What this guide covers
WatchZit is a free, open-source (AGPLv3) monitoring system for networks and servers. There are no licence keys and no price per device or per check; MaxZit sells optional support. This guide does not compare the two products. It walks through the parts a Windows-based IT team usually needs and says, for each, what WatchZit does and where it stops.
- What you need
- Install the server on Hyper-V or VMware
- Sign in with Active Directory or Entra ID
- Windows servers: the agent through GPO or Intune
- Switches, routers and firewalls
- Printers and UPSs
- The internet line and branch offices
- Alerts in Teams, by email and SMS
- Wall screens, reports and opening devices
- What WatchZit does not do
- Run WatchZit next to your current monitoring
- What to do next
What you need
- A virtual machine with 2 CPUs, 2 GB of memory and a 16 GB disk on Hyper-V, VMware ESXi or Proxmox VE. The WatchZit server runs on Linux; the appliance brings its own Debian, so you do not have to set up Linux yourself.
- SNMP read access (a v2c community or a v3 user) on switches, routers, firewalls, printers and UPSs.
- For Windows servers, a way to roll out an MSI (Group Policy, Intune or a script) — or WinRM, for servers where nothing may be installed.
- Optionally a service account that may read users and groups in Active Directory.
Install the server on Hyper-V or VMware
- Download
watchzit-appliance-vhdx.zip(Hyper-V) orwatchzit-appliance.ova(VMware) from the download page. - Hyper-V: unzip it and create a generation 1 VM with 2 GB of memory that uses the VHDX as its disk. VMware ESXi: Create / Register VM → Deploy a virtual machine from an OVF or OVA file.
- Start the VM. Its screen shows the console address and the setup token. On the VM's own screen you can sign in as
wzadmin(first passwordwatchzit, which must be changed) for a setup menu with a fixed address, the server name and the time zone. - Open
https://<address>:8443, enter the setup token, create the first administrator and set up two-factor sign-in with an authenticator app.
After that, the server is looked after from the browser: Settings → Updates and server installs signed WatchZit updates (with an automatic rollback if the new version does not start) and the operating system updates, and Settings → Certificate takes a certificate of your company or one from Let's Encrypt. Settings → Console backup backs up the whole console every day, optionally encrypted and copied to an SFTP folder.
Sign in with Active Directory or Entra ID
- Settings → Active Directory / LDAP: the domain controller over
ldaps://(orldap://with StartTLS), a service account, the base DN and the user filter. Groups in groups are followed. WatchZit still asks for its own second factor. - Settings → Single sign-on: OpenID Connect (Microsoft Entra ID, Google, Okta, Keycloak) or SAML 2.0 (ADFS, Entra ID, Okta). The sign-in page then gets a button.
- Directory groups decide the role at every sign-in: Administrator, Operator (devices, checks, maps, triggers, maintenance, dashboards) or Viewer. Local accounts keep working when the directory is down.
Every change in the console is written to an audit log whose entries are chained, so changes to the log show.
Windows servers: the agent through GPO or Intune
Under Agents, create an enrollment code (how long it is valid, how many computers may use it, the group new devices go to). Then install the MSI silently with Group Policy, Intune or a script:
msiexec /i watchzit-agent-x64.msi ENROLLCODE="WZ1-…" /qn
The agent runs as the service WatchZitAgent on Windows 10/11 and Windows Server (x86-64) and makes one outbound connection to port 8444 of the server; the servers need no open port. It reports every minute:
- CPU, memory, every disk and every network interface with traffic, as checks with graphs;
- services you choose under Watch services on the device page — a service that stops opens a problem;
- programs you name under Watch processes, for example
sqlservr.exeorjava: how many run, their CPU and memory, and a problem when none runs; - critical and error events of the Windows System and Application logs, which go to Logs, where a log trigger can open problems from them;
- on Hyper-V hosts, the virtual machines with their state, CPU, memory and addresses;
- an inventory: operating system, CPU model, memory, vendor, model and serial number.
Agents older than the server update themselves when they connect. For servers where nothing may be installed, the check Windows over WinRM (no agent) reads CPU, memory, the page file, the fullest disk and the uptime over WinRM — see Agentless Linux and Windows server monitoring.
Switches, routers and firewalls
- Discovery: create an SNMP profile under Settings → SNMP, then scan a range under Devices → Discovery (up to a /22, now, daily or weekly) and click Add as devices.
- SNMP: interfaces with link state and live traffic, CPU, memory, storage and any numeric OID; VLANs with tagged and untagged ports from most managed switches; the VRRP role of routers and firewalls.
- Firewalls and APIs: the JSON API value check reads one number from a REST API such as FortiGate's; MikroTik RouterOS 7 has a check of its own.
- Configuration backups: every night over SSH from MikroTik RouterOS, Cisco IOS, FortiGate or with your own command, with the changes against the version before.
