Thanks for the webex today, the project looks like it's coming along well.
One of the things I have not heard mentioned at all during this process is troubleshooting. Which can be especially complex in multi-tiered environments. I know that you're applying a strict unit testing approach to the development, are you expecting this to be the foundation for operational troubleshooting? Do you have a plan for how the unit testing will operate across the tiers? How do you envision members doing troubleshooting?
Thanks
While you are right that, in general, troubleshooting an application with more tiers is more difficult, there are a few things we've done to help us in that area. First, we have started and are maintaining an administrative dashboard application which can be used at all sites to actively monitor the health and connectivity of all the tiers. Second, we have introduced far more logging, both at the client and services layers, than we've had in the past. And we envision that this logging will be able to be consumed in several different ways--rolling files, automated alerts, perhaps even transmitted to Tessitura staff for aggregate analysis with other sites' logs. Finally, the web application itself is much easier to debug in real-time production use than is our current application with the use of tools such as Firebug, etc.
Thanks Chuck, it sounds like you're working in the right direction. Any aggregation of logging over the tiers (in a verbose mode) will be helpful. My experience with a similar architecture is that it's always harder than you think it will be. Especially for your users who are not deeply experienced with the system. Any tools you can create during the development will help.