Invalid security token

[atlekbai]

Hello! I am using traefik as my reverse-proxy and set its config as traefik.frontend.passHostHeader=true. Although, I am always getting Invalid security token error while logging into the backend. Any suggestion to fix this?

[petehalverson]

Hmm. I'm unable to recreate your issue. Can you offer some more specifics? E.g. your traefik.toml, October CMS container settings...

[atlekbai]

I followed this tutorial on how to setup docker swarm cluster using traefic: https://dockerswarm.rocks

This is the configuration for october cms:

version: '3.3'

volumes:
    web_data: {}

networks:
  net:
    driver: overlay
    attachable: true
  traefik-public:
    external: true

services:
  web:
    image: aspendigital/octobercms:latest
    networks:
      - net
      - default
      - traefik-public
    environment:
      - DB_TYPE=mysql
      - DB_HOST=mysql
      - DB_DATABASE=****
      - DB_USERNAME=****
      - DB_PASSWORD=****
      - CMS_LINK_POLICY=secure
    container_name: october
    volumes:
      - web_data:/var/www/html/storage/app
    deploy:
      mode: replicated
      replicas: 3
      labels:
        - traefik.frontend.rule=Host:example.com
        - traefik.enable=true
        - traefik.port=80
        - traefik.tags=${TRAEFIK_PUBLIC_TAG:-traefik-public}
        - traefik.docker.network=traefik-public
        # Traefik service that listens to HTTP
        - traefik.redirectorservice.frontend.entryPoints=http
        - traefik.redirectorservice.frontend.redirect.entryPoint=https
        # Traefik service that listens to HTTPS
        - traefik.webservice.frontend.entryPoints=https

Configuration for traefic:

version: '3.3'

services:
  ...
  # consul images
  ...
  traefik:
    image: traefik:v1.7
    ports:
      - 80:80
      - 443:443
    deploy:
      replicas: 3
      placement:
        constraints:
          - node.role == manager
        preferences:
          - spread: node.id
      labels:
        - traefik.frontend.rule=Host:traefik.${DOMAIN}
        - traefik.enable=true
        - traefik.port=8080
        - traefik.tags=traefik-public
        - traefik.docker.network=traefik-public
        # Traefik service that listens to HTTP
        - traefik.redirectorservice.frontend.entryPoints=http
        - traefik.redirectorservice.frontend.redirect.entryPoint=https
        # Traefik service that listens to HTTPS
        - traefik.webservice.frontend.entryPoints=https
        - traefik.frontend.auth.basic.users=${USERNAME}:${HASHED_PASSWORD}
    volumes:
      - /var/run/docker.sock:/var/run/docker.sock
    command:
      --docker
      --docker.swarmmode
      --docker.watch
      --docker.exposedbydefault=false
      --constraints=tag==traefik-public
      --entrypoints='Name:http Address::80'
      --entrypoints='Name:https Address::443 TLS'
      --consul
      --consul.endpoint="consul-leader:8500"
      --acme
      --acme.email=${EMAIL}
      --acme.storage="traefik/acme/account"
      --acme.entryPoint=https
      --acme.httpChallenge.entryPoint=http
      --acme.onhostrule=true
      --acme.acmelogging=true
      --logLevel=INFO
      --accessLog
      --api
    networks:
      - default
      - traefik-public
    depends_on:
      - consul-leader

volumes:
  consul-data-leader:
  consul-data-replica:

networks:
  traefik-public:
    external: true

[petehalverson]

It looks like your problem is related to your October CMS stack config. You aren't defining a shared cache or session driver, which is required when using the admin across replicas. Try this example:

version: '3.3'
services:
  web:
    image: aspendigital/octobercms:latest
    ports:
      - 80:80
    environment:
      - CACHE_STORE=redis
      - DB_DATABASE=octobercms
      - DB_HOST=mysql
      - DB_PASSWORD=root
      - DB_TYPE=mysql
      - DB_USERNAME=root
      - INIT_OCTOBER=true
      - SESSION_DRIVER=redis
    deploy:
      replicas: 3

  mysql:
    image: mysql:5.7
    environment:
      - MYSQL_ROOT_PASSWORD=root
      - MYSQL_DATABASE=octobercms

  redis:
    image: redis:5

I should also point out your web_data volume may not function as you are expecting. When using Swarm across multiple nodes, a volume will be created on each host. The data from each won't be synced across hosts.

You can overcome this by leveraging introducing cloud storage for October CMS Filesystem.

[atlekbai]

Thanks!