6.5 Full Failover

A Full Failover is the scenario where a tenant only relies on the copy of his infrastructure stored at the service provider to run his virtual machines, as his local infrastructure is not up and running. Two differences with the Partial Failover can be listed here, from the point of view of Veeam Cloud Connect:

• only the NEA at the service provider side is involved, as the full failover assumes there is no infrastructure component left at the tenant side;
• the Full Failover has to be started using a Cloud Failover Plan. Any failover of a single VM starts a Partial Failover.

In a Full Failover situation, the Network Extension Appliance doesn’t act anymore as a VPN extension, but its two roles are to become the new default gateway for every network created in the Hardware Plan, and to publish services running in failed over VMs to internet, like a firewall.

There are three different ways to start a Full Failover:

• the tenant selects the Cloud Failover Plan from his Veeam installation and starts it. This is an unlikely situation as, again, a full failover is usually required because the tenant has lost his infrastracture;
• the service provider, upon a request by the tenant selects the corresponding Cloud Failover Plan and starts it on behalf of the tenant. This is a more likely scenario, especially for those tenants not confident in using the self-service capabilities of Veeam Cloud Connect;

6.39: A service provider can start a Cloud Failover Plan on behalf of a tenant

• the tenant can connect to the Veeam Cloud Connect Portal using the credentials received from the service provider upon subscribing to the service, and he can start here his Failover Plan without involving the service provider at all. We again suggest service providers to install and operate Veeam Service Provider Console, since its vast set of features makes it a way better choice than the simple Cloud Connect Portal. However, for some quick installations, the portal itself may still be useful.

Start the Failover Plan using the Cloud Connect Portal

The tenant starts the procedure by connecting to the Veeam Cloud Portal and logging in with the received credentials:

6.40: The login screen of the Veeam Cloud Connect Portal

After the tenant has successfully logged into the portal, he can see his failover plan(s), safely stored at the service provider:

6.41: The list of available Cloud Failover Plans

The tenant can now select one of the available cloud failover plans, and using the Start button on the top left, start it. The portal asks to the tenant the point in time the tenant wants to use as a restore point:

6.42: Select the point in time to be used as a restore point

Immediately after the selection, the failover plan is executed. After some time, depending on the amount and size of the different VM’s that need to be started, the plan is successfully completed:

6.43: Cloud Failover Plan is executed successfully via Veeam Cloud Connect Portal

Tenant can open the details of the two failed over VM’s, and check the IP publishing rules.

Here, we can see the two different publishing rules that were previously configured during the creation of the Failover Plan:

203.0.113.151:80 -> 172.27.217.27:80 (NGINX webserver running on DummyLinux VM)

203.0.113.151:3389 -> 172.27.217.26:3389 (Remote Desktop server running on DummyWin VM)

The Failover Plan has been correctly executed, and all the needed publishing rules have been applied successfully. The tenant can then reach the published services over the listed IP address and ports.