ClusterControl for custom Docker databases

ClusterControl is an all-inclusive open source database management system for MySQL, MongoDB, and PostgreSQL with advanced monitoring and scaling features.

I created this repository to manage a databases cluster deployed with custom Docker images, in particular when you need to run a database with some extensions or already built-in configurations.

Requirements:

• Docker Engine
• Docker Compose 3.5+

Environment variables

All services require environment configuration, to accomplish this you must create three files in project root:

• clustercontrol.env
• postgres.env
• postgres_replicator.env

Each file contains environment configuration for its service.

Create a new file clustercontrol.env then copy and paste following variables and change default values.

CMON_PASSWORD_FILE=password1
MYSQL_ROOT_PASSWORD_FILE=password2

Create a new file postgres.env then copy and paste following variables and change POSTGRES_USER and POSTGRES_PASS default values.

AUTO_DEPLOYMENT=0
POSTGRES_USER=pguser
POSTGRES_DBNAME=postgisdb
POSTGRES_PASS=password3

Create a new file postgres_replicator.env then copy and paste following variables and set POSTGRES_USER and POSTGRES_PASS as you did in postgres.env

AUTO_DEPLOYMENT=0
POSTGRES_USER=pguser
POSTGRES_PASS=password3
ALLOW_IP_RANGE=0.0.0.0/0s
REPLICATE_FROM=pg_master

For the full list of available variables and default values please check respective projects on GitHub:

• severalnines/docker
• kartoza/postgis:9.6-2.4

How it works

First of all you have to build the PostGIS database Docker image, it starts from kartoza/postgis:9.6-2.4 image and then build process will add all required packages to work with ClusterControl.

I created a new entrypoint.sh as bash scripts wrapper in order to include all entrypoint files from kartoza/postgis:9.6-2.4 and clustercontrols Docker projects.

Build the master PostGIS Docker image:

docker-compose build pg_master 

When the new image postgis:9.6-2.4 is ready, you can run the ClusterControl databases manager and a cluster of PostgreSQL + PostGIS databases, cluster includes:

• 1 PostGIS Master node (read-write)
• 2 PostGIS Slave nodes (read-only)

Run Docker containers stack with Docker Compose command:

docker-compose -p dbc up

I used -p dbc argument to create a project name prefix (dbc) for my new cluster, the previous command runs containers stack in foreground and shows logs, to run in background add -d argument:

docker-compose -p dbc up -d

Stop and remove your all containers with:

docker-compose -p dbc down

All important data, as well PostgreSQL data, will be kept in Docker volumes and stored as persistent data.

You will find several Docker volumes in your system, check new volumes:

docker volume ls

The persistent volumes are:

• /etc/cmon.d - ClusterControl configuration files.
• /var/lib/mysql - MySQL datadir to host cmon and dcps database.
• /root/.ssh - SSH private and public keys.
• /var/lib/cmon - ClusterControl internal files.
• /root/backups - Default backup directory only if ClusterControl is the backup destination
• pg_master - Master PostgreSQL datadir
• pg_slave1 - Slave 1 PostgreSQL datadir
• pg_slave2 - Slave 2 PostgreSQL datadir

If you restart/delete cluster containers all data will be available as before, in case you want to wipe all data in persistent volumes you must delete all volumes.

Wipe all volumes (works when containers stack is down)

Pay attention: it removes all not used volumes, if you have volumes from other projects do not use it

docker volume prune

Instead using bulk volumes removal command, you may use this command to remove a single volume:

docker volume rm <volume-name>

Manage Databases with ClusterControl

Open to ClusterControl Dashboard through the web address: http://server-ip-or-hostname:5000/clustercontrol
Now you can configure a new cluster and import PostgreSQL nodes.

In order to configure your cluster please follow official guide on Severalnines website: https://severalnines.com/

---

MIT License