aws-swarm-leader-migrate

Docker image responsible for automatic migration of swarm leader in AWS using Dynamodb. Most of the logic lives in the script entry.sh.

In order to automate the leader migration,

1. Run swarm node ls command and check if there is a new leader
2. If there is a new leader update Dynamodb and sleep for x seconds and then go to step 1
3. If there is no new leader then sleep for x seconds and then go to step 1

This docker image performs the above steps ensuring the leader information is up to date in Dynamodb.

This docker image needs following environment variables

• NODE_TYPE: Should be either manager or worker
• DYNAMODB_TABLE: Name of the dynamodb table to be used for locking and passing cluster information
• REGION: AWS region in which dynamodb table was created (if not provided will default to region of the instance)
• CHECK_SLEEP_DURATION: time in seconds to sleep after each poll to check if leader has changed (defaults 300 or 5 mins)

docker-compose:

version: "2"
services:
    aws-swarm-leader-migrate:
      image: "srikalyan/aws-swarm-leader-migrate:version"
      container_name: "aws-swarm-leader-migrate"
      restart: "always"
      environment:
        NODE_TYPE: "<manager|worker>"
        DYNAMODB_TABLE: "<dynamodb_table>"
      volumes:
        - /var/run/docker.sock:/var/run/docker.sock
        - /usr/bin/docker:/usr/bin/docker
        - /var/log:/var/log

Note 1: This image needs a docker client but does not install one as this would create unnecessary versions which we could easily skip by mount the docker binary from host ubuntu machine.