This article applies to PRTG Network Monitor 15.2.17 or later

Remote Probes and Clustering

Remote Probes and the PRTG cluster follow different principles and are used for different scenarios.

Monitor Multiple Locations

If you want to monitor multiple locations (for example, networks of different customers), using Remote Probes are the means of choice if you do not have VPN connections into the target networks. A remote probe is a little piece of PRTG software running within the target network. It collects monitoring data from within the network and reports it to the central PRTG core server (for example, located at your data center) via one dedicated SSL-encrypted connection (see also our video about distributed monitoring).

Monitor Fail-Safely

A PRTG cluster setup consists of two, three, four, or five PRTG core server installations (we call them cluster nodes). All nodes share one configuration which contains the devices that will be monitored. Each cluster node monitors every device independently. This means, that in a single failover cluster with two nodes, each device will answer to a double number of monitoring requests, in a cluster with three nodes, the monitoring requests will be tripled, and so on.

If one of the nodes fails, the other nodes keep monitoring and make sure there is always live monitoring data available. The cluster also ensures that there is always exactly one Master node which sends out notifications, generates reports, and provides write access to the monitoring configuration. All failover nodes can be accessed in read-only mode in order to review data.

Remote Probes and Clustering

PRTG provides cluster support for remote probes. This means that all your probes can connect to all your cluster nodes, the primary master node as well as the failover node. Because of this you can still see monitoring data of remote probes and sensor warnings and errors even when your master node fails.

Note: You cannot cluster remote probes but connect them to your cluster nodes.

How Can I Achieve a Failsafe Setup _And_ Use Remote Probes?

For details about remote probes in a cluster, please see

• PRTG Manual: Failover Cluster Configuration—Remote Probes in Cluster
• and the blog article Cluster Support for Remote Probes: Failover Nodes Show Remote Probe Data

Other options (for example, for PRTG versions previous to 15.2.17 when cluster support for remote probes was introduced) are, for example,

• use the best metal for the PRTG core server you can get
• make sure your PRTG core server is built fail-safely using hardware which cares for interruption-free operation (using fail-safe disks, etc.)

We also have one customer who put the PRTG database on a high-performance fail-safe SAN (you can freely define the location of the database in the PRTG settings). The database holds the monitoring data as well as the current configuration. This customer runs one core server installation of PRTG on one server, and keeps available another server with exactly the same configuration as the first one (incl. the same IP address), as a (hot) standby server. In case the first server fails, they put it off the network and spin up the second server. The second server then connects to the SAN and takes over the current database and configuration (this is possible because the second server is an exact copy of the first one). This would only cause a short downtime in case of a server failure.

Please note that we did not test this scenario ourselves, so we do not officially recommend or support it. We just want to show a way to a possible solution.

Apart from that, an additional fail-safe single-failover PRTG cluster installation (2 nodes) (running, for example, at your central data center) would enable you to create a fail-safe monitoring of all your external (customers') locations from "the outside". With this installation you could monitor everything which is reachable from the location your PRTG- nodes are running at—which should be, at least, a Ping so you will see immediately when one location goes offline.