PiHole Web Interface keeps going back to an older version

gh0sti commented 10 months ago

This is a: Run Issue

Details

Web Interface will revert to an older version of 5.18 when the latest web interface is 5.21. I am using docker build 2023.11.0. I have to consistently run update checker to have the web interface update to show the latest version when signing in.

Related Issues

[x] I have searched this repository/Pi-hole forums for existing issues and pull requests that look similar

How to reproduce the issue

Environment data
- Operating System: root@raspberrypi:/home/pi# cat /etc/os-release PRETTY_NAME="Raspbian GNU/Linux 12 (bookworm)" NAME="Raspbian GNU/Linux" VERSION_ID="12" VERSION="12 (bookworm)" VERSION_CODENAME=bookworm ID=raspbian ID_LIKE=debian
- Hardware: Raspberry Pi 4
- Kernel Architecture: Arm 32bit
- Docker Install Info and version:
- Software source: 2023.11.0 Official Docker Github
- Supplimentary Software: Portainer 2.19.4
- Hardware architecture:
docker-compose.yml contents, docker run shell command, or paste a screenshot of any UI based configuration of containers here

version: "3"

services:
  pihole:
    container_name: pihole
    image: pihole/pihole:latest
    hostname: homepi  
    ports:  
     - "54:53/tcp"  
     - "54:53/udp"    
     - "85:80/tcp"    
    volumes:  
     - './etc-pihole:/etc/pihole'  
     - './etc-dnsmasq.d:/etc/dnsmasq.d'

any additional info to help reproduce

These common fixes didn't work for my issue

[ ] I have tried removing/destroying my container, and re-creating a new container
[ ] I have tried fresh volume data by backing up and moving/removing the old volume data
[x] I have tried running the stock docker run example(s) in the readme (removing any customizations I added)
[ ] I have tried a newer or older version of Docker Pi-hole (depending what version the issue started in for me)
[ ] I have tried running without my volume data mounts to eliminate volumes as the cause

If the above debugging / fixes revealed any new information note it here. Add any other debugging steps you've taken or theories on root cause that may help.

rdwebdesign commented 10 months ago

When this happens again, DO NOT run updatechecker and generate a debug log.

gh0sti commented 10 months ago

When this happens again, DO NOT run updatechecker and generate a debug log.

Here is the screen shot of it reverting. I will generate a debug log next

gh0sti commented 10 months ago

https://tricorder.pi-hole.net/CaJWYFA2/

yubiuser commented 10 months ago

This is interesting. It does not only revert the web repo, but all Pi-hole components, except the docker tag.

** [ DIAGNOSING ]: Core version
[✓] Version: v5.15
[i] Remotes: origin https://github.com/pi-hole/pi-hole (fetch)
             origin https://github.com/pi-hole/pi-hole (push)
[i] Branch: master
[i] Commit: v5.15-0-g9048429b

*** [ DIAGNOSING ]: Web version
[✓] Version: v5.18.1
[i] Remotes: origin https://github.com/pi-hole/AdminLTE.git (fetch)
             origin https://github.com/pi-hole/AdminLTE.git (push)
[i] Branch: master
[i] Commit: v5.18.1-0-g30668d4d

*** [ DIAGNOSING ]: FTL version
[✓] Version: v5.20.1
[i] Branch: master
[i] Commit: 47128aaf

rw-r--r-- 1 www-data pi 379 Dec 30 14:21 /etc/pihole/versions
   CORE_BRANCH=master
   WEB_BRANCH=master
   FTL_BRANCH=master
   CORE_VERSION=v5.15
   WEB_VERSION=v5.18.1
   FTL_VERSION=v5.20.1
   GITHUB_CORE_VERSION=v5.17.2
   GITHUB_WEB_VERSION=null
   GITHUB_FTL_VERSION=v5.23
   CORE_HASH=9048429b
   GITHUB_CORE_HASH=60b6a101
   WEB_HASH=30668d4d
   GITHUB_WEB_HASH=be05b0f6
   FTL_HASH=47128aaf
   GITHUB_FTL_HASH=1a114133
   DOCKER_VERSION=2023.11.0
   GITHUB_DOCKER_VERSION=2023.11.0

What is also strange is the already taken port

2023-11-28 04:14:46.891 4249M] Successfully accessed setupVars.conf
   [2023-11-28 04:14:46.893 4249M] FATAL ERROR in dnsmasq core: failed to create listening socket for port 53: Address already in use
   [2023-11-28 04:14:46.915 4249M] ########## FTL terminated after 107ms  (code 1)! ##########

There is no way to downgrade the components within the container so I suspect an issue with Portainer. How did you upgrade the image to the latest version? Did you remove the container completely?

gh0sti commented 10 months ago

This is interesting. It does not only revert the web repo, but all Pi-hole components, except the docker tag.


