Alerting for failed distributed traces

[Peter_Youssef]

I need to create an alert for the failed distributed traces for this service as attached below 

I've specified the failed only and the amount of time it exceeds to be alerted on.

appreciate your support and guidance.

[Peter_Youssef]

I've created it through creating a custom metric then create a custom events for alerting.

It's very interesting, Thanks.
If there any extra methods to configure it, please share it with me.

[AurelienGravier · Accepted solution]

Hello @Peter_Youssef 

The best way for me :

1- Create a request naming rule on the service settings to group all your traces on the same request (request naming rules help to adjust how your requests are tracked). example of naming rule for your use case :

2- It takes a while for newly added key requests to show on the top request list, the new request will appear on your service, opening it and mark it as key request. It will help to apply different thresholds for this specific request.

3- Go on the services settings on the menu, anomaly detection and you will be able to specify a custom threshold for this key request to detect anomalies on error rate for example :

For example, this configuration will generate a problem when the error rate on the request "/b2b/trnprocessor/identifiers/no-id/sales-ext" will exceed 5% :

Regards Aurelien.

[Peter_Youssef]

after these steps, I've created problem alerting profile and it works perfectly now.

[Peter_Youssef]

Thanks @AurelienGravier, It worked fine.

[AurelienGravier]

hello, @Peter_Youssef 

Did you try the solution ?

[Peter_Youssef]

Tested ,Thanks for guidance @AurelienGravier