Background
Three of the biggest pain points of a single-pass Bare Metal Restore are
- The time it takes to perform the full image recovery process
- The requirement that the protected system be shut down during this time and
- The loss of any changes or new backups made to the protected system after we begin
What is incremental recovery?
Incremental recovery is the process of updating only the blocks that have changed between the last recovery pass and a more recent backup recovery point.
Since the changed blocks in this case are typically a tiny fraction of the total system data, performing an incremental recovery pass can happen quickly.
Why should I use incremental recovery?
- Performing incremental recovery with the Bare Metal Restore process allows you to eliminate the lengthy downtime required when performing Bare Metal Restore in a single pass.
- Your protected system can be temporarily recovered using appliance virtualization or by using Axcient's Virtual Office, which immediately restores access to the affected system for your client.
- Once the underlying hardware issues are resolved and you are ready for recovery back to the client system, you can start the Bare Metal Restore base full image recovery process without having to shut down your virtual machine instance of the system. Your clients can continue to access the system, and new backups will continue to run, while the potentially lengthy initial full image restore is taking place.
OVERVIEW: What are the basic steps for incremental recovery?
First, restore client access to services:
|
In the event that your client has a protected system which experiences total failure, the quickest route to recovery is to perform a virtualization from the backup recovery point. ==== For clients using an appliance, this means using StartVM for the protected system on the appliance. ====For clients using Direct-to-Cloud, this means starting a Virtual Office and virtualizing the protected system in the cloud.
Note: Ensure that at least one backup is completed after virtualizing the protected system. This first backup will be recorded as a ‘Full’ backup, which means all blocks on the system will be scanned for changes. Only blocks that have actually changed will be written to disk. (i.e. this will NOT generate a duplication of data on the backup server.) |
Next, review this summary of required steps:
Once your protected system is virtualized and your client is back up and running, you may begin the Bare Metal Restore recovery process.
1. Begin by performing the initial full image recovery of the protected system, which may take a considerable amount of time. You client may continue to run their system as a virtual machine throughout this process.
2. Once the initial full image recovery is completed, you may add one or more incremental recovery passes. This will apply any changes to the system from more recent backups than the one that was used for the initial base image.
3. Once the system is fully updated with the running virtual machine, you are ready to fail back from the virtual machine to the original environment. Take one last backup and then shut down the virtual machine.
4. Apply the last backup by performing one more incremental recovery pass.
5. Select 'Finish Recovery Now' in the incremental restore wizard to complete the process.
6. Boot the recovered system and return it to your client environment.
Now, follow the detailed instructions below for performing each of these steps.
Setup the Bare Metal Recovery workflow: Boot and select a method
1. Boot the x360Recover Bare Metal Restore ISO.
2. Once the Linux Live CD environment has booted, launch the Recovery Wizard from the desktop icon.
3. Accept the License Terms and click Next
4. Select an authentication method.
We highly recommend that you select x360Portal User.
Note: Only select Local Appliance/Vault user if you are recovering from a local appliance backup server and have no Internet connection.
Click Next
5. On the next pages, enter your username, password, and MFA code when prompted.
6. Choose your recovery mode, based on where your data is located.
- If you are recovering from a local appliance, choose Recover from Appliance
- If you are recovering from a vault (Direct-to-Cloud)
- If you HAVE a Local Cache repository for the protected system, choose Recover from Vault (with Local Cache)
- If you DO NOT HAVE a Local Cache repository for the protected system, choose Recover from Virtual Disk Images
Let's get started
A. Begin your Bare Metal Recovery
Important:
|
1. Boot your protected system from the Bare Metal Recovery ISO.
- For a complete overview of the Bare Metal Recovery process, see the Bare Metal Restore Guide.
- For help downloading the ISO and creating a bootable USB drive, see: Create a Bootable USB Flash Drive.
2. Once you have booted the Recovery ISO. you have several recovery options from the Recovery Wizard:
- Recover from appliances
- Recover from vaults (with Local Cache)
- Recover from Virtual Disk Export
It’s important to note that, with the Incremental Bare Metal Restore process, you may continue to operate your client’s protected system as a Virtual Machine throughout the recovery process.
Although it may take several days to download and/or recover the initial full image of the system, your client will not be down and will be able to access their system, as it will be available as a virtual machine.
