Automated VM Failover – Transfer Workloads to VM Replicas in a Few Clicks

Often, a disaster can cause the failure of both your physical infrastructure and your virtual environment. These situations are significantly mitigated if you have backed up and replicated your VMs to the different locations. Running VM replicas at another site is the usual solution considered when a disaster affects the primary site. Manual failover of virtual machines (VMs) can take a long time because the different sites have different networks. Manually reconfiguring the network settings of each VM is a time-consuming process if you have a large virtual infrastructure – and if you are in DR mode, you likely have other pressing matters vying for your time and attention. For this reason, NAKIVO Backup & Replication v.7.4 features new Automated VM Failover functionality that can help you make the failover process faster and easier.

How Automated VM Failover Works

First, you must have VM replicas, which can be created with NAKIVO Backup & Replication. A VM replica is an identical copy of a source VM that can be created with a replication job and stored in the target location. Multiple replication jobs can be created in NAKIVO Backup & Replication, and one job can replicate multiple VMs. A VM replica is a regular VM with snapshots that represent recovery points. Having standby replicas at a secondary site allows you to create an automated VM failover job to run them at the secondary site when a failure at the primary site makes the source VMs unreachable. One failover job can be linked with multiple replication jobs, simplifying organization of the failover process.

Network Mapping allows you to select a network for reconnection of VMs. VMs residing on ESXi or Hyper-V hosts are connected by virtual switches to specific networks. These networks would likely differ for hosts located at the primary site vs. secondary site.

Suppose that a VM group with a source VM is connected to the network named Network1 at the primary site, but at the secondary site, the name of the network is Network2. Because replicas are identical copies of a source VM, the network to which the replica would attempt to connect is Network1 – but there is no such network at the secondary site. The replica must be connected to Network2. Network Mapping helps you easily change the connection settings for the VM replicas in this situation. Moreover, Network Mapping lets you create multiple rules for different networks that can be applied to entire groups of VMs. Thus, source networks are mapped to target networks. You don’t need to worry about altering the network configuration manually for each VM, because Network Mapping handles the changes automatically. The product checks the network settings for the VM replica during the Failover job, and if the network of the VM matches a source network specified in one of the rules, then the target network is remapped for that VM.

Example

There are 2 ESXi hosts. The first ESXi host is located at the primary site, and has Virtual Switch 1 configured. The second ESXi host is at the secondary (DR) site, with Virtual Switch 2 configured. The VMs residing on ESXi1 are connected to Network1 by their virtual switch. The VMs residing on ESXi2 are connected to Network2 accordingly. You need to map Network2 for VMs replicated from ESXi host 1 for running them with a network connection on ESXi host 2.

Note that networks can have names like 10.10.10.0, 10.10.20.0, 192.168.1.0, etc. Don’t confuse network names with network addresses (such as 10.0.0.0/8, 10.10.20.0/24, 192.168.1.0/24, etc.). Network names are used for creating network mapping rules, while network addresses are used for creating Re-IP rules in NAKIVO Backup & Replication.

Two virtual switches with different networks on different ESXi hosts

# Source network name Target network name
1. Network1 Network2
2. 10.10.10.0 10.10.20.0

Re-IP allows you to change the IP network settings for VM replicas that are used for a failover. If the IP configuration for virtual network adapters used by virtual machine is set manually (not obtained automatically via DHCP), Re-IP can change the IP address, network mask, default gateway, and DNS servers. Re-IP rules can be created similarly to the way Network Mapping rules can.

Note: In order to use Re-IP, VMware Tools must be installed on VMware VMs and Hyper-V Integration Services must be installed on Hyper-V VMs. This is necessary in order for Re-IP to access the VM IP network settings.

Example

Network1 is 192.168.11.0/24 and Network2 is 192.168.17.0/24.

The asterisk (*) represents any number from 1 to 254. If you create a rule for changing IP addresses from 192.168.11.* to 192.168.17.*, the octet covered by the * symbol would keep the original value. Hence, source IP addresses 192.168.11.10, 192.168.11.200, and 192.168.11.253 are changed to 192.168.17.10, 192.168.17.200, and 192.168.17.253, respectively, in the VM replicas. The values used to create the Re-IP rule in this walkthrough are displayed in the table below:

Parameter Source network (Network1) Target network (Network2)
IP address 192.168.11.* 192.168.17.*
Net mask 255.255.255.0 255.255.255.0
Default gateway 192.168.11.2 192.168.17.2
Primary DNS server 192.168.11.2 192.168.17.2
Secondary DNS server 192.168.11.3 192.168.17.254

The last octets in the IP addresses of the secondary DNS servers are different. You can create a second Re-IP rule for the VM replica that has the secondary DNS server role. That rule should look as follows:

Parameter Source network (Network1) Target network (Network2)
IP address 192.168.11.3 192.168.17.254
Net mask 255.255.255.0 255.255.255.0
Default gateway 192.168.11.2 192.168.17.2
Primary DNS server 192.168.11.2 192.168.17.2

When Can Automated VM Failover Be Used?

