Toll Brothers (XNYS:TOL) is an American real estate company and builder of luxury homes in major metropolitan areas in the United States.
Production Monitoring Ecosystem
About the Team
Our team develops an application that offers an order processing engine along with operating a product catalog system. While our application is internal, it’s being used by over 7,000 employees along with third party contractors, vendors and consultants.
Key challenges in error handling
Our main challenge was identifying that a critical error has occurred, and then trying to find the information needed to solve it by sifting through the application logs. In many cases, we only learned about errors from our end users and customers who encountered an issue.
Once a critical error arose, we had to talk with the users to get the exact conditions that caused the error and try and reproduce it on our end. This process required a massive amount of research, which could take hours and sometimes even days.
In certain cases, the users might just ignore that error considering it to be just a hiccup and move on without reporting it. In such cases, they lose confidence on the quality of our product.
How does OverOps help you solve issues?
After installing OverOps, we now know when and where production code breaks, along with the complete source code and variable state that caused it. We can pinpoint the exact location of the user and the values that were sent in a specific transaction, seeing the root cause and fixing the error within minutes.
How are you integrating OverOps with your daily workflow?
We use the OverOps dashboard on a daily basis to proactively identify and detect when new errors are introduced across our servers and applications. Whenever we detect a critical issue, we immediately create a JIRA ticket with the relevant information, instead of manually researching logs and copy-pasting the variable values ourselves.
Since OverOps can also provide our DEBUG statements, we can quickly understand what happened by looking at the error analysis view. We don’t have to manually set up the data and recreate the scenario to debug it.