v0.28.0

Adding capacity-rebalance as a new option to aws_autoscaling_group @leonardomn (#73)

what

• When Capacity Rebalancing is enabled for an Auto Scaling group, Amazon EC2 Auto Scaling attempts to proactively replace Spot Instances in the group that have received a rebalance recommendation, providing the opportunity to rebalance your workload to new Spot Instances that are not at elevated risk of interruption.

why

• We use this module and this feature is needed for our own project. Since we have our internal fork, I don't see why not push it to the open-source version and give everyone the new ability to use when needed the capacity_rebalance.

references

• https://docs.aws.amazon.com/autoscaling/ec2/userguide/capacity-rebalance.html
• This new feature/input isn't related to any issues at all, only a nice have to option.

v0.27.0

Fix instance profile not present bug @cdobbyn (#72)

what

• This block should not be defined if no instance profile name is present.

why

• If ec2_instance_profile variable is not present terraform will want to continually apply the following block on the launch template every time terraform apply occurs. Additionally terraform wants to auto-increment the latest_version.

iam_instance_profile {}

references

closes #71

v0.26.0

Add `warm_pool` @nitrocode (#70)

what

• Add warm_pool

why

A warm pool gives you the ability to decrease latency for your applications that have exceptionally long boot times, for example, because instances need to write massive amounts of data to disk. With warm pools, you no longer have to over-provision your Auto Scaling groups to manage latency in order to improve application performance.

references

• https://docs.aws.amazon.com/autoscaling/ec2/userguide/ec2-auto-scaling-warm-pools.html

notes

All inputs have to be provided e.g.

This works

  warm_pool = {
    pool_state                  = "Running"
    min_size                    = 1
    max_group_prepared_capacity = null
  }

This fails

  warm_pool = {
    pool_state                  = "Running"
    min_size                    = 1
  }

The only way around this is to use the optional() variable feature which is still experimental.

v0.25.0

Add autoscaling policy arn to outputs @linhkikuchi (#69)

what

• To add autoscaling policy scale up and down arn to outputs

v0.24.1

🐛 Bug Fixes

Remove duplicate tags from launch template @Nuru (#68)

what

• Remove duplicate tags from launch template

why

• Apparently due to merge errors and oversights, #56 restored a static tag specification removed by #34 and replaced with dynamic tag specifications. The result is that launch templates were getting created with duplicate "volume" tags.

v0.24.0

set desired_capacity in aws_autoscaling_group as optional @ByJacob (#67)

what

• set desired_capacity in aws_autoscaling_group as optional

why

• When auto-scaling is set to 2 and updating asg, the number of instances drops to the minimum value

references

• #66

v0.23.1

🐛 Bug Fixes

Support Terraform 0.15 @Nuru (#65)

what

• Document that this module can only create SimpleScaling autoscaling policies. Closes #55
• Remove obsolete call to list function. Closes #64

v0.23.0

Normalize metadata service, revert option to use fixed name @Nuru (#63)

what

• Provide Cloud Posse standard interface to IMDS configuration
• Remove var.use_name_prefix parameter
• Set lifecycle to ignore changes to desired_capacity
• Update module boilerplate to current

why

• IMDS:
  • Replaces #56 to keep interfaces standard across Cloud Posse modules
  • Use secure settings (as defined by Bridgecrew) by default
  • Expose all IMDS configuration options
  • Provide sensible default with option to override and make more (or less) secure
• var.use_name_prefix
  • This module has always set create_before_destroy on the autoscaling_group
  • Terraform does not provide a way to make create_before_destroy configurable, so this module only supports it being set
  • You must use name_prefix with create_before_destroy or else a replacement operation will fail
  • The implementation of var.use_name_prefix in #60 left the name with a trailing delimiter and did not support directly specifying the name, making it unlikely to be suitable for the purported use case #28
• It is normal and expected for an external source to change the desired capacity of the autoscaling group and Terraform should not override that. If desired, you can set the size of the group by changing min_size and max_size to both be the desired size. #29 #37 #58

v0.22.0

Update example subnet module @nitrocode (#62)

what

• Upgraded subnet module to latest https://github.com/cloudposse/terraform-aws-dynamic-subnets

why

• Looks like it's causing issues with bridgecrew

references

• N/A

v0.21.0

Expose metadata HTTP tokens @alexjurkiewicz (#56)

Expose aws_launch_template -> metadata_options -> http_tokens.