Link to home
Start Free TrialLog in
Avatar of klsphotos
klsphotos

asked on

Hyper V Cluster/Storage Question

Hi Experts,

I hope you can help me and I hope this question makes sense:

We recently had a critical storage issue documented in this thread:
https://www.experts-exchange.com/questions/28998074/Hyper-V-virtual-disk-error.html

The issue ended up being the 2 TB limitations on that particular drive, 2 different back ups being kicked off at the same time.

When this cluster was originally built, we had only one ISCI NAZ and 2 nodes.
I have several shared cluster volumes built from this (smb) NAZ inside the cluster.

The problem is we are a year later and I would like to have fail over for the nodes but we have utilized all available resources on both nodes and our storage for several different servers and they went into production.  We had projects that happened so rapidly, what was testing is now live and critical.  

I am at a loss at where to go from here to correct this.

WE do not have the resources currently to host this cluster the way it's suppose to work, they would seem to work better working as stand alone Hyper V host with the virtual currently on them, but then can I still share the storage that isn't CSVs?

Can I remove the nodes and storage from the cluster without killing the VMS or just remove this cluster?

I feel like I am going in circles trying to keep resources online that are going down related to cluster events that are not configured properly.

I hope that made sense, thank you for your help.
Avatar of Philip Elder
Philip Elder
Flag of Canada image

Add more nodes to the cluster.
Avatar of klsphotos
klsphotos

ASKER

I don't understand, my question was "Can I remove the nodes and storage from the cluster without killing the VMS or just remove this cluster?"

I want to use the servers as stand alone and still use the ICSCI storage for the VM's not clustered volumes.  We have removed all data off the clustered volumes.
ASKER CERTIFIED SOLUTION
Avatar of Philip Elder
Philip Elder
Flag of Canada image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Make sure to remove the shared LUN setup on the storage so that only one node has access to each LUN to avoid data corruption.

We have been having a lot of weird storage issues, currently both Nodes have access to each LUN, we already have had data loss, is this why?  I have since set up new targets that only one node has access to but the early ones have all nodes connected.
Yes. Without the cluster service to arbitrate storage calls between nodes connected to the same LUN bad things can happen.