I wanted to use the archiving feature, with log shipping / warm-standby server. No problem on setting up so far. However, reading the documentation brought me one question that I'd like to share with the list.
In the documentation, as far as I understood, it's said you need to make a base backup and copy the data cluster area to the destination you choose. Use pg_start_backup(), copy the data/ directory (or the cluster area), and pg_stop_backup(). It created the first checkpoint (correct?) and copied the wal segments to the area you specified. You can even delete the files under pg_xlog, because you will rely on the ones shipped out. The question is: once the base backup has been made and I change any setting on this primary server, on postgresql.conf for example, would I have to redo the procedure and set Pgsql to backup mode if I want to apply these changes I made updated on standby server or it's just a matter of transferring the file to the standby server?
