Overview
If you experience unexpected behavior with Datadog RUM, use this guide to resolve issues quickly. If you continue to have trouble, contact Datadog Support for further assistance.
Android
Check if Datadog RUM is initialized
Use the utility method isInitialized to check if the SDK is properly initialized:
if (Datadog.isInitialized()) {
// your code here
}
Debugging
When writing your application, you can enable development logs by calling the setVerbosity method. All internal messages in the library with a priority equal to or higher than the provided level are then logged to Android’s Logcat:
Datadog.setVerbosity(Log.INFO)
Set tracking consent (GDPR compliance)
To be compliant with the GDPR regulation, the SDK requires the tracking consent value at initialization.
Tracking consent can be one of the following values:
TrackingConsent.PENDING: (Default) The SDK starts collecting and batching the data but does not send it to the
collection endpoint. The SDK waits for the new tracking consent value to decide what to do with the batched data.TrackingConsent.GRANTED: The SDK starts collecting the data and sends it to the data collection endpoint.TrackingConsent.NOT_GRANTED: The SDK does not collect any data. You are not able to manually send any logs, traces, or
RUM events.
To update the tracking consent after the SDK is initialized, call Datadog.setTrackingConsent(<NEW CONSENT>). The SDK changes its behavior according to the new consent. For example, if the current tracking consent is TrackingConsent.PENDING and you update it to:
TrackingConsent.GRANTED: The SDK sends all current batched data and future data directly to the data collection endpoint.TrackingConsent.NOT_GRANTED: The SDK wipes all batched data and does not collect any future data.
Sending data when device is offline
RUM ensures availability of data when your user device is offline. In cases of low-network areas, or when the device battery is too low, all the RUM events are first stored on the local device in batches.
Each batch follows the intake specification. They are sent as soon as the network is available, and the battery is high enough to ensure the Datadog SDK does not impact the end user’s experience. If the network is not available while your application is in the foreground, or if an upload of data fails, the batch is kept until it can be sent successfully.
This means that even if users open your application while offline, no data is lost. To ensure the SDK does not use too much disk space, the data on the disk is automatically discarded if it gets too old.
Migrating to 2.0.0
If you’ve been using the SDK v1, there are some breaking changes introduced in version 2.0.0. See the migration guide for more information.
iOS
Check if Datadog SDK is properly initialized
After you configure Datadog SDK and run the app for the first time, check your debugger console in Xcode. The SDK implements several consistency checks and outputs relevant warnings if something is misconfigured.
Debugging
When writing your application, you can enable development logs by setting the verbosityLevel value. Relevant messages from the SDK with a priority equal to or higher than the provided level are output to the debugger console in Xcode:
Datadog.verbosityLevel = .debug
If all goes well you should see output similar to this saying that a batch of RUM data was properly uploaded:
[DATADOG SDK] 🐶 → 17:23:09.849 [DEBUG] ⏳ (rum) Uploading batch...
[DATADOG SDK] 🐶 → 17:23:10.972 [DEBUG] → (rum) accepted, won't be retransmitted: success
Recommendation: Use Datadog.verbosityLevel in DEBUG configuration, and unset it in RELEASE.
Using DDURLSessionDelegate with your own session delegate
If you want to automatically track network requests with DDURLSessionDelegate but your app already implements its own session delegate, you can use either inheritance or composition patterns and forward calls to DDURLSessionDelegate.
