Sakai Docker Deployment Examples

A collection of tools and examples to demonstrate building and deploying Sakai using Docker and/or Docker Swarm.

Features

Features of the files you will find here:

• Example Swarm stacks
  • Sakai + Mysql
  • Sakai + Mysql + Elasticsearch + Cerebro + Mailcatcher

Table of Contents

• Quick install Docker (Linux, new server/workstation)
• Running Sakai
  • Swarm enabled Docker
    • Sakai+Mysql
    • Sakai+Mysql+Elasticsearch+Mailcatcher
    • Monitoring Examples
    • Management Examples
  • Standalone Docker
    • Monitoring Examples
    • Management Examples

Quick install Docker (Linux, new server/workstation)

Docker provides an installation script for most Linux distributions, With Ubuntu and RHEL/CentOS being the most used. This script is located at https://get.docker.com/ and has the following instructions at the top of the file:

This script is meant for quick & easy install via:
$ curl -fsSL https://get.docker.com -o get-docker.sh
$ sh get-docker.sh

This will install and prepare everything needed to build images and run them in standalone mode. Additionally, docker has a built-in container orchestration platform called "Docker Swarm". Swarm needs to be enabled to use any of these deployment example scripts, this can be enabled by simply running:

 $ docker swarm init 

Running Sakai

You will need to provide at a minimum a sakai.properties configuration containing a minimal database configuration. In the examples here, a MySQL configuration is provided.

Swarm enabled Docker

These are basic examples, that do not include mounting storage volumes to persist data.

Sakai+Mysql

To start the example Sakai and Mysql stack follow these steps:

1. From this folder.
2. Execute docker stack deploy -c sakai_docker.yml sakai
   • This creates a stack named sakai using the compose file sakai_docker.yml
   • Services in the stack are named <StackName>_<ServiceName> (e.g. sakai_mysql and sakai_sakai)
   • The configuration file conf/sakai.properties is mounted inside the container in /usr/local/sakai/properties
   • Sakai will be located at http://<dockerhost>:8080/portal

Sakai+Mysql+Elasticsearch+Cerebro+Mailcatcher

To start the example Sakai, Mysql, Elasticsearch, Cerebro, and Mailcatcher stack follow these steps:

1. From this folder.
2. Execute docker stack deploy -c sakai_es_docker.yml sakai
   • This creates a stack named sakai using the compose file sakai_es_docker.yml
   • Services in the stack are named <StackName>_<ServiceName> (e.g. sakai_mysql and sakai_sakai)
   • The configuration file conf/sakai.properties is mounted inside the container in /usr/local/sakai/properties
   • The secrets file secrets/security.properties is deployed using Docker Secrets and placed in /usr/local/tomcat/sakai/
   • Sakai will be located at http://<dockerhost>:8080/portal
   • Mailcatcher will be located at http://<dockerhost>:8081/
   • Cerebro (Elasticsearch Management) will be located at http://<dockerhost>:8082/
   • Use ES Node address: http://elasticsearch:9200/

Monitoring Examples

Now that the stack has started you can monitor with the Swarm commands:

1. List running stacks with docker stack ls
2. List services running in the sakai stack with docker stack services sakai
3. List containers running in the sakai service with docker service ps sakai_sakai
4. View (tail) the logs from the Sakai container with docker service logs -f sakai_sakai

Management Examples

You can manage using Swarm commands:

1. Restart the sakai service with docker service update --force sakai_sakai
2. Stop and remove the stack with docker stack rm sakai

Standalone Docker

In this example you will start a standalone MySQL container and a linked standalone sakai container.

1. From this folder.
2. Start a MySQL 5.5 container with:

   • docker run -d \
     --name mysql \
     -e MYSQL_ROOT_PASSWORD=examplerootpassword \
     -e MYSQL_DATABASE=sakai -e MYSQL_USER=sakai \
     -e MYSQL_PASSWORD=examplepassword \
     mysql:5.5 \
     --character-set-server=utf8 \
     --collation-server=utf8_general_ci
     

3. Monitor MySQL startup with docker logs -f mysql and hit Ctrl-C once MySQL has finished starting.
4. Start a Sakai Container with the MySQL container linked to it:

   • docker run -d \
     --name sakai \
     -p 8080:8080 \
     -v $(pwd)/config/sakai.properties:/usr/local/sakai/properties/sakai.properties \
     --link mysql:mysql \         
     sakai
     

5. Monitor Sakai startup with docker logs -f sakai and hit Ctrl-C once Sakai has finished starting.
6. Sakai should be available on port 8080 of the Dockerhost.

Monitoring Examples

You may need to monitor the containers

1. View logs of the sakai container with docker logs -f sakai
2. Inspect the sakai container configuration with docker inspect sakai

Management Examples

You may need to manage the containers

1. Stop the running sakai container with docker kill sakai
2. Remove the sakai container with docker rm sakai