Argo workflows cleanup options

To avoid cluster overload with completed pods Argo Workflow Controller has a cleanup feature. Cleanup parameters could be defined in a workflow and also in workflow defaults subsection of helm values in application.yaml:

          workflowDefaults:
            spec:
              podGC:
                strategy: OnWorkflowSuccess
              ttlStrategy:
                secondsAfterCompletion: 43200
                secondsAfterSuccess: 1800
                secondsAfterFailure: 21600

As shown above the declaration is self-explaining: podGC strategy is a pod garbage collector parameter, and ttl values define the time to keep workflows after execution. After the complete cleanup performed, the only option to observe the results of workflows is to look at collected artifacts as well as archived logs which in their tur are also kind of an artifact.