I have a feeling there is going to be a very simple answer to this one, but I'm going round in circles!
Basically I have a basic RMQ producer POC using both the standard .NET Core RMQ client as well as EasyNetQ. All is working fine, where as per the examples, a connection is created within the scope of the function and then disposed of after.
I then started doing some performance tests of the code and found that there is a serious performance issue with regard to throughput, and I think it may be because a new TCP connection is being created for every request.
So, I was wondering what the best method for creating a connection to RMQ was (which contains logic for detecting if the connection has died etc), maybe on application startup which was then available to controllers etc to be able to create channels and produce messages.
Can anyone help?! Thanks.