** [ DIAGNOSING ]: Core version

[✓] Version: v5.15

[i] Remotes: origin   https://github.com/pi-hole/pi-hole (fetch)

             origin   https://github.com/pi-hole/pi-hole (push)

[i] Branch: master

[i] Commit: v5.15-0-g9048429b

*** [ DIAGNOSING ]: Web version

[✓] Version: v5.18.1

[i] Remotes: origin   https://github.com/pi-hole/AdminLTE.git (fetch)

             origin   https://github.com/pi-hole/AdminLTE.git (push)

[i] Branch: master

[i] Commit: v5.18.1-0-g30668d4d

*** [ DIAGNOSING ]: FTL version

[✓] Version: v5.20.1

[i] Branch: master

[i] Commit: 47128aaf

rw-r--r-- 1 www-data pi 379 Dec 30 14:21 /etc/pihole/versions

   CORE_BRANCH=master

   WEB_BRANCH=master

   FTL_BRANCH=master

   CORE_VERSION=v5.15

   WEB_VERSION=v5.18.1

   FTL_VERSION=v5.20.1

   GITHUB_CORE_VERSION=v5.17.2

   GITHUB_WEB_VERSION=null

   GITHUB_FTL_VERSION=v5.23

   CORE_HASH=9048429b

   GITHUB_CORE_HASH=60b6a101

   WEB_HASH=30668d4d

   GITHUB_WEB_HASH=be05b0f6

   FTL_HASH=47128aaf

   GITHUB_FTL_HASH=1a114133

   DOCKER_VERSION=2023.11.0

   GITHUB_DOCKER_VERSION=2023.11.0

What is also strange is the already taken port


2023-11-28 04:14:46.891 4249M] Successfully accessed setupVars.conf

   [2023-11-28 04:14:46.893 4249M] FATAL ERROR in dnsmasq core: failed to create listening socket for port 53: Address already in use

   [2023-11-28 04:14:46.915 4249M] ########## FTL terminated after 107ms  (code 1)! ##########

There is no way to downgrade the components within the container so I suspect an issue with Portainer. How did you upgrade the image to the latest version? Did you remove the container completely?

I use watchtower to do any updates when a new container comes out. I actually have a pi5 that I want to image up on 64bit Debian and set up portainer on that with watchtower.

rdwebdesign commented 10 months ago

What is also strange is the already taken port

2023-11-28 04:14:46.891 4249M] Successfully accessed setupVars.conf
   [2023-11-28 04:14:46.893 4249M] FATAL ERROR in dnsmasq core: failed to create listening socket for port 53: Address already in use
   [2023-11-28 04:14:46.915 4249M] ########## FTL terminated after 107ms  (code 1)! ##########

This log is from a month ago (2023-11-28).

Apparently this container is using volumes (even for the logs) and something is overwriting volumes content with older versions of the files.

@gh0sti

~~Did you configure volumes for this container? Can you please post the compose file or docker run command used to start the container?~~

I just noticed your original comment has a compose file.

gh0sti commented 10 months ago

What is also strange is the already taken port
2023-11-28 04:14:46.891 4249M] Successfully accessed setupVars.conf
   [2023-11-28 04:14:46.893 4249M] FATAL ERROR in dnsmasq core: failed to create listening socket for port 53: Address already in use
   [2023-11-28 04:14:46.915 4249M] ########## FTL terminated after 107ms  (code 1)! ##########
This log is from a month ago (2023-11-28).

Apparently this container is using volumes (even for the logs) and something is overwriting volumes content with older versions of the files.

@gh0sti

~Did you configure volumes for this container? Can you please post the compose file or docker run command used to start the container?~

I just noticed your original comment has a compose file.

Is there anything else I should try before I just switch over to using my Pi5 I got and setting up brand new?

rdwebdesign commented 10 months ago

I don't have a suggestion, because I have no idea how the container files are reverting to previous versions. Apparently even the logs are older files.

As a test, you can try to recreate the container using Portainer interface (stop the stack to remove the container and start it again), create a debug log with the initial state and later (when the issue happens again) you create a second debug log, for comparison.

gh0sti commented 10 months ago

I don't have a suggestion, because I have no idea how the container files are reverting to previous versions. Apparently even the logs are older files.

As a test, you can try to recreate the container using Portainer interface (stop the stack to remove the container and start it again), create a debug log with the initial state and later (when the issue happens again) you create a second debug log, for comparison.

I finished setting up my pi5 with portainer and using this container and it auto updated just fine with watchtower to the latest version and shows current versions. Probably my pi4 container is corrupted and I will be decommissioning it and just using my pi5 going forward. For now I will close this if it happens again on my pi5 I'll bring up more logs.

pralor-bot commented 7 months ago

This issue has been mentioned on Pi-hole Userspace. There might be relevant details there:

https://discourse.pi-hole.net/t/update-error/68973/2

pi-hole / docker-pi-hole