I am developing a web application, and I was asked if it will work with a load balancer. My initial reaction is yes, as there is no state being tracked between requests anywhere in the system. However, there is a certain state of the application loaded when the application starts (configuration settings from the database basically.)
This data is read-only. Is it enough to rely on the usual cache dependency mechanisms to manage this and invalidate these objects in all cluster applications, or will I need to switch to a common caching system like App Fabric to ensure reliability / consistency?
With diagnostics enabled, I have numerous logging calls using EventSource.Write and exiting the process logger. I assume that in this case I will need one registrar installed on each of the servers in the cluster to pick up events, each of which is triggered. I haven’t been too fussed about this, but what is a good way to determine which server in the cluster was serving the request?
source share