Storage Space Direct S2D using dissimilar physical nodes
Please, feedback about the limitation with S2D. Has anyone been able to configure S2D with nodes with different physical configuration? I am moving to a cluster with 4 servers that up to this day been online in production independently. Now, I wanted to configure a high availability using 3 of these units under a Storage Space Direct configuration and the 4rth server running the Active directory . I have been strugling with this and wanted to get feedback on my scenarios. The nodes are describes as follow. 2 units Dell R730XP and 1 node Dell R640. The 3 are running 2 CPU but not exarlty with the same amount of memory (526,480 and 256 GByte RAM). They also have desimilar HD disks in the local storage. We have configured the S2D following the limited documentation and Youtube. The 4th server is a Dell R620 running Active Directory . Everythign is Windows server Data Center edition These cluster run 6 Virtual machines running with HyperV and replicated with Fail Over manager. At first , the VM are running ok. But the problem comes when I try to do a test that enages moving a VM from one node to another node. This operation disrupts the status of the other VM to Pause or fail status. Only few of the VM return to Running status after several minutes. This is a terrible scenario and not ideal to provide the high availability environment. Please, feedback if there is a limitation with S2D. Has anyone here able to configure S2D with nodes with different physical configuration?
[Bug] Can't connect to cluster or node running WAC
Our environment is a Hyper-V hyper-converged S2D cluster running Windows Server 2016 Datacenter edition. WAC has been installed as a high availability deployment option called wac. WAC is currently running on node 1 in the cluster. When it is on node 1, connecting to node 1 in WAC is really slow and connecting to node 2 in WAC is really fast. If I failover WAC to node 2, then connecting to node 1 in WAC is really fast and connecting to node 2 is really slow. If I try and add the cluster by it's FQDN it says it cannot verify it's network. If I click the blue hyper-link to the cluster from one of the nodes via their WAC page it adds the cluster but it fails to load. It prompts for credentials, using my windows credentials or supplying alternative credentials does not work. It says that the process could not be completed. Kerberos delegation has been done as documented by MSFT for the SSO to work. Winrm is enabled and listening on HTTP on all IPs on all nodes. We have another cluster - cluster2 that is also Hyper-V hyper-converged running Windows Server 2016 Data edition. If I try and access that cluster or it's nodes from the same WAC install it works absolutely fine.Can't connect to Hyper-V Converged Cluster
I've raised a bug here - https://windowsserver.uservoice.com/forums/295071-management-tools/suggestions/37729852--bug-can-t-connect-ti-hyper-v-converged-cluste I've installed Windows Admin Center with high availability following https://docs.microsoft.com/en-us/windows-server/manage/windows-admin-center/deploy/high-availability. I can connect to nodes in the cluster fine through WAC, however when I try to add the cluster it says it can't find the name. If I click on one of the nodes it gives the FQDN of the cluster, I click it and it adds it as a connection but when I try to access it, it complains that WINRM couldn't not process the operation. I've tried it with two different clusters on our network. Any ideas?