Beyond Monitoring: Why Your Network Needs a Source of Truth

You’ve got monitoring. Maybe it’s LibreNMS or Zabbix, maybe PRTG’s free tier, maybe Checkmk, or perhaps an RMM like Action1 handling the endpoints. Something pages you when a link drops or a host stops responding. That problem is solved.

So why does it still take half a morning to answer, “What’s actually plugged into port 14 on the third-floor switch?” or “Which VLAN does the new POS gear live on?” or “What was this /24 reserved for again?” Why does onboarding the next hire mean walking them through a spreadsheet that was last accurate in March?
That gap isn’t a monitoring problem. Monitoring tells you whether things are up. It was never built to tell you what you have, how it’s wired, or what it’s supposed to look like. That’s a different layer, and it’s the one small shops almost always skip.

The Three Layers (and the One Nobody Names)

If you’ve read any “what should I run for my small network” thread, you’ve watched this play out. Someone asks for a cheap network management system (NMS), and the replies split fast and instinctively into two camps:
• Monitoring: Is it up, is it slow, or is something on fire? This includes LibreNMS, Zabbix, Checkmk, PRTG, OpenNMS, and Observium.
• Managing / RMM: Push patches, run scripts, remote in, and control endpoints. This includes Action1, NetLock RMM, your UniFi controller, or standard RMM platforms.
People will even split it cleanly — “Action1 for managing, Checkmk for monitoring” — and that’s the right instinct. But there’s a third layer that almost never gets named in those threads, and it’s the one doing the quiet damage:
• The Source of Truth: What do we have, where is it, how is it connected, who owns it, and what is it supposed to look like?

Monitoring and management are well-served by free and cheap tools. The source-of-truth layer isn’t missing because it doesn’t matter; it’s missing because, for most small shops, the category is invisible. Nobody pitches it to you the way monitoring gets pitched, so it never makes the shortlist, and the spreadsheet limps on for another year.

Is a Source of Truth Overkill for a Small Shop?

Honest answer: Sometimes, yes.

If you’re running ten devices in one rack and one person touches all of it, a well-kept spreadsheet or a Markdown file in a repository is genuinely fine. Don’t let anyone upsell you. The source-of-truth problem was never really about device count for its own sake; it’s about how much infrastructure you can no longer hold in your head.

The spreadsheet stops being fine somewhere around the point where:
• More than one person touches the network, and your edits start clobbering theirs.
• You’ve got a second site, and “where is that” now has a non-obvious answer.
• An audit, a cyber-insurance renewal, or a compliance requirement demands evidence of what you have and who can change it.
• You’re growing fast enough that the gap between the spreadsheet and reality is permanently a few weeks wide.
• Something breaks, and the first 20 minutes go to figuring out what’s even involved before you can start fixing it.
If two or three of those are true, you don’t have a documentation-habit problem. You’ve outgrown the spreadsheet, and adding more tabs is just deferring the inevitable.

What a Network Source of Truth Actually Is

A network source of truth isn’t just a document; it’s a model.
Concretely, it’s a structured, queryable record of your infrastructure—every device, rack, site, IP range, VLAN, cable, circuit, and virtual machine—and the relationships between them. Instead of a flat list, it models how things actually connect. Questions like “What’s on this switch?”, “What’s in this subnet?”, and “What depends on this circuit?” become queries you answer in seconds instead of reconstructing by hand.

The open-source project that defined this category is NetBox. It’s the de facto standard, boasting north of a hundred thousand deployments, and for good reason: the data model is excellent. Devices, racks, sites, IPAM, DCIM, cabling, circuits, and virtualization are all related, all under change control, and all accessible behind an API you can build against.

The part that earns it the name: A real source of truth holds the intended state, not just the observed state. That’s the core difference between “this is what’s running” (monitoring) and “this is what’s supposed to be running” (the model). Once the intended state is written down in a structured way, you can do things monitoring can’t: surface config drift when a device stops matching its design, review changes before they ship, and hand a new hire a map instead of a campfire story.

Why Doesn’t Everyone Just Run NetBox?

Because the model being right and the project being free doesn’t make it free to run.
Self-hosting NetBox is a real commitment. It’s a Django app with a PostgreSQL database, Redis, a web stack, plugins, backups, and an upgrade path you have to stay on top of. Anyone who’s done it knows the drill: you stand it up in an afternoon, it’s great for six months, and then you’re the one babysitting the upgrades, chasing a breaking change in a plugin, and quietly realizing your documentation tool now needs its own documentation.

For a one-person IT shop that already lacks time — the exact shop that needs a source of truth most—the requirement to operate and maintain a web application forever is precisely the upkeep that makes the whole idea feel like overkill. The model isn’t overkill; the operational tax is. Those are two different problems, and they almost always get blamed on the same thing.

The Managed Middle: Introducing Palpabl

This is the gap Palpabl is built for, so here’s the plain version of what it is and isn’t.

Palpabl provides the NetBox data model, fully hosted and managed. It offers the same way of thinking about your network (devices, racks, sites, IPAM, DCIM, circuits, and virtualization) without requiring you to stand up or maintain anything. There is no PostgreSQL database to back up, no upgrades to babysit, and no plugins breaking on a Tuesday. You get the model; we run the plumbing.

What It Deliberately Does

• IPAM and DCIM on the open NetBox data model: Nothing is locked in a proprietary schema, so your data stays fully portable.
• Topology mapping: Built directly from your modeled data to draw out the relationships you’ve recorded.
• Configuration drift detection: Surfacing where a device’s running config no longer matches the intended config in your source of truth.
• Inventory dashboards: Ensuring the model is something you can read at a glance, not just store.
• A REST API: Because a source of truth you can’t query from your own scripts isn’t much of one.

What It Is Not

Just as important for anyone who’s been burned by a vendor deck, here is what Palpabl is not:
Palpabl is not a monitoring tool. It does not replace LibreNMS, Zabbix, Checkmk, PRTG, or your RMM, and it isn’t trying to. It is not a real-time alerting system, and it does not auto-discover your network by scanning the wire. You model your infrastructure deliberately so the source of truth reflects what you’ve recorded, not whatever happened to answer a ping.

Keep the monitoring stack you already trust. Palpabl is the documentation-and-model layer those tools were never meant to be, answering “what do we have and how is it wired,” while your monitoring keeps answering “is it up.” If a tool can’t tell you plainly what it doesn’t do, assume it doesn’t do the things it claims either. That is the line, drawn on purpose.
Transparent Pricing with Real Numbers
The other reason this category stays invisible to small shops is that most of it is sales-gated. Palpabl handles this differently:
• Free up to 20 devices: No credit card, no sales call. This is enough to model a small site and find out whether the tool is useful before anyone spends a dollar.
• $39/month flat above that: Unlimited users included. Getting your whole team into the model shouldn’t require a per-seat upsell.
That’s the entire pricing page. Compare that to most of the source-of-truth and management space, where discovering what it costs means booking a call and waiting for a quote. NetBox Labs’ managed tier and Auvik both gate their pricing behind sales. For a shop where the IT lead is also the budget holder and decisions happen fast when value is obvious, a published flat rate isn’t a gimmick. It’s the difference between trying it this afternoon and never getting to it.

When It’s Worth It (and When It’s Not)

The straight version, because the skeptical reader is the one still here:
• Skip it if you have a handful of devices, one site, one person, and the spreadsheet still fits in your head. A managed source of truth won’t change your life at that scale. Bookmark this page and come back when it stops fitting.
• Look at it if you’ve outgrown the spreadsheet, you’ve considered NetBox but decided the self-hosting upkeep isn’t worth it, and you want the model without the maintenance, sitting perfectly alongside the monitoring you already run instead of replacing it.

The free tier exists precisely so you don’t have to take any of this on faith. Model a site, wire up a few racks, point your scripts at the API, and decide for yourself in an afternoon.