This episode puts observability at the center of backup operations: Veeam ONE (real-time monitoring, alarms, AI anomaly detection, and reporter), job status from the VBR console, compliance reports (backup success & retention), and alerting via email and SNMP. You will know the state of your backups before users report it.

In episode 19 your backup architecture expanded to the cloud: object storage, Cloud Connect, and DR to cloud. The wider the deployment, the more important the answer to one question: is everything working? In episode 20 we build observability — Veeam ONE (monitoring, alarms, AI anomaly detection, and reporting), job status directly from the VBR console, compliance reports, and alerting via email and SNMP.
The goal is simple: a backup state that is measurable and observed — not "hopefully running", but "proven running".
Veeam ONE is the monitoring, reporting, and capacity planning platform for the Veeam Data Platform. It collects data from VBR, agents, hypervisors (vSphere/Hyper-V/Proxmox), storage, and cloud services — then presents it as a single source of truth. Its two main sides:
The Monitor presents real-time dashboards: backup job status, proxy and repository health, storage performance, and hypervisor connectivity. Alarms appear with a severity (Error/Warning/Info) and can trigger automatic notifications:
Get-VEAlarm | Where-Object { $_.AlarmState -eq "Triggered" } |
Select-Object Name,Object,State,TriggerTimeSince v13 (episode 17), Veeam ONE has AI-based anomaly detection: "normal" backup patterns are learned historically, then unusual shifts — backup size shrinking drastically, duration spiking, or suspicious encryption patterns — are flagged as anomalies. This is an early detection layer that complements CyberDCR (episode 13).
Note
Alarms are only input for human decisions. An error on an archive backup job can have a different context from the same error on a production job. Adjust thresholds to your environment's baseline, not raw defaults — avoiding alarm fatigue that makes teams ignore real alarms.
Before opening Veeam ONE, the VBR console itself already gives a quick picture:
The PowerShell version for checking the last session results of all jobs:
Get-VBRJob | ForEach-Object {
$s = Get-VBRJobSession -Job $_ | Select-Object -First 1
[PSCustomObject]@{ Job = $_.Name; Status = $s.Result; Date = $s.EndTime }
} | Format-Table -AutoSizeThe backup job success report answers the most frequently asked audit question: what percentage of jobs succeeded this week, which failed, and why. Veeam ONE Reporter provides templates that can be scheduled (daily/weekly/monthly) and exported to PDF/CSV.
The retention/3-2-1 report maps how long restore points are kept per job, whether GFS (episode 7) is running, and whether off-site copies (episode 19) are in place. For DR, the failover plan status report verifies that the failover plan (episode 9) is still valid.
$report = Get-VEReport -Name "Backup Compliance"
Start-VEReport -Report $report -Format PDFVeeam ONE Reporter also projects growth in storage and VM counts — valuable input for planning a new SOBR (episode 18) before disks fill up, not after.
Email is the most basic notification channel. VBR stores global SMTP settings used for job and infrastructure notifications — enable them at General Options → E-mail Notifications and verify with a test email:
Get-VBREmailNotificationOptions | Format-ListWarning
An alert without a response is not monitoring. Every alarm must have an owner and a runbook: who receives it, the response SLA, and triage steps (episode 16). Better 10 important alarms that are all processed than 100 alarms that are all ignored.
After this episode, make sure:
Key takeaways:
In the next episode, episode 21, we cover roadmap & community — Veeam's direction in the Broadcom era (a data protection platform: B&R + M365 + Kasten + AI), the certification path (VCP → VMCE → VMCA), and the official learning ecosystem: Veeam Community, VeeamHub, and Veeam Technical Academy. You will know where the product and your skills are growing!