Open skearnes opened 2 years ago
Confirming that this reproduces on AWS but not locally; likely something to do with the Aurora Postgres configuration?
I ran some queries directly on an EC2 instance, and now the web interface seems to be performing better. Maybe Aurora lazily creates its indices?
Confirming that this issue still exists with the new ORM backend; we may need a polling mechanism for long-running queries.
@bdeadman and @skearnes to check if bug is still there.