Currently you can only differentiate by message. So having, say multiple camera nodes running means you can't pinpoint which one failed.