Bucardo provides a very capable control mechanism in bucardo_ctl. Unlike Slony, which depends on an arcane programming language to create new replication sets and subscriptions, Bucardo is much more straightforward. As with Slony, we still want to copy data to other servers to avoid overwhelming our primary server.
In this recipe, we will utilize bucardo_ctl to create what Bucardo refers to as a herd. Bucardo herds contain one or more tables, and they are the basis of its synchronization system.
Let's begin.
We will be continuing where we left off in the Setting up Bucardo recipe. Please make sure that you have completed that recipe before continuing. As usual, we will use the pgbench utility to create an initial set of tables. Execute this command on the primary PostgreSQL server as the postgres user if you haven't already done so:
pgbench -i postgres