Monitoring UPS Units From Multiple Vendors

Five years of procurement, four different brands

Hardly any infrastructure that's grown organically bought all its UPS units on the same day and never touched the fleet again. A site joins through an acquisition and brings its own UPS with it, a supply shortage with the usual vendor leads to a stopgap purchase elsewhere, a replacement goes to whichever brand is cheapest at the time rather than the one already installed. After a few years, the same infrastructure often runs a mix of APC, Eaton, CyberPower, Vertiv/Liebert, Riello or Socomec — each with its own management card and its own bundled software.

The mix of hardware itself is rarely the real problem — UPS units are long-lived, and replacing one purely for consolidation's sake rarely pays off. The actual problem is what happens with the mix of software: APC PowerChute Network Shutdown, Eaton Intelligent Power Manager (or Intelligent Power Protector), CyberPower PowerPanel Business Edition, Vertiv/Liebert MultiLink, Riello NetMan or Socomec NetVision then run side by side — each with its own login, its own configuration logic, its own update cycle, with no instance aware of any of the others.

SNMP is the common denominator almost nobody uses

For UPS units without a usable SNMP card, or with a plain USB/serial connection, NUT (Network UPS Tools) fills the same role — an open-source de facto standard with its own driver library that, combined with SNMP, covers 100+ vendors, from CyberPower to Riello, Legrand and Powerware. The same rule applies here: the protocol doesn't know about vendors, only the vendor's own software draws that line.

One installation for the entire fleet

1. SNMP v1/v2c/v3 and NUT on equal footing, detected automatically
   A single installation of the UPS Hyper-V Shutdown Monitor reads the vendor-specific OIDs of APC (PowerNet), Eaton (XUPS and MGE), Socomec (Netvision 5 and 6), Vertiv/Liebert, Huawei (UPS2000/UPS5000) and CyberPower automatically, with the RFC 1628 standard as a universal fallback. NUT covers everything beyond that. Combined, that reaches 100+ vendors, including Riello, Legrand and Powerware.
2. Multiple UPS brands monitored at once
   Multiple Hyper-V hosts sitting behind different UPS brands run through the same configuration, the same protocol layer and the same alerting — instead of as many separate consoles as there are brands in inventory (multiple UPS devices from the Pro edition up; see also the Hyper-V cluster power outage guide for cross-host coordination).
3. Large fleets at a glance
   With two or more UPS devices, the dashboard switches from tiles to a compact table: one row per UPS with status, battery, runtime, load and model, topped by a summary of how many devices are on mains, on battery or unreachable. The remote web interface (Enterprise) moves problem devices to the top and can hide everything that's fine. The PDF report (Pro) covers every UPS in one document, including model and serial number, which are stored during setup — so a UPS that's briefly unreachable still shows up complete.
4. No rip-and-replace required
   SNMP-/NUT-based monitoring can be set up and observed alongside the existing vendor software before any decision is made about replacing it — nothing already installed has to come out first.

That doesn't necessarily replace every function of the respective vendor software — diagnostics reaching deep into the UPS's own firmware, or firmware updates themselves, stay the vendor tool's job. For the core task — monitoring UPS state and shutting down gracefully when needed — the shared SNMP/NUT layer is usually enough on its own.

Same protocol, different meaning

RFC 1628 makes sure every UPS answers. What a value means, though, is defined by each vendor in its own MIB, and the differences aren't cosmetic. Monitoring that just passes raw values through will, at worst, shut down for no reason — or not at all when it matters. Three examples from practice:

The UPS Hyper-V Shutdown Monitor handles these cases per vendor. Its parsing is checked against the official vendor MIBs and the Network UPS Tools drivers; the latest corrections are listed in the version 1.10.0 release notes. Whatever software you use: after setup, compare the displayed values once against the management card's own web interface, and run the shutdown once in test mode before relying on it in an actual outage.

Check before consolidating

1. How many UPS brands are actually in inventory? Count what grew in historically through acquisitions, site takeovers or replacement purchases — not just the current shopping list.
2. How many separate vendor agents or consoles run in parallel? And across how many different servers, including the update upkeep each one needs.
3. Do the existing management cards speak SNMP? Most models do — check the card's own web interface under Network → SNMP or similar; older or USB-connected units need a NUT server instead.
4. Is there a shared view of battery charge and remaining runtime across all UPS units today? Or does every vendor console have to be opened separately?
5. Do multiple Hyper-V hosts sit behind different UPS brands? Then coordinated, cross-host shutdown ordering is the real issue — not just UPS monitoring on its own.
6. Can SNMP-/NUT-based monitoring run as a trial alongside the existing software? That answers the consolidation question risk-free, before anything currently in place gets switched off.
7. Are the values right for each brand? Compare status, battery charge and runtime for every UPS brand once against its management card's web interface — ideally during a short battery test as well. Discrepancies show up exactly where vendors encode values differently.