{"id":8294,"date":"2020-03-18T09:41:44","date_gmt":"2020-03-18T16:41:44","guid":{"rendered":"https:\/\/www.couchbase.com\/blog\/?p=8294"},"modified":"2025-06-13T23:42:47","modified_gmt":"2025-06-14T06:42:47","slug":"kubernetes-vpc-peering","status":"publish","type":"post","link":"https:\/\/www.couchbase.com\/blog\/kubernetes-vpc-peering\/","title":{"rendered":"Inter-Kubernetes Networking via VPC Peering"},"content":{"rendered":"\r\n

1. Introduction<\/h2>\r\n\r\n\r\n\r\n

It is often desirable by enterprise customers to have standby database cluster\/s for data locality and high performance, disaster recovery and\/or for mere data backups. Couchbase Cross Data Center Replication (XDCR) requires no introduction as customers have been using this feature for long to achieve these goals in their environment.<\/p>\r\n\r\n\r\n\r\n

However, with more and more Couchbase deployments on the Cloud lately using Couchbase Autonomous Operator (CAO) for Kubernetes, customers have requested for some guidance in setting up networking for their cloud platform.<\/p>\r\n\r\n\r\n\r\n

This article goes into depth about configuring networks between Couchbase Clusters in two different regions, using a step-by-step approach. We have used the Amazon AWS platform to deploy the Couchbase cluster in the Amazon EKS environment but we believe the high-level approach would more or less be the same irrespective of the platform of your choice.<\/p>\r\n\r\n\r\n\r\n

2. Prerequisites<\/h2>\r\n\r\n\r\n\r\n

It is highly recommended that you read through my previous blog on how to deploy Couchbase Cluster on EKS<\/a>. Most of the detailed steps on deploying Couchbase Autonomous Operator are just going to be referenced in this article, so we can focus on networking aspects to setup cross-datacenter replication.<\/p>\r\n\r\n\r\n\r\n

3. Deploy EKS Clusters in Two Regions<\/h2>\r\n\r\n\r\n\r\n

Let’s begin by deploying two Amazon Elastic Container Service for Kubernetes (Amazon EKS) in two separate regions (Virginia and Ohio).<\/p>\r\n\r\n\r\n\r\n

\"\"<\/figure>\r\n\r\n\r\n\r\n


Figure 1<\/strong>: EKS cluster deployed in Ohio and Virginia regions.<\/p>\r\n\r\n\r\n\r\n

Each Amazon EKS cluster will have a minimum of three worker nodes, which will be used to host Couchbase pods as described in section 4. Our goal at the end of this article is that you have two Couchbase Clusters deployed on these two Amazon EKS clusters, networking configured and active XDCR established from source to the target cluster.<\/p>\r\n\r\n\r\n\r\n

3.1. Deploy EKS in Virginia region<\/h3>\r\n\r\n\r\n\r\n

We will use eksctl<\/code> command to deploy Amazon EKS, where we will create a new nodegroup called bluegroup<\/code> with minimum of three m5.large<\/code> instances and a maximum of six.<\/p>\r\n\r\n\r\n\r\n

$ eksctl create cluster \\\r\n--name blueEKS \\\r\n--version 1.14 \\\r\n--region us-east-1 \\\r\n--zones us-east-1a,us-east-1b,us-east-1c \\\r\n--nodegroup-name bluegroup \\\r\n--node-type m5.large \\\r\n--nodes 3 \\\r\n--nodes-min 3 \\\r\n--nodes-max 6 \\\r\n--node-ami auto \\\r\n--vpc-cidr 172.16.0.0\/24\r\n\r\n[\u2139]  using region us-east-1\r\n...\r\n[\u2714]  EKS cluster \"blueEKS\" in \"us-east-1\" region is ready<\/code><\/pre>\r\n\r\n\r\n\r\n
Once eksctl<\/code> finishes deploying the Kubernestes (K8s) cluster, login to AWS console to note down the VPC-ID and CIDR block. Here are the steps to find out these details, which will later use it in setting up VPC peer.<\/p>\r\n\r\n\r\n\r\n
\"\"<\/figure>\r\n\r\n\r\n\r\n

Figure 2<\/strong>: EKS console showing resource dashboard by region.<\/p>\r\n\r\n\r\n\r\n
    \r\n
  1. Select Virginia<\/code> region from the dropdown menu.<\/li>\r\n
  2. If you see the VPC Dashboard then click Your VPCs<\/code> option from the left pane. If you have a different page open then search for VPC service first and then click Your VPCs<\/code> option.<\/li>\r\n<\/ol>\r\n\r\n\r\n\r\n
    \"\"<\/figure>\r\n\r\n\r\n\r\n

    Figure 3<\/strong>: VPC detail page.<\/p>\r\n\r\n\r\n\r\n
      \r\n
    1. As can be seen above, you need to select the VPC that just got created and then copy the VPC ID<\/code> into the table below:<\/li>\r\n<\/ol>\r\n\r\n\r\n\r\n
      Table 1. Blue EKS Cluster Attributes<\/h4>\r\n\r\n\r\n\r\n
      \r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\n
      Attributes<\/th>\r\nBlue Cluster<\/th>\r\nGreen Cluster<\/th>\r\n<\/tr>\r\n<\/thead>\r\n
      Region<\/td>\r\nVirginia<\/td>\r\nOhio<\/td>\r\n<\/tr>\r\n
      CIDR block<\/td>\r\n172.16.0.0\/24<\/td>\r\n\u00a0<\/td>\r\n<\/tr>\r\n
      VPC-ID<\/td>\r\nvpc-0c8d3d5919e79659d<\/td>\r\n\u00a0<\/td>\r\n<\/tr>\r\n<\/tbody>\r\n<\/table>\r\n<\/figure>\r\n\r\n\r\n\r\n
      3.2. Deploy EKS in Ohio region<\/h3>\r\n\r\n\r\n\r\n
      Next, deploy the Kubernetes cluster in the Ohio region with 3 worker nodes of m5.large<\/code> type. The count and type of these worker nodes can be different depending on the size of the cluster but in this setup, we are going to deploy a 3 node Couchbase cluster on these 3 worker nodes.<\/p>\r\n\r\n\r\n\r\n
      $ eksctl create cluster \\\r\n--name greenEKS \\\r\n--version 1.14 \\\r\n--region us-east-2 \\\r\n--zones us-east-2a,us-east-2b,us-east-2c \\\r\n--nodegroup-name greengroup \\\r\n--node-type m5.large \\\r\n--nodes 3 \\\r\n--nodes-min 3 \\\r\n--nodes-max 6 \\\r\n--node-ami auto \\\r\n--vpc-cidr 10.0.0.0\/24\r\n\r\n[\u2139]  using region us-east-2\r\n...\r\n[\u2714]  EKS cluster \"greenEKS\" in \"us-east-2\" region is ready<\/code><\/pre>\r\n\r\n\r\n\r\n
      After the Green cluster is ready, open up a new browser tab and login to AWS console.<\/p>\r\n\r\n\r\n\r\n
      \"\"<\/figure>\r\n\r\n\r\n\r\n

      Figure 4<\/strong>: AWS console connected to Ohio region.<\/p>\r\n\r\n\r\n\r\n