chore: add EKS 1.33-1.34 support with AL2023 AMIs (#1811)

* chore: upgrade EKS clusters to support k8s 1.29-1.34, remove 1.27-1.28

* increase concurrency

* fix: use AL2023 AMIs for EKS 1.33+ clusters

* fix: add retry logic for go mod download

Author: Kausik-A

cdk_infra/Makefile

@@ -1,3 +1,3 @@
 .PHONY: deploy-infra
 deploy-infra: 
-	npm run cdk deploy -- --all --require-approval never --concurrency 4
+	npm run cdk deploy -- --all --require-approval never --concurrency 8

cdk_infra/lib/cluster-deployment.ts

@@ -56,16 +56,31 @@ export function deployClusters(
     if (clusterInterface.launch_type === 'ec2') {
       const ec2Cluster = cluster as ec2ClusterInterface;
       validateInterface(ec2Cluster);
+      
+      // Determine AMI type based on Kubernetes version and architecture
+      // K8s 1.33+ only supports AL2023, earlier versions use AL2
+      const k8sVersion = parseFloat(clusterInterface.version);
+      const isArm64 = ec2Cluster.name.match('-arm64-');
+      let amiType: NodegroupAmiType;
+      
+      if (k8sVersion >= 1.33) {
+        amiType = isArm64 
+          ? NodegroupAmiType.AL2023_ARM_64_STANDARD
+          : NodegroupAmiType.AL2023_X86_64_STANDARD;
+      } else {
+        amiType = isArm64 
+          ? NodegroupAmiType.AL2_ARM_64
+          : NodegroupAmiType.AL2_X86_64;
+      }
+      
       clusterStack = new EC2Stack(app, `${ec2Cluster.name}EKSCluster`, {
         name: ec2Cluster.name,
         vpc: vpc,
         version: versionKubernetes,
         instanceTypes: [
           new InstanceType(ec2Cluster.instance_type.toLowerCase())
         ],
-        amiType: ec2Cluster.name.match('-arm64-')
-          ? NodegroupAmiType.AL2_ARM_64
-          : NodegroupAmiType.AL2_X86_64,
+        amiType: amiType,
         env: envInput
       });
     } else {

cdk_infra/lib/stacks/eks/ec2-cluster-stack.ts

@@ -18,6 +18,10 @@ function getReleaseVersion(amiType: string, clusterVersion: string, scope: Const
     parameterName = `/aws/service/eks/optimized-ami/${clusterVersion}/amazon-linux-2/recommended/release_version`;
   } else if (amiType === NodegroupAmiType.AL2_ARM_64) {
     parameterName = `/aws/service/eks/optimized-ami/${clusterVersion}/amazon-linux-2-arm64/recommended/release_version`;
+  } else if (amiType === NodegroupAmiType.AL2023_X86_64_STANDARD) {
+    parameterName = `/aws/service/eks/optimized-ami/${clusterVersion}/amazon-linux-2023/x86_64/standard/recommended/release_version`;
+  } else if (amiType === NodegroupAmiType.AL2023_ARM_64_STANDARD) {
+    parameterName = `/aws/service/eks/optimized-ami/${clusterVersion}/amazon-linux-2023/arm64/standard/recommended/release_version`;
   } else {
     throw new Error(`Unsupported amiType: ${amiType}`);
   }

cdk_infra/lib/stacks/vpc/vpc-stack.ts

@@ -7,7 +7,7 @@ export class VPCStack extends Stack {
   constructor(scope: Construct, id: string, props?: StackProps) {
     super(scope, id, props);
     this.vpc = new ec2.Vpc(this, 'EKSVpc', {
-      cidr: '10.0.0.0/16',
+      ipAddresses: ec2.IpAddresses.cidr('10.0.0.0/16'),
       natGateways: 3,
       maxAzs: 3,
       vpnGateway: false,

mocked_servers/grpc_metrics/Dockerfile

@@ -6,7 +6,8 @@ RUN apt-get update && apt-get install -y ca-certificates && update-ca-certificat
 WORKDIR $GOPATH/main
 COPY . .
 RUN go env -w GOPROXY=direct
-RUN GO111MODULE=on go mod download
+# Retry logic for go mod download to handle transient network issues
+RUN for i in 1 2 3; do GO111MODULE=on go mod download && break || sleep 5; done
 RUN GO111MODULE=on CGO_ENABLED=0 GOOS=linux go build -o=/bin/main .
 
 FROM scratch

mocked_servers/grpc_trace/Dockerfile

@@ -6,7 +6,8 @@ RUN apt-get update && apt-get install -y ca-certificates && update-ca-certificat
 WORKDIR $GOPATH/main
 COPY . .
 RUN go env -w GOPROXY=direct
-RUN GO111MODULE=on go mod download
+# Retry logic for go mod download to handle transient network issues
+RUN for i in 1 2 3; do GO111MODULE=on go mod download && break || sleep 5; done
 RUN GO111MODULE=on CGO_ENABLED=0 GOOS=linux go build -o=/bin/main .
 
 FROM scratch

mocked_servers/https/Dockerfile

@@ -1,8 +1,13 @@
 FROM golang:1.23 AS build
+
+# Update CA certificates to fix TLS handshake issues on ARM64
+RUN apt-get update && apt-get install -y ca-certificates && update-ca-certificates && rm -rf /var/lib/apt/lists/*
+
 WORKDIR $GOPATH/main
 COPY . .
 RUN go env -w GOPROXY=direct
-RUN GO111MODULE=on go mod download
+# Retry logic for go mod download to handle transient network issues
+RUN for i in 1 2 3; do GO111MODULE=on go mod download && break || sleep 5; done
 RUN GO111MODULE=on CGO_ENABLED=0 GOOS=linux go build -o=/bin/main .
 
 FROM scratch

sample-apps/prometheus/Dockerfile

@@ -6,7 +6,8 @@ RUN apt-get update && apt-get install -y ca-certificates && update-ca-certificat
 WORKDIR $GOPATH/main
 COPY . .
 RUN go env -w GOPROXY=direct
-RUN GO111MODULE=on go mod download
+# Retry logic for go mod download to handle transient network issues
+RUN for i in 1 2 3; do GO111MODULE=on go mod download && break || sleep 5; done
 RUN GO111MODULE=on CGO_ENABLED=0 GOOS=linux go build -o=/bin/main .
 
 FROM scratch