Kubernetes
Overview
Our Kubernetes Service provides you with the ability to deploy and manage Kubernetes clusters without worrying about the setup and maintenance of master nodes. We offer a reliable and highly available infrastructure for your cluster, which is hosted across three cloud providers: Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP).
Here on the "Kubernetes clusters" page, you can conveniently view a list of all your clusters. Each cluster is accompanied by a brief overview, showcasing essential information such as its allocated resources, associated cost, and current status.
Key Features
- Independence from Cloud Providers: Our service ensures independence from a single cloud provider. We deploy your cluster across the networks of three providers, ensuring the resilience and reliability of your cluster.
- Easy Cluster Creation: Creating a Kubernetes cluster has never been easier. Simply select a location and node configuration, and your cluster will be ready for use with just a few clicks.
- Management of the Cluster for You: We take care of managing the Kubernetes control planes, allowing you to focus on deploying and managing your applications.
What are the network options?
When working with Kubernetes clusters in a multi-cloud environment, you can choose between:
- Direct Connect – Transfers data over dedicated network ports between cloud providers, offering lower latency, higher stability, and greater bandwidth. To use dedicated ports for a specific region and provider, enable the Direct connect option in the Network tab. This additional setting increases costs and is recommended for large-scale or high-traffic workloads (>17 TB).
- Internet Connect – Transfers data between cloud providers over secure public internet channels. Suitable for workloads that do not require maximum bandwidth or ultra-low latency and may be more cost-effective.
Both options are part of the multi-cloud architecture, enabling workloads to operate across AWS, Azure, and GCP with high availability.
Current limitations
- Cluster Location: Currently, you can place clusters in European and North American (on demand) regions of cloud providers.
- Node Configuration Changes: After creating a cluster, you can add or remove worker nodes via the Edit cluster option. However, resizing an existing node is not supported — to change a node's configuration, remove it and add a new one with the desired parameters. Individual nodes within an autoscaling group cannot be removed manually.
- Gradual Feature Rollout: We are actively developing our service, and we plan to introduce more features in the near future. If you have any ideas for improvements, please feel free to share them with our support team.
Statuses
Each Kubernetes cluster exposes statuses at three levels: the control plane, individual worker nodes, and autoscaling groups. Together, these statuses give you a complete picture of your cluster's health and help you track progress during creation, scaling, and deletion operations.
Control Plane Status
The Control Plane status provides information about the health and operational state of the core Kubernetes management components.
| Status | Description |
|---|---|
CREATING | The control plane is being initialized. Cluster management operations are not yet available. |
ACTIVE | The control plane is fully operational. All components are running and healthy. |
UPDATING | A control plane update is in progress (e.g., a Kubernetes version upgrade). Management operations may be temporarily limited. |
DEGRADED | One or more control plane components are impaired. The API may still be accessible, but reliability is reduced. |
UNHEALTHY | Critical control plane components have failed. Cluster management operations are unavailable. |
FAILED | The control plane deployment or update has ended in an unrecoverable error. Manual intervention is required. |
DESTRUCTION | The control plane is being deleted as part of cluster teardown. |
Worker Node Status
The node lifecycle status shows where each worker node is in the process of joining and operating within your cluster.
| Status | Description |
|---|---|
PROVISIONING | The underlying VM is being created and configured at the infrastructure level. |
JOINING | The VM is provisioned and the node is registering with the Kubernetes cluster (kubelet startup, certificate bootstrapping). |
ACTIVE | The node is registered with the cluster and ready to accept workloads. |
PENDING | The node has lost contact with the cluster or is experiencing a health issue. Recovery may be attempted automatically. |
DELETING | The node is being removed from the cluster and the underlying VM is being destroyed. |
FAILED | The node provisioning or joining process has failed. The node is not available for workloads. |
Autoscaling Group Status
The autoscaling group status shows the current runtime state of each node group. Returns null for static (non-autoscaling) node groups.
| Status | Description |
|---|---|
CREATING | The autoscaling group is being initialized and initial nodes are being provisioned. |
ACTIVE | The group is running and all nodes are in the expected state. |
READY | The group is created with no added nodes and no load. |
SCALING_UP | The autoscaler has detected increased workload and is provisioning new nodes. |
SCALING_DOWN | The autoscaler is removing underutilized nodes from the group to reduce cost. |
UNHEALTHY | One or more nodes in the group are in a failed or unresponsive state. Automatic remediation may be in progress. |
DESTRUCTION | The autoscaling group and all its nodes are being deleted. |
Pricing
Note that the Kubernetes service is initially free. However, we may introduce fees in the future for control plane management. We'll give notice and communicate changes to users.
Now you only pay for compute instances in which your nodes are situated and for logs collection.
How can I delete a Kubernetes cluster? - FAQ
In addition to viewing your clusters, you have the option to delete a cluster. To do this, simply click on the "Delete" button. Please note that initiating cluster deletion requires "Manage" level access rights. The platform will prompt you for confirmation, and once confirmed, the cluster will be permanently removed from the clusters page.
Deleting a cluster will result in the termination of all the worker nodes and the master node that was part of the cluster. All the data stored on the nodes, including the container images, logs, and application data, will also be lost.
It is important to note that deleting a cluster is a permanent action, and there is no way to recover the data once it has been deleted. Therefore, it is critical to make sure that you have backups of your data and configurations before deleting a cluster.
Before deleting a cluster, it is recommended to ensure that all the resources running on the cluster have been terminated and that there are no dependencies on the cluster. You should also make sure that any data stored on the cluster has been backed up to avoid losing critical data.
In summary, deleting a Kubernetes cluster is a significant action that should be taken with caution. It is important to have a proper plan and backup strategy in place to avoid losing critical data and resources.
Last updated on 1 Aug 2026