Skip to main contentDeploy CortexOps
Deploy CortexOps on Kubernetes in minutes.
Supported Platforms
- Kubernetes (Vanilla, 1.28+)
- GKE (Google Kubernetes Engine)
- EKS (Elastic Kubernetes Service)
- AKS (Azure Kubernetes Service)
- Kind (Kubernetes in Docker, for local testing)
Deployment Methods
CortexOps can be deployed using either Helm Charts (recommended for production) or pure Kubernetes Manifests (for evaluation).
Requirements
- kubectl installed and configured
- Helm v3.x installed
- At least 4 CPU cores and 8GB RAM available in the cluster
- Storage class supporting persistent volume claims (for PostgreSQL and NATS)
Installation via Helm
# Add the CortexOps Helm repository helm repo add cortexops https://charts.cortexops.amshithnair.in helm repo update # Install the control plane helm install cortexops cortexops/control-plane \ --namespace cortexops-system \ --create-namespace
Verification
Ensure all components are running correctly before injecting telemetry:
kubectl get pods -n cortexops-system # You should see: # - cortexops-collector # - cortexops-correlator # - nats-cluster # - temporal-server # - postgresql # - qdrant
Troubleshooting
If pods are failing to start, check the operator logs or ensure your cluster has adequate resources allocated to the `cortexops-system` namespace.
For detailed diagnostic commands, refer to the Operations Runbooks.