WatchZit

A LibreNMS and Observium alternative for SNMP network monitoring

If you are looking for an alternative to LibreNMS or Observium, this guide shows how WatchZit monitors a network that mostly speaks SNMP — switches, routers, firewalls, printers, UPSs and MikroTik — from discovery to port graphs, maps that draw themselves and alerts. Step by step, with what WatchZit does not do and how to run it next to the tool you have.

What this guide covers

WatchZit is a free, open-source (AGPLv3) monitoring system for networks and servers, with one web console on your own server. There are no licence keys and no price per device or per port; MaxZit sells optional support. This guide does not compare products. It follows the SNMP side of WatchZit and says, for each part, what it does and where it stops.

What you need

  • A virtual machine with 2 CPUs, 2 GB of memory and a 16 GB disk on VMware, Proxmox VE or Hyper-V — or a 64-bit Debian 12+, Ubuntu 22.04+, AlmaLinux or Rocky Linux 9+ server with PostgreSQL 13+.
  • SNMP read access on the devices: a v1 or v2c community, or a v3 user.
  • Optionally: devices that can send syslog, SNMP traps or NetFlow/IPFIX/sFlow to one more destination, and an SSH user that may read the configuration for configuration backups.

Install the server

Import the appliance from the download page (OVA for VMware, qcow2 for Proxmox VE and KVM, VHDX for Hyper-V generation 1, or an ISO), or run the install script on a Linux server of your own:

curl -fsSL https://github.com/Max-Zit/watchzit/releases/latest/download/install.sh | sudo sh

Open https://<server>:8443, enter the setup token that the VM's screen or the script shows, create the first administrator and set up two-factor sign-in with an authenticator app.

Discover the network

  1. Create an SNMP profile under Settings → SNMP (see the next section). Discovery asks SNMP with all your profiles.
  2. Under Devices → Discovery, add a range such as 10.10.0.0/23 (up to a /22, or one address) and scan it now, every day or every week.
  3. WatchZit pings every address, tries common ports (SSH, Telnet, web, RDP, WinBox, SMB, printers, VNC), asks SNMP for the name, description and maker, and looks up the DNS name. Hosts that are not devices yet are listed with a guessed kind and what answered.
  4. Tick the hosts and click Add as devices, optionally into a group. Each gets a ping check (or a TCP check when ping is blocked) and the SNMP profile that answered. Hide keeps a host out of the list.

With a schedule, new hosts that appear on the network show up in the list by themselves. When a device's SNMP description names a model of the picture library (MikroTik, for example), that picture is chosen by itself when the device is added to a map.

SNMP profiles and what is read

A profile under Settings → SNMP holds the version (v1, v2c or v3), the port and the community, or the v3 user with authentication and privacy (SHA-2 and AES are supported). Communities and passwords are stored encrypted. To give many devices a profile at once, filter the device list, tick them and choose the profile under Change selected:.

When a device gets a profile, WatchZit opens SNMP discovery: tick the interfaces, CPU, memory and disks to watch; loopbacks and interfaces without link are left out. Discovery also finds what a device of its kind reports:

  • Switches, routers and firewalls: interfaces with link state and traffic, CPU, memory and storage.
  • Printers: toner, ink, drums and waste boxes as a level in percent.
  • UPSs (UPS-MIB, or the APC PowerNet MIB of APC network cards): battery charge, runtime left and load. The check is down while the UPS runs on battery or its battery is low.
  • MikroTik: temperatures, voltages, fans and power, each as a check with its unit.
  • Anything else: any numeric OID as a check of its own.

Every check runs on its own interval (10 seconds to an hour; a minute by default).

Ports, traffic graphs, VLANs and HA roles

  • Ports: the device page shows the ports with link state and live traffic, and a graph of every value. Drag on a graph to zoom in; the time range buttons go from one hour to a year.
  • History: every value is kept raw for 30 days, as 5-minute values for 90 days and as hourly values for two years by default (Settings → Metric data), with minimum, average and maximum.
  • Your own graph layout: Edit graphs puts several values on one graph (both directions of an uplink, or several ports), stacks them and draws warning and critical lines.
  • VLANs are read over SNMP once an hour (Q-BRIDGE-MIB on HP/Aruba, Cisco small business, Ubiquiti, Netgear, D-Link and most managed switches; CISCO-VTP-MIB on Cisco Catalyst). The device page lists each VLAN with its tagged and untagged ports and each port as Access, Trunk or Hybrid.
  • HA role: devices that run VRRP report whether they are master or backup (VRRP-MIB and VRRPv3-MIB); the device page shows the role and since when.

MikroTik in detail

Besides SNMP, MikroTik RouterOS 7 has a check of its own, MikroTik RouterOS (API), over the REST API: CPU, memory, uptime and temperature. With it, WatchZit also reads the VLANs of MikroTik bridges and the VXLAN interfaces every ten minutes, the VRRP role, and the MikroTik neighbour table for the maps. The built-in trigger High temperature warns when RouterOS reports more than 70 °C for 5 minutes.

From a device page or a map, Open with starts WinBox, SSH or the web interface on your Windows PC through the small MaxZit Launcher. The guide A modern alternative to MikroTik The Dude covers MikroTik networks and maps step by step.

