Google cloud networking with Kubernetes Engine clusters can be complex. Assigning optimal CIDR ranges for the relevant VPC subnets and the Kubernetes Engine clusters' reserved IP ranges from the start is very important since VPC subnets are not always easy to resize and the cluster's reserved IP ranges are immutable. Using the correct method to expose the applications in the cluster is important as every method was designed for a different set of use cases.
The demos in the project demonstrate the following best practices:
- Connecting two GCP networks using VPC peering and Cloud VPN containing two Kubernetes Engine clusters each.
- Deploying the nginx pods.
- Exposing the pods using Kubernetes Engine services
- Validating pod-to-service communication across the Kubernetes Engine clusters within the same region and the different regions.
If you use Google Cloud Shell the prerequistite tools are installed. Otherwise install:
- Install gcloud from https://cloud.google.com/sdk/downloads
- Install kubectl with "gcloud components install kubectl"
macOS, Linux and using Google Cloud Console are supported.
- gcloud cli ( >= Google Cloud SDK 200.0.0 )
- bash
- kubectl - ( >= v1.10.0-gke.0 )
- Kubernetes Engine >= 1.10.0-gke.0
- The gke-to-gke-vpc-peering and gke-to-gke-vpn folders each contain a project.
- README files exist for the above examples; gke-to-gke-vpc-peering/README.md and gke-to-gke-vpn/README.md.
- The network folder contains the manifest files and deployment manager templates to setup networks.
- The clusters folder contains the manifest files and deployment manager templates to create Kubernetes Engine clusters.
- The manifests folder contains the manifest files to create Kubernetes Engine services.
This is not an officially supported Google product