JUSTPING INFOTECHReliable IT Always

From bare-metal servers to a Proxmox cluster for a diamond manufacturer

Applications, databases and Active Directory ran on individual bare-metal servers, and a single failure could take a day or two to recover. We moved everything onto a Proxmox cluster with verified central backup — without downtime — and now look after the company as its IT partner.

1–2 days
to restore a failed server before the project
No downtime
during the move from bare metal to Proxmox
2 weeks
of analysis to trace the CCTV network's root cause
AMC renewed
after year one, with more branches handed over

The situation

The company ran its applications, databases and Active Directory on individual bare-metal servers, with one or two machines on VMware's free edition. When a server failed, restoring it from backup took a day or two, sometimes longer — and while it was down, the departments that depended on it, and production, were affected.

Two other problems had been around for a long time: intermittent ping drops on a large CCTV network, and a NAS on HPE hardware running Windows, which lacked features a dedicated NAS operating system provides, such as a network recycle bin.

Before

Separate machines, slow recovery.

  • Applications, databases and Active Directory on individual bare-metal servers
  • One or two servers on VMware's free edition
  • A failed server took a day or two or more to restore
  • Windows-based NAS with no network recycle bin
  • Recurring ping drops on the CCTV network

After

One managed platform.

  • Workloads running as VMs and LXC containers on a Proxmox cluster
  • More RAM and enterprise SSDs on selected servers
  • Central Proxmox Backup Server with backup verification
  • NAS running OpenMediaVault
  • CCTV network issue traced and permanently fixed
The environment before and after the project.

What we changed

  1. Assessed the whole environmentWe analysed every server and application before proposing anything, and recommended Proxmox VE virtualization.
  2. Upgraded selected hardwareWe added RAM and installed enterprise-grade SSDs in the servers chosen to host the virtual machines.
  3. Built a Proxmox clusterThe hosts were set up as a Proxmox cluster rather than as separate machines.
  4. Migrated step by step, without downtimeEach bare-metal installation moved to a virtual machine or an LXC container, depending on what the application needed.
  5. Centralised backupWe configured a central Proxmox Backup Server, with backup verification.

The CCTV network problem

The CCTV network had suffered ping drops for a long time. It is a large network and the cause was not obvious: it took two weeks of detailed analysis to trace the root cause, after which we put a permanent fix in place.

A proper NAS operating system

We replaced Windows on the company's HPE NAS server with OpenMediaVault, an operating system built for network storage. That brought features the Windows setup lacked, including a network recycle bin.

Result

  • Servers run as virtual machines and containers on a Proxmox cluster, not as individual bare-metal installations
  • Backups are centralised on Proxmox Backup Server and verified
  • The long-standing CCTV ping-drop problem is permanently resolved
  • The NAS runs a purpose-built NAS operating system

An ongoing partnership

We continue to support the company on a monthly AMC covering support, monitoring, updates and security monitoring. The client renewed the AMC after the first year and has since handed us more of its branches, in India and abroad. We are now their single IT partner.

Facing a similar problem?

Tell us what's happening in your environment. We'll explain what we'd look at first and what the practical options are.

Talk to an IT Engineer Send an enquiry