selfhosted-apps-docker/nextcloud

Nextcloud in docker

guide by example

Purpose

File share & sync.

• Official site
• Github
• DockerHub

Files and directory structure

/home/
└── ~/
    └── docker/
        └── nextcloud/
            ├── nextcloud-data/
            ├── nextcloud-db-data/
            ├── .env
            ├── docker-compose.yml
            └── nextcloud-backup-script.sh

docker-compose

Official examples here

Four containers to spin up

• nextcloud - nextcloud app that stores uploaded files and runs apache web server with php as a module.
• nextcloud-db - mariadb database where files-metadata and users-metadata are stored
• nextcloud-redis - in memory file caching and more reliable transactional file locking
• nextcloud-cron - for being able to run maintenance cronjobs

docker-compose.yml

version: '3'
services:

  nextcloud-db:
    image: mariadb
    container_name: nextcloud-db
    hostname: nextcloud-db
    command: --transaction-isolation=READ-COMMITTED --binlog-format=ROW
    restart: unless-stopped
    volumes:
      - ./nextcloud-data-db:/var/lib/mysql
    environment:
      - MYSQL_ROOT_PASSWORD
      - MYSQL_PASSWORD
      - MYSQL_DATABASE
      - MYSQL_USER

  nextcloud-redis:
    image: redis:alpine
    container_name: nextcloud-redis
    hostname: nextcloud-redis
    restart: unless-stopped

  nextcloud:
    image: nextcloud:apache
    container_name: nextcloud
    hostname: nextcloud
    restart: unless-stopped
    depends_on:
      - nextcloud-db
      - nextcloud-redis
    volumes:
      - ./nextcloud-data/:/var/www/html
    environment:
      - MYSQL_HOST
      - REDIS_HOST
      - MAIL_DOMAIN
      - MAIL_FROM_ADDRESS
      - SMTP_SECURE
      - SMTP_HOST
      - SMTP_PORT
      - SMTP_NAME
      - SMTP_PASSWORD

  nextcloud-cron:
    image: nextcloud:apache
    container_name: nextcloud-cron
    hostname: nextcloud-cron
    restart: unless-stopped
    volumes:
      - ./nextcloud-data/:/var/www/html
    entrypoint: /cron.sh
    depends_on:
      - nextcloud-db
      - nextcloud-redis

networks:
  default:
    external:
      name: $DEFAULT_NETWORK

.env

# GENERAL
MY_DOMAIN=blabla.org
DEFAULT_NETWORK=caddy_net
TZ=Europe/Bratislava

# NEXTCLOUD-MARIADB
MYSQL_ROOT_PASSWORD=nextcloud
MYSQL_PASSWORD=nextcloud
MYSQL_DATABASE=nextcloud
MYSQL_USER=nextcloud

# NEXTCLOUD
MYSQL_HOST=nextcloud-db
REDIS_HOST=nextcloud-redis

# USING SENDGRID FOR SENDING EMAILS
MAIL_DOMAIN=blabla.org
MAIL_FROM_ADDRESS=nextcloud
SMTP_SECURE=ssl
SMTP_HOST=smtp.sendgrid.net
SMTP_PORT=465
SMTP_NAME=apikey
SMTP_PASSWORD=SG.asdasdasdasdasdasdsaasdasdsa

All containers must be on the same network.
If one does not exist yet: docker network create caddy_net

Reverse proxy

Nextcloud official documentation regarding reverse proxy.

Caddy v2 is used, details here.
There are few extra directives here to fix some nextcloud warnings.

Caddyfile

nextcloud.{$MY_DOMAIN} {
    reverse_proxy nextcloud:80
    header Strict-Transport-Security max-age=31536000;
    redir /.well-known/carddav /remote.php/carddav 301
    redir /.well-known/caldav /remote.php/caldav 301
}

First run

Nextcloud needs few minutes to start, then there is the initial configuration, creating admin account and giving the database details as set in the .env file

The domain or IP you access nextcloud on this first run is added to trusted_domains in config.php. Changing the domain later on will throw "Access through untrusted domain" error. Editing config.php and adding the new domain will fix it.

