[Network Load Balancer] Is there a way to make it aware of Application failures ?

Copper Contributor

Hello dear Tech Community,

I just started to explore NWLB feature of Windows Server so please forgive my very poor knowledge on this IT-oriented topic!

My goal is to load balance TCP IP traffic over a cluster of, say, 2 servers for a particular Windows service using port 1234.

This service is a custom one developed in my company to expose a particular server application and using it on a single server is not an ideal when the number of consuming clients is exceeding 20 or so.

I was able to test LB nominal cases (where both nodes/hosts are up and running), and also to verify that, when one node is down, the LB is still running and serving clients with the other node.

But what if the node is up but the application "behind the port" is failing to respond?

It seems to me NWLB is totally blind about this case and continue to wait forever.

So my question is, is there a way to provide a kind of application-custom health probe to NWLB so that it can discover the node is actually not available for further requests?

Many thanks for reading me and looking forward to get your insights !

1 Reply

Update to my previous post:

I have seen that, using WMI and NWLBProvider classes, I might be able to influence NWLB nodes/cluster "in real time" but at the cost of a ad-hoc piece of software development, that would regularly check if the application windows service is started/running.

I am wondering if there is no "on-the-shelf" solution for this windows service monitoring purpose?