Homarr•3mo ago

500 Internal Error | 1.0 Beta

```

message.txt

Solution:

Has been fixed in v1.0.0-beta.4

50 Replies

Thank you for submitting a support request. Depending on the volume of requests, our team should get in contact with you shortly.

⚠️ Please include the following details in your post or we may reject your request without further comment: - Log (See https://homarr.dev/docs/community/faq#how-do-i-open-the-console--log) - Operating system (Unraid, TrueNAS, Ubuntu, ...) - Exact Homarr version (eg. 0.15.0, not latest) - Configuration (eg. docker-compose, screenshot or similar. Use ``your-text`` to format) - Other relevant information (eg. your devices, your browser, ...)

❓ Frequently Asked Questions | Homarr documentation

Can I install Homarr on a Raspberry Pi?

Mx•3mo ago

Same error here

HeavensongOP•3mo ago

looks like it's trying to create /db and /redis folders

Mx•3mo ago

looks like it created the appdata dir, just nothing under it

HeavensongOP•3mo ago

the /appdata folder has "nobody" user so maybe need to use PUID and PGID? 1000 I think in Unraid for "nobody" but that's specific to me

lupine•3mo ago

same error here

Mx•3mo ago

> ll
total 4.0K
drwxr-xr-x 2 root root 4.0K Dec 17 16:19 appdata
> ll appdata
total 0

> ll
total 4.0K
drwxr-xr-x 2 root root 4.0K Dec 17 16:19 appdata
> ll appdata
total 0

Mine has root user

Manicraft1001•3mo ago

Can you try https://docs.docker.com/engine/containers/run/#user instead? The image no longer supports PUID and PGID

Docker Documentation

Running containers

Running and configuring containers with the Docker CLI

Manicraft1001•3mo ago

1000 should be correct for Unraid. It will depend on your system.

lupine•3mo ago

uhm have to find the argument in compose

Mx•3mo ago

Shouldn't default root user still work then?

Manicraft1001•3mo ago

services: 
  app: 
    user: 1001:1001

services: 
  app: 
    user: 1001:1001

Meierschlumpf•3mo ago

Probably not because we change the ownership of the files used to run homarr and /appdata to gid=1001 and uid=1001. So because of that it's not the same and does not work with root probably

ArcaneWater•3mo ago

Okay with user i got it through this db error

HeavensongOP•3mo ago

HeavensongOP•3mo ago

tried that with both 1000 and 1001, no joy same errors

lupine•3mo ago

same here

Meierschlumpf•3mo ago

Send the raw docker command coming out of this

lupine•3mo ago

services: homarr: container_name: homarr image: ghcr.io/homarr-labs/homarr:beta restart: unless-stopped user: 1001:1001 environment: - TZ=Europe/Copenhagen - "NODE_TLS_REJECT_UNAUTHORIZED: 0" volumes: - /home/homarr/appdata:/appdata - /var/run/docker.sock:/var/run/docker.sock ports: - 7575:7575 thats the compose i run

HeavensongOP•3mo ago

docker run
  -d
  --name='homarr'
  --net='farrosphere'
  --pids-limit 2048
  -e TZ="Europe/London"
  -e HOST_OS="Unraid"
  -e HOST_HOSTNAME="farrosphere"
  -e HOST_CONTAINERNAME="homarr"
  -l net.unraid.docker.managed=dockerman
  -l net.unraid.docker.icon='https://cdn.jsdelivr.net/gh/walkxcode/dashboard-icons/png/homarr.png'
  -p '7575:7575/tcp'
  -v '/mnt/cache/appdata/homarr/appdata':'/appdata':'rw'
  -v '/var/run/docker.sock':'/var/run/docker.sock':'rw'
  -u="1000" 'ghcr.io/homarr-labs/homarr:beta'
b2cb9654f8caba49ef8f80b8a808c9e109de6bea64973eaa35f6342f719cb059

docker run
  -d
  --name='homarr'
  --net='farrosphere'
  --pids-limit 2048
  -e TZ="Europe/London"
  -e HOST_OS="Unraid"
  -e HOST_HOSTNAME="farrosphere"
  -e HOST_CONTAINERNAME="homarr"
  -l net.unraid.docker.managed=dockerman
  -l net.unraid.docker.icon='https://cdn.jsdelivr.net/gh/walkxcode/dashboard-icons/png/homarr.png'
  -p '7575:7575/tcp'
  -v '/mnt/cache/appdata/homarr/appdata':'/appdata':'rw'
  -v '/var/run/docker.sock':'/var/run/docker.sock':'rw'
  -u="1000" 'ghcr.io/homarr-labs/homarr:beta'
b2cb9654f8caba49ef8f80b8a808c9e109de6bea64973eaa35f6342f719cb059

Manicraft1001•3mo ago

Please use code blocks, it's incredibly messy: https://support.discord.com/hc/en-us/articles/210298617-Markdown-Text-101-Chat-Formatting-Bold-Italic-Underline#h_01GY0DAKGXDEHE263BCAYEGFJA The command is wrong, remove the = from the -u

