Containerized End-2-End Testing – automate it!

Setting up unit- and integration tests are tasks which developers have learned to deal. But testing UIs from the perspective of an end user is mostly still a challenge. The key question usually to answer is: How it is possible to make these end-2-end tests stable, scalable and reproducible? The upcoming container technologies bring the hope of managing and automating your UI tests as easily as your container deployment. A live demo will show how it is possible to test and monitor a web- or a rich-client application in a containerized Linux UI.

Target Audience: Architects, Developers, Tester, DevOps, Monitoring Teams
Prerequisites: none
Level: Practicing

Extended Abstract
The end-2-end test example will be demonstrated by the open source tools "Docker" and "Sakuli". The demo shows how to do a blackbox UI test of a complex application, which is build on purpose of the current microservice pattern. The results will be interpreted and visualized in the Jenkins CI build pipeline and we will take a look into the UI activity of the testing container during the test execution.
In an alternative scenario the previous build end-2-end tests will be executed with the objective to forward there results to a monitoring system like OMD Nagios. This use case makes it possible to continuously measure the runtime of different test execution steps and create alerts on broken thresholds. This approach enables a new perspective of monitoring. It is now possible to check constantly if the functionality of an application, like the user login or the search engine is up and running - not just the sever healthy state.