** Stretch goals are set by Helium, these will be worked on within the sprint if other planned tickets are completed They are noted here for benefit and visibility of application teams, but are not part of the release commitment i.e delivery of the stretch goals is not guaranteed.
Helium 1.4.1 is the first in a series of releases building up to the Helium 1.5 release. We aim to use Helium 1.5 as the first official Private Platform-as-a-Serviceoffering. Private PaaS in practice means that developers, contracted by Vodacom, will be enabled to build products on Helium.
This naturally leads us to place more emphasis on ensuring all apps in the multi-tentant environment behave well and that any misbehaving apps don't negatively impact other apps sharing the same environment.
In order to achieve this we will introduce process monitoring and management features in the Helium 1.4.1 release. These process monitoring and management features will allow Helium administrators to monitor long running processes that were started by apps and to - if necessary - interrupt the processes, which may be creating a negative impact on the platform.
Helium 1.4.1 will also have automated process management that will - based on configuration settings - interrupt any long running processes. The default time-out for any single invocation of a process within an app will be 5 minutes, meaning that if the process is still running after 5 minutes it will be interrupted by the Helium app management feature. These values are configurable on a per-app and per-function basis, meaning that Helium administrators can set shorter or longer time-out values for specific apps or even for specific functions within apps.
The purpose of these features are not to stop developers from executing long running processes, but rather to ensure that such processes are anticipated and managed responsibly. JIRA tickets should be created for DevOps to extend the time-out for specific functions to values that the developers of the app see as reasonable. Teams can create a single JIRA ticket per app, with an .xls attachment of any functions that require an extended time-out (ie. longer than 5min). In this way we hope to create more visibility around functions that are actually behaving outside of the expected parameters.
The Helium teams will keep an eye on these configurations and use the information for resource planning and further improvements to the monitoring and management process.