One command to install
A Docker container, external to the nodes: one command and under two minutes from the first line to the first screen. If it does not convince you, stop the container and nothing is left on the nodes.
cv4pve-admin
Proxmox VE manages the hypervisor
cv4pve-admin manages the infrastructure
Written by sysadmins for sysadmins. Snapshots with retention rules, backup and replication analytics, diagnostics, permissions and an audit trail. External to the nodes, which stay as they are.
- Free forever
- Open source AGPL-3.0
- Nothing installed on the nodes
One command and it is running
It ships as a Docker container, external to the nodes. Under two minutes from the first line to the first screen.
curl -fsSL https://raw.githubusercontent.com/Corsinvest/cv4pve-admin/main/install.sh | bashThe problem it solves
Proxmox VE does its own job very well: the hypervisor, the nodes, the machines. What it does not do is the job of the people who run those clusters every day. A backup that has not run in three weeks does not stand out, a stalled replication does not either, and on Monday morning nobody is certain whether Saturday’s ran. Failures turn up when somebody rings. And when something stops working the question is always the same: what changed since yesterday?
What it does
The state over time, not only now
Carry on running the machines from Proxmox VE. What you see here is the trend: every cluster together, what changed, what does not add up.
Consoles without exposing the nodes
The VNC proxy gives access to machine consoles without opening Proxmox VE’s own ports: users connect to cv4pve-admin and nothing else.
Who can do what
Role-based permissions, two-factor authentication, application tokens and a record of who did what. Not everybody needs to be able to stop a machine.
Snapshots and node backups
Automatic snapshots with retention rules, and node configuration backup pushed to Git: the part that machine backups do not save.
Backup and replication analytics
Which backups ran, which did not, which have not started in weeks. The same for replication, for storage and for virtual machine performance.
It updates itself
When a new version is out it says so, and updates with one click. Nothing to download, nothing to redo by hand.
Diagnostics
Automated health checks across the cluster, catching bad configuration before it becomes an outage.
Reports
Nodes, machines and storage in a document you can export: for when an audit arrives, and for deciding what to buy next.
The modules
Every feature is a module of its own: you switch on what you need and the rest is not there. These are the ones available today.
Fifteen modules, these are the ones available today.
- ResourcesLive cluster, node and machine status
- DiagnosticHealth checks, issues found and compliance checks
- System ReportThe same in a document: clusters, nodes, machines and storage
- PortalMulti-tenant portal: each sees their own and nothing else
- Backup AnalyticsWhich backups ran, and how they trend
- Replication AnalyticsReplication status and synchronisation
- VM PerformanceMachine performance, IOPS included
- Metrics ExporterMetrics out to your monitoring systems
- Update ManagerSystem updates, scanned and applied
- AutoSnapScheduled snapshots with retention rules
- Node ProtectNode configuration backup, with versions and restore
- UPS MonitorUninterruptible supplies, with automatic shutdown
- WorkflowAutomations drawn as boxes, started by an event or a schedule
- BotsCluster management from Telegram
- AI ServerAn MCP server: query the infrastructure in words as you would a database, and act on it
Finding out before somebody rings you.
- ResourcesLive cluster, node and machine status
- DiagnosticHealth checks, issues found and compliance checks
- System ReportThe same in a document: clusters, nodes, machines and storage
- PortalMulti-tenant portal: each sees their own and nothing else
- Backup AnalyticsWhich backups ran, and how they trend
- Replication AnalyticsReplication status and synchronisation
- VM PerformanceMachine performance, IOPS included
- Metrics ExporterMetrics out to your monitoring systems
- Update ManagerSystem updates, scanned and applied
What saves you when something goes wrong.
- AutoSnapScheduled snapshots with retention rules
- Node ProtectNode configuration backup, with versions and restore
- UPS MonitorUninterruptible supplies, with automatic shutdown
What runs on its own, once you have said how.
- WorkflowAutomations drawn as boxes, started by an event or a schedule
- BotsCluster management from Telegram
- AI ServerAn MCP server: query the infrastructure in words as you would a database, and act on it
Editions
Community
Open sourceFree forever. AGPL-3.0: it covers the core features, does not expire and does not cap the number of clusters. Support is the community’s, on GitHub.
Enterprise
With supportEverything in Community, plus the advanced modules and notifications to over a hundred and forty services. Support comes from us and our partners, with agreed response times.
What we do
Installation
We put it in place on your infrastructure, connected to the clusters you have, with permissions for the people who will use it.
Support
Configuration and production problems, with agreed response times. Whoever answers wrote the program, rather than reading a manual.
Development on request
Modules and integrations you need that are not there. If they make sense for everyone they end up in the public version.
Training
Half a day or a full one, for people who will use it daily and have no time to discover it a piece at a time.
What people say
Think of it as vCenter reimagined for the Proxmox ecosystem: except lighter, faster, and built with modern .NET technologies.
This isn’t a side project. It’s production infrastructure tooling built by people who’ve lived through 3 AM cluster emergencies.
Shall we put it in place for you?
We start from what you have today: how many clusters, how many people, what gets past you most often.
Coming from VMware?Leaving vSphere means losing vCenter along with everything else. We do the migration, machine by machine, and this is what you are left running.There is a course tooOn the whole suite: cv4pve-admin on one side, PowerShell and the command line on the other. We work on your own clusters, not on a lab.
