Docker-compose traefik PathPrefixStrip do not work correctly - docker

Question

I've a problem who drive me mad, please help me.
I want to access all my apps from one free no-ip subdomain, so I think traefik can do that for me, I want to access all my apps like that:
mysubdomain.no-ip.com/emby
mysubdomain.no-ip.com/pydio
mysubdomain.no-ip.com/adminer...
Here is my docker compose:
version: "2"
services:
db:
image: linuxserver/mariadb
restart: always
environment:
MYSQL_ROOT_PASSWORD: "test"
labels:
- "traefik.enable=false"
volumes:
- ./config/mariadb:/etc/mysql/
ports:
- '3306:3306'
adminer:
image: adminer
restart: always
labels:
- "traefik.enable=true"
- "traefik.backend=adminer"
- "traefik.frontend.rule=PathPrefixStrip:/dbadmin"
- "traefik.backend.port=8080"
volumes:
- ./config/adminer:/config
emby:
image: emby/embyserver:latest
restart: always
labels:
- "traefik.enable=true"
- "traefik.backend=emby"
- "traefik.frontend.rule=PathPrefixStrip:/media"
- "traefik.backend.port=8096"
volumes:
- ./config/emby:/config
cloud:
image: linuxserver/pydio:latest
restart: always
environment:
PGID: "1000"
PUID: "1000"
labels:
- "traefik.enable=true"
- "traefik.backend=cloud"
- "traefik.frontend.rule=PathPrefixStrip:/cloud"
- "traefik.backend.port=443"
- "traefik.protocol=https"
volumes:
- ./config/cloud:/config
- ./data/test:/data
organizr:
image: lsiocommunity/organizr
restart: always
environment:
PGID: "1000"
PUID: "1000"
TZ: "Europe/Paris"
labels:
- "traefik.enable=true"
- "traefik.backend=organizr"
- "traefik.frontend.rule=PathPrefixStrip:/"
- "traefik.backend.port=80"
volumes:
- ./config/organizr:/config
- ./data/organizr:/data
traefik:
image: traefik:1.3.3
command: --web --docker --docker.domain=traefik --logLevel=DEBUG #-c /dev/null --web --docker --logLevel=INFO
restart: always
ports:
- '80:80'
- '443:443'
- '8080:8080'
labels:
- "traefik.enable=false"
volumes:
- /var/run/docker.sock:/var/run/docker.sock
- ./traefik.toml:/traefik.toml
- ./config/traefik/data:/data
- ./config/traefik/sslcerts:/ssl
My traefik.toml
# defaultEntryPoints must be at the top because it should not be in any table below
defaultEntryPoints = ["http", "https"]
InsecureSkipVerify = true
[web]
# Port for the status page
address = ":8080"
# Entrypoints, http and https
[entryPoints]
# http should be redirected to https
[entryPoints.http]
address = ":80"
[entryPoints.http.redirect]
entryPoint = "https"
# https is the default
[entryPoints.https]
address = ":443"
[entryPoints.https.tls]
[[entryPoints.https.tls.certificates]]
CertFile = "/ssl/tls.crt"
KeyFile = "/ssl/tls.key"
[retry]
# Enable ACME (Let's Encrypt): automatic SSL
# [acme]
# # caServer = "https://acme-staging.api.letsencrypt.org/directory"
# email = "test#gmail.com"
# storage = "acme.json" # or "traefik/acme/account" if using KV store
# entryPoint = "https"
# onDemand = false
# OnHostRule = true
[docker]
endpoint = "unix:///var/run/docker.sock"
domain = "localhost"
watch = true
exposedbydefault = false
So I have only 1 app who work like a charm: emby.
Adminer semms to work correctly, load css and others assets, but when I submit the form, it send me to localhost/server=db&username=test it should send me to localhost/dbadmin/server=db&username=test
When I access to localhost/cloud/ it load me a blank page, if I open console of Chrome:
pydio.material.min.css Failed to load resource: the server responded with a status of 404 () pydio.boot.min.js Failed to load resource: the server responded with a status of 404 ()
cloud:18 Uncaught ReferenceError: PydioBootstrap is not defined at cloud:18 pydio.material.min.css Failed to load resource: the server responded with a status of 404 ()
In fact it try to load plugins from localhost/plugins and not from localhost/cloud/plugins ...
I see a lot of issue on github related to this but it seems to be corrected in 1.3.3 version, I try 1.3.3, latest...
Do pydio and adminer needs to support reverse-proxy?
Sorry for my bad english.

Answer 1

In traefik, PathPrefixStrip and redirects in the entrypoint do not currently work together. So if your request goes to http instead of https, you'll get an error.
In my own demo, I just setup nginx on port 80 to send the redirect as a short term workaround until the above issue gets fixed.