B. Perform incremental recovery passes
Remember, there are multiple ways you can recover:
- from appliances
- from vaults (with Local Cache)
- from Virtual Disk Export
|
From appliances: Look for the checkbox named Start an Incremental Bare Metal Restore Recovery. (You will see this checkbox when selecting the Recovery Point.)
Note: If you unselect this option, you are choosing to perform a single-pass Bare Metal Recovery. This option cannot be changed later. |
 |
|
From a vault with local cache: Look for the checkbox named Start an Incremental Bare Metal Restore Recovery. (You will see this checkbox when selecting the Recovery Point.)
Note: If you unselect this option, you are choosing to perform a single-pass Bare Metal Recovery. This option cannot be changed later. |
 |
|
Using Recover from Virtual Disk: The checkbox for Start an Incremental Bare Metal Restore Recovery is on the Virtual Disk Location selection screen: |
|
Did you choose appliance, vault or virtual disk image?
How to run an incremental BMR recovery from an appliance
1. Select Recover from Appliance and click next
Assuming you previously selected x360Portal User authentication, enter the IP address of the appliance.
Note: If you selected Local User authentication, you will also be prompted to enter the local device username (usually admin) and password on this page
Click Next
2. Select the protected system you wish to recover.
Click Next
3. Select the snapshot to use for the initial full image recovery pass.
Note: To begin an incremental recovery (which will allow for additional incremental data copy passes to be applied), be sure to select Start an Incremental Bare Metal Restore Recovery on this page:
Click Next
4. Wait for the Recovery Wizard to complete staging the recovery on the appliance.
The selected snapshot will be enabled for iSCSI export, and the iSCSI disk volumes will be mounted to the Live CD environment.
Click Next
5. Select the original protected system disk volumes you wish to recover and map them to the local disks to be overwritten.
Click Next
6. Confirm you wish to continue and overwrite data on this device. Once you click next, any existing data on disks selected for recovery will be erased and overwritten.
Click Next
The initial full image recovery data copy pass will begin.
Wait for it to complete.
How to run an incremental BMR recovery from a vault (with local cache)
1. Select Recover from Vault (with Local Cache) and click Next.
2. Select the option which describes where the local cache data is located.
- If the local cache is on a network share, enter the Windows Networking or Linux NFS connection information.
Click Next
3. Browse to select the local cache folder.
Select the protected system to be recovered.
Click Next
4. The FQDN address of the vault should already be populated.
If you previously chose Local User authentication, enter the vault local username and password on this page.
Click Next
5. Select the snapshot to use for the initial full image recovery pass.
Note: To begin an Incremental recovery which will allow for additional incremental data copy passes to be applied, be sure to select Start an Incremental Bare Metal Restore Recovery on this page.
Click Next
6. Wait for the Recovery Wizard to finish staging the recovery point on the vault.
The protected system will be exported via the Disaster Recovery Access Layer (DRAL) remote transport service, and the exported disk volumes will be iSCSI mounted on the Live CD environment.
Click Next
7. Select the original protected system disk volumes you wish to recover and map them to the local disks to be overwritten.
Click Next
8. Confirm you wish to continue and overwrite data on this device. Once you click next, any existing data on disks selected for recovery will be erased and overwritten.
Click Next
The initial full image recovery data copy pass will begin. Wait for it to be completed.
How to run an incremental BMR recovery from a Virtual Disk image
Download all required disk image files
Using this recovery method requires that you have already downloaded your virtual disk images from the vault.
See: Export Full Disk Images for help exporting and downloading virtual disk images.
Important note:
- Only VMDK and VHDX disks are supported for recovery via virtual disk images.
We recommend you choose VHDX Dynamic mode when exporting, as this will provide a thin-provisioned image containing only ‘used’ blocks, which will provide a smaller download.
|
Please note: You MUST make note of the exact recovery point, date and time selected to export and download. You will be asked later during the Recovery Wizard to identify this snapshot.
The entire basis for identifying which blocks have been changed during incremental recovery passes is based on the specified base image. If you select a different recovery point than what you exported, YOUR RECOVERY WILL BE CORRUPT |
Start the Recovery Wizard
Once you have completed the download of all required disk image files, you may start the Recovery Wizard.
1. Select Recover from Virtual Disk Image(s) and click Next
2. Select the option that matches the location of your Virtual Disk images.
- If the virtual disks are on a USB, make sure the USB drive is attached to the system being recovered.
