I referred to this post to update both the standalone.xml and web.xml to add ReadOnly datasources as both had a common exception.

Currently we recommend using Connection Pooling for the ExternalMKTData DB.  To use this, there must be a datasource configured which defaults to the schema where the Marketing customer data resides (e.g. customerdata).  Depending on the database type, this can be done either by using a database user which defaults to the schema, or by specifying the default schema in the datasource. For Tomcat this is configured in either the context.xml or prweb.xml files.

Here is an example for a Postgres DB with Tomcat, specifying the default customerdata schema:

<Resource name="jdbc/mktDataSource" auth="Container" type="javax.sql.DataSource" driverClassName="org.postgresql.Driver" url="jdbc:postgresql:thin://<hostname>:5432:pega" username="<someuser>" password="<password>" connectionProperties="currentSchema=customerdata"/>

Appropriate DB privileges must be granted to the user to allow full privileges on the Marketing customer data schema.

For the ExternalMKTData DB rule you would select "Use JDBC Connection Pool", and in the JNDI name field you would enter: java:comp/env/jdbc/mktDataSource. 

More information can be found in the PegaMarketing 8.3 Installation Guide

Except the below issue, everything else is resolved once the hard coded datasources are removed from the standalone.xml. I am able to connect to customer data schema and external mkt data schema using JDBC URL connection in the Database instance level.

1) When I create a DB instance and choose "use configuration in preference" as the type of connection, I get the following error message.

"The database is configured to get SQL connection information from preferences, but the preferences have no "url" or "name" setting for this database"