Error setting auto scale using custom metric

Question

Hello.

An error occurred when I tried to set up auto scale using custom metrics with the following settings.

   {  
       "location": "Japan East",  
       "tags": {},  
       "properties": {  
           "name": "hogehoge",  
           "enabled": true,  
           "targetResourceUri": "hogehoge",  
           "profiles": [  
               {  
                   "name": "hogehoge",  
                   "capacity": {  
                       "minimum": "1",  
                       "maximum": "3",  
                       "default": "1"  
                   },  
                   "rules": [  
                       {  
                           "scaleAction": {  
                               "direction": "Increase",  
                               "type": "ChangeCount",  
                               "value": "1",  
                               "cooldown": "PT5M"  
                           },  
                           "metricTrigger": {  
                               "metricName": "metrics_un_d",  
                               "metricNamespace": "metrics_un_d",  
                               "metricResourceUri": "hogehoge",  
                               "operator": "GreaterThan",  
                               "statistic": "Average",  
                               "threshold": 1,  
                               "timeAggregation": "Minimum",  
                               "timeGrain": "PT1M",  
                               "timeWindow": "PT10M",  
                               "Dimensions": [],  
                               "dividePerInstance": false  
                           }  
                       }  
                   ]  
               }  
           ],  
           "notifications": [],  
           "targetResourceLocation": "Japan East"  
       },  
       "id": "hogehoge",  
       "name": "hogehoge",  
       "type": "Microsoft.Insights/autoscaleSettings"  
   }  

The error is as follows.

   {"code": "UnsupportedMetric", "message": "Exception of type 'Microsoft.WindowsAzure.Management.Monitoring.MonitoringServiceException' was MonitoringServiceException' was thrown."}  

The custom metric used is the value sent by the Rest API (python) from the App Service (flask) built on the Azure server.

Is it possible to tell us why an error occurs when we set up auto scale using the custom metric?

Answer 1

Your app service is a function app on the consumption plan. Autoscale for app service plans are only supported for SKUs Standard and above. By default, the Azure Function Consumption plan will scale automatically in periods of high load, there's no need for an auto scale rule.