The problem still exists. I removed the exposed ports and put the container in the same network. Nothing changes the behaviour.
This is the actual yml for the cryptpad compose.
version: '3.8'
services:
cryptpad:
image: "cryptpad/cryptpad:version-5.6.0"
restart: unless-stopped
container_name: cryptpad
labels:
- traefik.enable=true
- traefik.http.routers.cryptpad.rule=Host(`office.intranet`, `sandbox.office.intranet`) && ClientIP(`192.168.0.0/16`)
environment:
- CPAD_MAIN_DOMAIN=https://office.intranet
- CPAD_SANDBOX_DOMAIN=https://sandbox.office.intranet
- CPAD_CONF=/cryptpad/config/config.js
# - CPAD_HTTP2_DISABLE=true
volumes:
- /media/nfs/docker/cryptpad/data/blob:/cryptpad/blob
- /media/nfs/docker/cryptpad/data/block:/cryptpad/block
- /media/nfs/docker/cryptpad/customize:/cryptpad/customize
- /media/nfs/docker/cryptpad/data/data:/cryptpad/data
- /media/nfs/docker/cryptpad/data/files:/cryptpad/datastore
- /media/nfs/docker/cryptpad/config/config.js:/cryptpad/config/config.js
# ulimits:
# nofile:
# soft: 1000000
# hard: 1000000
networks:
default:
external: true
name: traefik_proxy
Is there another thing what I can check?
Update:
I test this with a traefik/whoami service. This works. There must be a problem with the cryptpad specially. Could it be that the image (cryptpad) affected the behaviour of traefik labels?