Disaster Recovery Overview
Disaster recovery is performed when the source site is unreachable and it is necessary to fail over the VMs and the protected groups to the target cluster. The process recovers the VM on the target cluster. The following bullet points describe the commonly used operations for disaster recovery:
Testing VM recovery: Testing VM recovery gives the user the ability to test recovery without breaking replication. It can bring up the user VM workload on the target to verify the contents of the VM.
Recovering virtual machines: Recovering virtual machines means restoring the most recent replication snapshot from the target (recovery) cluster. Once recovery starts, all the scheduled replication is stopped.
Planned migration: Performing a planned migration pauses the replication schedule, replicates the most recent copy, recovers on the target, switches the ownership from the source to the target, and resumes replication on the target that is now the new source.
Compatibility Matrix for Disaster Recovery Operations
The compatibility matrix in Table 6-2 lists the disaster recovery operations that are supported when an HX Data Platform version 3.5(x) cluster is paired with an HX Data Platform version 3.5(x) or 3.0(1x) cluster.
Table 6-2 Disaster Recovery Operation Compatibility Matrix
Feature |
3.5(x) Paired with 3.5(x) |
3.5(x) Paired with 3.0(1x) |
Replication |
✓ |
✓ |
Cluster pairing |
✓ |
✓ |
Datastore mapping |
✓ |
✓ |
Protection |
✓ |
✓ |
Planned migration (single-click using HX Connect) |
✓ |
✗ |
Planned migration (Multistep stcli or WebCLI and HX Connect for recovery) |
✓ |
✓ |
Test recovery using HX Connect |
✓ |
✓ |
Recover using HX Connect |
✓ |
✓ |
Re-protect using HX Connect |
✓ |
✗ |
Re-protect (multistep stcli or WebCLI) |
✓ |
✓ |
Testing Virtual Machine Recovery
Testing recovery does not disrupt the running clusters. The intent is to verify that, in the event of an actual disaster, the VMs will be recoverable. Using the HX Connect user interface to test VM recovery, you can run a maximum of 10 tasks in a sequence without waiting for the previously submitted task to complete. The following section describes the steps needed to properly test virtual machine recovery:
Step 1. Log in to HX Connect on the target cluster as administrator.
Step 2. Navigate to the Replication page and select the Remote VMs tab. Figure 6-20 shows the Replication page for remote VMs for test recovery.
Figure 6-20 Replication Test Recovery
Step 3. To test the recovery process, select a protected VM and click the Test Recovery button. Figure 6-21 shows the test recovery parameters, which are as follows:
){kind=link}
)
Figure 6-21 Test Recovery Parameters
Step 4. When you have finished setting the parameters in the Test Recovery Parameters dialog, click Recover VM.
Step 5. When the job completes, verify that the VM has been recovered to the HxRecoveryTest folder or the folder you designated.
Step 6. Power on the recovered VMs via vSphere Web Client or HTML5 vSphere Client to test their functionality.
Step 7. Repeat steps 3 through 6 for each VM you want to test.
Step 8. When the testing is completed, delete the test recovery VMs.
Recovering Virtual Machines
In the case of a site outage or the failure of a cluster, VMs can be recovered to their state as of the last successfully transmitted snapshot, running on the secondary or target cluster as part of a disaster recovery operation. This recovery operation assumes that the primary, or source, site and cluster is either offline or isolated in such a way that it can no longer communicate with the secondary, or target, site, and it cannot be managed. A recovery operation stops all replication between the two clusters; replication can be reestablished at a later time, after the faults or outages have been repaired. The following section describes the steps needed to properly recover virtual machines:
Step 1. Log in to HX Connect as administrator.
Step 2. Navigate to the Replication page and select the Remote VMs tab (see Figure 6-22).
Figure 6-22 Virtual Machine Recovery
Step 3. Select a protected VM and click the Recover button.
Step 4. To recover the VM and build a new VM on the local cluster, set the following options in the Recover VM on This Cluster dialog (see Figure 6-23):
){kind=link}
)
Figure 6-23 Virtual Machine Recovery Parameters
Step 5. When you have finished setting the parameters in the Recover VM on This Cluster dialog, click Recover VM.
Step 6. Wait for the recovery to complete. View the recovered VM in the target vCenter.
Plan Migration
Performing planned migration pauses the replication schedule, replicates the most recent copy, recovers on the target, switches the ownership from the source to the target, and resumes replication on the target that is now the new source. The following section describes the steps needed to properly perform a planned migration:
Step 1. Log in to HX Connect as administrator.
Step 2. Navigate to the Replication page and select the Remote VMs tab.
Step 3. Select a protected VM and click Migrate.
Step 4. To migrate a VM, set the following options:
Resource Pool: Specify the location for the new VM to be stored.
Folders: Specify the location for the new VM to be stored.
Power On/Off: Select the status of the recovery VM.
Map Networks: Create a map between the source and the target cluster networks. These are the options:
Source Network: Indicates the network name at the source side on which the VM is connected.
Target Network: Select the target network where the VM has to be connected.
Step 5. Monitor the progress on the Activity page.