#3843 [Core feature] Add grace period before declaring ImagePullBackoff a failure Issue created by hamersaw Motivation: Why do you think this is important? Currently when FlytePropeller detects

ImagePullBackoff

on a

Pod

it immediately declares the task a failure. In most scenarios this may not be the most efficient behavior because FlytePropeller will then retry with a new

Pod

(given retries are enabled) which will just attempt to pull the same image(s). By introducing a configurable grace period before FlytePropeller declares a task failure we allow k8s to continue attempting to pull the image. A grace period mechanism has precedent in other issues as well. Goal: What should the final outcome look like, ideally? FlytePropeller should wait a configurable period before declaring an

ImagePullBackoff

a failure rather than immediately on first detection. Describe alternatives you've considered Leaving this as is causes unnecessary failures / retries. 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