[azure arm endpoint creation] too many 500 error responses

Hello,

I am trying to use service_endpoint_client to create a new azure ARM service connection, here's the code I am using:

core_client = self.__get_connection().clients_v6_0.get_service_endpoint_client()
        try:
            azure_arm = core_client.create_service_endpoint(
                {
                    "authorization": {
                        "parameters": {
                            "tenantid": "TENANT_ID",
                            "scope": "/subscriptions/SUB_ID/resourcegroups/RG_NAME"
                        },
                        "scheme": "ServicePrincipal"
                    },
                    "data": {
                        "environment": "AzureCloud",
                        "scopeLevel": "Subscription",
                        "subscriptionId": "SUB_ID",
                        "subscriptionName": "SUB_NAME",
                        "creationMode": "Automatic"
                    },
                    "name": "TEST_ME",
                    "type": "azurerm",
                    "isReady": True,
                    "url": "https://management.azure.com/",
                    "serviceEndpointProjectReferences": [
                        {
                            "name": "TEST_ME",
                            "description": "test",
                            "projectReference": {
                                "id": "AZ_DEVOPS_PROJECT_ID"
                            }
                        }
                    ]
                }
            )

The service connection was created successfully but I got this exception:

msrest.exceptions.ClientRequestError: Error occurred in request., RetryError: HTTPSConnectionPool(host='****', port=443): Max retries exceeded with url: /_apis/serviceendpoint/endpoints (Caused by ResponseError('too many 500 error responses'))

Could you please help to identify the issue?

Inizia con il service_endpoint_client dell’Azure DevOps Python API e la chiamata create_service_endpoint mostrata nel report. Riproduci la richiesta usando i campi dell’endpoint forniti e analizza la risposta 500 risultante; il lavoro è completato quando viene identificato il comportamento della richiesta o del client che causa l’eccezione e questo viene documentato o corretto.