As a Java developer, my least favorite word in the world is EXCEPTION.
Errors happen, and when they do you have to know in order to be able to act. Sometimes log collection is not enough on its own; where timing is critical and SLA are a Damocles sword, you need dedicated error reporting.
What is this ?
Rollbar is a centralized error collecting and reporting service.
It collects error reports from various applications, is language agnostic, counts occurrences, collects stacktraces and traces deploys and fixes. It is a terribly useful tool that enables efficient error detection and an agile bugfixing process.
How did I do this ?
To configure centralized error management I just had to:
- sign up to the service
- integrate an error collection agent with appropriate SDK for every specific language (I only did Java and Python but more are available)
- configure exception filtering to avoid false positives
- configure dashboards filtering by applications
- configure alerting policies
- integrate with Jenkins deploys to get deploy-tracking
- integrate with BitBucket and GitLab CI/CD to get deploy-tracking
(Deploy tracking integration is highly recommended to get the most out of error tracking).
How would this be instrumental in a business environment ?
Error management is a very widespread activity that we work on daily.
Configuring a centralized error collecting service gave me a more track-your-errors oriented mindset. I never forget to track uncaught errors anywhere now!
Can I test this?
You can test the application at Rollbar but I can’t give you access to my account because of security reasons. Write me an email and I will happily discuss it with you.