After updating to DataDog 2.20.0, seeing lot many App Hang crashes

[PSinha1202]

Question

Hi Team,

We recently updated to DataDog 2.20.0 and seeing lot many errors coming our way. One of them are App hang - with crash. First we thought this is real issue and even we believe now, but when we looked to the attached report, there were no crashes seen to the developer and also all of them were reported exactly at same time (check the time stamp).

This makes us think to learn more about this error report. So here are our questions.

1. Are all these reports are user facing or sometime can be internal sdk related and can be caught.
2. If they are real crash, how can they be reported at same time. Since once app crash it takes time to launch and crash again. So technically they can not have same crash.
3. Would love to more about the crash reporting in details, specially on app hang and crash.

[PSinha1202]

[mariedm]

Hi @PSinha1202,

Let me clarify how it works:

• All crashes are reported as RUM Errors. As seen in your screenshot, they are distinguished from regular errors with the “(!) Crash” label on the right.
• Non-fatal App Hangs (introduced in v2.8.0, #1685) are reported as both RUM Errors and RUM Long Tasks.
• Fatal App Hangs (introduced in v2.10.0, #1763) are also reported as RUM Errors.

Regarding your problem, we have previously observed duplicate events, which might be the issue you’re reporting. To help us narrow this down, could you please let us know:

• What SDK version were you using before upgrading to v2.20?
• If you have observed duplicate events with other versions of the SDK?
• How often are you seeing these duplicated events?

If you'd like us to take a closer look at your data, feel free to reach out to your support contact.

Hope this helps!

[maxep]

Closing for inactivity, feel free to add further comment to re-open.

[backslash-f]

We are using Datadog 2.23.0 and see duplicated app hangs being reported. Look at the timestamp:

This messes up metrics, as we have only one crash, not five (the four app hangs + the actual crash). It's getting out of hand as the user base grows, to a point where we just disabled app hangs as errors (as long as memory warnings, but that's a different matter):

RUM.enable(
    with: RUM.Configuration(
        ...
        errorEventMapper: { event in
            let errorCategory = event.error.category

            // Disables categorizing memory warnings as errors. Workaround provided by
            // Datadog: https://github.com/DataDog/dd-sdk-ios/issues/2020#issuecomment-2577054755
            //
            // Disables categorizing app hangs as errors, as *the same hang* will be reported
            // multiple times in Datadog, sometimes 4 times in a row (with the same timestamp).
            // Datadog: https://github.com/DataDog/dd-sdk-ios/issues/2158
            if errorCategory == .memoryWarning || errorCategory == .appHang {
                return nil
            }
            return event
        }
    )
)

Any suggestions?