Helm chart: can't install due to missing resources

[nlamirault]

According to the documentation:

helm repo add klt https://charts.lifecycle.keptn.sh
helm repo update
helm upgrade --install keptn klt/klt -n keptn-lifecycle-toolkit-system --create-namespace --wait

but:

> helm template keptn klt/klt -n keptn-lifecycle-toolkit-system
Error: template: klt/templates/metrics-operator-webhook-service.yaml:9:18: executing "klt/templates/metrics-operator-webhook-service.yaml" at <.Values.metricsOperatorWebhookService.type>: nil pointer evaluating interface {}.type

There is not metricsOperatorWebhookService into the default Values file.
I try with these values:

metricsOperatorWebhookService:
  ports:
  - port: 443
    protocol: TCP
    targetPort: 9443
  type: ClusterIP

then another error:

❯ helm template  .
Error: template: klt/templates/metrics-operator-manager-config.yaml:12:40: executing "klt/templates/metrics-operator-manager-config.yaml" at <.Values.metricsOperatorManagerConfig.controllerManagerConfigYaml.health.healthProbeBindAddress>: nil pointer evaluating interface {}.controllerManagerConfigYaml

Also:

Error: template: klt/templates/metrics-operator-controller-metrics-service.yaml:10:18: executing "klt/templates/metrics-operator-controller-metrics-service.yaml" at <.Values.metricsOperatorControllerMetricsService.type>: nil pointer evaluating interface {}.type

another one:

helm template .
Error: template: klt/templates/klt-cert-manager-metrics-service.yaml:13:18: executing "klt/templates/klt-cert-manager-metrics-service.yaml" at <.Values.kltCertManagerMetricsService.type>: nil pointer evaluating interface {}.type

Could you try to use these Github Actions to validate the Helm chart:

• helm/chart-testing-action
• helm/chart-releaser-action

A final diff:

diff --git charts/keptn-lifecycle-toolkit/values.yaml charts/keptn-lifecycle-toolkit/values.yaml
index 020d242..e72c2cd 100644
--- charts/keptn-lifecycle-toolkit/values.yaml
+++ charts/keptn-lifecycle-toolkit/values.yaml
@@ -50,6 +50,28 @@ lifecycleManagerConfig:
       bindAddress: 127.0.0.1:8080
     webhook:
       port: 9443
+klcManagerConfig:
+  controllerManagerConfigYaml:
+    health:
+      healthProbeBindAddress: :8081
+    leaderElection:
+      leaderElect: true
+      resourceName: 6b866dd9.keptn.sh
+    metrics:
+      bindAddress: 127.0.0.1:8080
+    webhook:
+      port: 9443
+metricsOperatorManagerConfig:
+  controllerManagerConfigYaml:
+    health:
+      healthProbeBindAddress: :8081
+    leaderElection:
+      leaderElect: true
+      resourceName: 6b866dd9.keptn.sh
+    metrics:
+      bindAddress: 127.0.0.1:8080
+    webhook:
+      port: 9443
 lifecycleOperator:
   manager:
     containerSecurityContext:
@@ -114,6 +136,30 @@ lifecycleWebhookService:
     protocol: TCP
     targetPort: 9443
   type: ClusterIP
+metricsOperatorControllerMetricsService:
+  ports:
+  - port: 443
+    protocol: TCP
+    targetPort: 9443
+  type: ClusterIP
+kltCertManagerMetricsService:
+  ports:
+  - port: 443
+    protocol: TCP
+    targetPort: 9443
+  type: ClusterIP
+klcWebhookService:
+  ports:
+  - port: 443
+    protocol: TCP
+    targetPort: 9443
+  type: ClusterIP
+klcControllerManagerMetricsService:
+  ports:
+  - port: 443
+    protocol: TCP
+    targetPort: 9443
+  type: ClusterIP
 metricsManagerConfig:
   controllerManagerConfigYaml:
     health:
@@ -181,6 +227,12 @@ metricsWebhookService:
     protocol: TCP
     targetPort: 9443
   type: ClusterIP
+metricsOperatorWebhookService:
+  ports:
+  - port: 443
+    protocol: TCP
+    targetPort: 9443
+  type: ClusterIP
 scheduler:
   nodeSelector: {}
   replicas: 1

[agardnerIT]

Hi @nlamirault thanks for reporting this. Indeed I just hit this too.

Do you have code that could be created as a PR to resolve this? If so, please create a PR and unblock both of us :D

[thisthat]

this has been solved by #1082