Can't get multiple services running behind traefik on the same host - docker

I have two services as shown below:
version: '3'
services:
# reverse proxy
traefik:
image: traefik:1.7-alpine
ports:
- 8080:8080 # Access through port 8080 e.g. localhost:8080 -> traefik dashboard
- 9000:80 # Access through port 9000 e.g. localhost:9000 -> actual service (flask and any others)
- 9001:80 # Access through port 9001 e.g. localhost:9001 -> actual service (flask and any others)
- 443:443 # Access through port 443 e.g. localhost:443 -> unused until we get https going
networks:
- traefik
volumes:
- /var/run/docker.sock:/var/run/docker.sock
# - ./acme.json:/acme.json # ignored for production version
# - ./traefik.toml:/traefik.toml # ignored for production version
container_name: traefik
command: --docker --api --docker.domain=local.docker
restart: unless-stopped
# flask app 1
flaskapp1:
# build the Dockerfile
build:
context: ./flaskapp1
dockerfile: Dockerfile
container_name: flaskapp1
restart: always
command: >
gunicorn -b 0.0.0.0:5000
--access-logfile -
--timeout=1200
--reload
"flaskapp1.app:create_app()"
volumes:
- '.:/flaskapp1'
networks:
- traefik
# exposing a port to the other services (note, this is not publishing the port to the world)
expose: ['5000']
# these labels override the traefik configure file so we can setup traefik here, and not deal with a .toml file
labels:
- traefik.enable=true
- traefik.backend=flaskapp1
- traefik.docker.network=traefik
- traefik.frontend.rule=Host:localhost
- traefik.flaskapp1.port=5000
# flask app 2
flaskapp2:
# build the Dockerfile
build:
context: ./flaskapp2
dockerfile: Dockerfile
container_name: flaskapp2
restart: always
command: >
gunicorn -b 0.0.0.0:5000
--access-logfile -
--timeout=600
--reload
"hairByElli.app:create_app()"
volumes:
- '.:/flaskapp2'
networks:
- traefik
# exposing a port to the other docker services (note, this is not publishing the port to the world)
expose: ['5000']
# these labels override the traefik config file so we can setup traefik here, and not deal with a .toml file
labels:
- traefik.enable=true
- traefik.backend=flaskapp2
- traefik.docker.network=traefik
- traefik.frontend.rule=Host:localhost:9001
# - traefik.frontend.rule=Host:localhost;PathPrefixStrip:/app2
- traefik.flaskapp2.port=5000
networks:
traefik:
external: false
As you can see, this is my localhost testing machine. Ubuntu 18.04. So I'm currently trying to get to localhost:9000 and localhost:9001 as my entry points whilst I get this working properly.
My flaskapp1 works perfectly. My flaskapp2 is giving a 404 error.
Here is my toml:
defaultEntryPoints = ["http", "https"]
logLevel = "error"
debug = false
# WEB interface of Traefik - it will show web page with overview of frontend and backend configurations
[api]
dashboard = false
address = ":8080"
# Connection to docker host system (docker.sock)
[docker]
domain = "local.docker"
watch = true
# This will hide all docker containers that don't have explicitly
# set label to "enable"
exposedbydefault = false
# Force HTTPS
[entryPoints]
[entryPoints.http]
address = ":9000"
[entryPoints.http.redirect]
entryPoint = "https"
[entryPoints.http]
address = ":9001"
[entryPoints.http.redirect]
entryPoint = "https"
[entryPoints.https]
address = ":443"
[entryPoints.https.tls]
What am I doing wrong here? I mean, I don't mind if it comes in at 9000 / 9001 or some other entry point. It's just so I can test them on the same machine until going live. At which point I'll have separate domains pointing to them.
Any help would be greatly appreciated.

