Apr 24 17:17:14 dnsmasq[749]: failed to load names from /etc/pihole/black.list: No such file or directory
Apr 24 17:17:26 dnsmasq[749]: read /etc/pihole/gravity.list - 121666 addresses

here's the token for tricorder: 8o758ydtge interestingly it looks like the FTLDNS service isn't launching correctly at boot.

Nope you ar e not the only one... I've been battling similiar weird problems, but since I didn't do anything to the system, I can't help but wonder if my server for pi-hole hasn't been hacked somehow, or if perhaps someone has developed a pi-hole destroyer type ad? I don't know really, just guessing randomly at this point. I think I am going to completely reinstall ubuntu from the ground up. It was all working perfectly until a few days ago, since then I have no clue what has happened.

You are not on the FTLDNS beta, and are experiencing this issue:

3 Likes

well that was an easy enough fix! wow, ok that works now... and it appears that pi-hole is now blocking things correctly so that's fixed! thank you!

Sorry I havent been on here lately. I am glad you got it fixed though.

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.