Inconsistent Datadog Metric Datasets

[Jon Whitaker]

I've setup the Datadog metric integration and have added a metric that reports my system's uptime. The metric looks like the following:

A = sum:synthetics.http.response{check_id:checkid,step_id:stepid,status_code:200}.as_count()
B = sum:synthetics.http.response{check_id:checkid,step_id:stepid}.as_count()

C = 100 * (A/B)

I'm reporting metric C in my Statuspage monitor.

If I look at the last 1 month of timeseries data for metric C in Datadog I never see a value > 100%. However, the same timeseries data in Statuspage for the same 1 month period shows a spike at 101% a couple of days ago.

Why would the timeseries metric values in Statuspage report different values than those in Datadog? What should I try to troubleshoot this further?

[Rafael Meira]

Hey @Jon Whitaker 

It's possible that the issue could be caused by a discrepancy in the way that the metric is being collected and processed by Datadog and Statuspage. To troubleshoot this further, I recommend checking the following:

1. Make sure that the metric is being collected and processed correctly in Datadog. You can do this by checking the raw data for the metric in Datadog and comparing it to the values that are being displayed in the timeseries chart.

2. Check the settings and configurations for the integration between Datadog and Statuspage. Make sure that the integration is set up correctly and that all necessary permissions are in place.

3. Check the statuspage monitor settings to ensure that the monitor is configured to display the correct data. Make sure that the correct metric is being used and that the time range for the monitor is set correctly.

4. Check for any network or connectivity issues that could be causing delays or errors in the data being sent from Datadog to Statuspage.

I hope this helps! If you continue to have issues, it may be helpful to open a formal ticket with us by going to the  https://support.atlassian.com/statuspage/

😄