Cluster manifests

• Cluster addons/controller-manifests with relevant configurations using argocd applicationset

• Metadata needed by manifests are retrieved automatically from applicationset cluster secret bootstrapped via terraform

List of controller manifests

• Argocd

Enabled and configured argocd ingress to share ALB with other applications using ALB group

Enabled service monitor for prometheus

Increased (Queries Per Second) rate limit in Argo CD

Enabled horizontal pod autoscaling

Enabled argocd UI access via loadbalancer

Configured argocd RBAC

Configured to access ECR OCI registries

Configured health check for alustan.io/* custom resources

Created default argocd AppProject

• Argo rollouts

Configured to query cloudwatch for http request target

Enabled service monitor for prometheus

• Cloudwatch metrics

Configured serviceAccount; extracting role-arn from eks blueprint addons

• EBS CSI

GP3 enabled

• External DNS

Configured serviceAccount; extracting role-arn from eks blueprint addons

Policy: upsert-only

Enabled service monitor for prometheus

• External secret

Configured serviceAccount; extracting role-arn from eks blueprint addons

Enabled service monitor for prometheus

• Grafana

Enabled service monitor for prometheus

Datasources: prometheus and loki

kubernetes dashboards: gnetId: 10000

Exposed via Ingress

EBS persistence storage

• Karpenter

Configured serviceAccount; extracting role-arn from eks blueprint addons

Enabled service monitor for prometheus

Karpenter provisioner

• Kube cost

Dashboard exposed via Ingress

• Metrics-server

• Prometheus

Enabled application service monitor

Dashboard exposed via Ingress

Enabled EBS storage

• Kyverno

Enabled service monitor for prometheus

Kyverno policies

• ALB controller

configured serviceAccount; extracting role-arn from eks blueprint addons

Enabled service monitor for prometheus

• Loki stack

• Robusta

Enabled prometheus stack

Enabled argoRollouts

Slack integration

• Alustan

Configured service account arn

application manifests

• Argo Rollout

RollbackWindow revisions: 3

Strategy: canary

Pingpong and AntiAffinity implementation: Zero-Downtime with ALB IP target group

Analysis: cloudwatch

• Argo AnalysisTemplate

Error count: result[0] >= 0.05

provider: cloudwatch

AWS/ApplicationELB Metrics: rate of HTTPCode_Target_5XX_Count/RequestCountPerTarget

• ConfigMap

• Deployment

Secrets pulled from ExternalSecret

• ExternalName service

Expose DNS access to postgres DB

• Horizontal Pod Autoscaler

• Ingress

configured for both default rolling update and canary deployment

• Namespace

elbv2.k8s.aws/pod-readiness-gate-inject: enabled annotations for ip target zero downtime

• ClusterSecretStore and ExternalSecret

pulls secret from external secret operator

• Service

Deployment Service

Rollout Service

Prometheus Service with TargetPort: http-metrics