Quickstart

This walks through installing tuppr and running one Talos and one Kubernetes upgrade. It assumes you have granted Talos API access to the controller namespace - see Requirements.

Install

helm install tuppr oci://ghcr.io/home-operations/charts/tuppr \
  --namespace system-upgrade

Every chart value is documented in the generated Helm chart values reference.

Upgrade Talos

Create a TalosUpgrade. This is the minimal form - one target version applied to every node, sequentially:

apiVersion: tuppr.home-operations.com/v1alpha1
kind: TalosUpgrade
metadata:
  name: cluster
spec:
  talos:
    # renovate: datasource=docker depName=ghcr.io/siderolabs/talos
    version: v1.14.5

tuppr health-checks the cluster, then upgrades each node in turn: drain → talosctl upgrade → reboot → wait for the node to return Ready on the new version. Watch it:

kubectl get talosupgrade -w
kubectl describe talosupgrade cluster

Adding health checks, parallel batches, hooks, maintenance windows, and per-node overrides: Talos upgrades.

Single-node clusters

With one node, the upgrade pod runs on the node being rebooted. tuppr handles this automatically (issues the upgrade with --wait=false, disables talosctl’s own drain, and tracks completion by polling node readiness over the Talos API), then uncordons the node once the upgrade is verified. Set spec.drain.enabled if you want the node’s pods evicted and fully terminated before the reboot.

Upgrade Kubernetes

Create the (single) KubernetesUpgrade:

apiVersion: tuppr.home-operations.com/v1alpha1
kind: KubernetesUpgrade
metadata:
  name: kubernetes
spec:
  kubernetes:
    # renovate: datasource=docker depName=ghcr.io/siderolabs/kubelet
    version: v1.37.1

kubectl get kubernetesupgrade -w

There is exactly one KubernetesUpgrade per cluster; to upgrade again, edit spec.kubernetes.version on it. Details: Kubernetes upgrades.

Both at once

You can apply both resources; tuppr never runs them concurrently. Whichever starts first runs to completion while the other waits in Pending, then the second starts automatically. See Upgrade coordination.