fabmanager container keeps restarting after production install - socket.error reported errno.EACCES

[user90210]

Describe the bug

docker logs fabmanager-fabmanager-1
Error: Cannot open an HTTP server: socket.error reported errno.EACCES (13)
For help, use /usr/bin/supervisord -h
Error: Cannot open an HTTP server: socket.error reported errno.EACCES (13)
For help, use /usr/bin/supervisord -h
Error: Cannot open an HTTP server: socket.error reported errno.EACCES (13)
For help, use /usr/bin/supervisord -h
Error: Cannot open an HTTP server: socket.error reported errno.EACCES (13)
For help, use /usr/bin/supervisord -h

CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
c08d3310361c nginx:latest "/docker-entrypoint.…" 6 minutes ago Up 15 seconds 0.0.0.0:80->80/tcp, 0.0.0.0:443->443/tcp fabmanager-nginx-1
1c4011e5e37f sleede/fab-manager "/usr/bin/supervisor…" 6 minutes ago Restarting (2) 6 seconds ago fabmanager-fabmanager-1
8f4e79aae351 elasticsearch:5.6 "/docker-entrypoint.…" 6 minutes ago Up 16 seconds 9200/tcp, 9300/tcp fabmanager-elasticsearch-1
e63115e6e3fa postgres:9.6 "docker-entrypoint.s…" 6 minutes ago Up 16 seconds 5432/tcp fabmanager-postgres-1
16f159efe29e redis:6-alpine "docker-entrypoint.s…" 6 minutes ago Up 16 seconds 6379/tcp fabmanager-redis-1

To Reproduce
Steps to reproduce the behavior:
I followed this guide https://github.com/sleede/fab-manager/blob/master/doc/production_readme.md

Both with custom and default values.

Expected behavior

Screenshots
If applicable, add screenshots to help explain your problem.

Server (please complete the following information):

• debian
• Kernel: default/current
• Fab-manager version: default/current

[nilssjostrom]

I tracked this down to changes in filesystem permissions made in 6.3.7. It works when I add

chmod -R a+w /usr/src/app && \
chmod -R a+w /var/run

to the dockerfile and build it locally.
I totally understand why using low granularity is a bad idea though 😄

[remisarrailh]

Same issue here (Ubuntu)
I follow steps from the forum and it works:
https://forum.fab-manager.com/t/issues-with-production-installation-psql-rails-failed/801/3

Procedure:
Install as normal using scripts
Then

cd ~
git clone https://github.com/sleede/fab-manager.git

edit /fab-manager/Dockerfile so that the WEB app section ends like:

    mkdir -p /usr/src/app/tmp/pids && \
    chmod -R a+w /usr/src/app && \
    chmod -R a+w /var/run

Build images

docker build fab-manager
docker images #(copy image id for fab-manager, check that it was built recently)
docker tag <IMAGE ID> fab-manager

Change /apps/fabmanager/docker-compose.yml

fabmanager:
    image: fab-manager

docker compose down && docker compose up -d

[thomasxyz429]

I encountered the same problem, but removing the line
user: USER_ID from the docker-compose.yml file in your setup folder resolves the issue.
After making this change, run:
docker compose up --force-recreate -d
However,this solution is not a good practice. Instead, I think we need to adjust Supervisord's permissions to avoid using a root account.