Security & setup warnings

Nextcloud has a status check in Settings > Administration > Overview
There are likely several warnings on a freshly spun container.

The database is missing some indexes

docker exec --user www-data --workdir /var/www/html nextcloud php occ db:add-missing-indices

Some columns in the database are missing a conversion to big int

docker exec --user www-data --workdir /var/www/html nextcloud php occ db:convert-filecache-bigint

The "Strict-Transport-Security" HTTP header is not set to at least "15552000" seconds.

• helps to know what is HSTS
• fixed in the reverse proxy section above in caddy config
• the line header Strict-Transport-Security max-age=31536000;

Your web server is not properly set up to resolve "/.well-known/caldav" and Your web server is not properly set up to resolve "/.well-known/carddav".

• fixed in the reverse proxy section above in caddy config
• redir /.well-known/carddav /remote.php/carddav 301
• redir /.well-known/caldav /remote.php/caldav 301

Extra info

• check if redis container works
  at https://<nexcloud url>/ocs/v2.php/apps/serverinfo/api/v1/info
  ctrl+f for redis, should be in memcache.distributed and memcache.locking

  you can also exec in to redis container: docker exec -it nextcloud-redis /bin/bash
  start monitoring: redis-cli MONITOR
  start browsing files on the nextcloud, there should be activity in the monitoring

• check if cron container works
  in settings > administration > basic settings
  Background jobs should be set to Cron
  the last job info should never be older than 10 minutes
  

Update

• watchtower updates the image automatically

• manual image update
  docker-compose pull
docker-compose up -d
docker image prune

Backup and restore

• backup using borgbackup setup that makes daily snapshot of the entire directory

• restore
  down the nextcloud containers docker-compose down
  delete the entire nextcloud directory
  from the backup copy back the nextcloud directory
  start the container docker-compose up -d

Backup of just user data

User-data daily export using the official procedure.
For nextcloud it means entering maintenance mode, database dump and backing up several directories containing data, configs, themes.

For the script it just means database dump as borg backup and its deduplication will deal with the directories, especially in the case of nextcloud where hundreds gigabytes can be stored.

Create a backup script

Placed inside nextcloud directory on the host.

nextcloud-backup-script.sh

#!/bin/bash

# MAINTENANCE MODE ON
docker container exec --user www-data --workdir /var/www/html nextcloud php occ maintenance:mode --on

# CREATE DATABASE DUMP, bash -c '...' IS USED OTHERWISE OUTPUT > WOULD TRY TO GO TO THE HOST
docker container exec nextcloud-db bash -c 'mysqldump --single-transaction -h nextcloud-db -u $MYSQL_USER -p$MYSQL_PASSWORD $MYSQL_DATABASE > /var/lib/mysql/BACKUP.nextcloud.database.sql'

# MAINTENANCE MODE OFF
docker container exec --user www-data --workdir /var/www/html nextcloud php occ maintenance:mode --off

the script must be executable - chmod +x nextcloud-backup-script.sh

Cronjob on the host

crontab -e - add new cron job
0 2 * * * /home/bastard/docker/nextcloud/nextcloud-backup-script.sh - run it at 02:00
crontab -l - list cronjobs

Restore the user data

Assuming clean start, first restore the database before running the app container.

• start the containers: docker-compose up -d
  let it run so it creates its file structure
• down the containers: docker-compose up -d
• from backup copy the directories data, configs, themes in to nextcloud-data replacing the ones in place
• from backup copy the backup database in to nextcloud-db-data
• start the containers: docker-compose up -d
• set the correct user ownership of the directories copied:
  docker exec --workdir /var/www/html nextcloud chown -R www-data:www-data config data themes
• restore the database
  docker exec --workdir /var/lib/mysql nextcloud-db bash -c 'mysql -u $MYSQL_USER -p$MYSQL_PASSWORD $MYSQL_DATABASE < BACKUP.nextcloud.database.sql'
• turn off the maintenance mode:
  docker container exec --user www-data --workdir /var/www/html nextcloud php occ maintenance:mode --off
• update the systems data-fingerprint:
  docker exec --user www-data --workdir /var/www/html nextcloud php occ maintenance:data-fingerprint
• restart the containers: docker-compose restart
• log in