Skip to content

Latest commit

 

History

History
67 lines (50 loc) · 2.35 KB

appuio-specifics-public.rst

File metadata and controls

67 lines (50 loc) · 2.35 KB
Raw

APPUiO Public Platform Specifics

APPUiO is based on OpenShift Container Platform. This page describes APPUiO specific OpenShift configuration settings as well as features which were added to APPUiO that are not present in OpenShift.

Versions

  • Operating System: Red Hat Enterprise Linux (RHEL) 7.2
  • OpenShift Container Platform: 3.2.1.13
  • Docker: 1.10.3

Please note that currently only OpenShift Clients which have the same version as OpenShift Container Platform are guaranteed to work. You can download matching clients directly from APPUiO: :doc:`getting-started`.

URLs and Domains

Persistent Storage

APPUiO currently uses GlusterFS based persistent storage. For now volumes with the following sizes are available out of the box:

  • 1 GiB
  • 5 GiB
  • 20 GiB
  • 50 GiB

However you can contact us to get larger volumes: Contact. All volumes can be accessed with ReadWriteOnce (RWO) and ReadWriteMany (RWX) access modes. Please see Persistent Volumes for more information.

Quotas and Limits

The quotas are defined in the project size you ordered. The exact numbers can be found on the product site APPUiO Public Platform

Secure Docker Builds

Usually Docker builds from Dockerfile have to be disabled on multi-tenant platforms for security reasons. However, APPUiO uses it's own implementation to securely run Docker builds in dedicated VMs: :ref:`appuio_docker_builder`

Features

Most features are considered as stable. But there are some exceptions, mainly the following two features are not yet as stable as we want and therefore declare them as Beta:

  • Metrics: The metrics storage fails from time to time and looses all it's data. This is being fixed together with Red Hat.
  • Logging: This is not yet as stable as it should be, we're continuing to improve it.
  • Failover IP: The high availability of service IPs (application router and master loadbalancer) is not yet completely automated. We're also continue to work on that.