From 52427a42711671a745d170e0d61f32cc9f1800f0 Mon Sep 17 00:00:00 2001
From: Dalton Hubble <dghubble@gmail.com>
Date: Sun, 14 Aug 2022 21:37:26 -0700
Subject: [PATCH] Refresh instances in autoscaling group when launch
 configuration changes

* Changes to worker launch configurations start an autoscaling group instance
refresh to replace instances
* Instance refresh creates surge instances, waits for a warm-up period, then
deletes old instances
* Changing worker_type, disk_*, worker_price, worker_target_groups, or Butane
worker_snippets on existing worker nodes will replace instances
* New AMIs or changing `os_stream` will be ignored, to allow Fedora CoreOS or
Flatcar Linux to keep themselves updated
* Previously, new launch configurations were made in the same way, but not
applied to instances unless manually replaced
---
 CHANGES.md                                      | 6 ++++++
 aws/fedora-coreos/kubernetes/workers/workers.tf | 8 ++++++++
 aws/flatcar-linux/kubernetes/workers/workers.tf | 8 ++++++++
 3 files changed, 22 insertions(+)

diff --git a/CHANGES.md b/CHANGES.md
index 05b15bbb..a181cb09 100644
--- a/CHANGES.md
+++ b/CHANGES.md
@@ -32,6 +32,12 @@ version: 1.0.0
 
 ### AWS
 
+* [Refresh](https://docs.aws.amazon.com/autoscaling/ec2/userguide/asg-instance-refresh.html) instances in autoscaling group when launch configuration changes
+  * Changes to worker launch configurations start an autoscaling group instance refresh to replace instances
+  * Instance refresh creates surge instances, waits for a warm-up period, then deletes old instances
+  * Changing `worker_type`, `disk_*`, `worker_price`, `worker_target_groups`, or Butane `worker_snippets` on existing worker nodes will replace instances
+  * New AMIs or changing `os_stream` will be ignored, to allow Fedora CoreOS or Flatcar Linux to keep themselves updated
+  * Previously, new launch configurations were made in the same way, but not applied to instances unless manually replaced
 * Rename worker autoscaling group `${cluster_name}-worker` ([#1202](https://github.com/poseidon/typhoon/pull/1202))
   * Rename launch configuration `${cluster_name}-worker` instead of a random id
 
diff --git a/aws/fedora-coreos/kubernetes/workers/workers.tf b/aws/fedora-coreos/kubernetes/workers/workers.tf
index 2973845e..9508808b 100644
--- a/aws/fedora-coreos/kubernetes/workers/workers.tf
+++ b/aws/fedora-coreos/kubernetes/workers/workers.tf
@@ -22,6 +22,14 @@ resource "aws_autoscaling_group" "workers" {
     var.target_groups,
   ])
 
+  instance_refresh {
+    strategy = "Rolling"
+    preferences {
+      instance_warmup        = 120
+      min_healthy_percentage = 90
+    }
+  }
+
   lifecycle {
     # override the default destroy and replace update behavior
     create_before_destroy = true
diff --git a/aws/flatcar-linux/kubernetes/workers/workers.tf b/aws/flatcar-linux/kubernetes/workers/workers.tf
index fe298b4d..576de30b 100644
--- a/aws/flatcar-linux/kubernetes/workers/workers.tf
+++ b/aws/flatcar-linux/kubernetes/workers/workers.tf
@@ -22,6 +22,14 @@ resource "aws_autoscaling_group" "workers" {
     var.target_groups,
   ])
 
+  instance_refresh {
+    strategy = "Rolling"
+    preferences {
+      instance_warmup        = 120
+      min_healthy_percentage = 90
+    }
+  }
+
   lifecycle {
     # override the default destroy and replace update behavior
     create_before_destroy = true