PIHOLE works, until the ads come through and internet stops and i have to reboot

Expected Behaviour:

Blocking ads and handout ipadresses through DHCP

Core** [v6.1.2]
FTL** [v6.2.3]
Web interface** [v6.2.1]
RaspberryPi4

Actual Behaviour:

Internet stops after a while (days) , like if DHCP doesnt work anymore

  1. dnsmasq warning Ignoring query from non-local network
  2. dnsmasq warning: not using configured address 192.168.178.137 because it is in use by the server or relay
  3. https://i.imgur.com/RogFlCx.png here you see PIHOLE stops working after 23:00

nr. 2 is new today, never seen before. Pihole is set to act as DHCP with 192.168.178.137 as fixed adress......and helptopic says this is because i ran pihole -d so it will pop up. this is TRUE. so we can IGNORE this, i believe .
excuse my bad everything, i am not native english nor very good wizzkid but retired MCSE (windowsXP)

Debug Token:

/w3Wtz2nY/

In this forum, you can paste images directly into a post.

Your image shows a period of time where your Pi-hole hasn't received any DNS requests. This could happen if Pi-hole would be by-passed.

And indeed, your debug log shows that your (Sagemcom?) router is advertising a set of public IPv6 addresses as DNS servers:

   * Received 136 bytes from fe80::3a<redacted>5 @ eth0
     Hop limit: 64
     Stateful address conf.: No
     Stateful other conf.: Yes
     Mobile home agent: No
     Router preference: Medium
     Neighbor discovery proxy: No
     Router lifetime: 1800 s
     Reachable time: N/A
     Retransmit time: N/A
     Recursive DNS server 1/3: 2001:b88:1002::10
     Recursive DNS server 2/3: 2001:b88:1202::10
     Recursive DNS server 3/3: 2001:730:3e42:1000::53
     DNS server lifetime:300 sec
     - Prefix: 2001:<redacted>::/64
       Valid lifetime: 604800 sec
       Preferred lifetime: 604800 sec
       On-link: Yes
       Autonomous address conf.: Yes
     - Route: 2001:<redacted>::/56
       Route preference: Medium
       Route lifetime: 1800 sec
     Source link-layer address: 3<redacted>5

You'd have to find a way to configure your router to stop advertising IPv6 DNS server addresses, or to advertise your Pi-hole host machine's IPv6.

You'd have to consult your router's documentation sources on further details for its IPv6 configuration options.

If your router doesn't support configuring IPv6 DNS, you could consider disabling IPv6 altogether, provided you'd not depend on IPv6 for reasons.

If your router doesn't support that either, your IPv6-capable clients will always be able to bypass Pi-hole via IPv6.

If your router would only advertise its own IPv6 address, you could then try to mitigate this, by setting Pi-hole as the only upstream of your router, provided your router supports it.
But note that you won't be able to attribute DNS requests to original individual IPv6 clients in such a configuration.

EDIT: In your case, your router seems to advertises public DNS servers, so that part won't work.

Thankyou very much for your answer. I will look into my router to find the problem there. Indeed it all started with the upgrade to IPv6.

I looked into my ziggo modem and find the DNS page. It was set to "use the standard DNS (recommended)" . I tried to put in my IP of the PIHOLE (192.168.178.137) but the system refused it so I changed it into the numbers of OPENDNS.

In PIHOLE i used the same DNS of OPENDNS, and changed the setting from "allow only local" to "respond only on Eth0).

And the moment i changed that my system removed the yellow numbers after settings, indicatiing there are no problems.
Lets hope it stays that way for a week (DHCP lease is set to a week)

Your screenshot seems to show your router's upstream DNS server configuration.

You'd need adjust its local DNS configuration, i.e. NDP/RA/RDNSS related options for IPv6 (and DHCP ones for IPv4).

Thankyou for your answer. I dont think we can do that in our providers router.
All i got is DDNS. That is disabled. and if i enabled it , it will need password and username , so i dont think that is the right one.

You could contact Ziggo and ask them if its possible for them to disable advertising IPv6 DNS IP's on your LAN.
Some other users were able to with their ISP.
DHCPv6 should be disabled anyway as it's not supported on some devices like for example Android.

At your end you can check what Recursive DNS Servers (RDNSS) are advertised via the IPv6 Neighbor Discovery Protocol & Router Advertisement (NDP/RA) on your LAN if run below one on the Pi:

sudo pihole-FTL dhcp-discover

This is completely unrelated. Please keep DDNS disabled.

