Hi, I don't know if this problem is from the controller. We have a cluster with 3 cores and 1 replica. We have a transaction in python (multiple nodes were created) and then we have a query, but the query does not get the latest information. What happened here?? How can I solve this problem?
You have to make sure to
- either have both operations (write transaction and read) within the same session as it handles bookmarks for you (see causal consistency)
- or extract the bookmark from the session (see docs) and pass it to the next session that does the read (see docs).