In answer.sql, insert one row into replication_probe, with the ID 1 and the note replica check. run.sh sends the file to the primary, counts the rows there, and polls the streaming replica until the row shows up. From the root of your copy, run ./reps/05/run.sh <primary> <replica>; the top of run.sh shows the way for Postgres in containers.
Expected: primary=1 replica=1, and the replica lists the row with the note replica check. The replica may show 0 for a moment before it catches up. Running it a second time fails on the primary key: delete the row on the primary before a retry.