It's friday again and a week has passed. My pihole stopped working again, couldnt access it through it's IP nor through it's REALVNC.
I don't get it why it keeps hanging up. His IP adress is fixed, and coinsedence the DHCP lease time is a week. It all makes no sense.
Are there RaspberryPI logs i can upload?

it answers this:

$ sudo pihole-FTL dhcp-discover             
Scanning all your interfaces for DHCP servers and IPv6 routers
Timeout: 6 seconds

* Received 120 bytes from fe80::XXXX @ eth0
  Hop limit: 64
  Stateful address conf.: No
  Stateful other conf.: Yes
  Mobile home agent: No
  Router preference: Medium
  Neighbor discovery proxy: No
  Router lifetime: 1800 s
  Reachable time: N/A
  Retransmit time: N/A
  Recursive DNS server 1/2: 2620:119:35::35
  Recursive DNS server 2/2: 2620:119:53::53
  DNS server lifetime:300 sec
  - Prefix: 2001:XXXX::/64
    Valid lifetime: 604800 sec
    Preferred lifetime: 604800 sec
    On-link: Yes
    Autonomous address conf.: Yes
  - Route: 2001:XXXX::/56
    Route preference: Medium
    Route lifetime: 1800 sec
  Source link-layer address: 38:17:B1:XX:XX:XX

* Received 136 bytes from fe80::XXXX @ eth0
  Hop limit: 64
  Stateful address conf.: No
  Stateful other conf.: Yes
  Mobile home agent: No
  Router preference: Medium
  Neighbor discovery proxy: No
  Router lifetime: 1800 s
  Reachable time: N/A
  Retransmit time: N/A
  - Prefix: 2001:XXXX::/64
    Valid lifetime: 604800 sec
    Preferred lifetime: 604800 sec
    On-link: Yes
    Autonomous address conf.: Yes
  MTU: 1500 bytes (valid)
  Source link-layer address: B8:27:EB:XX:XX:XX
  Recursive DNS server 1/4: 2001:XXXX
  Recursive DNS server 2/4: 2001:XXXX
  Recursive DNS server 3/4: fe80::XXXX
  Recursive DNS server 4/4: fe80::XXXX
  DNS server lifetime:604800 sec

* Received 313 bytes from 192.168.178.137 @ eth0
  Offered IP address: 192.168.178.194
  Server IP address: 192.168.178.137
  Relay-agent IP address: N/A
  BOOTP server: (empty)
  BOOTP file: (empty)
  DHCP options:
   Message type: DHCPOFFER (2)
   server-identifier: 192.168.178.137
   lease-time: 604800 ( 7d )
   renewal-time: 302400 ( 3d 12h )
   rebinding-time: 529200 ( 6d 3h )
   netmask: 255.255.255.0
   broadcast: 192.168.178.255
   domain-name: "lan"
   hostname: "pihole"
   dns-server: 192.168.178.137
   dns-server: 192.168.178.137
   dns-server: 192.168.178.137
   router: 192.168.178.1
   --- end of options ---

Received 1 DHCP (IPv4) and 2 RA (IPv6) answers on eth0

How did you set this IP address as static and what is your DHCP address pool range?

1 Like

I think we got the problem there. the pool starts at 20 and i made an exclusion for 137....
So now i changed the IPadres of PIHOLE to 17, so outside the range. rebooted, and now i can't find it anymore. not at 17 nor at 137.
i feel so stupid. I think i will start over fresh tonight with a new clean installation and outside the DHCP range, and write down the logincredentials. a donkey dont hit his head twice on the same rock.

Thank you all for helping, thinking with me.

Thats not necessary I think.

Below is an IPv6 RA coming from the router with MAC 38:17:B1:XX:XX:XX:

$ dig +short -x 2620:119:35::35
dns.opendns.com.
resolver1.opendns.com.
dns.sse.cisco.com.
dns.umbrella.com.

Below is an IPv6 RA coming from the Pi with MAC B8:27:EB:XX:XX:XX

And below is an IPv4 DHCP reply coming from the Pi with IP 192.168.178.137:

Clients prefer IPv6 over IPv4 so those DNS IP's advertised by the router via IPv6 RA will allow the clients to bypass Pi-hole.
Thats an issue that I believe can only be resolved by disabling IPv6 support on the router ... if I search other Ziggo related issues.
Preferably on the LAN side only!
Or:

The other issue of not being able to connect to the Pi anymore is most likely related to below if its configured to do DHCP services for your LAN like in the dhcp-discover output:

Hint:

EDIT: Ps its a crappy job by Ziggo anyway when the clients need to query an IPv6 DNS server thats not in their own LAN segment but external on the Internet instead.
And those IPv6 DNS servers seem to be not on the own Ziggo network as well.
Most routers redirect queries to their own IPv6 address to allow caching for all of your LAN clients.
LAN is quick.
And if it needs to go to WAN/Internet is slow.

Thank you. I already set my static IPv4 address on the interface (in /etc/network/interfaces.d)

I was wondering how the OP had set it, at the OS or as an IP address reservation. I was thinking that maybe the static address was with the DHCP pool.

I would not recommend that on Raspbian/Pi-OS Bookworm.
Except maybe if you kicked NetworkManager from the Pi.

I use plain Debian, currently Trixie.

Well the OP didnt mention what distro so I'm assuming Pi-OS :wink:

1 Like

Ps, you do know that Trixie is still in testing?
Not ready yet to go live for a DNS server.

Yep. I went early for Bookwork too. It's been running fine for weeks hosting my docker containers, including Pi-Hole.

If it had all gone sideways, I could easily have switched DNS back to my OPNsense box until I reflashed the SD card.

1 Like

We need some folks scarifying to get past the testing phase :wink: