fix horizontal pod autoscaling rules (#140)

* fix horizontal pod autoscaling rules --------- Co-authored-by: ranchodeluxe <[email protected]>
developmentseed · Sep 4, 2024 · a7119de · a7119de
1 parent 8c1f812
commit a7119de
Show file tree

Hide file tree

Showing 8 changed files with 18 additions and 10 deletions.
diff --git a/docs/autoscaling.md b/docs/autoscaling.md
@@ -187,6 +187,10 @@ with the `release` name we installed the chart with below `<release-name>-grafan
    kubectl get svc -n eoapi-support
    ```
 
+3. Login and you should be default be able to see the eoapi-k8s grafana dashboard
+
+   ![](./images/gfdashboard.png)
+
 ### Install or Upgrade Autoscaling Changes to `eoapi` Chart
 
 1. If you haven't already decide which services (`vector` || `raster` || `stac`) you want to enable `autoscaling` on change your values yaml for these and redeploy

diff --git a/docs/images/gfdashboard.png b/docs/images/gfdashboard.png
diff --git a/docs/loadtesting.md b/docs/loadtesting.md
@@ -18,8 +18,6 @@ export INGRESS_ENDPOINT=$(kubectl -n ingress-nginx  get svc/ingress-nginx-contro
 # k8s-eoapi-ingressn-404721dbb4-e6dec70321c3eddd.elb.us-west-2.amazonaws.com
 ```
 
-*
-
 2. Then run some naive load testing against some static read-only endpoints in a couple different terminals 
 
 ```sh

diff --git a/helm-chart/eoapi-support/Chart.yaml b/helm-chart/eoapi-support/Chart.yaml
@@ -1,8 +1,8 @@
 apiVersion: v2
 name: eoapi-support
 
-appVersion: "0.1.5"
-version: "0.1.5"
+appVersion: "0.1.6"
+version: "0.1.6"
 
 dependencies:
   # Prometheus for collection of metrics.

diff --git a/helm-chart/eoapi-support/values.yaml b/helm-chart/eoapi-support/values.yaml
@@ -162,3 +162,8 @@ grafana:
 
   dashboardsConfigMaps:
     default: "eoapi-dashboards"
+
+
+metrics-server:
+  apiService:
+    create: true
diff --git a/helm-chart/eoapi/Chart.yaml b/helm-chart/eoapi/Chart.yaml
@@ -15,13 +15,13 @@ kubeVersion: ">=1.23.0-0"
 # This is the chart version. This version number should be incremented each time you make changes
 # to the chart and its templates, including the app version.
 # Versions are expected to follow Semantic Versioning (https://semver.org/)
-version: "0.4.8"
+version: "0.4.9"
 
 # This is the version number of the application being deployed. This version number should be
 # incremented each time you make changes to the application. Versions are not expected to
 # follow Semantic Versioning. They should reflect the version the application is using.
 # It is recommended to use it with quotes.
-appVersion: "0.3.14"
+appVersion: "0.3.15"
 
 dependencies:
   - name: postgrescluster

diff --git a/helm-chart/eoapi/templates/services/hpa.yaml b/helm-chart/eoapi/templates/services/hpa.yaml
@@ -25,7 +25,8 @@ spec:
       target:
         type: Utilization
         averageUtilization: {{ index $v "autoscaling" "targets" "cpu" }}
-  {{- else if or (eq (index $v "autoscaling" "type") "requestRate") (eq (index $v "autoscaling" "type") "both") }}
+  {{- end }}
+  {{- if or (eq (index $v "autoscaling" "type") "requestRate") (eq (index $v "autoscaling" "type") "both") }}
   # NOTE: 'Object' are custom metrics using third-party plugins such as prometheus + prometheus-adapter
   # SEE: ../../../docs/autoscaling.md
   - type: Object

diff --git a/helm-chart/eoapi/values.yaml b/helm-chart/eoapi/values.yaml
@@ -192,7 +192,7 @@ raster:
       # https://kubernetes.io/docs/tasks/run-application/horizontal-pod-autoscale-walkthrough/#quantities
       # so when the average unit among these pods is <requestRate>/1000 then scale
       # you can watch the actual/target in real time using `kubectl get hpa/<name>`
-      requestRate: 1000000m
+      requestRate: 100000m
   image:
     name: ghcr.io/stac-utils/titiler-pgstac
     tag: uvicorn-1.2.0
@@ -254,7 +254,7 @@ stac:
       # https://kubernetes.io/docs/tasks/run-application/horizontal-pod-autoscale-walkthrough/#quantities
       # so when the average unit among these pods is <requestRate>/1000 then scale
       # you can watch the actual/target in real time using `kubectl get hpa/<name>`
-      requestRate: 15000000m
+      requestRate: 100000m
   image:
     name: ghcr.io/stac-utils/stac-fastapi-pgstac
     tag: 2.4.9
@@ -303,7 +303,7 @@ vector:
       # https://kubernetes.io/docs/tasks/run-application/horizontal-pod-autoscale-walkthrough/#quantities
       # so when the average unit among these pods is <requestRate>/1000 then scale
       # you can watch the actual/target in real time using `kubectl get hpa/<name>`
-      requestRate: 1000000m
+      requestRate: 100000m
   image:
     name: ghcr.io/developmentseed/tipg
     tag: uvicorn-0.6.1
-Original file line number
+Diff line change
@@ Expand Up @@
     # k8s-eoapi-ingressn-404721dbb4-e6dec70321c3eddd.elb.us-west-2.amazonaws.com
     ```
-    *
 . Then run some naive load testing against some static read-only endpoints in a couple different terminals
     ```sh
@@ Expand Down @@