Closed bakayolo closed 3 years ago
Instead of implementing a "fail" workflow, we will just make sure the abort button always change the state of the release unless there is an issue with argo directly. Meaning that if a user click on abort while the workflow is not here anymore (it might happen in case of internal failure), then the state will be modified and they will not be blocked anymore
Is your feature request related to a problem? Please describe.
If a workflow fail to be created or if it fails due to an internal reason, the workflow state is not getting update in the configmap. Therefore, the service is "locked" with a PENDING/RUNNING release and we need to update it manually directly in kubernetes.
Describe the solution you'd like Similar to "abort", we should add an option to "fail" a release.