display_name description icon verified tags
AWS AMI Snapshot Create and manage AMI snapshots for Coder workspaces with restore capabilities ../../../../.icons/aws.svg false
aws
snapshot
ami
backup
persistence

AWS AMI Snapshot Module

This module provides AMI-based snapshot functionality for Coder workspaces running on AWS EC2 instances. It enables users to create snapshots when workspaces are stopped and restore from previous snapshots when starting workspaces.

module "ami_snapshot" {
  source  = "registry.coder.com/mavrickrishi/aws-ami-snapshot/coder"
  version = "1.0.0"

  instance_id    = aws_instance.workspace.id
  default_ami_id = data.aws_ami.ubuntu.id
  template_name  = "aws-linux"
}

Features

  • Automatic Snapshots: Create AMI snapshots when workspaces are stopped
  • User Control: Enable/disable snapshot functionality per workspace
  • Custom Labels: Add custom labels to snapshots for easy identification
  • Snapshot Selection: Choose from available snapshots when starting workspaces
  • Automatic Cleanup: Optional Data Lifecycle Manager integration for automated cleanup
  • Workspace Isolation: Snapshots are tagged and filtered by workspace and owner

Parameters

The module exposes the following parameters to workspace users:

  • enable_snapshots: Enable/disable AMI snapshot creation (default: true)
  • snapshot_label: Custom label for the snapshot (optional)
  • use_previous_snapshot: Select a previous snapshot to restore from (default: none)

Usage

Basic Usage

module "ami_snapshot" {
  source = "registry.coder.com/modules/aws-ami-snapshot"

  instance_id     = aws_instance.workspace.id
  default_ami_id  = data.aws_ami.ubuntu.id
  template_name   = "aws-linux"
}

resource "aws_instance" "workspace" {
  ami           = module.ami_snapshot.ami_id
  instance_type = "t3.micro"

  # Prevent Terraform from recreating instance when AMI changes
  lifecycle {
    ignore_changes = [ami]
  }
}

With Optional Cleanup

module "ami_snapshot" {
  source = "registry.coder.com/modules/aws-ami-snapshot"

  instance_id               = aws_instance.workspace.id
  default_ami_id           = data.aws_ami.ubuntu.id
  template_name            = "aws-linux"
  enable_dlm_cleanup       = true
  dlm_role_arn            = aws_iam_role.dlm_lifecycle_role.arn
  snapshot_retention_count = 5

  tags = {
    Environment = "development"
    Project     = "my-project"
  }
}

IAM Role for DLM (Optional)

If using automatic cleanup, create an IAM role for Data Lifecycle Manager:

resource "aws_iam_role" "dlm_lifecycle_role" {
  name = "dlm-lifecycle-role"

  assume_role_policy = jsonencode({
    Version = "2012-10-17"
    Statement = [
      {
        Action = "sts:AssumeRole"
        Effect = "Allow"
        Principal = {
          Service = "dlm.amazonaws.com"
        }
      }
    ]
  })
}

resource "aws_iam_role_policy_attachment" "dlm_lifecycle" {
  role       = aws_iam_role.dlm_lifecycle_role.name
  policy_arn = "arn:aws:iam::aws:policy/service-role/AWSDataLifecycleManagerServiceRole"
}

Required IAM Permissions

Users need the following IAM permissions for full functionality:

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "ec2:CreateImage",
        "ec2:DescribeImages",
        "ec2:DescribeInstances",
        "ec2:CreateTags",
        "ec2:DescribeTags"
      ],
      "Resource": "*"
    },
    {
      "Effect": "Allow",
      "Action": [
        "dlm:CreateLifecyclePolicy",
        "dlm:GetLifecyclePolicy",
        "dlm:UpdateLifecyclePolicy",
        "dlm:DeleteLifecyclePolicy"
      ],
      "Resource": "*",
      "Condition": {
        "StringEquals": {
          "dlm:Target": "INSTANCE"
        }
      }
    }
  ]
}

How It Works

  1. Snapshot Creation: When a workspace transitions to "stop", an AMI snapshot is automatically created (if enabled)
  2. Tagging: Snapshots are tagged with workspace name, owner, template, and custom labels
  3. Snapshot Retrieval: Available snapshots are retrieved and presented as options for workspace start
  4. AMI Selection: The module outputs the appropriate AMI ID (default or selected snapshot)
  5. Cleanup: Optional DLM policies can automatically clean up old snapshots

Considerations

  • Cost: AMI snapshots incur storage costs. Use cleanup policies to manage costs
  • Time: AMI creation takes time; workspace stop operations may take longer
  • Permissions: Ensure proper IAM permissions for AMI creation and management
  • Region: Snapshots are region-specific and cannot be used across regions
  • Lifecycle: Use ignore_changes = [ami] on EC2 instances to prevent conflicts

Examples

See the updated AWS templates that use this module: