Having finally received the actual details of what the OP actually is doing in email #37 of this thread, I was struck by a simple thought. I have re-read the whole thread, and don't think/hope that I am about to say anything completely stupid.
> "We develop software that routes millions of requests to dozens of Tomcat instances. " So you have your own software in front of many tomcats that is responsible for distributing the load between multiple tomcat instances > "Yes, we can and do support connection throttling at a slight cost to safeguard a single Tomcat from receiving more connections that it can, but if Tomcat was able to not reset connections at the TCP level - we can perform our task much better, and I do not think this will cause any problem to any other use cases of Tomcat - if we can just enable this behavior with a configuration parameter" My simple thought was that it sounds like your code isn't working. You have more load than one tomcat instance can handle, which overloads that instance. You are trying to write code to handle this situation, and seem convinced that the only solution is to alter tomcat such that you can detect/handle this occurrence in a way that is easier for your software. You also state that when this happens, you will simple route to there tomcat instances - the implicit assumption that they have spare capacity on the other instances. If this is the case, why didn't your code route to these other instances in the first place? Surely this would obviate the need for any changes to tomcat? What algorithm do you use to determine where to send the load? > "I do not understand the negativity here.." After writing comments such as "If you can, try to understand what I said better.. Its ok to not accept this proposal and/or not understand it.." you really can't understand the negativity? Really? Are you sure? Chris
