terraform-bluegreen

Terraform module to setup blue / green deployments

blue-green

Inputs

Name	Description	Type	Default	Required
associate_public_ip_address	Associate a public ip address with an instance in a VPC	bool	false	no
blue_ami	The EC2 image ID to launch in the Blue autoscaling group	string	n/a	yes
blue_desired_capacity	The number of Amazon EC2 instances that should be running in the blue autoscaling roup	number	n/a	yes
blue_disk_volume_size	The size of the EBS volume in GB for the Blue instances	number	8	no
blue_disk_volume_type	The EBS volume type for the Blue instances	string	"gp2"	no
blue_instance_type	The Blue instance type to launch	string	n/a	yes
blue_max_size	The maximum size of the blue autoscaling group	number	n/a	yes
blue_min_size	The minimum size of the blue autoscaling group	number	n/a	yes
blue_user_data	The user data to provide when launching the Blue instances	string	"# Hello World"	no
green_ami	The EC2 image ID to launch in the Green autoscaling group	string	n/a	yes
green_desired_capacity	The number of Amazon EC2 instances that should be running in the green autoscaling roup	number	n/a	yes
green_disk_volume_size	The size of the EBS volume in GB for the Green instances	number	8	no
green_disk_volume_type	The EBS volume type for the Green instances	string	"gp2"	no
green_instance_type	The Green instance type to launch	string	n/a	yes
green_max_size	The maximum size of the green autoscaling group	number	n/a	yes
green_min_size	The minimum size of the green autoscaling group	number	n/a	yes
green_user_data	The user data to provide when launching the Green instances	string	"# Hello World"	no
health_check_grace_period	Time (in seconds) after instance comes into service before checking health	number	300	no
health_check_type	The health check type to apply to the Autoscaling groups.	string	"ELB"	no
iam_instance_profile	The IAM instance profile to associate with launched instances	string	null	no
initial_lifecycle_hooks	One or more Lifecycle Hooks to attach to the autoscaling group before instances are launched. The syntax is exactly the same as the separate aws_autoscaling_lifecycle_hook resource, without the autoscaling_group_name attribute	list(map(string))	[]	no
key_name	The key name that should be used for the instance	string	null	no
loadbalancers	A list of load balancer names to add to the autoscaling groups	list(string)	[]	no
name	Name of the Auto Scaling Groups	string	n/a	yes
security_groups	A list of associated security group IDS	list(string)	[]	no
spot_price	Spot price you want to pay for your instances. By default this is empty and we will use on-demand instances	string	""	no
subnets	A list of subnet IDs to launch resources in	list(string)	[]	no
tags	List as a map of additional tags	list(map(string))	[]	no
target_group_arns	A list of aws_alb_target_group ARNs, for use with Application Load Balancing	list(string)	[]	no
termination_policies	Order in termination policies to apply when choosing instances to terminate.	list(string)	[]	no
wait_for_capacity_timeout	A maximum duration that Terraform should wait for ASG instances to be healthy before timing out. Setting this to 0 causes Terraform to skip all Capacity Waiting behavior.	string	"10m"	no

Outputs

Name	Description
blue_asg_id	Blue autoscaling group id
green_asg_id	Green autoscaling group id
nonbinding_blue_asg_name	This is the raw blue autoscaling group name, without creating a dependency with the actual autoscaling group resource
nonbinding_green_asg_name	This is the raw green autoscaling group name, without creating a dependency with the actual autoscaling group resource

Example

module "bluegreen" {
  source                 = "github.com/skyscrapers/terraform-bluegreen//blue-green"
  name                   = "app-${terraform.workspace}"
  blue_ami               = "ami-blabla"
  blue_instance_type     = "t3.micro"
  blue_max_size          = 5
  blue_min_size          = 2
  blue_desired_capacity  = 2
  green_ami              = "ami-blabla"
  green_instance_type    = "t3.micro"
  green_max_size         = 0
  green_min_size         = 0
  green_desired_capacity = 0
  loadbalancers          = ["myloadbalancers"]
  security_groups        = ["mysecuritygroups"]
}

scaling

Terraform module to setup alarms and autoscaling triggers for autoscaling

Inputs

Name	Description	Type	Default	Required
adjustment_down	The number of instances to remove when the alarm is triggered (the value has to be negative)	number	-1	no
adjustment_type	What typ of adjustment needs to happen	string	"ChangeInCapacity"	no
adjustment_up	The number of instances to add when the alarm is triggered	number	1	no
autoscaling_group_names	The names of the Auto Scaling Groups this config needs to be applied to	list(string)	n/a	yes
cooldown_down	The amount of time, in seconds, after a scaling activity completes and before the next scaling activity can start.	number	600	no
cooldown_up	The amount of time, in seconds, after a scaling activity completes and before the next scaling activity can start.	number	300	no
datapoints_to_alarm_down	The number of datapoints that must be breaching to trigger the scale DOWN alarm	number	n/a	yes
datapoints_to_alarm_up	The number of datapoints that must be breaching to trigger the scale UP alarm	number	n/a	yes
dimension_name	n/a	string	"AutoScalingGroupName"	no
dimension_value	n/a	string	n/a	yes
evaluation_periods	The number of samples to evaluate	number	4	no
metric_name	The metric the scaling is based upon	string	"CPUUtilization"	no
name	Name for the created resources	string	n/a	yes
namespace	The namespace of the cloudwatch metric	string	"AWS/EC2"	no
num_asg	The number of autoscaling groups passed	number	2	no
period_down	The period in seconds over which the selected metric statistic is applied.	number	120	no
period_up	The period in seconds over which the selected metric statistic is applied.	number	60	no
policy_type	The policy type, either SimpleScaling or StepScaling	string	"SimpleScaling"	no
statistic	The statistic to apply to the alarm's associated metric. Either of the following is supported:	string	"Average"	no
threshold_down	The metric value to scale down	number	30	no
threshold_up	The metric value to scale up	number	80	no

Example

module "scaling" {
  source                  = "github.com/skyscrapers/terraform-bluegreen//scaling"
  name                    = "app-${terraform.workspace}"
  autoscaling_group_names = ["my_asg_name1","my_asg_name2"]
}

Blue-green deployments

The bluegreen.py script performs a bluegreen deployment of the selected terraform stack. It only works with Python 2.7. The blue-green deployment script expects certain inputs and outputs in the Terraform project you want to deploy in a blue-green fashion.

Required outputs

output "blue_asg_id" {
  value = "${module.<blue-green-module-name>.blue_asg_id}"
}

output "green_asg_id" {
  value = "${module.<blue-green-module-name>.green_asg_id}"
}

Required variables

variable "blue_max_size" {
  description = "max instances blue"
}

variable "blue_min_size" {
  description = "min instances blue"
}

variable "blue_desired_capacity" {
  description = "desired instances blue"
}

variable "green_max_size" {
  description = "max instances green"
}

variable "green_min_size" {
  description = "min instances green"
}

variable "green_desired_capacity" {
  description = "desired instances green"
}

variable "blue_ami" {
  description = "blue ami"
}

variable "green_ami" {
  description = "green ami"
}

These input variables need to be assigned to your module. Make sure you do a terraform refresh to be sure that the terraform outputs are saved in the Terraform state file.

Usage

First you need to install the requirements:

pip install -r requirements.txt

Run ./bluegreen.py --help to see the available options.

Example:

./bluegreen.py -f stacks/test/application -a ami-xxxx -c apply -t 500