This repo contains a Terraform template for generating and managing clusters of AWS instances with the community edition of the Storidge CIO software.

The terraform code will look for an AWS image to launch the cloud instances. This image can be easily created by following the steps in the packer-cio repo.

Default node deployment:

• 1 Swarm Master Node
  • t3.large instance type
  • 3 100GB storage drives
• 3 Swarm Worker Nodes
  • t3.large instance types
  • 3 100GB storage drives on each worker
• Default cluster region: us-west-2

During cluster initialization, Storidge will reconfigured the Swarm cluster to have 3 manager nodes for redundancy.

Download Terraform

Download template into desired project repository:

git clone https://github.com/Storidge/terraform-aws-swarm-cio.git

Initialize Terraform for project directory:

terraform init

Add credentials

Add your AWS credentials (access id and secret key) to file terraform.tfvars:

cd terraform-aws-swarm-cio
cp terraform.tfvars.template terraform.tfvars

Configure instance

Check variables.tf uses desired region.

Check instances.tf uses desired instance type and AMI.

Terraform will use your local SSH key for the deployment. Start ssh-agent and add the deployment key:

eval $(ssh-agent -s)
ssh-add ~/.ssh/id_rsa
echo $SSH_AGENT_SOCK
ssh-add -l

Before deploying, check what Terraform will build:

terraform plan

If there are no errors, run the following command to build infrastructure:

terraform apply

Login to cluster at the IP addresses listed. For example below, ssh 18.237.46.2 to login master node:

null_resource.ansible_deploy (local-exec): PLAY RECAP *********************************************************************
null_resource.ansible_deploy (local-exec): swarm-master-00            : ok=10   changed=8    unreachable=0    failed=0
null_resource.ansible_deploy (local-exec): swarm-worker-00            : ok=6    changed=4    unreachable=0    failed=0
null_resource.ansible_deploy (local-exec): swarm-worker-01            : ok=5    changed=4    unreachable=0    failed=0
null_resource.ansible_deploy (local-exec): swarm-worker-02            : ok=5    changed=4    unreachable=0    failed=0

null_resource.ansible_deploy: Creation complete after 2m41s (ID: 5166922584995052323)

Apply complete! Resources: 41 added, 0 changed, 0 destroyed.

Outputs:

swarm_master_address = 18.237.46.2
swarm_worker_address = 54.70.9.178,35.167.209.220,34.219.180.109

Check current infrastructure state:

terraform show terraform.tfstate

If the infrastructure has been changed outside terraform, update the state information:

terraform refresh

To terminate the cluster run:

terraform destroy