Getting 503s after accidental high load

I accidentally sent a high volume of traffic to one of my services and am now getting very random 503 errors. The service itself has been under no load for the past 12 hours, shows good metrics, and no errors in logs yet it is still throwing 503 errors.

The service is

Here are the headers I’m getting back (the x-render-routing one seems suspicious to me):

  1. alt-svc:

h3=“:443”; ma=86400, h3-29=“:443”; ma=86400

  1. cf-cache-status:


  1. cf-ray:


  1. content-length:


  1. date:

Sat, 20 Aug 2022 17:16:08 GMT

  1. expect-ct:

max-age=604800, report-uri=“

  1. server:


  1. vary:


  1. x-render-routing:


Hey Aaron,

I’m sorry that you ran into this. It looks like your service triggered an edge case with our routing layer’s load shedding mechanism. We’ll have a fix out for this shortly.

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.