- Previously, when trying to allocate devices through DRA for a node timed out, scheduling would proceed with another node if any had the necessary resources. This potentially hid that a node was ignored. Worse, if scheduling was slow overall, the pod was incorrectly moved to "unschedulable" and only retried after a periodic sweep. Now timeouts are errors that are always visible as pod scheduling failures and get retried with...

  `I never went to him,' the Mock Turtle said with a sigh:  `he
taught Laughing and Grief, they used to say.'

  `So he did, so he did,' said the Gryphon, sighing in his turn;
and both creatures hid their faces in their paws.

  `And how many hours a day did you do lessons?' said Alice, in a
hurry to change the subject.

  `Ten hours the first day,' said the Mock Turtle: `nine the
next, and so on.'

  `I never went to him,' the Mock Turtle said with a sigh:  `he
taught Laughing and Grief, they used to say.'

  `So he did, so he did,' said the Gryphon, sighing in his turn;
and both creatures hid their faces in their paws.

  `And how many hours a day did you do lessons?' said Alice, in a
hurry to change the subject.

  `Ten hours the first day,' said the Mock Turtle: `nine the
next, and so on.'