It's probably cause native backup options are not as robust as 3rd Party (pgbackrest,barman). As an Oracle DBA we were always expected to know and use built-in RMAN backup. I observe Postgres users to be relatively casual about backups.

Depesz article lists most problems in setting up logical replica at scale. Apart from these "initial sync snapshot too large" is also common at large size.

www.depesz.com/2023/06/04/a...

Pg_upgrade with link mode or clone option is not that painful, it's preparing logical replica that's most time consuming and full of nightmare part.

I assumed it would be via bidirectional logical replication. So it's not using postgres logical replication protocol?

What's the largest cluster have you upgraded using this method? Logical replication is PITA at >10TB size and >10K tables.

Update: In response to a trademark notice from the PostgreSQL Community Association of Canada, domain has been changed from “Postgres.Is” to pgscorecard.com