I’m a new haproxy user, and we have our config running doing roundrobin load balancing working well.
We have several application instances that need to send http requests upon discovery of certain conditions. Sometimes these discoveries are made too close together, and we need to throttle the outbound backend requests to only one session every 30 seconds per backend server to stay within limits.
I have looked through the docs and googled a few rate limit articles, and see where I can set maxconn per server, but still feel lost about the rate limiting. Can someone provide a sample config to make this work? If such rate limiting is possible, are sessions that go over over the rate limit queued or rejected?
Thank you for helping my understanding.