Add Support for Custom Metrics (#196)

* add support for hpa custom scaling metrics

* update CI builder image to internal ci image

* update pre-commit setup to align with ci builder image

* update minikube version in ci

* revert minikube upgrade

* revert test specific minikube upgrade

Author: ryehowell

.circleci/config.yml

@@ -7,7 +7,7 @@ base_env: &base_env
   GRUNTWORK_INSTALLER_VERSION: v0.0.38
   TERRATEST_LOG_PARSER_VERSION: v0.40.6
   HELM_VERSION: v3.8.0
-  MODULE_CI_VERSION: v0.51.0
+  MODULE_CI_VERSION: v0.55.1
   MINIKUBE_VERSION: v1.28.0
   TERRAFORM_VERSION: NONE
   TERRAGRUNT_VERSION: NONE
@@ -76,25 +76,18 @@ jobs:
     environment:
       <<: *base_env
     docker:
-      - image: cimg/python:3.10.2
+      - image: 087285199408.dkr.ecr.us-east-1.amazonaws.com/circle-ci-test-image-base:go1.21.9-tf1.5-tg39.1-pck1.8-ci54.0
     steps:
       - checkout
       # Install gruntwork utilities
       - run:
           <<: *install_gruntwork_utils
-      # Export go install location to $PATH so it is available in subsequent steps
-      - run:
-          command: |
-            echo 'export PATH=/usr/local/go/bin:$PATH' >> $BASH_ENV
       - run:
           <<: *install_helm_client
       # Fail the build if the pre-commit hooks don't pass. Note: if you run pre-commit install locally, these hooks will
       # execute automatically every time before you commit, ensuring the build never fails at this step!
       - run:
           command: |
-            pip install pre-commit
-            # Go's location was added to the PATH above
-            go get golang.org/x/tools/cmd/goimports
             pre-commit install
             pre-commit run --all-files
       - persist_to_workspace:

charts/k8s-service/templates/horizontalpodautoscaler.yaml

@@ -12,22 +12,25 @@ spec:
   minReplicas: {{ .Values.horizontalPodAutoscaler.minReplicas }}
   maxReplicas: {{ .Values.horizontalPodAutoscaler.maxReplicas }}
   metrics:
-  {{ if .Values.horizontalPodAutoscaler.avgCpuUtilization }}
+  {{- if .Values.horizontalPodAutoscaler.avgCpuUtilization }}
     - type: Resource
       resource:
         name: cpu
         target:
           type: Utilization
           averageUtilization: {{ .Values.horizontalPodAutoscaler.avgCpuUtilization }}
   {{- end }}
-  {{ if .Values.horizontalPodAutoscaler.avgMemoryUtilization }}
+  {{- if .Values.horizontalPodAutoscaler.avgMemoryUtilization }}
     - type: Resource
       resource:
         name: memory
         target:
           type: Utilization
           averageUtilization: {{ .Values.horizontalPodAutoscaler.avgMemoryUtilization }}
   {{- end }}
+  {{- if .Values.horizontalPodAutoscaler.customMetrics }}
+{{ toYaml .Values.horizontalPodAutoscaler.customMetrics | indent 4 }}
+  {{- end }}
   {{- if .Values.horizontalPodAutoscaler.behavior }}
   behavior:
 {{ tpl (toYaml .Values.horizontalPodAutoscaler.behavior) $ | indent 4 }}

charts/k8s-service/values.yaml

@@ -739,6 +739,8 @@ serviceAccount:
 #   - maxReplicas          (int)    : The maximum amount of replicas allowed
 #   - avgCpuUtilization    (int)    : The target average CPU utilization to be used with the metrics
 #   - avgMemoryUtilization (int)    : The target average Memory utilization to be used with the metrics
+#   - customMetrics        (list)   : List of objects that define custom resource metrics for scaling.
+#                                     See: https://kubernetes.io/docs/tasks/run-application/horizontal-pod-autoscale-walkthrough/#autoscaling-on-multiple-metrics-and-custom-metrics
 #   - behavior             (object) : Configures the scaling behavior of the target in both Up and Down directions 
 #                                     (scaleUp and scaleDown fields respectively). If not set, the default 
 #                                     HPAScalingRules for scale up and scale down are used.