On comment:31:ticket:20499 , Teor notes: > > If we really do want every failure to result in a schedule increment, we have to remove the following code: > > download_status_increment_failure: > {{{ > /* only count the failure if it's permanent, or we're a server */ > if (status_code != 503 || server) { > if (dls->n_download_failures < IMPOSSIBLE_TO_DOWNLOAD-1) > ++dls->n_download_failures; > } > }}} > > Because combined with this code in download_status_schedule_get_delay, it causes a reset on every 503: > {{{ > /* Check if we missed a reset somehow */ > if (dls->last_backoff_position > dls_schedule_position) { > dls->last_backoff_position = 0; > dls->last_delay_used = 0; > } > }}} > > Which is exactly what we don't want when relays are busy - imagine clients doing an automatic reset every time they DoS a relay...