News for hackers. Stuff that matters.
Re:Jepsen: NATS 2.12.1 by senderista on Monday December 08, 2025 @05:53PM (#46198785)
In practice, there must be a delay (from batching) if you fsync every transaction before acknowledging commit. The database would be unusably slow otherwise.
Re:Jepsen: NATS 2.12.1 by millipede on Tuesday December 09, 2025 @02:48PM (#46209675)
Right, I think the lazy thing implies that it would happen post "commit" being returned to the client, but it doesn't need to be. The commit just needs to be wait for "an" fsync call, not its own.