What is this?

This knowledgebase contains questions and answers about PRTG Network Monitor and network monitoring in general.

Learn more

PRTG Network Monitor

Intuitive to Use. Easy to manage.
More than 500,000 users rely on Paessler PRTG every day. Find out how you can reduce cost, increase QoS and ease planning, as well.

Free Download

Top Tags


View all Tags

Exchange 2010 NLB

Votes:

0

I have a Exchange 2010 environment and just had an issue with one of my CAS servers. This caused a problem for my users.

My question to you guys is how can I monitor an NLB (Network Load Balancer) for Exchange 2010? I have 2 CAS servers in my NLB. I want to be able to set up an alert so that if the NLB leaves a converged state I get notified.

Is this possible to set up?

Thanks, Jon

exchange smtp wmi

Created on Feb 12, 2012 1:45:06 PM



Best Answer

Accepted Answer

Votes:

0

I just went thru this, and got it working. Here's what I did.

First, create a "NLBSNodeCheck.wql" file in the PRTG "WMI WQL Scripts" folder, that contains this script: SELECT StatusCode FROM MicrosoftNLB_Node WHERE ComputerName= '<#PH1>'

Next, create a custom WMI sensor (not the string version), for the server that is running the Windows NLBS node that you want to monitor.

In the sensor settings:

  • Set the namespace to "root\microsoftnlb".
  • Select the NLBSNodeCheck.wql file in the WQL File drop-down list.
  • In Placeholder <#PH1>, enter the fully qualified name of the server on which the node is running, like "[email protected]".
  • Set "If Value Changes" to "Trigger change notification".

In the sensor notifications, add 2 Threshold triggers. Set one trigger to When Primary channel is Equal to 1005 and the other one to When Primary channel is Equal to 1017. 1005 means the node is stopped, and 1017 means the node is suspended. All other values are either OK, or 0 if the node doesn't exist.

That's it.

Here's a helpful PowerShell command you can run to read all the NLBS node values using WMI: Get-WmiObject -Namespace "root\microsoftnlb" -Class MicrosoftNLB_Node -Impersonation 3 -ComputerName mywebserver You can use this to read the actual server name NLBS is looking for, as that must match exactly for the PRTG sensor to work. To use this PowerShell command, change the ComputerName to be correct for you, and paste it into PowerShell. You must have administrative rights on the remote server for this to work.

Created on Nov 9, 2012 6:00:26 PM

Last change on Nov 9, 2012 6:00:26 PM



7 Replies

Votes:

0

Dear Jon,

I would search for MIB files for the load balancer, then with our MIB Importer you should be able to choose the desired OIDs to monitor the NLB over SNMP. But this depends on what device you are using in particular and if this device provides this information over SNMP.

Kind regards Tobias

Created on Feb 14, 2012 8:02:12 AM by Tobias Lange [Paessler Support]



Votes:

0

I am not using a hardware load balancer. I am using Microsoft's NLB that is built into Windows Server 2008.

When I access the NLB Manager it shows the status of the two CAS servers. When everything is normal they show a status of converged. I need to know when that status has changed. I didn't know if there was a way to monitor that using a PS command or WMI sensor.

For now I have set up a WMI Event Log sensor looking for source WLBS. But was hoping there was a better way.

Created on Feb 14, 2012 1:54:29 PM



Votes:

0

so if it's a Microsoft NLB maybe you could use our WMI custom sensor. You will have to find some WMI querys which will return the needed information from the NLB and place that query into an .wql file which then can be used by the sensor.

Created on Feb 17, 2012 9:50:32 AM by Tobias Lange [Paessler Support]



Accepted Answer

Votes:

0

I just went thru this, and got it working. Here's what I did.

First, create a "NLBSNodeCheck.wql" file in the PRTG "WMI WQL Scripts" folder, that contains this script: SELECT StatusCode FROM MicrosoftNLB_Node WHERE ComputerName= '<#PH1>'

Next, create a custom WMI sensor (not the string version), for the server that is running the Windows NLBS node that you want to monitor.

In the sensor settings:

  • Set the namespace to "root\microsoftnlb".
  • Select the NLBSNodeCheck.wql file in the WQL File drop-down list.
  • In Placeholder <#PH1>, enter the fully qualified name of the server on which the node is running, like "[email protected]".
  • Set "If Value Changes" to "Trigger change notification".

In the sensor notifications, add 2 Threshold triggers. Set one trigger to When Primary channel is Equal to 1005 and the other one to When Primary channel is Equal to 1017. 1005 means the node is stopped, and 1017 means the node is suspended. All other values are either OK, or 0 if the node doesn't exist.

That's it.

Here's a helpful PowerShell command you can run to read all the NLBS node values using WMI: Get-WmiObject -Namespace "root\microsoftnlb" -Class MicrosoftNLB_Node -Impersonation 3 -ComputerName mywebserver You can use this to read the actual server name NLBS is looking for, as that must match exactly for the PRTG sensor to work. To use this PowerShell command, change the ComputerName to be correct for you, and paste it into PowerShell. You must have administrative rights on the remote server for this to work.

Created on Nov 9, 2012 6:00:26 PM

Last change on Nov 9, 2012 6:00:26 PM



Votes:

0

Thank you so much Jeffrey. Great job and I really appreciate the information.

Created on Nov 12, 2012 2:13:01 PM



Votes:

0

Thanks for this, it's great, I'm going to try it. We have a prtg-corenode cluster, and I've put the "LBSNodeCheck.wql" in the WMI WQL Scripts folder, but I'm yet unable to select it.

Questions: I'm assuming the service will need a restart to be able to pick it up? Question is, which of the two services would need restarting? PRTG Core Server Service PRTG Probe service Assuming the Core Server Service.

Anything I need to take into account restarting this service? We have a two-node PRTG setup (cluster) Should I place the current node in maintenance mode first before restarting it? I'm assuming it's going to failover to the failover node when I do.

Created on Feb 27, 2015 1:50:41 PM



Votes:

0

Dear JD

As first step, please check if you copied the script to both the master and the failover core. Please check if you can select the script when it is physically present on both machines.

Created on Mar 3, 2015 5:13:48 PM by Arne Seifert [Paessler Support]




Disclaimer: The information in the Paessler Knowledge Base comes without warranty of any kind. Use at your own risk. Before applying any instructions please exercise proper system administrator housekeeping. You must make sure that a proper backup of all your data is available.