問題文
A new revision of a container app fails to start. The pod events show that the image could not be pulled, and the registry's own diagnostics show no incoming request at all. Which cause is most consistent with both observations?
選択肢
- The identity assigned to the app lacks permission to pull, so the registry refuses the request before it is recorded as a successful operation
- The image reference names a registry or repository that does not exist, so no request reaches the intended registry
- The tag was moved to a different digest after the revision was created, so the pull resolves to an image that is no longer present
- The image is too large for the node's available disk, so the pull times out before the registry finishes streaming the layers