Good afternoon,
There is a problem which I have noticed only occurs during the weekends.
The problem: All or just some CSVs go into the offline state only on the weekends and this of course brings down all virtual machines as they depend on the vhds that are stored on those CSVs.
This is the 2nd weekend where the CSVs have gone offline. There are no problems bringing the CSVs back online and it takes me about 20 minutes to get all the VMs running again but this is horrible if I have deal with this every weekend.
I have looked through the Event logs and there is nothing which points to an actual problem. Errors in EV include live migration failed errors and hyper-v host and cluster errors. These errors all take place at 3am and are related to the CSVs going offline. The Hyper-V witness is one the CSVs that go offline.
I have a feeling that there might be something happening on the Compellent which hosts the CSVs but I would like to know your thoughts on this. Have you experienced something like this, where the CSVs go offline during the weekends but not during the work day?
This could also be a networking error between the cluster nodes and the Compellent or drivers? Maybe or maybe not.
The cluster nodes are all R720 with exactly the same hardware configurations and they are all running Windows Server 2012 R2 and all OS updates have been applied.
Any assistance and suggestions would be helpful.
Thanks in advance!