Related

Traefik cannot Issue Lets Encrypt Cert for gitlab container on different port

im running a gitlab-ee docker container behind a traefik v1 docker container. My gitlab is supposed to run on the domain gitlab.dev.example.com:65443 the port 65443 is being forwarded to 443 on my server within my router (i have other stuff running on my 443 and 80 port). my traefik dashboard is running on traefik.dev.example.com:65443/dashboard/
Now when i want to get a lets encrypt certificate with traefik it tries to get it for the domain gitlab.dev.example.com and fails with "Unable to obtain ACME certificate for domains "gitlab.dev.example.com" [...]"
if i visit https://gitlab.dev.example.com:65443 it opens my gitlab container but with the "Traefik Default Cert". I cannot push or pull from these gitlab repositories because the SSL Certificates are self signed hence why i want to get a lets encrypt one.
i swapped out the actual domain with "example.com" obviously
my traefik.toml:
debug = false
logLevel = "ERROR"
defaultEntryPoints = ["https", "http"]
[web]
address = ":80"
[docker]
endpoint = "unix:///var/run/docker.sock"
domain = "traefik.dev.example.com" //swapped the url out
watch = true
exposedByDefault = false
# Force HTTPS
[entryPoints]
[entryPoints.http]
address = ":80"
[entryPoints.http.redirect]
entryPoint = "https"
[entryPoints.https]
address = ":443"
[entryPoints.https.tls]
# Enable retry sending a request if the network error
[retry]
# Let's encrypt configuration
[acme]
email="e#mail.com" //swapped the email out
storage="acme.json"
entryPoint="https"
acmeLogging=true
OnHostRule=true
[acme.httpChallenge]
entryPoint = "http"
now to my docker-compose files. i have a seperate docker-compose.yml for each container
my docker-compose.yml for the traefik container:
version: "3.2"
services:
reverse-proxy:
image: traefik:alpine
command: --api --docker --logLevel=error
restart: unless-stopped
container_name: docker-traefik
volumes:
- /var/run/docker.sock:/var/run/docker.sock
- ./traefik.toml:/etc/traefik/traefik.toml
- ./acme.json:/acme.json
environment:
- "TZ=Europe/Berlin"
networks:
- traefik_proxy
- default
ports:
- "443:443"
- "80:80"
logging:
driver: "json-file"
options:
max-file: "3"
max-size: "5m"
labels:
- traefik.backend=traefik-proxy
- traefik.frontend.rule=Host:traefik.dev.loropserver.de
- traefik.docker.network=traefik_proxy
- traefik.port=8080
- traefik.enable=true
- traefik.frontend.auth.basic=lorop:$$apr1$$dHnqprRX$$DjIWIaE97EnMoxwu6o/14.
networks:
traefik_proxy:
external:
name: traefik_proxy
# default:
# driver: bridge
my docker-compose.yml for the gitlab container:
version: '3.5'
services:
gitlab:
image: 'gitlab/gitlab-ee:latest'
container_name: gitlab
restart: unless-stopped
hostname: 'gitlab.dev.example.com'
environment:
GITLAB_OMNIBUS_CONFIG: |
external_url 'https://gitlab.dev.example.com';
//some more configs
ports:
- '22:22'
- '5005:5005'
volumes:
- './volumes/gitlab/config:/etc/gitlab'
- './volumes/gitlab/logs:/var/log/gitlab'
- './volumes/gitlab/data:/var/opt/gitlab'
- /etc/localtime:/etc/localtime:ro
- './certs:/etc/gitlab/trusted-certs'
networks:
- traefik_proxy
labels:
- 'traefik.enable=true'
- 'traefik.port=65443'
- 'traefik.docker.network=traefik_proxy'
- 'traefik.backend=gitlab'
- 'traefik.frontend.rule=Host:gitlab.dev.example.com'
- 'traefik.http.routers.entrypoints=websecure'
networks:
traefik_proxy:
external: true

