Hi, I had a scenario where my Traefik 2.7.1 running in a Docker container became so overloaded due to many concurrent requests (180% sustained CPU usage) that it was only able to trickle a very small amount of requests to the services (service nodes were only at 1-2% CPU so it appears like none of the requests were making it through). Unfortunately I don't have the logs for that event but what could cause this and how can I prevent this from happening in the future? I would prefer it if the requests at least made it to the services. Is it something to do with max_open_files?
Thanks, looking forward to any insight on this.