Re: Trying to understand a failed upgrade in AWS RDS

[Alan Hodgson <ahodgson@xxxxxxxxxxxxxxx>]

On Sun, 2023-05-21 at 07:56 -0700, Mike Lissner wrote:

As far as I know it's impossible to reliably pg_upgrade a node that has subscriptions and eventually resume logical replication. 

Should this go in the documentation somewhere? Maybe in the pg_upgrade notes? I still don't understand the mechanism. You also say that:

 

It's possible to make it work with some efforts in some basic configurations and / or if no changes happen on the publications

But that kind of surprises me too, actually, because it seemed like pg_upgrade wiped out the LSN locations of the subcriber, making it start all over.

Upgrading a subscriber seems like something that could/should work, so it should be documented if pg_upgrade is incompatible with maintaining a subscription, shouldn't it?

The docs are strangely silent on this. AFAIK pg_upgrade on either the publisher or subscriber breaks logical replication, which does make sense since pg_upgrade basically makes a new database cluster as it runs.

There is a way to manually set the LSN position of an enabled=false replication slot, but I've failed to make that work right in tests so far.