What information needs to be saved in a Lamport clock system to deduce message causality?

Question

By using Lamport timestamps, we can determine the causality of messages. It is however unclear to me how one would deduce this simply from the lamport timestamp of the message.

Take the following model situation: there is a casuality break when it comes to process 2:

Now, we add Lamport clocks (hopefully I have done that correctly)

I'm at loss when it comes to understanding how I can read out the correct casuality of received messages by process 2 from just the timestamps. Unless I haven't determined the timestamps properly.

It looks like there's some additional information needed to do that. Maybe to save the timestamp of the message itself, and the id of the sending process?

Answer 1

Lamport clocks are not designed to allow you to infer all causality. No one ever claimed that knowing the Lamport timestamp is enough to read out the causality of all messages.

Rather, Lamport clocks provide a partial ordering that is consistent with causality: if event $a$ happens before event $b$ (causally), then $a$'s timestamp will be smaller than event $b$. The converse does not necessarily hold.

See https://en.wikipedia.org/wiki/Lamport_timestamps and https://en.wikipedia.org/wiki/Lamport_timestamps#Implications.

If you want to read out the full causality, use vector clocks or other methods that are designed for that.