Questa pagina non esiste ancora in Italiano.
 

icon how to white

How to set up a failover cluster
in PRTG in 6 steps

 


PRTG comes with a built-in failover cluster for high-availability monitoring. A failover cluster consists of two separate PRTG core server installations, where one server acts as the master node and one server as the failover node.

PRTG includes a single failover cluster with each license, no matter if you use the freeware edition or any commercial edition of PRTG Network Monitor. The cluster feature is not available in PRTG Hosted Monitor.

In this how-to guide, we show you how you can set up a single failover cluster in 6 steps.

 

Requirements for the failover cluster

Before you set up your cluster, you need to consider the following requirements:

PRTG core server

  • Make sure that you have two separate PRTG core server installations.
  • The PRTG core server systems can be physical machines or virtual machines.
  • The PRTG core servers must both run on 32-bit Windows systems or on 64-bit Windows systems.

License key

Use the same license key for both PRTG core server installations.

PRTG version

Make sure that both PRTG core servers have exactly the same PRTG version and build number. If necessary, install updates.

Port

  • Communication between the PRTG core servers must be possible in both directions.
  • Make sure that no firewall blocks the communication. If necessary, define Network Access Translation (NAT) rules.
  • The default TCP port is 23570.

Performance

  • The monitoring load doubles with each cluster node.
  • We recommend that you stay below 2,500 sensors in a cluster for best performance. More than 5,000 sensors are not supported.
  • For each additional failover node, divide the number of recommended sensors by two.

 


For more detailed requirements and in-depth information, see the PRTG Manual: Failover Cluster Configuration.

 

1. Configure the master node

 

Choose which of your PRTG core servers will act as the master node of the cluster. If you have a PRTG installation that has been running for some time, make this PRTG core server your master node to keep your monitoring configuration.

 

  1. On the PRTG core server that acts as your master node, open the Windows Start menu and click All apps.
  2. Scroll down to the PRTG Network Monitor folder and click it.
  3. Double-click PRTG Administration Tool.
  4. Confirm the Windows User Account Control with Yes.
  5. Go to the Cluster tab.
 
  1. Click Create a Cluster and confirm the warning message with Yes.
 
  1. Enter a Cluster Port or leave the default value. The port must be open on both the master node and the failover node.
  2. Enter a Cluster Access Key or leave the default value. The cluster access key must be unique. Both cluster nodes must use the same cluster access key.
  3. Save the cluster access key for later use.
 
  1. Click OK and confirm the message that appears with OK to restart the Windows services and convert the PRTG core server to act as the master node of the cluster.
 

2. Configure the failover node

 

After you have configured the master node, you need to configure the failover node as well.

  1. On the PRTG core server that acts as your failover node, open the Windows Start menu and click All apps.
  2. Scroll down to the PRTG Network Monitor folder and click it.
  3. Double-click PRTG Administration Tool.
  4. Confirm the Windows User Account Control with Yes.
  5. Go to the Cluster tab.
 
  1. Click Join a Cluster and confirm the warning message with Yes.
 
  1. Under Master Node (IP address/DNS name), enter the IP address or DNS name of the master node that you previously configured.
  2. Make sure that the Cluster Port matches the port that you previously configured for the master node.
  3. Enter or paste the Cluster Access Key that you previously configured for the master node.
 
  1. Click Join and confirm the message that appears with OK to restart the Windows services and convert the PRTG core server to act as the failover node of the cluster.
 

3. Confirm the failover node

 

Now that you have configured the master node and the failover node, you need to confirm the failover node by setting it to the Active status in the settings of the master node.

  1. Go back to the PRTG core server that acts as your master node and log in to the PRTG web interface.
  2. From the main menu, select Setup | System Administration | Cluster to open the cluster settings.
  3. In the Cluster Node Setup list, you can see your master node in the first line and your failover node in the second line. Set your failover node to Active.

 

  1. Click Save for the cluster nodes to connect. It might take a few minutes until the cluster nodes start to exchange configuration data.
 

4. Check the cluster connection status

 

Make sure that the master node and the failover node are connected by checking the cluster status.

  1. From the main menu, select Setup | PRTG Status | Cluster Status.
  2. Check that under State, both cluster nodes show the cluster status Connected.
  3. Go to the PRTG core server that acts as your failover node and log in to the PRTG web interface there.
  4. Repeat steps 1 and 2.
 

If the two cluster nodes have connection issues, check the following log files for any related error messages:

  • the Cluster Log below the Cluster Status section
  • the PRTG core server log file that you can find in the \Logs\core subfolder of the PRTG data directory

Also make sure that no firewall blocks the connection between the master node and the failover node via the specified cluster port on both sides. If necessary, define NAT rules for your network. See also How to connect PRTG through a firewall in 4 steps.

 

5. Move monitoring objects to the cluster probe

 

After you set up the failover cluster, you need to move your monitoring objects from the local probe to the cluster probe. This is necessary because PRTG can only monitor objects that are located on the cluster probe via all cluster nodes.

  1. On the PRTG core server that acts as your master node, select Devices from the main menu to open the device tree.
  2. Move all groups, devices, and sensors from the local probe to the cluster probe. The easiest way to do so is to make use of the object hierarchy and move only the groups that are directly underneath the local probe. This way, all subobjects are moved to the cluster probe as well.
    See the following example:
    • To move a group to the cluster probe, right-click the group name and select Move | To Other Group from the context menu.
    • In the dialog that appears, select Cluster Probe as the target group.
    • Click OK to move the group and its subobjects to the cluster probe. PRTG automatically transfers these objects including all settings to the failover node.
 

6. Move custom files to the failover node

 

PRTG automatically transfers maps and custom lookup files to the failover node when the master node starts up. However, you need to manually reload custom lookup files on both the master node and the failover node. See also our guide How to work with lookups in PRTG.

In addition, you must manually copy other custom files from the following subfolders of the PRTG program directory on the master node to the respective subfolders of the PRTG program directory on the failover node:

Custom content

 

Subfolder of the PRTG program directory

Device templates

 

\devicetemplates

Custom sensors

 

\Custom Sensors

MIB files

 

\MIB

SNMP libraries

 

\snmplibs

Notifications

 

\notifications

 

Add further failover nodes (optional)

 

You can add more failover nodes to the cluster at any time. Keep in mind, however, that you need an additional license key if you want to run two or three failover nodes, and two additional license keys if you want to run four failover nodes.

All in all, a failover cluster can consist of a maximum of 5 cluster nodes: one master node and up to four failover nodes.

Also keep in mind the performance considerations mentioned at the beginning of this how-to guide.

 

This how-to guide only scratches the surface?
Get more deep-dive information here!