Network Discovery and Scanner Agents
How Plenix finds every device on a network β agents, scanner agents, ARP and port sweeps, SNMP, LLDP and switch tables β and how to control when it runs.
How devices are found
Network maps and IPAM are filled in from several sources at once:
| Source | What it finds |
|---|---|
| Plenix agents | Each computer's own addresses, subnets, gateway and status |
| Scanner agents | Everything else on their network, including devices that can't run an agent |
| SNMP | Model and description of switches, routers, printers and other network gear |
| LLDP | Which switch or router is cabled to which |
| Switch MAC tables | Which device is plugged into which switch port |
| Printers | Printers you manage in Plenix |
| IPAM | Subnets and VLANs you've documented |
Devices seen by more than one source are merged into one, matched by MAC address and IP address.
Scanner agents
A scanner agent is a normal Plenix agent that also scans its network. Any Windows, macOS or Linux agent can be one. To turn one on, open the device in RMM, go to its Network Devices tab and turn on Gateway Mode. Pick a machine that's always on, such as a server. Rescan Network on the same tab scans from it straight away.
When a scanner runs, it:
- Sweeps its own subnet using ARP. Every device that answers is found, even firewalled machines that ignore ping and have no open ports.
- Checks common ports on each device to work out what it is. These include web admin, SSH, Windows file sharing and remote desktop, printing, cameras, VoIP and hypervisors.
- Asks SNMP-capable devices for their model, their neighbours and their switch port tables.
- Looks up names from DNS and NetBIOS.
- Also sweeps other small subnets you've documented in IPAM for the same customer. Routed subnets are checked with a lighter ping-and-port pass, because ARP doesn't cross routers.
Each scanner covers up to 8 subnets. Very large ranges are limited to 1,024 addresses (a /22) so a scan finishes in a couple of minutes.
The Open topology link on the Network Devices tab opens that device on its network map.
Working out what a device is
Plenix gives every device a role: router, switch, access point, firewall, server, hypervisor, NAS, workstation, laptop, printer, camera, phone, UPS, IoT device and so on. It uses:
- the manufacturer, worked out from the MAC address;
- the open ports;
- SNMP descriptions;
- the operating system, for devices with the agent;
- the hostname.
If Plenix gets a role wrong, change it in the device panel on the map. Your choice is kept and never overwritten. If AI is set up, it can identify devices the rules can't. See AI network review.
When discovery runs
Go to Network Maps β Discovery. You can:
- click Scan all networks at the top of the hub to scan everywhere now;
- click Scan now next to a scanner to scan from just that one;
- click Scan site on a site's map.
The settings (administrators only):
| Setting | Default | What it does |
|---|---|---|
| Scheduled scans | On | Scanner agents sweep their networks on a schedule |
| Scan every | 6 hours | How often automatic discovery runs |
| Keep IPAM up to date | On | Adds what's found to IPAM twice an hour |
| Create subnets automatically | On | Adds subnets that aren't in IPAM yet, marked Discovered |
| Mark stale after | 30 days | When unseen devices are marked stale and their discovered addresses are freed |
| AI review | On | Lets AI identify unknown devices after each scheduled scan (needs the AI module and AI set up on the server) |
The Discovery tab also lists every scanner agent with whether it's connected and when it last reported.
Good to know
- Scans only look at private network ranges. They never scan the internet.
- A scanner only sees networks it can reach. For a site with several VLANs, either put a scanner on each VLAN, or document the VLANs in IPAM so scanners sweep them as routed subnets.
- SNMP uses the scanner agent's SNMP community (
publicunless you've set another). Switches with SNMP turned off still appear, but without their cabling.
Related
Was this article helpful?