Distributed Application Redundancy

New web developer here. I have a web application that I want to build redundancy into. Currently we have a setup involving two web/application servers, one being active and the other passive. The active is set to replicate to the passive, which normally occurs about every 15 minutes.

I am looking to implement a solution that in the case of a failure, will allow a client to continue onto the passive server. The issue is currently if someone is on the active server, and it fails, then any information they were working there will not be immediately available on the backup server.

What is the best way to approach this? I understand that if the information hasn't been submitted on a specific page when the server goes down, then the DB will not like that/or the information might become corrupt, but I want to get a place where the information is more recent and if at all possible more relevant than from 15 mins ago. Can I cluster the application servers, point those to clustered db servers and setup the requests to the clustered names? Or does anyone have a suggestion as to somewhere where this is working in action that I can delve into?