This Terraform module sets up everything necessary for dynamically setting hostnames following a certain pattern on instances spawned by AWS Auto Scaling Groups (ASGs).
Learn more about our motivation to build this module in our blog post Dynamic Route53 records for AWS Auto Scaling Groups with Terraform.
This repository and the module it houses are maintained Foundation Missions A-Team. Should you encounter issues or require changes to code maintained in this repository, please reachout through an issue that is part of this project.
- Terraform 0.12+ and Terraform AWS provider 2.0+ use v2+ releases
- Terraform 0.11 and below, Terraform AWS provider 2.0.14 and below use v1.x releases
Create an ASG and set the asg:hostname_pattern
tag for example like this:
asg-test-#instanceid.asg-handler-vpc.testing@Z3QP9GZSRL8IVA
#instanceid
is converted by a Lambda function within this module to the actual AWS instance_id that corresponds to the launched instance. The @
symbol is used to split the FQDN from the Route 53 zone_id.
This could be interpolated in Terraform like this:
tag {
key = "asg:hostname_pattern"
value = "${var.hostname_prefix}-#instanceid.${var.vpc_name}.testing@${var.internal_zone_id}"
propagate_at_launch = true
}
Once you have your ASG set up, you can just invoke this module and point to it:
module "clever_name_autoscale_dns" {
source = meltwater/asg-dns-handler/aws"
version = "~> 2.0"
# use_public_ip = true
# route53_record_ttl = 300
autoscale_handler_unique_identifier = "clever_name"
autoscale_route53zone_arn = "ABCDEFGHIJ123"
vpc_name = "my_vpc"
}
The module sets up these things:
- A SNS topic
- A Lambda function
- A topic subscription sending SNS events to the Lambda function
The Lambda function then does the following:
- Fetch the
asg:hostname_pattern
tag value from the ASG, and parse out the hostname and Route53 zone ID from it. - If it's an instance being created
- Fetch internal IP from EC2 API
- Create a Route53 record pointing the hostname to the IP
- Set the Name tag of the instance to the initial part of the generated hostname
- If it's an instance being deleted
- Fetch the internal IP from the existing record from the Route53 API
- Delete the record
Add initial_lifecycle_hook
definitions to your aws_autoscaling_group
resource , like so:
resource "aws_autoscaling_group" "my_asg" {
name = "myASG"
vpc_zone_identifier = var.aws_subnets
min_size = var.asg_min_count
max_size = var.asg_max_count
desired_capacity = var.asg_desired_count
health_check_type = "EC2"
health_check_grace_period = 300
force_delete = false
launch_configuration = aws_launch_configuration.my_launch_config.name
lifecycle {
create_before_destroy = true
}
initial_lifecycle_hook {
name = "lifecycle-launching"
default_result = "ABANDON"
heartbeat_timeout = 60
lifecycle_transition = "autoscaling:EC2_INSTANCE_LAUNCHING"
notification_target_arn = module.autoscale_dns.autoscale_handling_sns_topic_arn
role_arn = module.autoscale_dns.agent_lifecycle_iam_role_arn
}
initial_lifecycle_hook {
name = "lifecycle-terminating"
default_result = "ABANDON"
heartbeat_timeout = 60
lifecycle_transition = "autoscaling:EC2_INSTANCE_TERMINATING"
notification_target_arn = module.autoscale_dns.autoscale_handling_sns_topic_arn
role_arn = module.autoscale_dns.agent_lifecycle_iam_role_arn
}
tag {
key = "asg:hostname_pattern"
value = "${var.hostname_prefix}-#instanceid.${var.vpc_name}.testing@${var.internal_zone_id}"
propagate_at_launch = true
}
}
module "autoscale_dns" {
source = "meltwater/asg-dns-handler/aws"
version = "2.1.7"
autoscale_handler_unique_identifier = "my_asg_handler"
autoscale_route53zone_arn = var.internal_zone_id
vpc_name = var.vpc_name
}
Please read CONTRIBUTING.md to understand how to submit pull requests to us, and also see our Code of Conduct.
Lifecycle_hook can have CONTINUE
or ABANDON
as default_result. By setting default_result to ABANDON
will terminate the instance if the lambda function fails to update the DNS record as required. Complete_lifecycle_action
in lambda function returns LifecycleActionResult
as CONTINUE
on success to Lifecycle_hook. But if lambda function fails, Lifecycle_hook doesn't get any response from Complete_lifecycle_action
which results in timeout and terminates the instance.
At the conclusion of a lifecycle hook, the result is either ABANDON or CONTINUE. If the instance is launching, CONTINUE indicates that your actions were successful, and that the instance can be put into service. Otherwise, ABANDON indicates that your custom actions were unsuccessful, and that the instance can be terminated.
If the instance is terminating, both ABANDON and CONTINUE allow the instance to terminate. However, ABANDON stops any remaining actions, such as other lifecycle hooks, while CONTINUE allows any other lifecycle hooks to complete.
This project was built at Meltwater. It is licensed under the Apache License 2.0.