Terraform module to setup blue / green deployments
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 |
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 |
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"]
}
Terraform module to setup alarms and autoscaling triggers for autoscaling
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 |
module "scaling" {
source = "github.com/skyscrapers/terraform-bluegreen//scaling"
name = "app-${terraform.workspace}"
autoscaling_group_names = ["my_asg_name1","my_asg_name2"]
}
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.
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}"
}
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.
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