Triton Provider

​

Requirements

• A Triton account with API access
• An OpenSSH private key as a Control Plane opaque secret
• A private network configured in your datacenter
• triton-moirai images available in your datacenter

​

Supported Add-ons

• Dashboard: Provides a Kubernetes dashboard UI for the cluster.
• Headlamp: The recommended dashboard for the cluster
• Local Path Storage: Create PVCs backed by local volumes.
• Logs: Enable logging for pods and cluster auditing.
• Metrics: Enable metrics collections.

​

Load Balancing

​

Logging

• externalSyslog — Forward the log sink to a syslog server outside the cluster, specified as host:port.
• nodePort — If you run a syslog service inside the cluster and expose it via a Kubernetes NodePort, point the log sink to that port number (valid range: 30000–65535).

​

Step 1 — Prepare Your Triton Environment

1. Datacenter capacity: Verify that your datacenter has enough capacity for the number of nodes you’re planning to run.
2. Private network: Set up at least one private network for backend traffic between worker nodes.
3. Instance types: Make sure the instance types you need are available for both the node pool and the load balancer.
4. triton-moirai images: Have your triton-moirai machine images ready and accessible in your datacenter.

​

Step 2 — Create a Managed Kubernetes Cluster Using a Manifest File

1. Update the manifest below: Modify the following triton-mk8s-template.yaml YAML manifest with actual values. Replace the $ values in the YAML below with your actual values before applying.

kind: mk8s
name: ${NAME}
spec:
  provider:
    triton:
      connection:
        account: ${ACCOUNT}
        privateKeySecretLink: //secret/${SECRET}
        url: ${URL}
      imageId: ${IMAGE_ID}
      loadBalancer:
        manual:
          tags:
            my-key-sample: my-value-sample
          cnsInternalDomain: ${CNS_INTERNAL_DOMAIN}
          cnsPublicDomain: ${CNS_PUBLIC_DOMAIN}
          count: 1
          imageId: ${LB_IMAGE_ID}
          metadata:
            my-meta-key: my-meta-value
          packageId: ${LB_PACKAGE_ID}
          privateNetworkIds:
            - ${LB_PRIVATE_NETWORK_ID}
          publicNetworkId: ${LB_PUBLIC_NETWORK_ID}
      location: ${LOCATION}
      networking:
        podNetwork: ${POD_NETWORK_CIDR}
        serviceNetwork: ${SERVICE_NETWORK_CIDR}
      nodePools: # At least one node pool is required.
        - name: private
          labels:
            cpln.io/nodeType: core
            internet: false
          maxSize: 4
          minSize: 1
          packageId: ${PRIVATE_PACKAGE_ID}
          tritonTags:
            hello: world
      privateNetworkId: ${PRIVATE_NETWORK_ID}
  addOns:
    headlamp: {}
  firewall:
    - description: public access for testing purposes
      sourceCIDR: 0.0.0.0/0
  version: 1.34.2

2. Create the Cluster: Deploy the cluster by applying the manifest.
   • Console: Apply the triton-mk8s-template.yaml file using the cpln apply >_ option in the upper right corner.
   • CLI: Execute cpln apply -f triton-mk8s-template.yaml --org YOUR_ORG_HERE.
   Wait until the cluster is initialized.

​

Step 3 — Accessing the Cluster

​

1. Using the Terminal

1. Obtain the Cluster’s Kubeconfig File: Execute the command cpln mk8s kubeconfig ${NAME} -f /tmp/triton-mk8s-conf.
2. Access the Cluster with kubectl: Use the obtained kubeconfig file by running export KUBECONFIG=/tmp/triton-mk8s-conf for the current shell session.

​

2. Using Headlamp

1. Navigate to Control Plane Console: Visit the Control Plane Console.
2. Access Headlamp: In the Control Plane Console, navigate to Kubernetes in the left sidebar panel and click on Open under Headlamp for the cluster.

​

Advanced Configuration Options

​

Instance Rotation/Refresh