Sentry Review!
April 16, 2018
Sentry Review!
Score 9 out of 10
Vetted Review
Verified User
Overall Satisfaction with Sentry
We use Sentry for monitoring all production web systems. Sentry solves a really important problem for us: in production, we'd like our services to automatically recover if (when) they go down, have errors, or throw exceptions. That's a problem from an error reporting perspective because we then lose failure state on reboot. Sentry helps us bridge this by logging and reporting production errors, so we can actively work to fix them, but don't have to worry about systems staying down while we work to triage.
- Sentry doesn't just provide a list of errors and exceptions, but instead gives us a full stack trace to work from
- Internally, Sentry handles a lot of code versioning checks so we can see where errors originated from, and if deployed fixes are in fact working
- Sentry interfaces with existing reporting tools so we can respond to production exceptions in minutes, instead of when we find out about them in viewing logs
- Very much a one-off tool; Sentry would be a lot nicer to use if it built out additional monitoring tools so we could run the entirely of our logging and reporting through their software
- Some of their documentation, particularly for the open source projects that drive error collection, can be lacking or inadequately communicated
- Sentry has gotten us ahead of tackling any production errors as they happen, as a result, it's far more unusual for a customer to find out about a problem before we do -- normally, we can triage and deploy a fix before they even notice
Sentry is really a tool to be used in combination with other things, like Pingdom and PagerDuty. For those applications, Sentry is a far more full-features offering that lets you see why errors happened, not just be alerted to their occurrence. We chose it over other error monitoring applications based on flexibility and ease of use.