Installing and configuring Virtual Machines in Google Cloud

The following instructions explain how to set up servers using Virtual Machines set up in Google Cloud. For more information, see here.

There are several other methods that you can use to set up your servers, including using Microsoft Azure (Windows Azure).

CloudM recommends setting up Virtual Machines using Google Cloud when migrating to Google (and Microsoft Azure when migrating to Microsoft) to limit network traffic costs.

All CloudM Migrate secondary servers must be on the same VM network as the primary server. The default network configuration for new VM instances in Google Cloud does not need to be modified to meet this requirement.

You will also need a Google Cloud billing account and associated Google Cloud project for creating VM instances and to be mindful of the associated costs.

The instructions differ at various points throughout this guide depending on the type of server being created and configured (primary or secondary), and these steps are highlighted in bold.

Creating a VM instance

When creating the VM in Google Cloud, you will see the estimated monthly cost of running the Virtual Machine at the current configuration. As you make the changes advised in this article, this estimated cost will also change.

It is also advised that you shut down any Virtual Machines that are not in use to avoid high monthly costs. 

1. Access the Google Cloud Platform console
2. Open the Google Cloud project to be used for the CloudM Migrate server/s, using the dropdown icon at the top of the screen.
3. Navigate to Compute Engine > VM instances

4. Select CREATE INSTANCE

5. Enter a name for the VM instance.
• Keep the name to less than 15 characters to avoid connection issues in multi-server environments. See Resource naming conventions.
• Standardize the naming convention for all instances, using numbering (1,2,3) or lettering (a,b,c) to help you differentiate between instances.

7. Select the ‘Machine Type’ dropdown menu and then, select ‘Custom’
   • In larger migrations, where numerous secondary servers will be utilized, you should change the Virtual Memory of the Primary VM from the default Custom, which has a low threshold, to 'System Managed'. This is to improve performance of the Primary Server as it communicates with the REDIS database.
   • Alternatively, use the Custom machine type and set either the primary or secondary core and memory specifications from the article here.
8. Select ‘Change’ in the ‘Boot disk’ section:

9. On the Public Images tab, change the:
• ‘Operating system’ to ‘Windows Server’.
• ‘Version’ to ‘Windows Server 2019 Datacenter’ (or another supported version of your choice).
• ‘Size (GB)’ to ‘300’ (the combined minimum requirements for both CloudM Migrate and Windows Server).
10. Click Select at the bottom of the page
11. If you are planning on configuring a public facing URL, in the ‘Firewall’ section,  check the box to ‘Allow HTTPS traffic’. This allows easy access to download the installation files required, but is optional. Both HTTP and HTTPS need to be enabled for the primary and secondary server for the purpose of building a cluster by cloning.

12. Select Create to create the VM instance.

Connecting to a VM instance

1. Go to the VM instances screen. Under the Connect column, select the small down arrow and then Set Windows password:

2. Leave the Username field as the default, and then select SET:

3. Copy the password and make a note of it for later:

4. Under the Connect column, select the small down arrow and then Download the RDP file:

5. Using the RDP (Remote Desktop Protocol) file, establish a connection and login to the VM instance using the password noted down earlier:
• If it fails to connect, disable your local environment Windows firewall and any other network firewalls.