Skip to content
master
Switch branches/tags
Code

Latest commit

 

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
 
 
 
 
 
 

Docker Compose based Databases

This repository contains a single docker-compose.yml that has every single dockerized database or similar service I've used, along with parameters used to setup the said service.

A new project maybe easily started by chopping services off this file and then running it with only whatever is necessary.

Since I have 16GB of RAM, I sometimes use this repo to bootstrap projects by just running shared databases here. Luckily data is namespaced so collections, databases, indexes, prefixes or whatever a DB chooses to call them don't tend to conflict.

There's Makefile included that you can refer to if you can't remember docker-compose commands!

Services and Ports

Please refer to the docker-compose.yml file directly! It's human readable.

Initial Setup

  • Run sudo ./host-setup.sh to ensure that the swap is turned off (for stable performance) and addressable virtual memory is set to 262144 (to stop ElasticSearch from crashing).
  • Run docker-compose pull or make pull to pull in the latest versions of all images defined in this docker-compose.yml file.

Manage Containers

Start containers via docker-compose up -d or via make or via make up.

Check what containers are running via docker-compose ps or via make ps.

Check logs of a single container via docker-compose logs mongo -f. The -f flag follows new logs as they come. Skip it to not follow and just print the logs. Omit the container name to get logs of all containers. docker-compose logs -f

Stop containers via docker-compose down. This preserves your volumes so that your data is exactly as you left it earlier.

Tips and Gotchas

Prune Docker Volumes

Run docker volume prune to get rid of all volumes created by docker to give yourself a fresh start if you screwed up somewhere in the configuration.

You could run docker-compose down -v or make clean to only remove volumes from THIS repository.

PGAdmin4 Credentials Change

If you changed username/password for default admin on PGAdmin4, you need to stop all containers and then run docker volume rm dcdbs_pgadmindata to remove the PGAdmin4 volume. Otherwise the updated user/pass from the environment variables passed through docker-compose.yml doesn't seem to be picked up.

MongoDB Connection Setup

You should first login to 127.0.0.1:7017 to open mongo-express and create a new database for a project you are starting.

The URI for MongoDB under this docker-compose.yml should look something like this:

mongodb://mongo:mongo@127.0.0.1:27017/projectname?connectTimeoutMS=3000&authSource=admin"

Note the additional authSource=admin most docs seem to leave out. This tells MongoDB that you want the user mongo to be authenticated against the admin database (that is created by default and has declared users inside).

Desktop Admin Interfaces

Usually you'd want to run without admin GUIs within the Docker image, so that you can use ONE desktop admin GUI per database but use this to manage multiple instances of one database. I'm listing names of admin GUIs I've found to be nice. Except for RedisInsight, all others are auto-installed if you use Rhino flavoured Ubuntu.

PostgreSQL - PgAdmin 4

The desktop app is just a system tray icon that launches the server on localhost and lets you open it on a browser. But it works!

Redis - RedisInsight

This isn't opensource but it's free use, by RedisLabs. Get it here. You need to jump through a few hoops to get it though. So if Commander works just fine, don't bother!

General Purpose SQL - DBeaver

I usually keep this around just to help me visualize someone else's database schema quickly. It creates a useful ER diagram of all tables and works with PostgreSQL, MySQL, MariaDB and SQLite3.

MongoDB - Compass

It's the official GUI by MongoDB, Inc. Released under SSPL, it's free to use as a standalone management tool. I recommend the Compass Isolated edition, since it has all the features of Compass but doesn't dial back home.

Check out the versions here, the source code here and finally, the download page is here.

MySQL - MySQL Workbench

Probably the ONE reason (alongside sharding & clustering support from way back) MySQL has way more users than PostgreSQL. I never used MySQL personally, but this is the app I've seen open on every MySQL user's desktop since forever!

Download it here.