NavigationContentFooter
Jump toSuggest an edit

Moving Kubernetes nodes to routed IPs

Reviewed on 03 June 2024Published on 22 April 2024

As part of our ongoing efforts to enhance infrastructure capabilities and provide better services to our customers, Scaleway is introducing routed IPs for all products, including Kubernetes (K8s) worker nodes.

One of the standout benefits of routed IPs is their support for IP mobility, streamlining IP management and movement within the Scaleway ecosystem. By simplifying the process of reallocating IPs across different products, routed IPs empower users with unprecedented flexibility and control over their network infrastructure.

Additionally, routed IPs unlock the ability to attach multiple IPv4 addresses to a single Instance, enabling specific functionalities such as IP failover. While this feature may have limited applicability within Kubernetes environments, it offers invaluable flexibility and resilience across other Scaleway products.

Note

Routed IPs provide primary support for IPv6, ensuring direct connectivity for IPv6 addresses with Instances. However, IPv6 support for Kapsule is not yet available due to compatibility limitations with existing Container Network Interfaces (CNIs).

Important

NAT IPs will reach end-of-life on October 1st 2024. After this date, Scaleway will organize migrations for all clusters to transition to routed IPs. Migrations will be planned region by region, with specific dates to be announced.

Moving your Kubernetes nodes to routed IPs

Before you start

To complete the actions presented below, you must have:

  • A Scaleway account logged into the console
  • Owner status or IAM permissions allowing you to perform actions in the intended Organization
  • Set up a Kubernetes Kapsule cluster
Note

Kapsule and Kosmos clusters created after April 11th 2024 are deployed with IP mobility enabled by default. For existing clusters, you can manually launch the migration to routed IPs.

Using the Scaleway console

  1. Click Kubernetes in the Containers section of the Scaleway console side menu. The Kubernetes overview displays.

  2. Click the name of the cluster you want to access. The cluster overview page displays.

  3. Click the Network tab. The cluster network information displays.

  4. Click Move to routed IP in the IP address type section of the page. A pop-up displays.

  5. Confirm the move of your worker nodes to routed IPs by typing MOVE, then click Move to routed IPs to launch the process.

    Important
    • This operation is similar to a cluster upgrade. Each node in your cluster will reboot. Ensure that you have more than one node ready in order to avoid downtime.
    • Kubernetes clusters use dynamic IPs, meaning nodes may receive new public IPs upon reboot. Users relying on node IPs for filtering/whitelisting should take necessary precautions.

    The worker nodes of your Kubernetes cluster will reboot during the process and a new flexible IP will be assigned to each worker node. Once the migration has completed, the IP address type of your cluster will change to Routed flexible IPs.

Using the API

Before you start

To complete the actions presented below, you must have:

  • A Scaleway account logged into the console
  • Owner status or IAM permissions allowing you to perform actions in the intended Organization
  • Created a Kubernetes cluster that uses NAT IPs
  • A valid API key
  1. Configure the following variables for your Kubernetes cluster:
    REGION=fr-par # The region of your cluster. Can be either "fr-par", "nl-ams", or "pl-waw"
    CLUSTER_ID=b69106db-e212-4fff-9584-5da4a47217ed # The ID of your Kubernetes cluster
    SCW_SECRET_KEY=196dadd9-5685-42c8-aa71-a34d46861303 # Your Scaleway secret key
  2. Run the following command to launch the migration of existing Kubernetes clusters to routed IPs:
    curl --location "https://api.scaleway.com/k8s/v1/regions/$REGION/clusters/$CLUSTER_ID/migrate-to-routed-ips" --header "x-auth-token: $SCW_SECRET_KEY" --header 'Content-Type: application/json' --data '{}'
    Important
    • Activating routed IPs on Kubernetes nodes will require draining each node before reconfiguration and reboot. Single-node clusters will experience downtime during this process. It is recommended to have at least two nodes and proper configuration to minimize downtime.
    • Kubernetes clusters use dynamic IPs, meaning nodes may receive new public IPs upon reboot. Users relying on node IPs for filtering/whitelisting should take necessary precautions.
Docs APIScaleway consoleDedibox consoleScaleway LearningScaleway.comPricingBlogCarreer
© 2023-2024 – Scaleway