traefik send http over wrong port to gitlab

I'm setting up a gitlab server behind traefik proxy, but my gitlab sshd logs says traefik send http request over 22 port to gitlab, causing Internal server error.
==> /var/log/gitlab/sshd/current <==
gitlab | 2019-08-27_03:39:15.42508 Bad protocol version identification 'GET / HTTP/1.1' from 192.168.144.2 port 33462
I found there is a similar discuss here, however the answer did not work, is there any way to change gitlab backend from 22 to 80?
The following is my code and configs.
# traefik docker-compose.yaml
version: '3'
services:
traefik:
container_name: traefik
image: traefik
command: --api
volumes:
- /var/run/docker.sock:/var/run/docker.sock
- ./traefik.toml:/traefik.toml
networks:
- web
ports:
- 8080:8080
- 80:80
- 443:443
networks:
web:
external: true
# traefik.toml
debug = false
logLevel = "INFO"
defaultEntryPoints = ["http","https"]
[entryPoints]
[entryPoints.http]
address = ":80"
[docker]
endpoint = "unix:///var/run/docker.sock"
domain = "mydomain.com"
watch = true
exposedByDefault = false
# gitlab docker-compose.yaml
version: '3'
services:
# expose 22 80 443
gitlab:
container_name: gitlab
image: gitlab/gitlab-ee
volumes:
- /nsrv/gitlab/config:/etc/gitlab
- /nsrv/gitlab/logs:/var/log/gitlab
- /nsrv/gitlab/data:/var/opt/gitlab
labels:
- traefik.enable=true
- traefik.frontend.rule=Host:gitlab.mydomain.com
- traefik.prot=80
networks:
- web
restart: always
networks:
web:
external: true
Looks like you have a typo in your gitlab docker-compose.yaml file.
Replace traefik.prot by traefik.port and it should work better.

Bad gateway on traefik docker

I build traefik with cloudflare CDN. I used docker container run command to execute my docker container execute by Drone CI. I have an issue when I successfully built docker container which leads to bad gateway on subdomain.
docker-compose.yml
version: '3'
services:
traefik:
image: traefik:latest
container_name: traefik
restart: always
domainname: ${DOMAINNAME}
networks:
- traefik_proxy
ports:
- "80:80"
- "443:443"
- "8080:8080"
environment:
- CF_API_EMAIL=${CLOUDFLARE_EMAIL}
- CF_API_KEY=${CLOUDFLARE_API_KEY}
labels:
- "traefik.enable=true"
- "traefik.backend=traefik"
- "traefik.frontend.rule=Host:monitor.${DOMAINNAME}"
- "traefik.port=8080"
- "traefik.docker.network=traefik_proxy"
- "traefik.frontend.headers.SSLRedirect=true"
- "traefik.frontend.headers.STSSeconds=315360000"
- "traefik.frontend.headers.browserXSSFilter=true"
- "traefik.frontend.headers.contentTypeNosniff=true"
- "traefik.frontend.headers.forceSTSHeader=true"
- "traefik.frontend.headers.SSLHost=example.com"
- "traefik.frontend.headers.STSIncludeSubdomains=true"
- "traefik.frontend.headers.STSPreload=true"
- "traefik.frontend.headers.frameDeny=true"
- "traefik.frontend.auth.basic.users:${HTTP_USERNAME}:${HTTP_PASSWORD}"
volumes:
- /var/run/docker.sock:/var/run/docker.sock:ro
- /etc/traefik:/etc/traefik
- /etc/docker/shared:/shared
networks:
traefik_proxy:
external:
name: traefik_proxy
Traefik.toml
nsecureSkipVerify = true
defaultEntryPoints = ["https", "http"]
# WEB interface of Traefik - it will show web page with overview of frontend and backend configurations
[api]
entryPoint = "traefik"
dashboard = true
address = ":8080"
# Force HTTPS
[entryPoints]
[entryPoints.http]
address = ":80"
[entryPoints.http.redirect]
entryPoint = "https"
[entryPoints.https]
address = ":443"
[entryPoints.https.tls]
# Let's encrypt configuration
[acme]
email = "example#gmail.com" #any email id will work
storage="acme.json"
entryPoint = "https"
acmeLogging=true
onDemand = false #create certificate when container is created
[acme.dnsChallenge]
provider = "cloudflare"
delayBeforeCheck = 300
[[acme.domains]]
main = "example.com"
[[acme.domains]]
main = "*.example.com"
# Connection to docker host system (docker.sock)
[docker]
endpoint = "unix:///var/run/docker.sock"
domain = "example.com"
watch = true
# This will hide all docker containers that don't have explicitly
# set label to "enable"
exposedbydefault = false
Command I used to run the docker container execute by Drone:
docker container run -d --name example-development --restart=unless-
stopped --label "traefik.backend=example-development" --label
"traefik.frontend.rule=Host:subdomain.example.com" --label
"traefik.enable=false" --label "traefik.port=6611" --expose 6611
cloud.canister.io:5000/username/repo
My docker container is listening to http://127.0.0.1:6611
Above codes examples lead to Error 504 Gateway Timeout.
Traefik needs to have a common network with the containers it is connecting to. In this case, you need to run containers with --net=traefik_proxy.
If you're container is on multiple networks, you'll also need the label traefik.docker.network=traefik_proxy to tell traefik which of those networks to use.

