I think so too, perhaps something with the blade requesting work by RPC call before the previous request closes... anyone have this same problem?