Make pod status update every 5 seconds instead of every 30 to mask the issue #3952. I suspect (but haven't confirmed) that #3927 doesn't completely address the issue since it will still happily cache a Pending status for up to 30 seconds even though the pod may have transitioned to Running already.