Traefik HTTPS entry points configurations not working via lets encrypt

Having some real issues getting HTTPS entrypoints working from my Traefik docker container.
I was trying to follow the guidelines over at https://docs.traefik.io/user-guide/docker-and-lets-encrypt/
Also have tried various bits from https://docs.traefik.io/user-guide/examples/ and https://ian-says.com/articles/traefik-proxy-docker-lets-encrypt/ and various other places
It all works fine in http, but https gives me a connection refused, and no https entrypoint is active in the traefik web portal.
When I check port 443 externally it says its closed, however there is no firewall, and its not already in use (see docker ps below), so all I can think is that the traefik container itself isn't setup correctly for https?
My question: How do I get HTTPS working on traefik? Also how do I see the traefik logs to get more info? docker-compose logs -f reverse-proxy is empty
traefik.toml
defaultEntryPoints = ["https","http"]
debug = true
[entryPoints]
[entryPoints.http]
address = ":80"
[entryPoints.https]
address = ":443"
[entryPoints.https.tls]
[acme]
email = "email#domain.com"
storage = "/home/project/acme.json"
caServer = "https://acme-staging-v02.api.letsencrypt.org/directory"
entryPoint = "https"
[acme.httpChallenge]
entryPoint = "http"
[[acme.domains]]
main = "domain.com"
sans = ["app.domain.com", "api.domain.com"]
docker-compose.yml
version: '3'
services:
reverse-proxy:
image: traefik
command: --api --docker
ports:
- "80:80"
- "443:443"
- "8080:8080" # The Web UI (enabled by --api)
networks:
- web
volumes:
- /var/run/docker.sock:/var/run/docker.sock
- /opt/traefik/traefik.toml:/home/project/traefik.toml
- /opt/traefik/acme.json:/home/project/acme.json
api:
build:
context: ./api/
dockerfile: Dockerfile
volumes:
- ~/api:/var/www/html
networks:
- web
restart: always
labels:
- "traefik.enable=true"
- "traefik.frontend.rule=Host:api.domain.com"
- "traefik.docker.network=web"
webapp:
build:
context: ./webapp/
dockerfile: DockerfileNode
volumes:
- ~/webapp:/app
networks:
- web
labels:
- "traefik.frontend.rule=Host:app.domain.com"
- "traefik.docker.network=web"
- "traefik.enable=true"
expose:
- 8188
networks:
web:
external: true
screengrab of my traefik portal
(note: only http entry points, no https)
docker ps
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
d956a9d25ace webapp "/usr/bin/kafka-sock…" 18 minutes ago Up 16 minutes 8188/tcp webapp_1
e46693c8ca3e api "docker-php-entrypoi…" 21 minutes ago Up 16 minutes 80/tcp api_1
321c5efc720b traefik "/traefik --api --do…" 2 hours ago Up 16 minutes 0.0.0.0:80->80/tcp, 0.0.0.0:443->443/tcp, 0.0.0.0:8080->8080/tcp reverse-proxy_1

