howto/postgresql: puppetization (team#41401)

this keeps the previous manual instructions for future reference, mainly
in case surprises with the new expanded puppet profile

manual bits should probably be removed eventually once puppetization is
proven reliable, to cut down on the level of clutter in here

howto/postgresql.md

@@ -1436,13 +1436,18 @@ recovery](#indirect-backup-recovery), above.
 
 ## Installation
 
-PostgreSQL servers are installed by hand on TPO servers, from Debian
-packages. For example, in Debian bullseye:
+The `profile::postgresql` Puppet class should be used to deploy and manage
+PostgreSQL databases on nodes. It takes care of installation, configuration and
+setting up the required role and permissions for backups.
+
+### Manual installation
+
+To install PostgreSQL manually on a server, first install the package:
 
     apt install postgresql-13
 
 To ensure backups are properly performed, the server must also have
-the `profile::postgres` class. The storage backup server also needs
+the `postgres::backup_source` class. The storage backup server also needs
 `REPLICATION` access to make full backups. To set this up, follow the
 steps below:
 
@@ -1506,8 +1511,12 @@ correctly:
 
 ### Prometheus PostgreSQL exporter deployment
 
-Prometheus metrics collection is not enabled by default. To enable it,
-include the following line in `/etc/postgresql/11/main/pg_hba.conf`:
+Prometheus metrics collection is configured automatically when the Puppet class
+`profile::postgresql` is deployed on the node.
+
+#### Manual deployment
+
+First, include the following line in `pg_hba.conf`:
 
     local   all             prometheus                              peer