Re:Crsql – Multi-writer and CRDT support for SQLite
by lifty on Tuesday November 15, 2022 @10:40AM
(#33610487)
Marmot looks interesting, but I am curious how it deals with conflicts. What happens if two writes for the same field happen on two different dbs, and they are conflicting?
[ Parent
| on HN ]
Re:Crsql – Multi-writer and CRDT support for SQLite
by maxpert on Tuesday November 15, 2022 @02:05PM
(#33613451)
Since each write is logged into a JetStream. The last writer will win for conflicting changes.
Re:Crsql – Multi-writer and CRDT support for SQLite
by lifty on Tuesday November 15, 2022 @04:13PM
(#33615093)
But since you don't have NATS message ordering (from different sources), it's possible that different DBs in the cluster will end up out of sync, right?
Re:Crsql – Multi-writer and CRDT support for SQLite
by maxpert on Tuesday November 15, 2022 @06:46PM
(#33616879)
So per-node there is an order of changes (read internals doc), and all of changes are sent to JS in that commit order. These changes when replayed will always eventually result in same values.
Re:Crsql – Multi-writer and CRDT support for SQLite
by lifty on Wednesday November 16, 2022 @04:48AM
(#33620790)
I am talking about writes that come from 2 different nodes. If 2 different nodes, each making a different write at the exact same time, you would need some global ordering or consensus in order to order them. Otherwise you might end up in a different state on different nodes.