Service 06 · Monitoring and support, day and night
Managed Proxmox: your servers looked after by the team that runs its own cluster.
High availability set up from replication, backups that are checked, patches kept up to date and a person on call for incidents, on your Proxmox servers.
First conversation free.
A Proxmox cluster that runs, and nobody to watch over it
- 01
Nobody at night. Your applications run on Proxmox, but nobody keeps watch at night or at weekends. We take over the watch.
- 02
High availability that was never tested. The box is ticked, the failover has never been seen. An untested cluster is a hypothesis, not a guarantee.
- 03
Backups nobody checks. They run every night, but nobody checks them. We check them automatically, and the monthly report accounts for it.
Who it is not for
Managing a fleet of workstations, office software licences or company email: an IT support provider is better placed.
Proxmox HA: failover is designed from replication
We run our own Proxmox cluster: three dedicated OVHcloud servers, in Strasbourg, Roubaix and Gravelines, which carry our products, our clients' applications and this site. Your Proxmox servers get the same care.
High availability restarts a machine on another server when its own goes down. On local ZFS storage, it starts again from its latest replica: a machine without a replica has no place in high availability, and the replication interval sets how much data can be lost. An LXC container is stopped and restarted elsewhere; a virtual machine migrates live.
Three settings are checked before production, not during an incident: machines returning after maintenance, placement priority, where the highest number wins, and stopping a resource managed by high availability. The details, on the blog: Proxmox HA in production.
Covered by managed services
- Continuous monitoring, one check per minute
- Security patches tracked on every machine
- Nightly backups, checked automatically
- A person on call for critical incidents
- Monthly report: incidents, availability, actions
Three levels, one goal
Essential, Standard or Premium, detailed in the quote and described on the managed services page. You choose the one each application's criticality calls for; the monthly order of magnitude is published on the pricing page.
Our Proxmox cluster, measured and written up
The measurements come from our own cluster, each with its instrument and its date. The case study of our Proxmox infrastructure describes what produces them; the blog, what it has taught us.
- Virtual machines and containers hosted on the cluster
- 37Virtualisation cluster (guest inventory)
- OVHcloud data centres housing our hypervisors
- 3 in FranceOVHcloud public API (dedicated server records)
- Backups completed successfully over the last seven days
- 322 out of 322Dedicated backup server (Proxmox Backup Server)
- Backup integrity checks over the last seven days
- 7 out of 7Dedicated backup server (Proxmox Backup Server)
Measurements taken on 5 September 2026, values recorded by our own instruments.
- Proxmox HA in production31 resources in high availability: replication, automatic return, placement priorities, stopping a resource.
- A disk failure that did not show itselfTwo and a half months of slowdowns without a single “dead” disk. Since then, every disk is read in full once a month.
- The Proxmox firewall pitfallMachines with the firewall “enabled”, actually wide open: it needs two conditions, not one.
- Backing up with Proxmox Backup ServerCompression, multi-level retention and a lock against deletion.
Before handing over your Proxmox
Is high availability enough to protect my data?
No. It restarts your machines elsewhere when their server goes down, from their latest replica: with replication every fifteen minutes, up to fifteen minutes of data. It does not protect against ransomware or human error, which are replicated too: you need separate backups, encrypted and checked.
LXC container or virtual machine?
A container does not migrate live: it is stopped and restarted, with ten to thirty seconds of downtime. A virtual machine migrates live. A service that tolerates no interruption goes in a virtual machine; the rest is in our KVM or LXC guide.
How do I choose the level?
By criticality: what an hour of downtime costs, and at what time of day. An internal application used during the day often falls under Essential; a site that takes payments at night, under Standard or Premium. We decide it with you during scoping.
And if I would rather not run servers any more?
We also host your applications on our cluster, in France, encrypted backups included: that is hosting in France.
Timelines, quotes, code ownership, end of contract: the answers are in the FAQ.
Quoted per project. Price ranges are published on the pricing page, along with what makes a budget vary.
Before the next failure
Who watches over your Proxmox cluster at night?
Tell us about your Proxmox servers and what runs on them. We will tell you what we monitor, what we take over, and at what price.
Reply within 24 working hours, from a founder.