pub async fn run_replication<SF, KF>(
job_id: &str,
streams: SF,
sinks: KF,
resources: &ResourcesConfig,
metrics: Arc<Metrics>,
cancel: CancellationToken,
progress: ProgressHandle,
) -> Result<Checkpoint>Expand description
Reconnecting driver: opens a stream and a sink, runs the apply loop, and on transient failures reloads the checkpoint from the sink (the source of truth) and reconnects.