Unable to log into web interface #409
Comments
you can use the environment Why you set the pihole fix to the TAG 4.0.0-1? This Image is 4 Month old. Please try this example below, and check if your environment docker-compose.yml |
|
I'm having the exact same problem. Running pihole/pihole:latest via docker-compose with ---
version: '3.6'
services:
pihole:
image: pihole/pihole
container_name: pihole
restart: unless-stopped
hostname: pihole
volumes:
- ./pihole/pihole/:/etc/pihole/
- ./pihole/dnsmasq.d/:/etc/dnsmasq.d/
- ./pihole/log/:/var/log/
environment:
- ServerIP=10.0.25.2
- TZ=America/New_York
- DNS1=10.0.25.1
- DNS2=no
- INTERFACE=brteam0.10
- VIRTUAL_HOST=pihole.rulerof.net
- WEBPASSWORD=${WEB_PASSWORD}
network_mode: "host"Running interactively shows the same This repeats in the logs on the same interval as the kill loop: I might try to troubleshoot this later, but I wanted to add my two cents since I just discovered that my pihole DNS wasn't working. I pull images once a day, so this has likely been happening since at least this morning. |
fix this issue with follow adjustment. >> docker-pi-hole-v411-important-upgrade-notes
maintainer note: updated for accuracy. DNS environment variables were confused with |
|
@mr-bolle I can confirm that your changes resolved my issue. While I don't currently use version pinning, the changes introducing these requirements should have entailed a major or at least minor version bump, to 5.0 or 4.2. For what it's worth, I'm not using the DHCP features of pihole (and honestly find them excessive), and am only using host networking because I don't want to obscure my DNS client IPs behind the docker bridge masquerade—this additional privilege is technically unnecessary. The commits and PRs that I was able to find that reference the issue aren't terribly helpful in explaining what circumstances require it, but I'm assuming it has something to do with listening to DHCP traffic. Is there any discussion somewhere about this requirement? Maybe I should open a separate issue? Thanks for your help! |
|
I agree major version are proper release procedure for breaking changes but the benefit of keeping the version mirrored between upstream Pi-hole projects and docker-pi-hole's is it prevents confusion by users thinking Pi-hole is on v5 and having to keep a mapping of 5.0 containers Core v4.1, FTL v4.1.2, and Web 4.1.1. I'd do just about anything to avoid having to do that but it poses the tricky question of if we had caught this before 4.1's release should Pi-hole it's self have been 5.0 even though this was mostly a docker exclusive bug. I'm hoping this is one of the few times a new docker run argument is introduce as a breaking change, it was caught a week or so after the initial v4.1 release was done so it was a bit late to even entertain the thought of a version bump of the upstream versioning in this case unfortunately. For all future releases we're aiming to test docker thoroughly before the official releases and have done that with the v4.2 release today so that should help prevent a re-occurrence of this weird 4.1 docker run breaking change. |
|
@mr-bolle's suggestion fixed the repeating error message, but not the password issue. Passwords are still not being stored between restarts of the container, so I have to read from the logs each time the container is rebooted. |
|
That might have something to do with your volume saving the password (the image tries not to over aggressively overwrite already set passwords). Try changing your password with the CLI |
|
If I change the password with |
|
Here's my |
|
@rschuetzler add |
|
Please check first your variable ip And check your pihole log, it should be like this. Below you find your example with the environment WEBPASSWORD. Addidional i remove the host volume to the pihole.log. Could you try it if you have now success
|
|
I was trying to avoid setting the password in the |
You can set the variable in a In the snippet from my
When I tested it, it had to be set every time, and would never stick. Every time the container comes up, it either generates a new password or pulls one set from the environment. |
|
Looks like #418 is fixing the only remaining issue I've had of webpassword getting set. |
This is a...
Description
This might be two separate issues, but it's hard to tell. FTL doesn't seem to be starting correctly in my Docker-enclosed pihole. Also, I can't seem to get it to keep a password, despite the fact that I have mapped the config directory to a volume (related to #364 ).
I've tried on latest and v4.2_rc1, and both have this issue. Going back to 4.0.0-1 does not.
Expected Behavior
docker-compose up -ddocker-compose run hole pihole -a -pActual Behavior
docker-compose up -ddocker-compose logs hole | grep passwordto get the current randomly set passwordPossible Fix
On the lastest versions of pihole, I get the following messages repeated over and over in the logs. It never stops as long as the pihole is running:
On 4.0.0-1, I don't get those error messages, but I still can't set a password that will persist.
Steps to Reproduce and debugging done
e.g. your docker run command, pages to visit, CLI commands you ran
1.
2.
3.
4.
Debug steps I have tried
docker runexample in the readme (removing any customizations I added)Context and extra information
docker-compose.yml:
Your Environment
The text was updated successfully, but these errors were encountered: