Thank you for your valuable suggestion!
I have a question regarding the process:
When we shut down the standby, upgrade it, and then start it back up, will the replication automatically resume from the primary to the standby?
Looking forward to your clarification.
2) What do you mean by resource-intensive? If it means network bandwidth, then read the pg_basebackup man page.
No, it’s not about pg_basebackup consuming resources. What I meant is that in the event of a failover, if we need to bring the standby back online, the process of running pg_basebackup takes a significant amount of time. However, if using a cron job for this purpose is a viable option, then that would be acceptable.
I understand your point and appreciate the clarification.
I have reviewed the references and now have a better understanding of the minor upgrade process.
However, my concern lies in the fact that we are working with production servers, where downtime is not acceptable.
Additionally, if a failover occurs due to a network issue or any other disaster, setting up replication again requires running the pg_basebackup command. For large databases, this process becomes a significant challenge, as running pg_basebackup for the entire cluster can be time-consuming and resource-intensive.
A comment and a question:
1) pg_basebackup runs just fine from cron. Thus, "time-consuming" (which you described as 2-3 hours) isn't that critical.
2) What do you mean by resource-intensive? If it means network bandwidth, then read the pg_basebackup man page.
-- Death to <Redacted>, and butter sauce.
Don't boil me, I'm still alive.
