I'm currently experimenting with Googles Cloud SQL and I'm facing the following problem:
Goal
I want to download a base-backup and after that the upcoming WAL-files to implement on-prem failover strategies for my Cloud-SQL hosted PostgreSQL database.
What I've tried & what went wrong
The commands pg_dump
and pg_dumpall
worked fine after some edits, but I can't get the pg_basebackup
command to deliver files.
I'm receiving the following error:
pg_basebackup: could not connect to server: FATAL: no pg_hba.conf entry for replication connection from host "[local]", user "postgres", SSL off
Thoughts
It seems to me that Google doesn't allow doing "Continous Archiving" on my own, which would be fatal for on-prem failover strategies and would explicitly force me to use Google Cloud SQL's archiving methods.
Question
Does anyone use PITR techniques on their Cloud SQL Postgres instances and has a workaround for this?