Great Moments in Application Monitoring
The consequences of poor application performance are both real and terrifying: lost customers, lost trust from your team, and lost confidence in your abilities. It’s why every developer remembers those moments where their solution improves both process and performance. We’ve seen this happen firsthand with our customers, and want to share their Great Moments in Application Monitoring:
When you’re connecting hungry customers with hurried drivers, you can’t afford to have debugging eat up development time. Unfortunately, whenever a DoorDash developer tried something new without informing the larger team, resources would be wasted debugging their unknown logic. With Sentry’s automated stack traces, however, DoorDash developers could finally — and clearly — understand their colleagues’ code. By including $request_id in their access logs, passing that value downstream to their application servers, and then modifying the application’s entry point, DoorDash was able to utilize
$request_id as a logging variable. Now, whenever there’s an error, DoorDash can search
$request_id and immediately see the exact issue and exception. The result? Scalable production, the ability to trace every request end‑to‑end, and a process that’s as cool as a cucumber bisque.
As the one service that’s consistently able to catch these types of NGINX variables, Sentry is really valuable at tracing exceptions of errors. And by providing automatic feedback, we’re able to see things clearly and gain a better understanding of what’s going wrong.
Zhaobang Liu, DevOps Engineer, DoorDash
For event marketer Eventbrite, their errors were talking over one another, as their team was constantly being distracted by alerts on duplicate issues. So they hired Sentry to emcee their errors. After instrumenting Sentry’s grouping and merging features, the Eventbrite team was able to create a list of errors prioritized by user impact and frequency. Unlike triaging in the past — where the Eventbrite team had to parse through countless duplicate issues — now they were only focused on solving what mattered most. What’s more, using Release Health, Eventbrite improved the quality of their releases by gauging adoption levels as well as the percentage of both crash-free sessions and crash-free users.
Sentry helps our team fix the most important issues in each release. We can track how a release is trending by percent of crash-free sessions. With this data, we can remediate issues that impact the most users and move on to building more features.
Jaylum Chen, Staff Software Engineer at Eventbrite
The year was 2019. And Cloudflare’s release process consisted of manually tagging a release and then manually running through their dashboard to find errors. Not only was this process tedious, it made the Cloudflare team susceptible to errors and redundant work. But with Sentry’s Releases, the Cloudflare team was able to part the skies of gloomy manual work. By setting up an automated release plan, the Cloudflare team could now safely deploy code, monitor every step of their deployment process, and link all releases back to the Releases page. When automated tools — not manual brains — do the work, both Cloudflare’s code and their customers win.
Sometimes errors on the front-end have roots on the backend. We use Sentry’s tags and metadata about a request that comes in to pass along a version of distributed tracing to link these back.
Tony Stuck, Engineering Manager, Cloudflare
With trivia games like Song Quiz, Yes Sire, and Are You Smarter Than a 5th Grader?, Volley entertains millions of families on smart speaker devices every month. But with all those players blurting out answers at their voice-controlled devices, Volley engineers knew that they needed complete visibility into their code so they could properly scale their product.
Using our performance monitoring offering, the Volley team observed that their voice-controlled LaunchRequest was waiting for an API call to update the user’s subscription status. Once Volley moved this call to only those sections of the code which needed it, they noticed an improved Apdex score, faster user interactions, and their time to launch was cut in half.
Sentry has become part of my daily routine. Every morning, I check to see if there are any new issues, and look at yesterday’s performance to see if any of our latest changes affect it negatively or positively.
Devin Ozel, Sr. Software Engineer, Volley
With Great Moments in Application Monitoring, we recognize those intrepid developers who were able to get their application to perform the way it was intended - and the way their customers demand.