It's a little bit concerning to note that the last GitLab outage was also caused by Postgres replication issues, which eventually culminated in some pretty bad

It's a little bit concerning to note that the last GitLab outage was also caused by Postgres replication issues, which eventually culminated in some pretty bad data loss and a realisation that the most recent backups they had weren't usable. Maybe they should reach out to someone from Postgres to help figure out why their replication appears to have some sort of systemic issue...