- If the virtual disks are on a network share, specify the Windows Networking or Linux NFS connection properties
- Be sure to select ‘Start an Incremental Bare Metal Restore Recovery’ to allow for additional incremental data passes to be performed
Click Next
3. Provide the address of the system you downloaded the virtual disk image files from (presumably the FQDN of a vault)
If you previously selected Use Local User authentication, you will also need to provide the username and password to the vault on this page.
Click Next
4. Select the protected system to be recovered.
Important: This must match the protected system for which the Virtual Disk Images were downloaded.
Click Next
5. Select the Recovery Point that was used to export the virtual disk images.
|
IMPORTANT: (In case you missed it the first time.) You MUST make note of the exact recovery point, date and time, that you selected to export and download. You must identify this recovery point to the Recovery Wizard to receive a valid recovery. THE SNAPSHOT YOU SELECT MUST BE CORRECT. The entire basis for identifying which blocks have been changed during incremental recovery passes is based on the specified base image. If you select a different recovery point than what you exported, YOUR RECOVERY WILL BE CORRUPT. If you don’t know which recovery point was used to export the Virtual Disk images, you will have to start over and download a new set of virtual disks. |
Click Next
6. For each disk being recovered, select the browse gadget and browse to the virtual disk image file.
- Choose a local disk from the Target Disks dropdown menu.
- Click Add to add it to the list of disks being recovered.
Once all image files have been mapped to local disk volumes, click Next
7. Confirm you wish to continue and overwrite data on this device.
Once you click next, any existing data on disks selected for recovery will be erased and overwritten.
Click Next
The initial full image recovery data copy pass will begin. Wait for it to complete.
Finish up an incremental BMR recovery
Regardless of which recovery method you have chosen for the initial base image recovery, if you have selected the Start an Incremental Bare Metal Restore Recovery option, you will be presented with the Incremental Recovery screen after the full image copy is completed
This screen identifies the current status of the recovery system.
- Base Restore Point is the date/time of the most recent recovery point to be fully applied to the system.
- Available Restore Points list all recovery points at or newer than the Base Restore Point and defaults to the most recent recovery point.
1. If no action is taken, the Recovery Wizard will automatically apply the most recent recovery point if any newer points exist, after a 5 minute timeout.
2. If no newer recovery points exist, the Recovery Wizard will wait for new backups to occur.
Data is refreshed every minute.
3. If you choose to finish the recovery later (or if the Recovery Wizard is otherwise exited or restarted) the current state of the recovery for this protected system will be retained in the cloud.
You will be asked to resume or abort the recovery the next time you run the Recovery Wizard for this protected machine.
4. The next time you run the Recovery Wizard, proceed through the options until you have selected the protected system. (You should select the same recovery path and options as you used in the initial restore, i.e. appliance, local cache, virtual disk mode, etc.)
- Perform next: If a previous unfinished Incremental Recovery job is found for this protected system, you will be prompted to continue or abort it.
- Abort: If you wish to start over, or you don’t remember having a recovery job in progress for this system, select abort to clear the recovery record from the cloud.
5. Once your system is completely up to date with the most recent recovery point, you are ready to finish the recovery process.
- Pick a time when you can briefly interrupt the users of this protected system
- Perform one last backup of the system after users have been logged out to catch any data changes
-
Stop and shut down the running failover system
- Stop the VM on the appliance or shutdown the VM in Virtual Office
- Perform one last incremental recovery pass to catch changes in the last backup performed above
-
Select Finish recovery now and click Finish in the Recovery Wizard
- Complete the recovery process and clear the recovery status records in the cloud
- Reboot the recovered system and remove the Bare Metal Recovery ISO/USB
- Reboot the recovered system normally and restore access to users
- Notify users that they may resume using the recovered system once it has booted in the client environment
Head Start Restore / Virtual Standby Service
The Incremental Bare Metal Recovery process is fully automated once the initial full image pass is completed.
You may leverage this process to provide a ‘Virtual Standby’ or ‘Headstart Restore’ type functionality.
Incremental recovery will continue to apply new recovery points indefinitely, to keep the recovery system up to date with new backups of the running production or failover system
SUPPORT | 720-204-4500 | 800-352-0248
- Contact Axcient Support at https://partner.axcient.com/login or call 800-352-0248
- Have you tried our Support chat for quick questions?
- Free certification courses are available in the Axcient x360Portal under Training
- Subscribe to Axcient Status page for updates and scheduled maintenance
2038