Hi,
I have a question about designing a highly available file server; we are aiming at providing high availability for general documents, design files, spreadsheets and so on. Our environment uses a mix of Windows 7 and 8 and I am aware of the limitations with SMB transparent failover for those clients. Each node in the FS Cluster will be a virtual server (guest clustering). The host and the guests will be 2012 R2. I understand that there is now a feature for using a shared VHDX between two guest cluster nodes, I can see this being an advantage but my main concern is how to achieve high availability if that shared VHDX becomes corrupt.
Our host is configured using a virtual SAN which synchronises synchronously across the two physical nodes, this shared VHDX could exist on that virtual SAN sure, but what if the VHDX itself gets corrupted and can’t be “attached” to the virtual machine or is formatted?
Is there a way I can utilise both DFS and HA for file server availability, struggling to come up with a sensible design which doesn’t go crazy with the number of duplicated VHDX’s for resiliency (I think 2 VHDX is enough in this scenario)
Many thanks for any suggestions
Steve