lupine•3mo ago

services:
  homarr:
    container_name: homarr
    image: ghcr.io/homarr-labs/homarr:beta
    restart: unless-stopped
    user: 1001:1001
    environment:
      - TZ=Europe/Copenhagen
      - "NODE_TLS_REJECT_UNAUTHORIZED: 0"
    volumes:
      - /home/homarr/appdata:/appdata
      - /var/run/docker.sock:/var/run/docker.sock
    ports:
      - 7575:7575

services:
  homarr:
    container_name: homarr
    image: ghcr.io/homarr-labs/homarr:beta
    restart: unless-stopped
    user: 1001:1001
    environment:
      - TZ=Europe/Copenhagen
      - "NODE_TLS_REJECT_UNAUTHORIZED: 0"
    volumes:
      - /home/homarr/appdata:/appdata
      - /var/run/docker.sock:/var/run/docker.sock
    ports:
      - 7575:7575

Mx•3mo ago

fwiw, even using user 1000 (my UID) or 1001, no dice - Still failling to perms on the subdirs. (Ubuntu) :slowfrog:

Meierschlumpf•3mo ago

Can you try extra parameters -u '1001:1001'?

HeavensongOP•3mo ago

same issue, tried -u "1001:1001" and -u '1001:1001' just in case the " or ' was important

Mx•3mo ago

someone gotta hop in that public chat and screenshare lol

lupine•3mo ago

cant do voice - house is sleeping 😉

HeavensongOP•3mo ago

and with user 1000 it's even worse, don't even get a 500 error

ArcaneWater•3mo ago

I have the same issue now that i fixed 500 with -user 🙂 ERR_CONNECTION_REFUSED and all green in homarr logs

Manicraft1001•3mo ago

If there is no error, the port is probably not exposed correctly. What is your port exposure?

ArcaneWater•3mo ago

ports:
      - 7575:7575

lupine•3mo ago

with user: 1000:1000 i also get a connection refused 1001:1001 gets a 500

HeavensongOP•3mo ago

okay so if I manually create db and redis folders, it does then write db.sqlite and I can create first user also, 99 is nobody, not 1000 1000 is one of the users I made for smb, 99 is the uid and 100 is the gid for 'nobody' in unraid, and if I use -u '99:100' I get the ERR_CONNECTION_REFUSED again

Mx•3mo ago

# gettext is required for envsubst
RUN apk add --no-cache redis nginx bash gettext
RUN mkdir /appdata
RUN mkdir /secrets

RUN addgroup --system --gid 1001 nodejs
RUN adduser --system --uid 1001 nextjs

RUN chown -R 1001:1001 /appdata
RUN chown -R 1001:1001 /secrets

VOLUME /appdata
VOLUME /secrets

# gettext is required for envsubst
RUN apk add --no-cache redis nginx bash gettext
RUN mkdir /appdata
RUN mkdir /secrets

RUN addgroup --system --gid 1001 nodejs
RUN adduser --system --uid 1001 nextjs

RUN chown -R 1001:1001 /appdata
RUN chown -R 1001:1001 /secrets

VOLUME /appdata
VOLUME /secrets

No luck here

Manicraft1001•3mo ago

We fixed this issue by chowning the directory. But this shouldn't be required, there is definetly some issue in the docker image. It's quite late for our time zone (23:20) and we will continue to investigate this tomorrow. I recommend all people, experiencing this problem, to wait until tomorrow instead of scewing with their permissions. See you tomorrow 👋

Mx•3mo ago

So - bad news https://github.com/moby/moby/issues/3206 doing VOLUME appdata is going to create the directory 'appdata' on the host as root if it doesn't exist. There is no way around that. It's created by the docker Daemon which is running as root. There is no 'automated' solution here via the dockerfile. Options - User creates directory prior to running compose - User chown's after (worse option considering it'll have failed once, and requires restarting container) - Pre-script which just mkdir's then can launch into docker compose. (Common option) This wouldn't really affect helm/K8s because it's a PVC and they do their own permissions. But for docker runtime, no solution. For docker, it's standard for directories to pre-exist with proper permissions before mounting to them. (Lots of OS projects require this, either manually, or via pre-script usually ran via curl) This is per my colleague is well versed in docker. There are some 'gotchas' like helm where this doesn't apply, like if you are using a GUi that wraps all docker commands, or has their own file perms. Otherwise, this is the summary

GitHub

Files created into a host mounted directory are owned by root user ...

