Something is killing Pihole DNS resolution

Help
1

Please follow the below template, it will help us to help you!

If you are Experiencing issues with a Pi-hole install that has non-standard elements (e.g you are using nginx, apache2 or another reverse proxy, or there is some other aspect of your install that is customised) - please use the Community Help category.

Expected Behaviour:

Setup:

OCI Micro Instance running Ubuntu 24.04

1 GB of RAM

Unbound 1.19.2

Wireguard (Split Tunnel)

Updated to the latest Pihole today by running sudo pihole -up

Expected Pihole to continue working normally

This Pihole instance is my secondary instance, with my primary instance running on a Raspberry Pi 4B locally, only on my LAN. This local instance is synced with the OCI using nebula-sync. I did modify the settings of the local instance to enable the “Advertise DNS server multiple times” option in DHCP. But that should not affect the OCI instance as DHCP is disabled on OCI instance AND DHCP config sync is “off” in nebula-sync. I’ve rechecked and the DHCP settings are not enabled on the OCI instance.

Actual Behaviour:

Since the update, twice the DNS server has gone down. The login page
displayed an error message stating that DNS server has failed
(Unfortunately, I did not take a screenshot or save the message.

I saw a similar error twice as the screenshot in the first post of this thread - Pi-hole v6 "DNS server failure" . This error hasn’t yet reappeared but the pihole behaviour was the same before this error appeared the previous two times - No DNS resolution.

Each time, the error went away when I restarted Pihole.

A similar ticket filed (But not the same as what’s happening with me):

On my local windows laptop, with Wireguard connection active (to use the OCI instance as the DNS server), I ran nslookup on a random domain and that failed as well:

PS C:\Users\AJ> nslookup croma.com
DNS request timed out.
timeout was 2 seconds.
Server: UnKnown
Address: 10.10.0.1

DNS request timed out.
timeout was 2 seconds.
DNS request timed out.
timeout was 2 seconds.
DNS request timed out.
timeout was 2 seconds.
DNS request timed out.
timeout was 2 seconds.
*** Request to UnKnown timed-out
PS C:\Users\AJ>

Debug Token:

Debug token is: https://tricorder.pi-hole.net/GrRdf4GG/

2

Uploading a teleporter backup will restart the DNS server:

2025-11-28 00:30:00.747 UTC [2026/T277404] INFO: Restarting FTL: Teleporter (ZIP) import
   2025-11-28 00:30:00.747 UTC [2026M] INFO: Asked to terminate by "/usr/bin/pihole-FTL -f" (PID 2026, user pihole UID 999)
   2025-11-28 00:30:00.789 UTC [2026/T277024] INFO: Terminating timer thread
   2025-11-28 00:30:00.818 UTC [2026/T277021] INFO: Terminating database thread
   2025-11-28 00:30:00.866 UTC [2026/T277059] INFO: No config changes detected
   2025-11-28 00:30:01.062 UTC [2026/T277022] INFO: Terminating GC thread
   2025-11-28 00:30:01.118 UTC [2026M] INFO: Finished final database update
   2025-11-28 00:30:01.119 UTC [2026M] INFO: Waiting for threads to join
   2025-11-28 00:30:01.119 UTC [2026M] INFO: Thread dns-client (2) is idle, terminating it.
   2025-11-28 00:30:01.119 UTC [2026M] INFO: Thread ntp-client (4) is idle, terminating it.
   2025-11-28 00:30:01.119 UTC [2026M] INFO: Thread webserver (7) is idle, terminating it.
   2025-11-28 00:30:01.119 UTC [2026M] INFO: All threads joined
   2025-11-28 00:30:01.119 UTC [2026M] INFO: PID file emptied
   2025-11-28 00:30:01.120 UTC [2026M] INFO: Stored 0 API sessions in the database
   2025-11-28 00:30:01.958 UTC [2026M] INFO: ########## FTL terminated after 59m 59s  (internal restart)! ##########

It appears that Pi-hole has been configured to use itself as the upstream, that would cause a loop that would eventually crash the DNS.

2025-11-28 08:33:45.909 UTC [288035/T289667] INFO: NTP server listening on 0.0.0.0:123 (IPv4)
   2025-11-28 08:33:45.909 UTC [288035/T289668] INFO: NTP server listening on :::123 (IPv6)
   2025-11-28 08:33:46.561 UTC [289646/T289657] INFO: Received 8/8 valid NTP replies from in.pool.ntp.org
   2025-11-28 08:33:46.561 UTC [289646/T289657] INFO: Time offset: -1.826027e+01 ms (excluded 1 outliers)
   2025-11-28 08:33:46.561 UTC [289646/T289657] INFO: Round-trip delay: 1.182692e+01 ms (excluded 1 outliers)
   2025-11-28 08:33:46.561 UTC [289646/T289670] ERROR: Error NTP server: Cannot bind to IPv6 address :::123 (Address in use), IPv6 NTP server not available
   2025-11-28 08:33:46.561 UTC [289646/T289669] ERROR: Error NTP server: Cannot bind to IPv4 address 0.0.0.0:123 (Address in use), IPv4 NTP server not available
   2025-11-28 08:33:47.549 UTC [288035/T289650] ERROR: Cannot receive UDP DNS reply: Timeout - no response from upstream DNS server
   2025-11-28 08:33:47.549 UTC [288035/T289650] INFO: Tried to resolve PTR "3.0.10.10.in-addr.arpa" on 127.0.0.1#53 (UDP)
   2025-11-28 08:33:49.597 UTC [288035/T289650] ERROR: Cannot receive UDP DNS reply: Timeout - no response from upstream DNS server
   2025-11-28 08:33:49.597 UTC [288035/T289650] INFO: Tried to resolve PTR "11.0.10.10.in-addr.arpa" on 127.0.0.1#53 (UDP)
   2025-11-28 08:33:51.645 UTC [288035/T289650] ERROR: Cannot receive UDP DNS reply: Timeout - no response from upstream DNS server
   2025-11-28 08:33:51.645 UTC [288035/T289650] INFO: Tried to resolve PTR "6.0.10.10.in-addr.arpa" on 127.0.0.1#53 (UDP)
   2025-11-28 08:33:53.693 UTC [288035/T289650] ERROR: Cannot receive UDP DNS reply: Timeout - no response from upstream DNS server
   2025-11-28 08:33:53.693 UTC [
3

Looks like a port conflict with the web interface as well, could there be two FTL binaries at the same time?

[2025-11-28 08:30:07.519 UTC 289599] cannot bind to 10.10.0.1:80: 98 (Address in use)
   [2025-11-28 08:30:07.519 UTC 289599] cannot bind to 443s: 98 (Address in use)
   [2025-11-28 08:30:07.519 UTC 289599] Failed to setup server ports
   [2025-11-28 08:33:41.490 UTC 288035] Initializing HTTP server on ports "10.10.0.1:80, 443s"
   [2025-11-28 08:33:42.409 UTC 289646] Initializing HTTP server on ports "10.10.0.1:80, 443s"
   [2025-11-28 08:33:42.410 UTC 289646] cannot bind to 10.10.0.1:80: 98 (Address in use)
   [2025-11-28 08:33:42.410 UTC 289646] cannot bind to 443s: 98 (Address in use)
   [2025-11-28 08:33:42.410 UTC 289646] Failed to setup server ports
   [2025-11-28 09:26:29.990 UTC 288035] Authentication required, redirecting to /admin/login
5

But why is this needed? I’m not entirely clear. What is causing the DNS server/FTL to crash?

It appears that Pi-hole has been configured to use itself as the
upstream, that would cause a loop that would eventually crash the DNS.

I double checked and the only DNS server was unbound:

pihole-oci-dns-config-2025-11-29_10-08
pihole-oci-dns-config-2025-11-29_10-08809×1505 90.9 KB

Looks like a port conflict with the web interface as well, could there be two FTL binaries at the same time?