Maps from LLDP, CDP and MAC tables

Every 15 minutes WatchZit reads from the devices with SNMP where things are plugged in: LLDP, Cisco CDP and MikroTik neighbour tables; the MAC address tables of switches (so printers, cameras and servers without SNMP are placed too); port descriptions that name another device; and saved configurations with two addresses in the same /30 or /31.

  1. Under Maps, add a map, click Edit and put a few devices on it.
  2. Open Suggested links: it lists the cables with the ports at both ends and the connected devices that are not on the map yet. Add all takes them.
  3. Turn on Draw by itself and Also add the devices they lead to, and the map keeps adding new links and devices every 15 minutes.
  4. Arrange lays the map out by its cables, the most connected device on top. Click Save.

On the map, devices are coloured by their state and cables get wider and change colour with utilisation (amber above 50 %, red above 80 %); a cable whose port has no link is red and dashed. Click a cable for its Traffic history. Choose a VLAN in the toolbar to see only the cables that carry it, or click two devices and Path to see the chain of cables between them. The picture library has some 3,500 real models of over 180 manufacturers, with their ports.

Syslog, traps, flows and configuration backups

  • Syslog and traps (Settings → Syslog and traps): syslog on UDP 514 and SNMP traps on UDP 162 (v1 and v2c; informs are answered), kept for 30 days and searchable under Logs.
  • Traffic flows (Settings → Traffic flows): NetFlow v5/v9 and IPFIX on UDP 2055 and sFlow v5 on UDP 6343. You see the traffic over time and the top sources, destinations, services, exporters and conversations, per exporter or for all.
  • Configuration backups in a device's Configuration backups section: every night over SSH from MikroTik RouterOS, Cisco IOS, FortiGate or with your own command. A new version is kept only when the configuration changed (the last 50), with the changes against the one before.

Alerts for network devices

A new console already has triggers under Problems → Triggers: Device down, Check down, Packet loss above 20 %, Toner or ink low below 10 %, UPS runtime short below 10 minutes, High temperature for RouterOS and HA group without redundancy for HA groups drawn on a map. For network devices these are useful too:

  • One port or one sensor: the bell button next to a check on the device page adds a trigger for that check alone — needed for OID values with their own unit (a temperature in °C, a voltage in V).
  • Traffic without a fixed threshold: Value unusual for the time (learned) learns from the last four weeks what is usual for each hour of the week and opens a problem when a value is far from that.
  • Messages: a trigger of the kind Syslog message or trap opens a problem when a message contains a text or matches a regular expression, such as /link (down|failure)/.
  • One message instead of fifty: open a map, click Dependencies and Set the ticked dependencies. When the router of a site is down, the devices behind it are only unreachable and their problems are not sent.

Problems go to the channels under Settings → Notifications: email, Slack, Microsoft Teams, Discord, Telegram, ntfy, PagerDuty, SMS and signed webhooks. The guide A Nagios and Checkmk alternative explains escalation, on-call hours and maintenance.

Sites behind NAT: remote collectors

Networks that the console cannot reach — a branch office, a closed management VLAN — get a remote collector: the WatchZit agent on any computer there, switched to Use as a remote collector. It runs the SNMP, ping and other checks of the devices set to Monitored from it, scans the ranges of its site under Devices → Discovery, reads interfaces and VLANs, and keeps measuring while the line to the console is down. A second collector can be the Backup. The guide Monitor branch offices with a remote collector shows every step.

What WatchZit does not do

  • No importer from other monitoring systems: devices come from discovery, the agent or by hand, and history starts on the day you add them.
  • No MIB files to load. Values that SNMP discovery does not offer are added by their numeric OID, and they are read as numbers.
  • Traps are SNMP v1 and v2c. They are kept as log messages with their text; a log trigger can open problems from them.
  • Reports are about availability (per device and against SLA targets of groups); there are no 95th-percentile or traffic billing reports.
  • Flows are summed per minute and kept for 7 days by default (up to 90); each minute keeps the 1,000 largest conversations of each exporter, the rest is one row without addresses.
  • A discovery range is at most a /22; add several ranges for a larger network. Links on maps are read every 15 minutes, not live.
  • Configuration backups know MikroTik RouterOS, Cisco IOS and FortiGate; other devices need your own command over SSH.
  • The server runs on Linux (x86-64) with PostgreSQL, or as the appliance.

Run WatchZit next to your current monitoring

  1. Install WatchZit on its own VM; nothing on the existing monitoring server changes.
  2. Give WatchZit its own read-only SNMP community or v3 user, so you can tell its queries apart and remove its access later. Most devices accept several of them.
  3. Start with one site: discover it, let a map draw itself, set the dependencies and send WatchZit's problems to a test channel first.
  4. Where devices can send syslog, traps or flows to a second destination, add WatchZit as one (UDP 514, 162, 2055 and 6343 by default) instead of moving them over.
  5. Compare for a few weeks which problems each system raised and which ones mattered, then decide.

What to do next

Questions and ideas are welcome in GitHub Discussions; companies can get commercial support from MaxZit.

LibreNMS and Observium are trademarks of their respective owners. MaxZit is not affiliated with, sponsored or endorsed by the owners of LibreNMS or Observium; the names are used only to describe what this guide is about.