Restart failed conformance tests in E2E tests #3287
Labels
kind/flake
Categorizes issue or PR as related to a flaky test.
sig/cluster-management
Denotes a PR or issue as being assigned to SIG Cluster Management.
Milestone
Description of the feature you would like to add / User story
We often face issue where a couple of conformance tests fail due to the unstable infrastructure. This fails the whole E2E job and we have to restart it which takes a lot of time and resources, and especially slows down the testing phase.
Solution details
If there are failed conformance tests in the E2E tests, those tests should be restarted once or twice to rule out potential infrastructure instability issues. A very basic implementation of this has been done in #3261
Alternative approaches
Use cases
Significantly speeds up the KubeOne testing phase (potentially talking about hours). 🚀
The text was updated successfully, but these errors were encountered: