Knowledge Base

Articles list
Scroll to top

Backup for the VPC

Virtual machines (henceforth VM) can be backed up with the help of Backup Agent, which is an additional service to the Virtual Private Cloud. The service allows you to create backup copies and restore your VM.

Technical specifications

The backup repository will be physically located in St. Petersburg at the Dubrovka-3 data center.

Backup is executed at operating system level with the help of Veeam® Agent for Windows and Veeam® Agent for Linux software. This software must be installed on each VM.

Backup files are replicated in 3 separate copies.

To improve accessibility, all elements of the Veeam® infrastructure are duplicated or deployed in fault-tolerant virtualization environments. The channel speed to Veeam® Backup&Rеplication™ servers is unlimited. The network is built on 10GbE interfaces.

Payment and billing

The Backup Agent service is paid for from your main balance. You can read more about how to top up your balance here.

NB: Payment for the Virtual Private Cloud service is taken from your VPC balance.

You can see which services have already been paid for in two ways: 

Billing

The cost of services is calculated from the number of virtual machines to be backed up and the rented disk space in the repository. There is no charge for traffic.

Prices for resources are listed on the services page.

Starting work

You can order the service on the Virtual Private Cloud project page on the Licenses tab. The service is only available in those locations with a Windows Server rental license.

Installing software

On each backup node you need to download and install backup agents (a management agent and a backup agent). The process is controlled via a self-service portal.

Configuring backup jobs

There are 3 possible methods for configuring backup jobs:

  • Using backup policies;
    In this case a backup policy is created via the VAC console. A template for the backup job can be centrally assigned to selected backup agents.
    The use of a backup policy for a backup agent automatically creates a backup job on it with corresponding settings. This is more convenient than assigning backup job for each backup agent manually. You can create several policies with different settings and attach them to the necessary backup agents.

  • Assigning jobs individually via VAC;
    Instead of creating or editing a policy and subsequently attaching it to backup agents, it is possible to individually configure a backup job via VAC, having selected the necessary backup agent. After individual configuration, the “Backup policy” column for the target backup agent will have the status “Custom.”
    This may be convenient when a policy has been used for a set of agents, after which changes have needed to be made to the backup settings of different agents, but the majority of backup agents remain under the original policy.

  • Local configuration from the backup agent application on the node.
    If Agent UI Mode is in read-only mode, then changes cannot be made locally to the backup agent settings in the application. This also concerns backup job settings.
    This way you can limit the ability of local node administrators to change backup jobs, by making changes possible only from the VAC console, which is accessible just to backup administrators.
    If Agent UI Mode is in full admin access mode, then the settings assigned in VAC can be changed locally. For this you need to launch the Configure Backup utility on the node: via Run — Programs — Veeam, or from the backup agent window.

image

Configuring backup policies

To manage backup policies you need to connect to VAC, click on ”Configuration” in the top-right corner, then select the Backup Policies section on the left.

image

To create a new policy click “+New.” Existing policies can be edited, copied and deleted.

If you don’t want to create a policy from scratch, you can use the preset policy Basic_bcp_policy, created at provider level and available to each client. This policy has optimal settings that are suitable in most circumstances.

It is recommended to use it as a template for the creation of your own policies. For this you need to create a copy of this policy with a new name and enter the necessary changes in its settings.

Let’s take a look at the basic settings for a policy during creation. For editing a copy of Basic_bcp_policy they look the same.

  1. Give your policy a name and complete the description if you like.
    image

  2. Choose your work mode: server or workstation, depending on the nodes to which the policy will be applied.
    Server mode is more functional. It maintains data consistency (concurrence) at application level and flexible configuration of the backup schedule.
    image

  3. Choose the backup mode: the whole computer, partition (volume), or files. The first two modes operate on block level, the last on file level.
    Computer backup allows you to restore individual partitions or the whole computer (all volumes). Partition backup allows you to restore that particular partition. Both modes allow you to restore files or folders from the necessary section.
    Backup on file level is significantly slower than block level, so it is not recommended for use to back up a large volume of data.
    For example, a volume (100GB) can be backed up many times faster than most of the files in that volume (40-60GB).
    image

  4. Choose where to store your backup data — the main option is the VCC Selectel cloud repository.
    image

  5. Configure backup retention policy. Specify the number of restore points to store, then go into Advanced Settings.
    image
    The recommended minimum is 7 restore points when using the Forever Forward Incremental method. To use it, the settings of the Advanced Settings Backup window should look like this:
    image
    When using this method it is not recommended to set more than 14 restore points. Since a large volume of backup data means restoring from a long chain of increments, this takes a long time and increases the chance of errors.

    In most cases, 7-14 increments are reasonable; more than 14 is already risky.
    If the volume of backup data is several hundred GB, it is best not to assign more than 7 increments. There is a description of how to do this below.
    If you need to store backup data for more than two weeks (more than 14 restore points), enter the necessary value in the “Restore points to keep on disk” field (14 and up) and assign weekly creation of active full backup, on any convenient day of the week:
    With these settings, weekly chains of full backups and the increments linked to them will be created. Backup policy will keep in the repository for no fewer than 2 and no more than 3 weekly chains:

    • After the creation of a fresh third chain, on the 21st day of the month, a new increment will be created;

    • The first and oldest chain will be deleted, leaving two fresh chains;

    • The next day a new full backup will be created, beginning a new third chain, this repeats cyclically.

  6. If you need to keep even more restore points when using this method, their quantity can only be increased in multiples of 7. For example:

    • 21 points — 3-4 weekly chains;

    • 28 points — 4-5 weekly chains, this option guarantees restoration for any day in the course of a month (28-35 days).

  7. Below are the recommended settings for Advanced Settings Storage.
    image
    If you need to encrypt backups you need to check the box, provide a password and an optional prompt.

  8. Assign an account and quota connection to the cloud repository. Below are the recommended settings.
    image
    If you uncheck the box, you can set the size of the User quota — a fixed quota (subquota), identical for each agent, their total should exceed the total quota of the client repository. This may be useful for certain scenarios, but in general it means a loss of flexibility. Without a subquota each agent takes up exactly as much space as it needs.
    The Use single tenant option… is not recommended for safety reasons, in this case each agent will have an identical login and password for connecting to the cloud. In the first option they are unique, and are created automatically after attaching the policy to the agents and can be changed manually via the VAC.

  9. There is no need to switch on the local cache, in the majority of cases it is not needed.
    image

  10. Configure ОС processing options.
    image
    When backing up any Windows node using Veeam® it is recommended to turn on application-aware processing (check the box).
    The use of this option allows you to create a transaction consistent backup for applications compatible with MS VSS (Volume Shadow Copy Service). Examples of such applications are: Microsoft Active Directory, Microsoft SQL Server, Microsoft SharePoint, Microsoft Exchange, Oracle.
    If you click on “Customize application handing options” a window will pop up with the option to configure settings for these applications.
    image
    You can read about these settings in more detail in the official documentation:

  11. Even if the applications listed above are not installed on Windows nodes, it is still recommended to turn on application-aware processing, since on any modern Windows OS there is a set of other operating VSS writers the Veeam agent knows how to work with. Using this option allows the risk of backup inconsistencies to be reduced even for ordinary file servers.
    If none of the above applications is installed on the node, then the “Customize application handing options” settings need to be left in default mode. Since the applications for which they are used are not present, they will not work in any case. Simply check the “Enable application-aware processing” box and leave everything else as it is.

  12. Set up a backup schedule.
    image
    The best option, which is suitable for the majority of cases, is a daily backup once a day at a specified time. To use this method, check the “Run the job automatically” box and choose a convenient time. If you do not require daily backup, then instead of “Everyday” you can choose “On weekdays” or “On these days” (select days of the week).
    Checking the “Automatic retry” box allows you to re-launch a failed backup job, (there is no point in re-launching more than once). If you uncheck the “Run the job automatically” box, a backup job can only be launched manually.
    Above are described basic recommendations for scheduling server backup. You can read about schedule settings for workstations and servers in more detail here.

  13. Completion.
    At the final stage you can view a summary of the policy settings you have assigned and click on “Finish” to create the policy.

Attaching policies to backup agents

After creating a policy you need to attach it to your chosen agent. To do this:

  1. Connect to the VAC.

  2. In the menu on the left, select Managed Computers.

  3. In the main part of the window, in the Backup Agents tab, select the agents you require by checking the boxes.

  4. In the top-left, click on “Assign” (opposite “Backup Policy”).
    image

In the window that appears, choose the policy you require and click “Assign.”
image

Restoring from a backup

Recovery is performed locally at backup application level on a specific node. To launch restore, transfer the backup agent on the required node into full admin access mode, this is executed via the VAC. In read-only UI mode the restore option is unavailable.

On the desired node, Veaam (via the Run menu) gives you the option of launching:

  • File Level Restore — restore master at file and folder level;

  • Volume Restore — restore master at section level;

  • Create Recovery Media — master for creation of Veeam bootable media.

File Level Restore

The recovery of files and folders is performed via File Level Restore from the target node. After launching the master (Run — Veeam — File Level Restore) you need to choose the required restore point and click on “Next”:
image

After this a window with the directory structure from the backup for your chosen restore point will open. Here you can select files or folders, which can be copied to a chosen location on the node or restored to their original location with the option of keeping or overwriting the original data.
image

Volume Restore

Recovery of partitions (volumes) is executed via Volume Restore. After launching the master (Run — Veeam — Volume Restore) you need to choose the required restore point and click on “Next.” A window will open in which you can choose partitions and configure accommodation (original or new partition of the node — Customize disk mapping) for their recovery.

image

Veeam Recovery Media

Veeam Recovery Media can be useful for the execution of the following tasks:

  • Full recovery of the node (all partitions);

  • Recovery of system partitions;

  • Recovery or migration of the node to a new platform (ВМ).

To create the Veeam Recovery Media you need to launch the corresponding master (Run — Veeam — Create Recovery Media):

image

After this you need to select the type of media from those available on the system (flash drive, optical disk or ISO file), assign additional settings and the media will be created.

When creating Veeam Recovery ISO you need to include drivers for network and SCSI controllers from VirtIO framework:

  • network — NetKVMw8amd64;

  • SCSI controller — vioscsi2k8R2amd64.

Booting and recovery

To execute recovery with the help of Veeam Recovery ISO follow these instructions:

  1. Before you begin, disconnect all disks from the VM.

  2. Create a new disk, for which Veeam Recovery ISO will be the source.

  3. Connect the new disk to the VM.

  4. After several seconds, connect the remaining disks to the VM one by one in the correct order.

  5. Launch the VM. Booting will be done from the Veeam Recovery ISO disk.

  6. Execute data recovery.

  7. Disconnect the Veeam Recovery ISO disk.