David Rowley writes:
> FWIW, here's a simple test case that shows the problem in current master.
This isn't telling the whole story. That first EXPLAIN did set the killed
bits in the index, so that subsequent ones are fairly fast, even without
VACUUM:
regression=# explain select * from a where
On Thu, 22 Apr 2021 at 00:03, Jeremy Schneider wrote:
>
> Two years later, I still remember this. And today I just confirmed
> someone hitting this on open source PG13.
The only thing that changed about get_actual_variable_range() is that
it now uses a SnapshotNonVacuumable snapshot. Previously a