Product Sheet and Services for Kubernetes Clusters

Kubernetes Cluster Service

teuto.net installs and operates a Kubernetes cluster on behalf of the customer. The Kubernetes clusters are hosted on a fully redundant cloud environment (OpenStack) by teuto.net in one of the data centers (Bielefeld BITel-Telehouse, Frankfurt Interxion.de). All data centers have security management and are ISO 27001 certified.

The Managed Kubernetes clusters from teuto.net are equipped by default with three Standard.2.1905 instances as control planes. The number and flavor of compute planes can be selected by the customer according to their requirements. The prices for the instances and all other listed services are recorded in our Kubernetes Service Catalog. Additionally, the clusters are delivered with a free load balancer for the API to enable normal operations.

Monitoring

teuto.net monitors the operational parameters around the clock (24x7) and guarantees API availability of 99.5% on a monthly average.

All nodes of the Kubernetes clusters are monitored using machine health checks. If a machine health check fails, the node is replaced with a new one after a timeout.

Furthermore, the following operational parameters are monitored:

  • Availability of cluster nodes
  • Availability of etcd nodes
  • Availability of API nodes

Operating an application within the Kubernetes cluster is the customer’s responsibility.

Updates

Kubernetes has a rapid upgrade cycle. teuto.net supports the latest minor version.

teuto.net regularly installs updates on the Managed Kubernetes clusters. Updates are always installed on the next business day after the second Monday of the month, provided an update is available. For updates to a new minor release, we wait at least for the second patch release of the new minor release; otherwise, we update to a new patch version of the existing minor release.

During updates, nodes are replaced one after another. A new node is always added to the cluster first before one of the existing nodes is replaced. In this way, your full resources are always available. teuto.net uses the normal Node drain process.

To perform updates smoothly, the customer must follow the best practices, especially the configuration of Pod Disruption Budgets. For a detailed description of the best practices, please refer to our documentation.

With incorrectly defined Pod Disruption Budgets, outages or a non-graceful shutdown of the affected application may occur during updates.

Basic Equipment for Managed Kubernetes

To facilitate configuration, teuto.net provides a Base Helm Chart to comfortably install components.

This allows you to easily install additional applications and benefit from prepared cross-configuration. For example:

  • Monitoring - Monitor the operation of your application.
  • Flux - Install your resources directly from a Git repository (GitOps).
  • Ingress Controller - Make your application available on the internet.
  • Cert Manager - Manage your certificates automatically.

Additional Packages

Interior Fit-Out

Upon request, teuto.net sets up typical cluster applications for you in your cluster. These include:

  • Ingress for external access to your applications.
  • Cert Manager for managing encryption certificates.
  • Database operators (PostgreSQL, MySQL) facilitate the management of your databases.
  • Cluster-internal monitoring provides you with detailed metrics and log data for your application and can also send you alerts about critical states if needed.
  • Descheduler ensures the best possible load distribution of your application across the cluster.
  • GitOps (Flux) manages your application based on Git and Helm repositories.

Application Support

We are happy to support you with the deployment of your application and offer the operation of additional supporting applications within your cluster.

  • Support with migrating your application to Cloud Native deployments.
    • Support with the design and implementation of a Helm Chart tailored to your application. Also consulting on the design of your application to prepare it for Cloud Native deployment (HA, Storage, Security, …).
    • One-time setup fee.
    • Billing for additional work is based on actual effort incurred.
  • A GitLab pipeline builds and publishes your application automatically on commits and can also test it.
    • One-time setup fee.
    • Billing for additional work is based on actual effort incurred.
  • A Harbor container registry in the cluster enables you to store your application images, make them available, and scan them for known security vulnerabilities if needed.
  • A Managed DB (PostgreSQL, MySQL) is an easy way to deploy and manage a database in your K8s cluster.

The basis for application support is the booking of the “Interior Fit-Out” additional package.

Access Security

Upon request, we additionally restrict access to your cluster to generate further security.

  • VPN connection to your cluster restricts access to the API and, upon request, also to the ingress.
  • Additional network configurations provide access from and, if applicable, to your internal network.

Extension Including Support Hours

If the included contingent of support hours is not sufficient for you, it can of course be extended.

  • 4 hours per month included. Unused hours expire at the end of the month.

Special Solutions

Additional extensions are available for booking; please feel free to inquire.

SLA

Basic

  • 24x7 recording of operationally relevant cluster parameters.
  • 99.5% availability of cluster services on a monthly average.
    • The cluster is considered available as long as n-1 of the control plane and worker nodes are available.
    • teuto.net will always take economically reasonable measures to avoid further infrastructure problems.
  • Support hours via ticket system on business days Mon-Fri 09:00 – 17:00.
    • Support billing is based on actual effort incurred.
  • Included free of charge monthly.

Advanced

Like Basic with the following extended services:

  • Restoration of failed nodes within one business day Mon-Fri 09:00 - 17:00.
  • Support hours via ticket system on business days Mon-Fri 09:00 – 17:00.
    • 1 hour of support per month included. Unused hours expire at the end of the month.
    • Support billing is based on actual effort incurred.

Premium

Like Advanced with the following extended services:

  • 99.5% availability of cluster services on a monthly average.
  • Restoration of failed nodes within 4 hours, 24x7.
  • In case of infrastructure problems, 24x7 on-call service is activated.
  • Support hours via ticket system on business days Mon-Fri 09:00 – 17:00.
    • 2 hours of support per month included. Unused hours expire at the end of the month.
    • Support billing is based on actual effort incurred.

Image Availability

To ensure the availability of frequently used container images, teuto.net operates a proxy. Container images from particularly frequently used registries are cached by this proxy on pull.

For more information, please refer to the Additional Information.

Costs

Please refer to our Kubernetes Service Catalog for prices.

Billing

The consumption invoice shows the number of instances and the calculated hours.

teuto.net performs updates to Kubernetes nodes at regular intervals. For these updates, a new instance is created each time, the pods on an old node are cleared, and the old instance is shut down. This increases the monthly number of instances on the invoice.

For prices, please refer to the Service Catalog Managed Kubernetes Cluster.