specifiing compatibility marshaller of distributed cache in console results in degrading cluster

in console go to an container -> an distributed cache -> click "configuration" (up right) -> compatibility -> set marshaller to e.g. newMarshaller -> click "applyChanges" -> "Update" -> "restart now". The server will restart without problem but console will show that all clusters are degraded. There should be some wait time after clicking "restart now" before the console fetches the state of the server/clusters.