>Update - primary failover briefly improved performance but did not fully mitigate, we've throttled inbound traffic and are investigating upstream Vitess issues
And now they're blaming their upstream vendor! Embarrassing stuff to be writing on a public page.