Kubernetes iconKubernetesCaaS

With our Kubernetes your developers get the best conditions for an effective and stable operation, this is Container as a Service.

Service Description

Date of version: 2020-11-10

Our Kubernetes service gives you access to a certified, accessible and managed Kubernetes platform entirely under Swedish regulations. We ensure that our platform meets the requirements for inclusion in CNCF's Certified Kubernetes Conformance Program. This means that you can take your solution and move it between our platform and any other platform that meets the same requirements. No lock-in! The platform is designed to be available even in the event of maintenance or operational disruptions in our underlying infrastructure. We take care of the basic operation of the platform so that you can focus on delivering benefits to your customers!

Scope

The service includes operation and monitoring of the components in your environment described under the headings "Control Plane Components" and "Node Components" in Kubernetes official documentation for the version of Kubernetes installed in your environment. The same applies to the DNS service in the cluster.

The service also includes the following selectable functions:

The service includes monitoring of cluster components. We fix critical errors that affect services and functions in these components, around the clock.

Availability and capacity

The service is built over three availability zones, each availability zone is a geographically separated data center. An accident in an entire availability zone shall not affect the availability of the service. Workloads in the crashed availability zone are scheduled if possible in the two remaining availability zones. Breakdowns in one or more availability zones affect the capacity of the service as the resources available in the affected zones become inaccessible. The service is running on our OpenStack platform and is dependent on the SLAs that apply to it. It also means that the same quotas apply.

Upgrades

For the components and services described above, Elastx undertakes to make ongoing updates available. Up to four upgrades per year are included in the service, additional upgrades can be performed for an additional charge. To fix critical security vulnerabilities, additional upgrades can be performed, these are not deducted from the above. The service includes upgrades during office hours, upgrades during non office hours can be performed against additional charges. Upgrades are scheduled in consultation with you and can be done on either Elastx or your initiative. Normally, one node is upgraded at a time, which does not affect the availability of the cluster. If an upgrade requires a deviation from this, you will be notified and appropriate measures will be taken in consultation.

Scalability

Kubernetes is designed to be flexible and scalable, this also applies to our Kubernetes service. We offer both horizontal and vertical upscaling and downscaling of worker nodes. For controlplane nodes, only horizontal scaling is offered. The service includes scaling during office hours, scaling during non office hours times can be performed against additional charges. Scaling is scheduled in consultation with you and takes place on your initiative, Elastx can make a recommendation for scaling if necessary. Normally, scaling does not affect the availability of the cluster.

Delimitations

In order to offer the service described in this document, your environment needs to consist of at least the following number of nodes:

  • Three control plane nodes spread over the availability zones
  • Three worker nodes, one in each of our availability zones, in a single nodegroup.

We offer smaller clusters for testing and development, but then these terms do not apply. The terms of the service apply provided that you do not make any changes to the configuration of the components listed under the heading Scope above. This also applies to configuration changes on the cluster nodes themselves. To make the demarcation inside the cluster clear, components that are part of the service are run either in the "kube-system" namespace or in namespaces with "elx-" prefix. In order to maintain a high quality and have the opportunity to act on any security issues, we support the two latest Elastx versions of the Kubernetes service. If extended support is needed, it is possible to agree separately. Storage resources are tied to the availability zone they were created in. Workloads that depend on specific storage resources can thus not be scheduled in another availability zone. Elastx is not responsible for other applications and services running in the customer's cluster. We also do not have the opportunity to take responsibility for problems that arise due to over-provisioning of resources in the cluster. However, we strive to proactively inform you if we see risks with resource utilization.

Consultation

Elastx has a number of employees who are Certified Kubernetes Administrators and have solid experience in operating complex infrastructure. We are happy to help with consultation in these areas and how your solution can best use our platform.

This website uses cookies to ensure you get the best experience on our website. Read more.