When using inheritance, use DDURLSessionDelegate as a base class for your custom delegate and make sure to call the super implementation from your overridden methods. For example:
class YourCustomDelegateURLSessionDelegate: DDURLSessionDelegate {
override func urlSession(_ session: URLSession, task: URLSessionTask, didCompleteWithError error: Error?) {
super.urlSession(session, task: task, didCompleteWithError: error) // forward to Datadog delegate
/* your custom logic */
}
}
When using composition, leverage Datadog’s __URLSessionDelegateProviding protocol to keep an internal instance of DDURLSessionDelegate and forward calls to ddURLSessionDelegate. For example:
private class YourCustomDelegateURLSessionDelegate: NSObject, URLSessionTaskDelegate, URLSessionDataDelegate, __URLSessionDelegateProviding {
// MARK: - __URLSessionDelegateProviding conformance
let ddURLSessionDelegate = DDURLSessionDelegate()
// MARK: - __URLSessionDelegateProviding handling
func urlSession(_ session: URLSession, task: URLSessionTask, didFinishCollecting metrics: URLSessionTaskMetrics) {
ddURLSessionDelegate.urlSession(session, task: task, didFinishCollecting: metrics) // forward to Datadog delegate
/* your custom logic */
}
func urlSession(_ session: URLSession, task: URLSessionTask, didCompleteWithError error: Error?) {
ddURLSessionDelegate.urlSession(session, task: task, didCompleteWithError: error) // forward to Datadog delegate
/* your custom logic */
}
func urlSession(_ session: URLSession, dataTask: URLSessionDataTask, didReceive data: Data) {
ddURLSessionDelegate.urlSession(session, dataTask: dataTask, didReceive: data) // forward to Datadog delegate
/* your custom logic */
}
}
Note: If using composition, ddURLSessionDelegate must receive all necessary calls listed in __URLSessionDelegateProviding protocol comments. Your delegate needs to:
- implement
URLSessionTaskDelegate and forward: - implement
URLSessionDataDelegate and forward:
Flutter
Cocoapods issues
If you have trouble building your iOS application after adding the Datadog SDK because of errors being thrown by Cocoapods, check which error you are getting. The most common error is an issue getting the most up-to-date native library from Cocoapods, which can be solved by running the following in your ios directory:
pod install --repo-update
Another common error is an issue loading the FFI library on Apple Silicon Macs. If you see an error similar to the following:
LoadError - dlsym(0x7fbbeb6837d0, Init_ffi_c): symbol not found - /Library/Ruby/Gems/2.6.0/gems/ffi-1.13.1/lib/ffi_c.bundle
/System/Library/Frameworks/Ruby.framework/Versions/2.6/usr/lib/ruby/2.6.0/rubygems/core_ext/kernel_require.rb:54:in `require'
/System/Library/Frameworks/Ruby.framework/Versions/2.6/usr/lib/ruby/2.6.0/rubygems/core_ext/kernel_require.rb:54:in `require'
/Library/Ruby/Gems/2.6.0/gems/ffi-1.13.1/lib/ffi.rb:6:in `rescue in <top (required)>'
/Library/Ruby/Gems/2.6.0/gems/ffi-1.13.1/lib/ffi.rb:3:in `<top (required)>'
Follow the instructions in the Flutter documentation for working with Flutter on Apple Silicon.
Set sdkVerbosity
If you’re able to run your app, but you are not seeing the data you expect on the Datadog site, try adding the following to your code before calling DatadogSdk.initialize:
DatadogSdk.instance.sdkVerbosity = Verbosity.verbose;
This causes the SDK to output additional information about what it’s doing and what errors it’s encountering, which may help you and Datadog Support narrow down your issue.
Not seeing Errors
If you do not see any errors in RUM, it’s likely no view has been started. Make sure you have started a view with DatadogSdk.instance.rum?.startView or, if you are using DatadogRouteObserver make sure your current Route has a name.
Issues with automatic resource tracking and distributed tracing
The Datadog tracking HTTP client package works with most common Flutter networking packages that rely on dart:io, including http and Dio.
If you are seeing resources in your RUM Sessions, then the tracking HTTP client is working, but other steps may be required to use distributed tracing.
By default, the Datadog RUM Flutter SDK samples distributed traces at only 20% of resource requests. While determining if there is an issue with your setup, you should set this value to 100% of traces by modifying your initialization with the following lines:
final configuration = DdSdkConfiguration(
//
rumConfiguration: RumConfiguration(
applicationId: '<RUM_APPLICATION_ID>',
tracingSamplingRate: 100.0
),
);
If you are still having issues, check that your firstPartyHosts property is set correctly. These should be hosts only, without schemas or paths, and they do not support regular expressions or wildcards. For example:
✅ Good - 'example.com', 'api.example.com', 'us1.api.sample.com'
❌ Bad - 'https://example.com', '*.example.com', 'us1.sample.com/api/*', 'api.sample.com/api'
React Native
No data is being sent to Datadog
Follow these instructions in order when the SDK has been installed and the app compiles, but no data is received by Datadog.
1. Check the configuration
Sometimes, no data is sent due to a small misstep in the configuration.
Here are some common things to check for:
- Ensure your
clientToken and applicationId are correct. - Make sure you have not set
sessionSamplingRate to something other than 100 (100 is the default value), or else your session might be sent. - If you’ve set up a
Proxy in the Datadog configuration, check that it has been correctly configured. - Check that you are tracking views (all events must be attached to a view) and sending events.
2. Review SDK logs in React Native
Set config.verbosity = SdkVerbosity.DEBUG, which imports SdkVerbosity from @datadog/mobile-react-native.
Logs start appearing in the JavaScript console, like the following output:
INFO DATADOG: Datadog SDK was initialized
INFO DATADOG: Datadog SDK is tracking interactions
INFO DATADOG: Datadog SDK is tracking XHR resources
INFO DATADOG: Datadog SDK is tracking errors
DEBUG DATADOG: Starting RUM View "Products" #Products-oaZlP_FVwGM5vtPoup_rT
DEBUG DATADOG: Adding RUM Action "RCTView" (TAP)
Note: In this example, the first four logs indicate that the SDK has been correctly configured and the last two lines are events that were sent.
