This query is sort of normal as I need to perceive the (safety) advantages of a ‘timeout’ typically.
For our Nginx proxy-server, we’ve got been utilizing HTTP/S timeouts the place we got here throughout the problem had been the Nginx server returned a time-out. Now, we’ve got solved this by merely rising the Nginx time-out. We hold upscaling the timeout, could it’s for a selected endpoint, it appears we hold pushing an underlying drawback. We see this drawback many times the place I requested the query: Why can we even need to have timeouts?
Fascinated with some malicious makes an attempt, like sending a bulk of load to the server, if Nginx offers a timeout (or any ‘timeout supervisor’) the server would nonetheless be processing the information.
So, why would we use server timeouts and what can be a greater approach to clear up the problems for reaching the timeout cap each time? Can be paradigm like WebSocket, SSE or (Lengthy-)Polling resolve this?