I'm having trouble understanding how the standard metric "Microsoft.AspNetCore.Hosting|Failed Requests (Azure.ApplicationInsights)" works in Azure. I'm hoping someone can shed some light on it as it seems no documentation on it exists either.
The alert is configured as shown in below picture. It's being triggered multiple times a day, while no failures are appearing in the relevant Application Insights instance.
I was expecting the chart from the above alert would match following kusto query:
| where timestamp > start and timestamp < end
| where client_Type != "Browser"
| summarize failedCount=sumif(itemCount, success == false) by bin(timestamp, timeGrain)
| extend request='Overall'
| render timechart