Hello folks, Recently, we have been facing issues where code is working well on our systems and all the tests are passing. However, for certain users the functionalities have been breaking on their devices. This is mostly related to Javascript meant to be executed on browsers. We checked if this is something device specific / browser specific but some of the affected users are using the same OS / browsers as us. We tried all device / browser combinations we could get our hands on but we could not reproduce the issue. What steps should we take to identify this issue? And how do you detect such issues early in the development process?

There are tools that let you record and replay error events in your front end application. You can look into these. https://www.rrweb.io/

You can use fullstory as well, and for testing we use browserStack, you can do cross browser testing easily

@limited-cartoon-72739 Could it be a race condition? Where some of the code is executed before all the JS is loaded? Maybe some of the JS libraries are on CDN and either not loading or blocked on the end-user's device. Have you looked at sentry.io ? Adding that to your front end could help identify the issue.

We use Sentry heavily at GitLab, and pretty much every place that throws any error is captured by Sentry (including ones within

try...catch

), so that should definitely help. Also, if it is Android webview where you’re facing this, check if they’re using the WebView as provided by OS (pre Android 6) or the one provided by Play Store (same as Chrome).