Learn Veeam - Backup Jobs: VM & Physical

#Introduction

In episode 4, your hypervisor is connected. Now comes the moment you've been waiting for: creating your first backup job. In episode 5 we assemble a VM job from scratch — choose the workload, repository, schedule, GFS retention, and guest processing — then discuss Veeam Agent for physical servers.

A job is Veeam's unit of work. Understanding each step of the job wizard means understanding Veeam's main operational language — from now on, almost everything (restore, copy, archive) will revolve around jobs.

#Backup Jobs: VMs & Containers

#Creating a New Job

From the console: Home → Backup Job → Virtual Machine. The wizard guides you through several steps:

1. Virtual Machines — select VMs or containers (folder/datastore) from the inventory.
2. Storage — select the destination repository (episode 6) and optimization (Local target / LAN / WAN).
3. Schedule — when the job runs (daily, weekly, hourly).
4. Retention policy — how many restore points are kept, including GFS (episode 7).
5. Guest Processing — application-aware processing (discussed below).
6. Summary — name the job and finish.

From PowerShell, jobs can be created with cmdlets. Here is a complete example:

$vm = Get-VBRViObject -Name "WebServer-01"
$repo = Get-VBRBackupRepository -Name "LocalRepo"
$job = Add-VBRJob -Name "Backup WebServer-01" -BackupType Vm `
    -Entity $vm -Repository $repo `
    -ScheduleOptions (Get-VBRJobScheduleOptions -Type Daily -DailyKind Days -Days Monday)
Start-VBRJob -Job $job

Start-VBRJob triggers the job immediately — use it for a test run after configuration.

#Storage, Schedule, and Retention

Three settings that determine data cost and security:

• Storage optimization: Local target (dedup at the repo), LAN (balanced), or WAN (higher compression for limited bandwidth).
• Schedule: when the job runs; for strict RPO you can run it hourly.
• Retention: the number of restore points; for example Keep 14 restore points for a 2-week window. GFS (Grandfather-Father-Son) for daily/weekly/monthly/yearly — we cover it in detail in episode 7.

#Synthetic Full and Reverse Incremental

Veeam's default is forever incremental + synthetic full — each day only the changes are stored, and Veeam assembles a synthetic full from the existing blocks. This saves storage while speeding up restore (no need to chain incremental files together).

#Guest Processing: Application-Aware Processing

#Why a VM Snapshot Alone Is Not Enough

A hypervisor-level VM snapshot produces a crash-consistent copy — consistent at the disk level, but it does not guarantee databases (MS SQL, Exchange, AD) are in a consistent state. Imagine a backup capturing a database in the middle of an uncommitted transaction: the disk files are intact, but the database is corrupt when restored. The solution is application-aware processing.

#How Application-Aware Processing Works

By enabling Guest Processing → Application-aware processing, Veeam installs VSS (Volume Shadow Copy) in the Windows guest:

1. Veeam requests a hypervisor snapshot.
2. Inside the guest, VSS is asked to quiesce — applications (SQL/Exchange/AD) complete transactions and bring files to a consistent state.
3. The snapshot is taken while the applications are consistent.
4. The VM is released and VSS finishes.

The result is an application-consistent backup, safe to restore to a specific point in time:

Hypervisor snapshot ──▶ VSS quiesce in guest ──▶ consistent snapshot ──▶ backup

#Guest Credentials and Reboot

For guest processing to run, Veeam needs guest credentials (a Windows admin account) stored in the job. If the guest does not use VSS (Linux or guest processing disabled), the backup still runs but is only crash-consistent.

#Physical Servers: Veeam Agent

#Veeam Agent for Windows / Linux

VMs are not the only workload. For physical servers and endpoints, Veeam provides Veeam Agent for Microsoft Windows and Veeam Agent for Linux — agents installed in the OS, centrally managed from the VBR console (in Managed by Veeam Backup & Replication mode). Each agent counts as one workload in the license.

#Protecting Agents from the Console

Add Backup Infrastructure → Managed Servers → Add Server → Microsoft Windows / Linux, then install the agent (push). Create an agent backup job: choose the protection level Entire computer, Volume, or Files & Folders, point it to a VBR repository, and set a schedule:

Get-VBRComputerBackupJob | Format-Table Name,PlatformType,State -AutoSize

The agent provides RPOs down to every 15 minutes and supports granular restore — useful for protecting physical servers (e.g. standalone databases) that cannot be snapshot at the hypervisor level.

#Verifying Your First Job

After the job finishes running:

Get-VBRJob | Select-Object Name,JobType,LastResult | Format-Table -AutoSize
Get-VBRBackupSession | Where-Object {$_.JobName -eq "Backup WebServer-01"} | Format-Table Name,Result

Make sure the status is Success and the first restore point is available in the repository.

#Closing

Key takeaways:

• A VM job consists of: workload → storage → schedule → retention → guest processing.
• Add-VBRJob and Start-VBRJob are the gateways to job automation.
• Application-aware processing (VSS) is mandatory for databases: the backup becomes application-consistent.
• Veeam Agent (Windows/Linux) protects physical servers and endpoints, managed from the console.
• Verify jobs with Get-VBRJob and Get-VBRBackupSession.

In the next episode, episode 6, we will cover repositories & storage — Windows/Linux repositories, SMB/NFS, S3 object storage, SOBR with capacity tier, as well as dedup & compression (None/Dedupe-friendly/Local target) and hardware deduplication appliances like HPE StoreOnce and ExaGrid. Your backup's home will get more sophisticated!