Cloudmon is made up of three core components that work together to provide unified IT observability: the Controller, the Probe, and the Agent.
A Cloudmon deployment always includes at least one Controller and one Probe. Agents are deployed on devices where agent-based monitoring is required. Understanding how these three components interact is essential for planning a Cloudmon deployment that fits your environment.
| Component | Role |
| Cloudmon Controller | The central server (deployed in cloud or on-premises) that provides the real-time dashboard, raises alarms when thresholds are crossed, and enables configuration and management capabilities for IT operators and administrators. Controller metrics and data are stored in a separate database (MongoDB). |
| Cloudmon Probe | Collects availability, performance, and visibility data from agentless devices and pushes it periodically to the Controller. Can be co-located with the Controller or placed at branch locations. Supports ICMP, TCP, SNMP, WMI, and NetFlow/IPFIX monitoring methods. In MSP deployments, separate probes are required per customer. |
| Cloudmon Agent | Collects metrics from servers, hypervisors, virtual machines, Docker containers, and end-user devices. Establishes a direct secure connection to the Controller. Enables liveness monitoring (within 25 seconds), automated remediation, high-frequency polling, Hyper-V monitoring, and Digital Experience Monitoring. Installed on Linux, Windows, macOS, and Android. |
The Controller is the central hub that everything connects back to. Probes sit in the network and actively poll agentless devices such as routers, switches, firewalls, and Windows servers via WMI, then push the collected data back to the Controller at regular intervals. Agents are installed directly on monitored devices and send their metrics to the Controller over a persistent secure connection, enabling faster detection and deeper visibility than agentless polling can provide.
In a distributed environment with multiple branch offices, a probe is deployed at each branch to ensure SNMP and WMI polling can reach local devices without traversing the WAN for every query. All probes feed data back to the single central Controller, maintaining a unified view of the entire environment.