On Tue, Jun 18, 2024 at 8:23 AM David G. Johnston
<david(dot)g(dot)johnston(at)gmail(dot)com> wrote:
> If this is indeed a regression from 16.2 odds are good you will need to work around the bug. Namely by ensuring that a generic plan for that query is never chosen. See: plan_cache_mode
Or enable_parallel_hash=off.