From 05112af15c027ed5ba627c04ce546d85c1442d19 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?St=C3=A9phane=20MERLE?=
 <95630726+smerle33@users.noreply.github.com>
Date: Wed, 26 Jun 2024 16:10:36 +0200
Subject: [PATCH] chore(infracijioagent1/cijioagent1): enable autoscaling for
 the system pool (#758)

during the upgrade to kubernetes 1.28
https://github.com/jenkins-infra/helpdesk/issues/4144, we saw a long
time to upgrade system pool and this may be due to the fact that the
system pools of those 2 clusters: `cijioagents1` and `infracijioagent1`
are not on autoscale.
This may improve next upgrade.
---
 ci.jenkins.io-kubernetes-agents.tf                | 5 +++--
 infraci.jenkins.io-kubernetes-sponsored-agents.tf | 5 +++--
 2 files changed, 6 insertions(+), 4 deletions(-)

diff --git a/ci.jenkins.io-kubernetes-agents.tf b/ci.jenkins.io-kubernetes-agents.tf
index 033f9547..82d3e953 100644
--- a/ci.jenkins.io-kubernetes-agents.tf
+++ b/ci.jenkins.io-kubernetes-agents.tf
@@ -54,8 +54,9 @@ resource "azurerm_kubernetes_cluster" "cijenkinsio_agents_1" {
     os_disk_size_gb      = 150 # Ref. Cache storage size athttps://learn.microsoft.com/fr-fr/azure/virtual-machines/dasv5-dadsv5-series#dadsv5-series (depends on the instance size)
     orchestrator_version = local.kubernetes_versions["cijenkinsio_agents_1"]
     kubelet_disk_type    = "OS"
-    enable_auto_scaling  = false
-    node_count           = 3 # 3 nodes for HA as per AKS best practises
+    enable_auto_scaling  = true
+    min_count            = 2 # for best practises
+    max_count            = 3 # for upgrade
     vnet_subnet_id       = data.azurerm_subnet.ci_jenkins_io_kubernetes_sponsorship.id
     tags                 = local.default_tags
     zones                = local.cijenkinsio_agents_1_compute_zones
diff --git a/infraci.jenkins.io-kubernetes-sponsored-agents.tf b/infraci.jenkins.io-kubernetes-sponsored-agents.tf
index 3c902ce9..393fd235 100644
--- a/infraci.jenkins.io-kubernetes-sponsored-agents.tf
+++ b/infraci.jenkins.io-kubernetes-sponsored-agents.tf
@@ -53,8 +53,9 @@ resource "azurerm_kubernetes_cluster" "infracijenkinsio_agents_1" {
     os_disk_size_gb      = 150 # Ref. Cache storage size athttps://learn.microsoft.com/fr-fr/azure/virtual-machines/dasv5-dadsv5-series#dadsv5-series (depends on the instance size)
     orchestrator_version = local.kubernetes_versions["infracijenkinsio_agents_1"]
     kubelet_disk_type    = "OS"
-    enable_auto_scaling  = false
-    node_count           = 3 # 3 nodes for HA as per AKS best practises
+    enable_auto_scaling  = true
+    min_count            = 2 # for best practises
+    max_count            = 3 # for upgrade
     vnet_subnet_id       = data.azurerm_subnet.infraci_jenkins_io_kubernetes_agent_sponsorship.id
     tags                 = local.default_tags
     zones                = local.infracijenkinsio_agents_1_compute_zones