main reason: table updates based on a call to backend (which we wait on), but the update to the table has some delay. This delay is not a problem local, but we've noticed flakiness when it runs on CI