Skip to content

Installation Enhancements

Online Upgrades for Kubernetes

Online upgrades are now supported for Kubernetes deployments, provided that the software is being upgraded by at most a single version (e.g. 6.15 -> 6.16), and that neither the export or import functionality of the installer is used. This is an extension of the online patch update support included with the 6.15 release, and is implemented in the same way, where the Clean Data action happens once the new application versions are running and the old versions have shut down.

The significant difference compared to patch updates is that database structural changes will also be made as part of the upgrade process. However, these changes are restricted to ensure that the existing software version can continue to run while the upgrade scripts perform the necessary changes. Kubernetes will then ensure that the new application version is rolled out in a controlled manner.

Note

It is recommended that software updates and upgrades are performed out of hours even when online, to minimise risk and mitigate any minor issues. The moving of datasets between the old and new versions of the applications can cause brief periods of service degradation.

Multi-Installation Systems for Kubernetes

Systems with multiple installations are now supported when using Kubernetes. Multiple instances of the application are installed across different clusters. This setup can be used to ensure high availability, load balancing, and disaster recovery. They can also be used to implement blue / green deployments for the Kubernetes infrastructure. All installations still connect to the same database. Network connectivity between installations is not needed. Each installation has a role where the role defines the function of the installation as follows:

  • Active: Receives general workbench and gateway traffic and actively processes datasets.
  • Passive: Keeps replica datasets in-memory ready to take over processing when converted to Active (when DatasetAvailability is at least 2).
  • Standby: A minimal installation scaled down to reduce resource usage. No datasets are processed.

See the Architecture and Sizing Guide - Multi-Installation Systems section for more information on installations. See the Administration Guide - Dataset Load Balancing section for more information on how datasets are processed in a multi-installation system.