Possible cause
If you are on iOS and see some DEBUG logs indicating that logs or RUM events were sent before the initialization logs, this may be why the SDK is not sending events.
You cannot send events before initialization, and attempting to do so puts the SDK in a state where it cannot send any data.
Solution
With DdSdkReactNative.initialize:
If you use DdSdkReactNative.initialize to start the Datadog SDK, call this function in your top-level index.js file so that the SDK is initialized before your other events are sent.
With DatadogProvider:
Starting from SDK version 1.2.0, you can initialize the SDK using the DatadogProvider component. This component includes a RUM events buffer that makes sure the SDK is initialized before sending any data to Datadog, which prevents this issue from happening.
To use it, see the Migrate to the Datadog Provider guide.
3. Review native logs
Reviewing native logs can give you more input on what could be going wrong.
On iOS
Open your project in Xcode by running xed ios.
Build your project for a simulator or a device.
Native logs start appearing on the bottom right corner:
You can filter logs by “DATADOG” and look for any error.
If you are indeed sending events, you should see the following logs:
[DATADOG SDK] 🐶 → 10:02:47.398 [DEBUG] ⏳ (rum) Uploading batch...
[DATADOG SDK] 🐶 → 10:02:47.538 [DEBUG] → (rum) accepted, won't be retransmitted: [response code: 202 (accepted), request ID: AAAABBBB-1111-2222-3333-777788883333]
The first log indicates that some data is being sent, and the second log indicates that the data has been received.
Possible cause
If you see the log below, it means that you have called a RUM method before initializing the SDK.
[DATADOG SDK] 🐶 → 10:09:13.621 [WARN] The `Global.rum` was called but no `RUMMonitor` is registered. Configure and register the RUM Monitor globally before invoking the feature:
Solution
With DdSdkReactNative.initialize:
If you use DdSdkReactNative.initialize to start the Datadog SDK, call this function in your top-level index.js file so the SDK is initialized before your other events are sent.
With DatadogProvider:
Starting from SDK version 1.2.0, you can initialize the SDK using the DatadogProvider component. This component includes a RUM events buffer that makes sure the SDK is initialized before sending any data to Datadog, which prevents this issue from happening.
To use it, see the Migrate to the Datadog Provider guide.
On Android
For a better debugging experience, Datadog recommends installing pidcat.
- pidcat filters the device logs (obtained by
adb logcat) to only show the one from your application. - See this issue for M1 users who don’t have Python 2.
Modify node_modules/@datadog/mobile-react-native/android/src/main/kotlin/com/datadog/reactnative/DdSdk.kt to enable verbose logging from the native SDK:
fun initialize(configuration: ReadableMap, promise: Promise) {
// ...
datadog.initialize(appContext, credentials, nativeConfiguration, trackingConsent)
datadog.setVerbosity(Log.VERBOSE) // Add this line
// ...
}
Run the app on a phone connected in debug mode to your laptop (should appear when running adb devices), or from an emulator.
Run pidcat my.app.package.name or adb logcat from your laptop.
Look for any error mentioning Datadog.
Pidcat output looks like this:
In this example, the last log indicates that the batch of RUM data was sent successfully.
Undefined symbols: Swift
If you see the following error message:
Undefined symbols for architecture x86_64:
"static Foundation.JSONEncoder.OutputFormatting.withoutEscapingSlashes.getter : Foundation.JSONEncoder.OutputFormatting", referenced from:
static (extension in Datadog):Foundation.JSONEncoder.default() -> Foundation.JSONEncoder in libDatadogSDK.a(JSONEncoder.o)
...
Open Xcode, go to the Build Settings of your project (not your app target), and make sure Library Search Paths have the following settings:
LIBRARY_SEARCH_PATHS = (
"\"$(TOOLCHAIN_DIR)/usr/lib/swift/$(PLATFORM_NAME)\"",
"\"/usr/lib/swift\"",
"\"$(inherited)\"",
);
Undefined symbols: _RCTModule
If you see an undefined _RCTModule symbol, it may be related to this change in the react-native v0.63 changelog.
You can make the following change to fix it:
// DdSdk.m
// instead of
#import <React/RCTBridgeModule.h>
// maybe that:
@import React // or @import React-Core
Infinite loop-like error messages
If you run into an issue where your React Native project displays a stream of error messages and significantly raises your CPU usage, try creating a new React Native project.
Further Reading
Additional helpful documentation, links, and articles:
