This message was deleted.

well, pulumi is trying to do a

replace

by creating the new DB first before deleting the old DB. The bigger question is, why is changing something like a username and password causing the RDS instance to need to be recreated? Are there any other changes that are part of this update? And if there isn't - and it's just behavior you can't get around - than is there a way for you to add new credentials without doing the replace? Presumably you don't want to lose your data by spinning up a new DB.

The only change is the master username and password on the cluster. Maybe that needs to be fixed ? i.e. immutable

@bitter-application-91815 master username doesn't support in place updates, that's on the aws api side

ah yea. So you can only rotate the password

Makes sense, thanks guys