When saving a file by running an application in a Docker container to an host mounted directory, that file results to be owned by the root user in the host file system. For example: docker run -v $...

Krispy•3mo ago

I just assumed permissions issues, hit terminal in the container, rain ID -u to determine what it was running as, and then just chowned the directory to match. Didn't even think if it as being a problem. Glad it's not running as root now

ArcaneWater•3mo ago

Ah i see more info for ERR_CONNECTION_REFUSED i guess

run.sh: line 43: can't create /etc/nginx/nginx.conf: Permission denied
homarr  | nginx: [alert] could not open error log file: open() "/var/lib/nginx/logs/error.log" failed (13: Permission denied)
homarr  | 2024/12/18 06:33:05 [warn] 19#19: the "user" directive makes sense only if the master process runs with super-user privileges, ignored in /etc/nginx/nginx.conf:3

Manicraft1001•3mo ago

Hi, another day started in our timezone and we will start to look into this again. Thanks for the patience

Meierschlumpf•3mo ago

Okay I think we fixed it, we now have a entrypoint.sh that runs as root and creates the two need directories, then changes the owner of it to the internal user and finally runs the application with the internal user. Can you try the following image out: ghcr.io/homarr-labs/homarr:fix-startup-permission-issue For me it worked on a ubuntu machine and when running ps I got the following:

PID   USER     TIME  COMMAND
    1 root      0:00 {entrypoint.sh} /bin/sh /app/entrypoint.sh sh run.sh
   10 nextjs    0:00 sh run.sh
   21 nextjs    0:00 nginx: master process nginx -g daemon off;
   22 nextjs    0:00 redis-server *:6379
   23 nextjs    0:04 node apps/tasks/tasks.cjs
   24 nextjs    0:01 node apps/websocket/wssServer.cjs
   25 nextjs    0:00 next-server (v14.2.20)
   26 nextjs    0:00 nginx: worker process

PID   USER     TIME  COMMAND
    1 root      0:00 {entrypoint.sh} /bin/sh /app/entrypoint.sh sh run.sh
   10 nextjs    0:00 sh run.sh
   21 nextjs    0:00 nginx: master process nginx -g daemon off;
   22 nextjs    0:00 redis-server *:6379
   23 nextjs    0:04 node apps/tasks/tasks.cjs
   24 nextjs    0:01 node apps/websocket/wssServer.cjs
   25 nextjs    0:00 next-server (v14.2.20)
   26 nextjs    0:00 nginx: worker process

lupine•3mo ago

just stops with mkdir: can't create directory '/appdata/db': Permission denied

services:
  homarr:
    container_name: homarr
    image: ghcr.io/homarr-labs/homarr:fix-startup-permission-issue
    restart: unless-stopped
    user: 1001:1001
    environment:
#      - PUID=1000
#      - PGID=1000
      - TZ=Europe/Copenhagen
      - "NODE_TLS_REJECT_UNAUTHORIZED: 0"
    volumes:
      - /home/homarr/appdata:/appdata
      - /var/run/docker.sock:/var/run/docker.sock
    ports:
      - 7575:7575
networks:
   default:
     name: containers
     external: true

services:
  homarr:
    container_name: homarr
    image: ghcr.io/homarr-labs/homarr:fix-startup-permission-issue
    restart: unless-stopped
    user: 1001:1001
    environment:
#      - PUID=1000
#      - PGID=1000
      - TZ=Europe/Copenhagen
      - "NODE_TLS_REJECT_UNAUTHORIZED: 0"
    volumes:
      - /home/homarr/appdata:/appdata
      - /var/run/docker.sock:/var/run/docker.sock
    ports:
      - 7575:7575
networks:
   default:
     name: containers
     external: true