<#4198 [Core feature] Introduce PhaseInfoFailureWi...
# flyte-github
a
#4198 [Core feature] Introduce PhaseInfoFailureWithCleanup for non-recoverable pod states Issue created by katrogan Motivation: Why do you think this is important? For non-recoverable pod states, such as flyte/flyteplugins/go/tasks/pluginmachinery/flytek8s/pod_helper.go Lines 516 to 520 in </flyteorg/flyte/commit/97a79c030555eaefa3e27383d9b933ba1fdc1140|97a79c0> We should introduce a
PhaseInfoFailureWithCleanup
method similar to flyte/flyteplugins/go/tasks/pluginmachinery/flytek8s/pod_helper.go Line 525 in </flyteorg/flyte/commit/97a79c030555eaefa3e27383d9b933ba1fdc1140|97a79c0> to ensure the pod is cleaned up Goal: What should the final outcome look like, ideally? Invalid pods should result in workflow failure and be eventually cleaned up Describe alternatives you've considered Current implementation: leaves pod hanging indefinitely Propose: Link/Inline OR Additional context No response Are you sure this issue hasn't been raised already? ☑︎ Yes Have you read the Code of Conduct? ☑︎ Yes flyteorg/flyte