On Sunday 25 October, between 02:01 and 02:06 CEST (00:01–00:06 UTC), a sudden spike in automated traffic overloaded the servers that route incoming traffic to Kundo. For about three minutes, some requests to Kundo returned errors. Traffic was back to normal at 02:06 CEST, and Kundo has worked as usual since then. You do not need to do anything.
We have identified the cause and are adding protection so that traffic spikes like this one do not affect the service in the same way. We are also improving our alerting for this type of event.