Latency and Privacy in Lightning

[carla]

@MattCorallo

These two statements appear to contradict each other

Indeed - statements are meant to apply in the context of each described attack, as they’re the two types of attacks I could find in the literature. Just meant to indicate that in [6] timing matters, in [7] it doesn’t.

@MattCorallo

the point of the forwarding delay is to create a batch such that a passive network monitor can’t trivially trace payments

My understanding of [6] is that just being able to construct these partial paths is insufficient to deanonymize senders and receivers (if we add some privacy-awareness to pathfinding).

Two questions here:

• Is there an off-path attack which does not require the attacker to run a ton of pathfinding to complete the path?
• Are you talking about the case where the attacker sees network messages for the full path (they’re all in the same malicious AS group)?

@MattCorallo

having no forwarding delays would mean that someone trying to get privacy by adding extra hops would be absolutely wrecked by an adversary running multiple nodes (post-PTLCs).

Could you explain this further? Based on the PTLC reference, I assume we’re talking about the on-path attack? It’s unclear to me why a receiver-side delay doesn’t help with a multi-node on-path attacker.