How to Fix “Error Running Load Balancer Syncing Routine” in GKE

Ensure we are using the correct annotation based on the load balancer type.

Here is an external Load Balancer example:

apiVersion: networking.k8s.io/v1

kind: Ingress

metadata:

name: external-ingress

annotations:

kubernetes.io/ingress.class: "gce"

spec:

rules:

- http:

paths:

- path: /

pathType: Prefix

backend:

service:

name: my-service

port:

number: 80

Here is an internal Load Balancer example:

apiVersion: networking.k8s.io/v1

kind: Ingress

metadata:

name: internal-ingress

annotations:

kubernetes.io/ingress.class: "gce-internal"

spec:

rules:

- http:

paths:

- path: /

pathType: Prefix

backend:

service:

name: my-service

port:

number: 80

This step is particularly relevant during the initial stages of Kubernetes cluster deployment on platforms like Proxmox 8, where misconfiguration can easily occur.

1. We need to create a dedicated proxy-only subnet. So, open Google Cloud Console.
2. Then, go to the VPC Network and select the “Subnets” section.
3. Next, click “Add Subnet” and configure the subnet with:
   • Purpose: Proxy-only subnet
   • Region: Match cluster region
   • IP range: Dedicated CIDR block
   • Network: Cluster’s VPC network

Here is a Subnet Configuration example:

gcloud compute networks subnets create proxy-subnet \

--network=my-vpc \

--region=us-central1 \

--range=10.200.0.0/24 \

--purpose=REGIONAL_MANAGED_PROXY_ONLY

Ensure the region and VPC match the GKE cluster.

We have to reserve a static IP and reference it properly in the Ingress manifest.

To reserve a static IP in Google Cloud:

gcloud compute addresses create my-static-ip --global –ip-version=IPV4

Here is the Ingress manifest configuration:

apiVersion: networking.k8s.io/v1

kind: Ingress

metadata:

name: static-ip-ingress

annotations:

kubernetes.io/ingress.global-static-ip-name: "my-static-ip"

spec:

rules:

- http:

paths:

- path: /

pathType: Prefix

backend:

service:

name: my-service

port:

number: 80

Furthermore, use the correct annotation for static IP:

kubernetes.io/ingress.global-static-ip-name: static-ip-name

Also, remember to verify that the IP address exists in the Google Cloud project

Verify that the BackendConfig resource exists and is correctly linked. If it doesn’t exist, create a BackendConfig resource, add a service annotation, and verify the resource configuration.

Here is a BackendConfig Example:

apiVersion: cloud.google.com/v1

kind: BackendConfig

metadata:

name: my-backendconfig

spec:

timeoutSec: 3600

connectionDraining:

drainingTimeoutSec: 60

Here is the Service Annotation:

apiVersion: v1

kind: Service

metadata:

name: my-service

annotations:

cloud.google.com/backend-config: '{"default": "my-backendconfig"}'

spec:

type: NodePort

selector:

app: my-app

ports:

- port: 80

Also, remember to check for spelling mistakes in BackendConfig references.

If you encounter errors like BackoffLimitExceeded during job execution, that can also stem from similar misconfiguration issues. Refer to our guide on Kubernetes BackoffLimitExceeded errors for additional troubleshooting tips.

In this case, add NEG annotation to the Service manifest:

cloud.google.com/neg: '{"ingress": true}'

Furthermore, ensure that proper Network Policy configurations are in place. We also need to verify the creation of NEG in shared VPC environments.

The NEG Configuration involves enabling container-native load balancing, adding NEG annotation to the service, and verifying network policies.

Here is the NEG Service Annotation:

apiVersion: v1

kind: Service

metadata:

name: my-service

annotations:

cloud.google.com/neg: '{"ingress": true}'

spec:

selector:

app: my-app

ports:

- port: 80

We can verify the NEGs with this command:

kubectl get networkendpointgroups -n <namespace>

Suppose you’re managing your GKE environments through dashboards. In that case, you may also find our walkthrough on setting up the OVH Kubernetes dashboard helpful, especially when visualizing network-related components such as NEGs and Ingresses.