On April 15th, from approximately 12:50pm - 1:30pm eastern, there was an incident which affected users ability to load the Roundtrip platform. The issue presented as users being unable to load the main dashboard pages - resulting in a blank page or incomplete page. Additionally, non-dashboard pages were loading in a state that was distorted and difficult to read.
The root cause of this incident was related to a failed step in the deployment of the application. during a routine deployment. In particular, the failed step did not load the "front-end" assets for the build properly. These "front-end" assets include images that are displayed in the application, styling information that define the presentation of the site information and scripting assets that control functionality on the web pages themselves. With these assets missing the site was unable to display correctly, thus causing the issue users experienced.
Once the team discovered the issue with the failed deployment step - we took immediate action to redeploy the application. The subsequent deployment worked as expected - and functionality was restored to the site. Root cause analysis of what caused the failed deployment was inconclusive. This was due to a lack of detailed logging of this step in the deployment process. This limited the data the team had available to analyze and diagnose the failed steps. We have corrected this logging, so that if this happens again we will be able to better diagnose the root failure. This additional logging will also alert us more quickly to a failed deployment step in the event this issue occurs again.
Although logging was not conclusive, we believe the most likely cause of this deployment issue was an temporary platform issue - either temporary network connectivity, temporary loss of an underlying component or something along those lines. We believe this because of two key facts:
1. Our build deployment steps have not changed since the 2024. 2. Deploying our software is something Roundtrip does frequently in order to manage risk of making changes. In fact we have deployed the application to production 208 times in 2025!
Because of this, we are certain that there is not an ongoing inherent risk in our deployment steps. But we do use this incident and any issue we encounter as an opportunity to improve our processes. We apologize for the interruption of service and will continue to work to make Roundtrip a reliable and safe service!
Final note - no back-end processes were affected with this issue. Functions like processing patient data, launching rideshare rides and sending notifications all continued to work as expected throughout the incident. Additionally, there was no data loss and there was no security exposure with this incident.