Configure Traefik in front of two docker containers, all on port 80

I try to run three docker containers on one host. Traaefik is one of the containers to proxy traffic to the other containers.
My first goal is to reach each container through a dedicated hostname on port 80. Traefik ui should be available only through a hostname and on port 80 also, having some sort of authentication.
Using only a docker-compose.yml, I can reach all three containers using the hostnames, all on port 80. But to add authentication, I guess I need to introduce a traefik.toml. But this gives me troubles.
Next goal would be to introduce SSL using let's encrypt on all three hosts.But first things first...
Working solution with three hosts, all on port 80, lacking authorization for Traefik UI:
version: "2"
networks:
web:
services:
prox:
image: containous/traefik:latest # The official Traefik docker image
command: --api --docker # Enables the web UI and tells Træfik to listen to docker
restart: unless-stopped
ports:
- "80:80" # The HTTP port
labels:
- "traefik.port=8080"
- "traefik.backend=traefikception"
- "traefik.frontend.rule=Host:traefik.test.com"
- "traefik.enable=true"
volumes:
- /var/run/docker.sock:/var/run/docker.sock # So that Traefik can listen to the Docker events
networks:
- web
seafile_1:
image: seafileltd/seafile
container_name: seafile_1
restart: unless-stopped
environment:
SEAFILE_ADMIN_EMAIL: me#test.com
SEAFILE_ADMIN_PASSWORD: ####
SEAFILE_SERVER_HOSTNAME: 1.test.com
labels:
- traefik.enable=true
- traefik.frontend.rule=Host:1.test.com
- traefik.port=80
- traefik.backend=seafile_1
- traefik.docker.network=web
volumes:
- /opt/seafile-data/ttt_1:/shared
networks:
- web
seafile_2:
image: seafileltd/seafile
container_name: seafile_2
restart: unless-stopped
environment:
SEAFILE_ADMIN_EMAIL: me#test2.com
SEAFILE_ADMIN_PASSWORD: #####
SEAFILE_SERVER_HOSTNAME: 2.test2.com
labels:
- traefik.enable=true
- traefik.frontend.rule=Host:2.test2.com
- traefik.port=80
- traefik.backend=seafile_1
- traefik.docker.network=web
volumes:
- /opt/seafile-data/ttt_2:/shared
networks:
- web
Adding the following traefik.toml:
defaultEntryPoints = ["http"]
[entryPoints]
[entryPoints.http]
address = ":80"
[entryPoints.proxy]
address=":80"
[entryPoints.proxy.auth]
[entryPoints.proxy.auth.basic]
users = [
"joh:$apr1$RKdHyOKO$QDK1EKB4UJbsda7CXfPfK0",
]
[api]
entrypoint="proxy"
I get lot's of the following errors in the log, none of the containers is reachable from outside:
prox_1 | time="2018-06-17T19:23:26Z" level=fatal msg="Error preparing server: listen tcp :8080: bind: address already in use"
prox_1 | time="2018-06-17T19:24:26Z" level=error msg="Error opening listener listen tcp :8080: bind: address already in use"
prox_1 | time="2018-06-17T19:24:26Z" level=fatal msg="Error preparing server: listen tcp :8080: bind: address already in use"
I am pretty sure I need to adapt my docker-compose.yml and move settings to traefik.toml, but I cannot get my head around how to to that.
Thanks in advance!!
With the help of traefik support on slack I was able to solve this.
you may not have more than one entryPoint per Port
The Authorization can be configured in the docker-compose.yml
add acme.json and configure https and Let's encrypt only in traefik.toml
In /opt/traefik put the following three files:
acme.json:
may be empty but must be well secured:
touch acme.json
chmod 600 acme.json
docker-compose.yml:
version: "2"
networks:
web:
services:
prox:
image: containous/traefik:latest # The official Traefik docker image
command: --api --docker # Enables the web UI and tells Træfik to listen to docker
restart: unless-stopped
ports:
- "80:80"
- "443:443"
# - "8080:8080" # Don't want this port open (on all hostnames!)
labels:
- "traefik.port=8080"
- "traefik.backend=traefikception"
- "traefik.frontend.rule=Host:traefik.example.me"