I’m pretty sure there isn’t, as I only installed Pihole once. Is it possible for me to check somehow? Is it possible that the latest update resulted in two pihole binaries somehow?

6

You are loading in a new configuration, FTL needs to restart in order to use the new configuration.

There are two instances running, one as root and one as pihole, the timestamps show they are both active:

*** [ DIAGNOSING ]: contents of /dev/shm
total 11M
-rw------- 1 pihole pihole 340K Nov 28 08:33 FTL-288035-clients
-rw------- 1 pihole pihole 4.0K Nov 28 08:33 FTL-288035-clients-lookup
-rw------- 1 pihole pihole  344 Nov 28 08:33 FTL-288035-counters
-rw------- 1 pihole pihole  48K Nov 28 08:33 FTL-288035-dns-cache
-rw------- 1 pihole pihole 8.0K Nov 28 08:33 FTL-288035-dns-cache-lookup
-rw------- 1 pihole pihole  52K Nov 28 08:33 FTL-288035-domains
-rw------- 1 pihole pihole  16K Nov 28 08:33 FTL-288035-domains-lookup
-rw------- 1 pihole pihole 556K Nov 28 08:33 FTL-288035-fifo-log
-rw------- 1 pihole pihole  120 Nov 28 08:33 FTL-288035-lock
-rw------- 1 pihole pihole 8.0K Nov 28 08:33 FTL-288035-overTime
-rw------- 1 pihole pihole 4.0K Nov 28 08:33 FTL-288035-per-client-regex
-rw------- 1 pihole pihole 3.1M Nov 28 08:39 FTL-288035-queries
-rw------- 1 pihole pihole 768K Nov 28 08:33 FTL-288035-recycler
-rw------- 1 pihole pihole  144 Nov 28 08:33 FTL-288035-settings
-rw------- 1 pihole pihole 120K Nov 28 08:33 FTL-288035-strings
-rw------- 1 pihole pihole  28K Nov 28 08:33 FTL-288035-upstreams
-rw------- 1 root   root   340K Nov 28 08:33 FTL-289599-clients
-rw------- 1 root   root   4.0K Nov 28 08:33 FTL-289599-clients-lookup
-rw------- 1 root   root    344 Nov 28 08:33 FTL-289599-counters
-rw------- 1 root   root    48K Nov 28 08:33 FTL-289599-dns-cache
-rw------- 1 root   root   8.0K Nov 28 08:33 FTL-289599-dns-cache-lookup
-rw------- 1 root   root    52K Nov 28 08:33 FTL-289599-domains
-rw------- 1 root   root    16K Nov 28 08:33 FTL-289599-domains-lookup
-rw------- 1 root   root   556K Nov 28 08:33 FTL-289599-fifo-log
-rw------- 1 root   root    120 Nov 28 08:33 FTL-289599-lock
-rw------- 1 root   root   8.0K Nov 28 08:33 FTL-289599-overTime
-rw------- 1 root   root   4.0K Nov 28 08:33 FTL-289599-per-client-regex
-rw------- 1 root   root   3.1M Nov 28 08:39 FTL-289599-queries
-rw------- 1 root   root   768K Nov 28 08:33 FTL-289599-recycler
-rw------- 1 root   root    144 Nov 28 08:33 FTL-289599-settings
-rw------- 1 root   root   120K Nov 28 08:33 FTL-289599-strings
-rw------- 1 root   root    28K Nov 28 08:33 FTL-289599-upstreams

The number value after FTL is the process ID.

7

No.

We are not saying there are 2 different binaries, but there are 2 instances (of the same binary) running at the same time on your machine.

Suggestion:

Restart the machine and check again. If you still have issues after the restart, generate a new debug log.

8

You are loading in a new configuration, FTL needs to restart in order to use the new configuration.

Sorry, I did not understand. Where and how am I loading a new configuration? That’s why I’m confused about your statement of uploading a teleporter backup. Do you mean the sync using nebula-sync?

Restart the machine and check again. If you still have issues after the restart, generate a new debug log.

Noted. I’ll try this and observe for a day or two.

9

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