LB Dynamic Backend-Pool isn't enabled when implementing CCM (Cloud Controller manager)

[RimDammak]

---

Hii guys, I am encountering an issue where the Load Balancer's Dynamic Backend-Pool feature is not enabled while implementing the Cloud Controller Manager (CCM) (Azure Provider). Below are the steps I've followed and the issue I'm facing.

1. I created Azure Cloud Provider Secret:

   kubectl create secret generic azure-cloud-provider --from-file=cloud-config=./cloud.json -n kube-system

2. I added Cloud Provider Azure helm repo:

   helm repo add cloud-provider-azure https://raw.githubusercontent.com/kubernetes-sigs/cloud-provider-azure/master/helm/repo

3. I installed Azure CCM helm-chart with the config i needed:

   helm install azure-ccm cloud-provider-azure/cloud-provider-azure \
   --set cloud-provider-azure.infra.clusterName="error-tf" \
   --set cloud-provider-azure.cloudControllerManager.cloudConfig='' \
   --set cloud-provider-azure.cloudControllerManager.cloudConfigSecretName="azure-cloud-provider" \
   --set cloud-provider-azure.cloudControllerManager.enableDynamicReloading="true" \
   --set cloud-provider-azure.cloudControllerManager.configureCloudRoutes="true" \
   --set cloud-provider-azure.cloudControllerManager.allocateNodeCidrs="true" \
   --set cloud-provider-azure.cloudControllerManager.imageRepository="mcr.microsoft.com/oss/kubernetes"

4. I edited the cloud controller manager arguments I DELETED --cloud-config=/etc/kubernetes/azure.json and added the secrets config:
   

After doing all of this, I tried to deploy an ngnix service with Load Balancer, the load balancer is created automatically with its public IP, but the issue is the LB's backend pool isn't associated to my nodes, can you help me resolve this issue? Also, I'm using Terraform, and when I destroy everything, the lb and public IP stay, any tips on how I can automatically delete them as well? Logs Messages:

   1 azure_loadbalancer.go:2259] reconcileLoadBalancer for service (default/nginx)(true): lb rules updated: [{"name":"xxxx-TCP-80","properties":{"backendAddressPool":{"id":"/subscriptions/xxxxx-/resourceGroups/rg-x/providers/Microsoft.Network/loadBalancers/kubernetes/backendAddressPools/kubernetes"},"backendPort":80,"disableOutboundSnat":false,"enableFloatingIP":true,"enableTcpReset":true,"frontendIPConfiguration":{"id":"/subscriptions/xxxxx/resourceGroups/xxxxxx-rg/providers/Microsoft.Network/loadBalancers/kubernetes/frontendIPConfigurations/xxxx"},"frontendPort":80,"idleTimeoutInMinutes":4,"loadDistribution":"Default","probe":{"id":"/subscriptions/xxxxxx/resourceGroups/xxxx/providers/Microsoft.Network/loadBalancers/kubernetes/probes/xxxxxxxx-TCP-80"},"protocol":"Tcp"}}]

[RimDammak]

Do I need to configure the azure-cloud-node-manager? (I am using a single node in my work)

[nilo19]

The master node will be ignored by default. Please try to add a worker node in your cluster.