We should change this awaitReplication to awaitLastOpCommitted to prevent the final majority read from timing out if the secondary is slow to update the primary with its replication progress.