On 14.12.21 02:31, Tomas Vondra wrote:
There's still one remaining problem, that I already explained in [1]. The problem is that with this:BEGIN; SELECT nextval('s') FROM generate_series(1,100); ROLLBACK; The root cause is that pg_current_wal_lsn() uses the LogwrtResult.Write, which is updated by XLogFlush() - but only in RecordTransactionCommit. Which makes sense, because only the committed stuff is "visible". But the non-transactional behavior of sequence decoding disagrees with this, because now some of the changes from aborted transactions may be replicated. Which means the wait_for_catchup() ends up not waiting for the sequence change to be replicated.
I can't think of a reason why this might be a problem.