- "traefik.enable=true"
- "traefik.frontend.auth.basic=admin:$$ert2$$RKdHyOKO$$QDK1EKB4UJbsda7CXfPfK0"
volumes:
- "/var/run/docker.sock:/var/run/docker.sock" # So that Traefik can listen to the Docker events
- "./traefik.toml:/traefik.toml"
- "./acme.json:/acme.json"
networks:
- web
seafile_org1:
image: seafileltd/seafile
container_name: seafile_org1
restart: unless-stopped
environment:
SEAFILE_ADMIN_EMAIL: mail#mail.me
SEAFILE_ADMIN_PASSWORD: ####
SEAFILE_SERVER_HOSTNAME: org1.example.me
labels:
- traefik.enable=true
- traefik.frontend.rule=Host:org1.example.me
- traefik.port=80
- traefik.backend=seafile_org1
- traefik.docker.network=web
volumes:
- /opt/seafile-data/org1:/shared
networks:
- web
seafile_org2:
image: seafileltd/seafile
container_name: seafile_org2
restart: unless-stopped
environment:
SEAFILE_ADMIN_EMAIL: mail#mail.com
SEAFILE_ADMIN_PASSWORD: ####
SEAFILE_SERVER_HOSTNAME: org2.example.com
labels:
- traefik.enable=true
- traefik.frontend.rule=Host:org2.example.com
- traefik.port=80
- traefik.backend=seafile_org2
- traefik.docker.network=web
volumes:
- /opt/seafile-data/org2:/shared
networks:
- web
get what you need to put as value to traefik.frontend.auth.basic issuing:
htpasswd -n admin
traefik.toml:
defaultEntryPoints = ["http", "https"]
[entryPoints]
[entryPoints.http]
address = ":80"
[entryPoints.http.redirect]
entryPoint = "https"
[entryPoints.https]
address = ":443"
[entryPoints.https.tls]
[retry]
[api]
dashboard = true
# Enable ACME (Let's Encrypt): automatic SSL.
[acme]
email = "you#mail.com"
storage = "acme.json"
entryPoint = "https"
# If true, display debug log messages from the acme client library.
# acmeLogging = true
# Enable certificate generation on frontends host rules.
onHostRule = true
# CA server to use.
# Uncomment the line to use Let's Encrypt's staging server,
# leave commented to go to prod.
caServer = "https://acme-staging-v02.api.letsencrypt.org/directory"
# Use a HTTP-01 ACME challenge.
# Optional (but recommended)
[acme.httpChallenge]
entryPoint = "http"
This uses Let's encrypt staging environment to get you three certs. Comment the line with caServer to get the real certs! Recreate an empty acme.json as well!
The seafile-data is stored in
/opt/seafile-data/org1
and
/opt/seafile-data/org2
respectively.
In /opt/traefik you can start the system:
docker-compose up -d
and watch the logs with
docker-compose logs
Startup takes some time on first run to setup seafile, get the certs,...
Your hosts should be reachable, giving no SSL errors or warnings on
http://traefik.example.me (Asking your credentials to see the page)
http://org1.example.me
http://org2.example.com
What's left to do is to edit the ccnet.conf file in each of the seafile installation directories (/opt/seafile-data/org1/seafile/conf/ccnet.conf) and change the protocol to "http" and remove the port ":8000" from SERVICE_URL so that shared links are correct for that setup as well. The line should read:
SERVICE_URL = https://org1.example.me
You can do it all in the Docker Stacks file:
version: "3.7"
services:
traefik:
image: traefik:1.7.13
command: >
--api
--docker
--docker.swarmmode
--docker.watch
--docker.exposedbydefault=false
# --debug=true
--loglevel=error # debug
--defaultentrypoints=https,http
--entryPoints="Name:http Address::80 Redirect.EntryPoint:https"
--entryPoints="Name:https Address::443 TLS"
--retry
--acme=true
--acme.entrypoint=https
--acme.httpchallenge
--acme.httpchallenge.entrypoint=http
--acme.domains="..."
--acme.email="..."
--acme.storage=/certs/acme.json
ports:
- 80:80 # HTTP
- 443:443 # HTTPS
- 8080:8080 # The Web UI (enabled by --api)
volumes:
- acme:/certs
- /var/run/docker.sock:/var/run/docker.sock:ro
whoami:
image: containous/whoami
deploy:
labels:
traefik.frontend.rule: Path:/whoami
traefik.enable: "true"
traefik.port: 80
volumes:
acme:

Resources