Huawei Cloud CCE to STACKIT Kubernetes Engine

A CCE cluster and node pool, assessed

A small Huawei Cloud estate written for this page and assessed by StackToIT on 4 November 2026. Left: the input. Right: the result StackToIT recorded for it, an excerpt of assessment.json with the values exactly as the run wrote them.

resource "huaweicloud_cce_cluster" "platform" {
  name                   = "platform"
  cluster_version        = "v1.31"
  flavor_id              = "cce.s2.small"
  vpc_id                 = huaweicloud_vpc.platform.id
  subnet_id              = huaweicloud_vpc_subnet.nodes.id
  container_network_type = "vpc-router"
  container_network_cidr = "172.16.0.0/16"
  service_network_cidr   = "10.247.0.0/16"
  kube_proxy_mode        = "ipvs"
}

resource "huaweicloud_cce_node_pool" "general" {
  cluster_id         = huaweicloud_cce_cluster.platform.id
  name               = "general"
  os                 = "Huawei Cloud EulerOS 2.0"
  flavor_id          = "c7.2xlarge.2"
  availability_zone  = "random"
  initial_node_count = 3
  scale_enable       = true
  min_node_count     = 3
  max_node_count     = 6

  root_volume {
    size       = 50
    volumetype = "SSD"
  }
  data_volumes {
    size       = 100
    volumetype = "SSD"
  }
}

{
  "id": "huaweicloud_cce_cluster.platform",
  "result": {
    "class": "REPLATFORM",
    "confidence": "MEDIUM",
    "target": {
      "service": "STACKIT Kubernetes Engine (SKE)"
    },
    "reason": "A CCE cluster becomes an SKE cluster. Workload manifests can stay, while node pools, load balancer and storage annotations, identity and add-ons of the source are replaced."
  }
}

• Manifests stay

  Deployments, services and the rest of the workload manifests move as they are. What is Huawei Cloud specific around them is replaced on SKE.

• Nine cluster settings named

  cluster_version, container_network_cidr, container_network_type, flavor_id, kube_proxy_mode, name, service_network_cidr, subnet_id and vpc_id are listed for your platform team.

• Node pool to your architects

  The node pool is decided by your architects. Its flavor c7.2xlarge.2, 3 to 6 nodes and the 50 GB and 100 GB volumes are the input for the SKE node pool they define.

• Clear ownership

  The cluster goes to the Platform Architect, VPC and subnet to the Network Architect, and the node pool gets an owner when the package is assigned.

How the CCE estate maps

Classes as the run decided them. The units move in four waves, one each: VPC, subnet, cluster, node pool.

The SKE version, checked for you

The run reads the SKE lifecycle pages and states each version as of the day it read them, 3 November 2026.

• Land on 1.36

  Kubernetes 1.36 on SKE is supported, released on 27 August 2026 with an expiry on 9 June 2027, and is marked suitable as a landing.

• Skip 1.32 to 1.34

  1.32, 1.33 and 1.34 have expired on SKE. The example cluster runs v1.31, so its move includes the upgrade to a supported version.

• Updates on expiry

  When a version reaches its expiration date, SKE starts a mandatory update. The run records this as STACKIT’s day-2 responsibility.

• Both regions

  SKE is in the STACKIT catalogues of eu01 and eu02, so the cluster can land in either region.

Questions platform teams ask

Sources

• Huawei Cloud Docs: What is CCE?
• STACKIT Docs: Kubernetes Engine (SKE)