There are some Windows Server failover clusters in our environment that are hosted in Azure. I have a problem that I do not seem to find information about.
Currently, the clusters work fine unless we can not connect to the cluster name (through PSRemoting, Failover Cluster Manager, and so on). I am aware that these names are not intended to be pingable, but should be available to the administration. This leads to a situation in which we can only manage the cluster if we already know the node names and establish a direct connection to one of them.
We probably have more than 40 clusters in our area, of which only four are in Azure. The Azure clusters are the only ones with this problem, and everyone has this problem, so this is a common denominator.
I'm sure this is a configuration issue that just was not done, but I do not know the Azure architecture well enough to understand what the problem is and what needs to be fixed. The next thing I found is on the clusteringformeremortals website, which points to a possible need for load balancers. However, this is specific to SQL clusters (our SQL cluster listeners work fine).
This problem increases management overhead and causes automation problems.
Any help in this matter would be fantastic.