Deployment with Microsoft.ServiceFabricMesh provider never completes

[tomkerkhove]

As of private previes 4 you can now use the Microsoft.ServiceFabricMesh ARM provider with the 2018-07-01-preview version .

When trying this on the Cloud Shell I've received the following error:

Deployment failed. Correlation ID: 2b54f110-9402-41ce-b87a-2a31e09783d4.

{
"status": "Failed",
"error": {
"code": "ResourceDeploymentFailure",
"message": "The resource operation completed with terminal provisioning state 'Failed'."
}
}

This is ok as it's marked as a known issue.

When using a regular ARM deployment in the portal it keeps on running forever but:

• Network is created according to the ARM deployment & portal but az mesh network cannot find it - Probably because of different resource type?
  • No way to get the public IP to call my API
• App is deployed according to the portal but az mesh app cannot find it and the deployment is still ongoing after 17 minutes.

My questions are:

• Will the cloud shell be updated for public preview so that it works with the new provider?
• Will the portal provide more information about the resource itself when going public?

The deployment that is stuck is tracked in another issue already AFAIK

ARM Deployment

Correlation Id: 4f3a6ffb-f6b5-454e-ac95-79162c8057e3

Template

{
    "apiVersion": "2018-07-01-preview",
    "name": "[parameters('Application.Network.Name')]",
    "type": "Microsoft.ServiceFabricMesh/networks",
    "location": "[parameters('Region')]",
    "dependsOn": [],
    "properties": {
        "description": "Public network for Containo",
        "addressPrefix": "10.0.0.4/22",
        "ingressConfig": {
            "qosLevel": "bronze",
            "layer4": [
                {
                    "publicPort": "1337",
                    "applicationName": "[parameters('Application.Name')]",
                    "serviceName": "Containo.Orders.API",
                    "endpointName": "HTTP-Public-APIs-Orders"
                }
            ]
        }
    }
}

[tomkerkhove]

Closing as version 0.8.0rc2 of the CLI fixes this.