Why is a snapshot not a Kubernetes backup?

A snapshot captures a point in time of a volume, not automatically the state of the control plane, the definition of the cluster or a consistent database. A backup protects several layers separately. Details are in the guide.

What does Velero back up, and what not?

Velero backs up Kubernetes resources and persistent volumes, can schedule backups, restore them and move resources to other clusters (Velero). Application consistency for databases additionally needs hooks or native dumps.

How often should the restore be tested?

The guide suggests a full restore of the critical layers at least once per quarter. In operations the interval is agreed.

Does etcd have to be backed up separately?

In self-operated clusters, yes: etcd holds the entire state of the cluster (etcd documentation). With the managed Kubernetes services of cloud providers, the provider handles this layer.

Kubernetes backup and restore with restore tests

In short: backups of etcd, resources, volumes and databases are monitored, and the restore is tested at the agreed interval. As operations, as a single engagement or as a check beforehand.

etcd, resources, volumes, databases
Restore test at the agreed interval
Backup jobs monitored
24/7 monitoring in operations

Request backup View scope

Reviews

IT Beratung Service Consulting Soforthilfe

Companies worldwide trust WZ-IT

Read client reviews

Starting point

When a backup is not enough

Typical situations in which a cluster cannot be restored reliably.

The restore was never tested
Backups exist, but nobody has played through a full recovery.
Snapshot instead of backup
A snapshot captures a point in time of a volume, not the state of the control plane and not a consistent database.
Replication as backup
Synchronous replication protects against the failure of a node. Deletions and faulty deployments are replicated as well.
etcd is not backed up
In self-operated clusters, backing up etcd is your own responsibility.

Services

What backup and restore covers

In operations, backup with restore tests is part of the scope; individually it is commissioned by effort.

Back up

Three layers
etcd, cluster resources and persistent volumes, stored separately from the cluster. Application-consistent database backups are agreed per application. Tools such as Velero cover resources and volumes.
Monitor backup jobs
Jobs are monitored and failures are handled.

Restore

Restore test
The restore is tested at the agreed interval. A restore that was never tested counts among the critical findings when taking over existing clusters.
Back up and restore etcd
In support, etcd backup and restore can be commissioned individually (etcd documentation).

Single engagements are contract work without a guaranteed response time. Server-level backups outside Kubernetes are described under managed backup.

Process

Four steps to a proven restore

Sequence as a single engagement. In operations, the restore test follows the agreed interval.

01
Record the current state
Which layers are backed up, with what, where to and how often.
02
Close gaps
Add missing layers, store the target separately from the cluster, back up databases in an application-consistent way.
03
Test the restore
Play through a recovery of the critical layers and record the result.
04
Document
Written result with findings and open items.

Assessment

In-house or supported backup

Anyone who practises the restore regularly needs no support.

In-house

In-house operation is enough when

all four layers are backed up and the target is separate from the cluster
a full restore of the critical layers is practised regularly
databases are backed up in an application-consistent way

Support

Support makes sense when

a restore has never been tested
a layer is missing, such as etcd or the database
nobody is responsible for the backup jobs and their failures

Entry point

Backup and restore tests in ongoing operations

Backups with restore tests are part of Managed Kubernetes, together with control plane and add-ons, upgrades, CVE assessment and 24/7 monitoring. Monitoring runs 24/7, the response to P1 incidents follows the service level.

Managed Kubernetesfrom €590excl. VAT per cluster and month, including 3 worker nodes, cancellable monthly

View Managed Kubernetes Request backup

All prices are net and exclude statutory VAT. The offers are addressed to businesses. As of October 2026.

Alternatively

Check beforehand

Cluster audit

Backup and restore path in the configuration review, not a penetration test, from €1,490 excl. VAT one-time.

View audit →Single engagement

Kubernetes Support

Back up and restore etcd, check volumes and storage, €120 excl. VAT per hour.

View support →

Frequently asked questions

Answers to the most important questions

Knowledge

Guides and sources on backup

Four layers, tools and adjacent services.

Guides

Primary sources

Related services

Reviews & projects

Operations that visibly relieve customers

Customer feedback on monitoring, updates, maintenance, support and stable production systems.

Networking & Provisioning

WZ-IT moved our studio infrastructure from decentralised individual devices to a central platform: every site is securely connected via VPN, new devices are onboarded automatically and an entire site is provisioned from a template, without manual steps on location. What impressed me most is the breadth and depth of their knowledge: Timo and Robin are not a typical IT provider who sets up a server and leaves. The two of them think their way into highly complex infrastructure and software topics, work through every requirement we put in front of them, and build networking, provisioning and operations so that everything fits together in the end. WZ-IT is an excellent partner for complex software, network and architecture projects.

Steve KirchnerManaging Director, nextGYM GmbH

View project

International

Built in Germany's Ruhr Valley. Running worldwide.

WZ-IT designs, develops and operates infrastructure and software for clients in Germany and internationally. We deliver projects remotely and continue supporting them in ongoing operations after go-live.