  • Recovery after a disaster. You can use Automated VM Failover to restore your virtual infrastructure at a secondary site using the VM replicas. The feature is especially useful for environments with large numbers of VMs. Note that any scheduled replications of VMs from the primary to the secondary site should be turned off during failover.
  • Planned migration to another site. Sometimes, changes in a company’s infrastructure can necessitate the migration of virtual machines from one site to another. If you know that the VMs at the primary site could go offline for reasons such as natural disasters (flood, typhoon, tornado, earthquake etc.) or scheduled blackouts, planned migration can also be considered. Automated VM Failover makes this migration process easier. The source VMs should be powered off to ensure that they do not interfere with the VM replicas.
  • Test migration. Allows you to check if your disaster recovery plan can be implemented smoothly before a disaster occurs.

Example VM Failover Job

Let’s demonstrate how to configure and run a VM failover job in NAKIVO Backup & Replication with a walkthrough. Open your browser and go to the address of your instance of NAKIVO Backup & Replication. Make sure that you have already created VM replicas. On the home page click Recover. If you use VMware, select VM failover to replica under the VMware Full Recovery section. If you use Hyper-V, select Failover to replica in the Microsoft Hyper-V Full Recovery section. In this example, a failover for VMware VMs is considered. The processes do not vary significantly between the two platforms.

Running VM failover to replica

The four-step VMware failover job wizard is launched.

1. Select the destination VM replicas that you want to use for failover. In the left pane of the Source section, tick the checkboxes corresponding to the appropriate VMs. In the right pane, you can drag items up and down to determine processing priority. Click Next when you are ready to proceed.

Selecting the VM replicas for failover2. In the Networks section, you can enable network mapping. Simply check the Enable network mapping box and click Create new mapping.

Enabling network mapping for VM failover

A popup menu is launched. Select a source network and target network from the appropriate drop-down lists.

Creating a new network mapping rule for VM failover

You can add multiple Network Mapping rules. After adding the rules, your screen should resemble the screenshot below. (Note that your network names may differ.) Click Next.

Created network mapping rules for VM Failover jobs

3. At this step you can enable Re-IP and set the Re-IP rules. If you want to do this, check the Enable Re-IP box and click Create new rule. In the popup box, you can set a new rule. Set the source IP address and subnet mask. Then set the target IP address, subnet mask, default gateway, and DNS servers according to your environment. Click Save to save the rule. You can add as many rules as you need.

Creating a new Re-IP rule for VM failover

Next, you should select VMs. Click Select VMs.

Selecting the VMs to manage credentials for Re-IP in VM failover

Select the checkboxes to indicate the VM replicas for which you want to configure Re-IP and click Manage Credentials.

Managing credentials for Re-IP in VM Failover jobs

Click the Add Credentials button.

Adding credentials for Re-IP in VM Failover jobs

Type the username and password of the account from the operating system inside the VM. The user must have sufficient permissions to change the network settings. Click Save.

Inputting the credentials for Re-IP in a VM Failover job

Now, select the credentials you recently added from the drop-down list and click Next.

Finalizing the Re-IP configuration

4. At the final stage of the Wizard, you can select the failover job options. Check the box Power off source VMs. This option powers off the source VM that was used for creating the replica. Click Finish & Run when ready.

Failover job options

Now you are familiar with automated failover and know how to configure this feature in the NAKIVO Backup & Replication web interface. Let’s summarize the advantages of this helpful feature.

Advantages of the Automated VM Failover Feature

  • You can restore VMs quickly after a disaster. The faster recovery time allows you to meet the shortest RTOs in the framework of your disaster recovery plan and business continuity plan. Overall, Automated VM failover makes a recovery process faster and easier.
  • Much less time and energy is required for administration. You don’t need to change the network settings manually on each VM. Network Mapping and Re-IP automate these procedures. The appropriate automation rules must only be created once, when you configure the failover job.
  • Multiple replication jobs can be linked to a single failover job. You don’t need to create separate failover jobs for each replication job, which proves extremely convenient. Any VM replicas created when new VMs are added to a replication job are also discovered automatically by the associated VM failover job and included therein.

Conclusion

This blog post has covered the Automated VM Failover feature included in the latest version of NAKIVO Backup & Replication, v7.4. Automated VM Failover helps you automatically reconfigure the network settings of VM replicas used for a failover to another site in the easiest way. One Failover job can link to multiple replication jobs. Replicas created when new VMs are added to replication jobs are automatically included in the associated failover jobs. As a result, you don’t need to spend time manually configuring your backup product and VM replicas. With NAKIVO Backup & Replication v7.4, which includes Automated VM Failover and other useful features, you can quickly and effortlessly restore your workloads after a disaster. Download the Free Trial to try in your own environment.

Automated VM Failover – Transfer Workloads to VM Replicas in a Few Clicks
4.8 (95%) 4 votes

Share:

LinkedIn Google+