ORACLE OKE → STACKIT SKE
Oracle OKE to STACKIT Kubernetes Engine
An OKE cluster moves to the STACKIT Kubernetes Engine (SKE), and your workload manifests can stay. StackToIT assesses the cluster and its node pools as a replatform, names what is rewritten around the workloads, and checks which SKE versions you can land on.
An OKE cluster and node pool, assessed
A small OCI estate written for this page and assessed by StackToIT on 4 November 2026. Left: the input. Right: the decision StackToIT recorded for it, exactly as it stands in assessment.json.
resource "oci_containerengine_cluster" "platform" {
compartment_id = var.compartment_id
name = "platform"
kubernetes_version = "v1.31.1"
vcn_id = oci_core_vcn.platform.id
endpoint_config {
subnet_id = oci_core_subnet.api.id
is_public_ip_enabled = false
}
}
resource "oci_containerengine_node_pool" "workloads" {
cluster_id = oci_containerengine_cluster.platform.id
compartment_id = var.compartment_id
name = "workloads"
kubernetes_version = "v1.31.1"
node_shape = "VM.Standard.E5.Flex"
node_shape_config {
ocpus = 4
memory_in_gbs = 32
}
node_config_details {
size = 3
placement_configs {
availability_domain = var.availability_domain
subnet_id = oci_core_subnet.nodes.id
}
}
}[
{
"id": "oci_containerengine_cluster.platform",
"class": "REPLATFORM",
"confidence": "MEDIUM",
"target": {
"service": "STACKIT Kubernetes Engine (SKE)"
},
"reason": "An OKE cluster becomes an SKE cluster. Workload manifests can stay, while node pools, load balancer and storage annotations, identity and images are rewritten."
},
{
"id": "oci_containerengine_node_pool.workloads",
"class": "REPLATFORM",
"confidence": "MEDIUM",
"target": {
"service": "STACKIT Kubernetes Engine (SKE)"
},
"reason": "An OKE node pool becomes an SKE node pool. The shape becomes a STACKIT flavor, and placement across zones is decided again."
}
]Manifests can stay
The assessment states it plainly: the workloads keep their manifests. What changes is around them: node pools, load balancer and storage annotations, identity and images.
Node pool to SKE node pool
The pool runs 3 nodes of VM.Standard.E5.Flex with 4 OCPUs and 32 GB. On SKE the shape becomes a STACKIT flavor, and placement across zones is decided again.
Private API endpoint noted
The architect report lists the endpoint settings of the cluster, its subnet and the disabled public IP, among the settings your architect carries over.
SKE versions, checked on the day
The report states each SKE version with its dates, as read from the STACKIT lifecycle pages on 3 November 2026. The source cluster runs v1.31.1.
Land on 1.36
SKE 1.36 is supported, released on 27 August 2026 and published to expire on 9 June 2027. The report marks it as the suitable landing version with a low timing risk.
Skip the expired ones
1.32, 1.33 and 1.34 have already expired on SKE, so the report advises against landing on them.
Updates on SKE
When a version expires, SKE starts a mandatory update. The report quotes the STACKIT documentation on this, so your team plans upgrades ahead of time.
How the OKE estate maps
| OCI resource | STACKIT target | Migration class |
|---|---|---|
| OKE cluster | SKE cluster | Replatform |
| OKE node pool | SKE node pool | Replatform |
| VCN | STACKIT networks, with a Network Area where routing between them is needed | Reconfigure |
| Subnets for nodes and API endpoint | STACKIT networks with their own address ranges | Reconfigure |
| NAT gateway | Egress design decided by the network owner | Architect decision |
| IAM policy for the workloads | STACKIT roles and service accounts, mapped by your architect | Architect decision |
Classes as the run decided them. The two subnets are two units of the run and share one row here.
The order of the move
The run puts the seven units in four waves from their dependencies.
Wave 1
The VCN and the IAM policy.
Wave 2
The NAT gateway and the two subnets.
Wave 3
The SKE cluster.
Wave 4
The node pool, which depends on the cluster and the node subnet.
Proven on a real AWS estate
The AWS Retail Store Sample App, a public Terraform project from AWS, assessed by StackToIT and published in full, exactly as the run wrote it.
View Case Study- 282
- Terraform resource declarations analyzed
- 44
- migration units identified
- 38
- target units generated, from 52 considered
- 331
- plan steps written
Questions platform teams ask
Do our Kubernetes manifests change when we move from OKE to SKE?
The workload manifests can stay. What is rewritten is OCI-specific: node pools, load balancer and storage annotations, identity and images.
Which Kubernetes version should the SKE cluster run?
In the run of 4 November 2026, SKE 1.36 was the supported landing version, published to expire on 9 June 2027. The report states the day its SKE lifecycle data was read, so your team knows how current the answer is.
What happens to the OCI policies our workloads use?
OCI policies have no document-level counterpart on STACKIT. Your architect maps the people and workloads behind them to STACKIT roles and service accounts, and the assessment marks each policy for that decision.
Sources
Your next move: STACKIT.
Tell us about your estate. We show you what StackToIT delivers for it: target architecture, Terraform artifacts and a structured migration plan.