- Syslog and traps (UDP 514 and 162) and Traffic flows (NetFlow v5/v9 and IPFIX on UDP 2055, sFlow on UDP 6343): top talkers, destinations and services per exporter.
On a map, the cables WatchZit found through LLDP, CDP and the MAC address tables of the switches are offered under Suggested links, and each cable shows the traffic of its port.
Printers and UPSs
Give printers and UPSs an SNMP profile and run SNMP discovery on their device page:
- Printers: toner, ink, drums and waste boxes as a level in percent; the trigger Toner or ink low warns below 10 %.
- UPSs with UPS-MIB or an APC network card (PowerNet MIB): battery charge, runtime left and load. The check is down while the UPS runs on battery or its battery is low, and UPS runtime short warns below 10 minutes.
To check all printers of a site at once, filter the device list, tick them and choose under Change selected: the SNMP profile or a template.
The internet line and branch offices
The check Network path (traceroute) traces the way to a device every five minutes, like MTR: each router on the way with its loss and its best, average and worst time, kept for 30 days, and the last ten route changes. Put it on a device behind your provider or at a branch office to see where loss or delay begins.
Branch offices behind NAT get a remote collector: the ordinary agent on one computer there, switched to Use as a remote collector. It runs the checks of the devices set to Monitored from that collector, scans the branch network and keeps measuring while the line to the head office is down; a second collector can take over. The guide Monitor branch offices with a remote collector shows every step.
Alerts in Teams, by email and SMS
Built-in triggers already watch for devices that are down, packet loss, high CPU load, full disks and expiring certificates. Under Settings → Notifications add channels: email over your SMTP server, Microsoft Teams, Slack, Discord, Telegram, ntfy, PagerDuty, SMS through Twilio or your own gateway, and signed webhooks. Each channel has a lowest severity.
- Escalation and hours of duty: Tell after (minutes) sends a problem to the team lead only when it is still open after, say, 15 minutes; Only on some days and hours (on-call) makes a channel the night shift.
- Dependencies: when the router of a branch is down, you get one message for the router, not one per device behind it.
- Maintenance under Problems → Maintenance, with Repeat for the monthly patch day: devices are still measured, but nothing is sent.
Wall screens, reports and opening devices
- Dashboards with graphs, current values, device tiles, open problems and maps; Settings → Kiosk displays shows maps full screen behind a secret link for a monitoring TV.
- Reports: the availability of every device for 7, 30 or 90 days or a month, downtime and problems, as CSV or PDF; with an SLA target (%) per group; by email every Monday or every 1st of the month.
- Open with: a device's web page, Remote Desktop, SSH (PuTTY), WinBox or VNC with one click, started on your Windows PC by the small MaxZit Launcher, which installs without administrator rights.
What WatchZit does not do
- The server does not run on Windows. It runs on Linux (x86-64) with PostgreSQL, or as the appliance VM on Hyper-V, VMware or Proxmox VE.
- No importer from other monitoring systems: devices come from discovery, the agent or by hand, and history starts when you add them.
- SQL Server is not a check type. The database checks are PostgreSQL and MySQL/MariaDB. For SQL Server, watch its service and the
sqlservr.exeprocess with the agent, and show its own tables on graphs as a data source (Settings → Data sources). - No check that runs your own scripts. Your own values come from an SNMP OID, a JSON API, or a data source on a graph.
- The agentless WinRM check covers the basics (CPU, memory, page file, the fullest disk, uptime). Every drive, services, processes and event logs need the agent.
- The agent for Windows is for x86-64; the MaxZit Launcher is a Windows program.
- Hypervisor checks (vSphere, Proxmox VE) run on the console, not through a remote collector.
Run WatchZit next to your current monitoring
- Import the appliance as a new VM; nothing on the existing monitoring server changes.
- Give WatchZit its own read-only SNMP community or v3 user, and its own service account in Active Directory.
- Install the WatchZit agent alongside what runs on the servers today; it needs only its outbound connection to port 8444. Start with one group of servers.
- If switches and firewalls can send syslog, traps or flows to a second destination, add WatchZit as one instead of moving them over.
- Send WatchZit's problems to a separate Teams channel at first, compare for a few weeks, then decide which system pages the on-call team.
What to do next
- See everything WatchZit does on the features page and get the appliance on the download page.
- Draw the network: A modern alternative to MikroTik The Dude shows maps step by step.
- Coming from another tool more generally? An open-source Zabbix alternative covers triggers, learned thresholds and templates.
- The complete reference is the user guide built into the console; see the documentation.
Questions and ideas are welcome in GitHub Discussions; companies can get commercial support from MaxZit.
PRTG is a trademark of its owner. MaxZit is not affiliated with, sponsored or endorsed by the owner of PRTG; the name is used only to describe what this guide is about.