• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Hyper-V Cluster Question

M

mattsnetwork

n00b
Joined
Apr 22, 2015
Messages
1
Hi All,

I am trying to help a small school re-use some hardware and build them an Hyper-V cluster.

I have two HP EasyStore NAS running Storage Server 2012.
I have two HP Proliant Servers Running Windows 2012 R2.

They are only running about 8 - 10 VM's. All very low usage.

I have only built Hyper-V cluster with shared storage using ISCSI. Any pointers on how to create one here? Is there a way to have the data on both of NAS and use a fail over File Server cluster to access them in case one of the NAS goes offline?

With 2012 it seems you can build a Hyper-V cluster using SMB, but it seems to get murky on how to setup the back end.

Has anyone done something similar or have any links I can look to?

Thanks
Matt
 
In order to create a Scale Out File Server cluster (SOFS) your two file server nodes would need shared disk. Sounds like they each have their own so you won't be able to create a SOFS cluster.

However, you can still have the VMs failover by clustering the two Hyper-V hosts together and pointing them to each separate EasyStore NAS for shared storage. If a Hyper-V host fails, the VMs on it will automatically fail over to the other node.

Another option would be to not cluster anything together and run half your VMs on Hyper-V host 1 and NAS 1 and the other half on Hyper-V host 2 and NAS 2 and replicate each to the other. This would double your storage needs, but would provide the ability to failover should a Hyper-V host or storage node go offline. The failover would not be automatic, it would require you to log in and initiate the failover but you get the added benefit of storage protection as well as host failure protection. Hyper-V replication can sync as often as every 30 seconds so that's the most amount of data you'd lose.
 
If you go the hyper-v replication route, just watch it closely. When the replication fails, it just stops. There may be an event log or something you can look for.

Once it fails, you have to tell it to continue replicating. Otherwise your VM on the replication partner will be a flashback to when replication stopped.
 
it's absolutely possible

you need some windows based virtual storage running on your sofs nodes and creating virtual luns with csv on top of them

starwind can do it and also datacore

http://www.slideshare.net/StarWindSAN/ms-scale-out-file-servers-without-sas-jbo-ds

however there's no point in basically wasting two servers

i'd build a 4 node hyper-v cluster and use two hypervisor nodes to run free starwind or datacore virtuaal san software

in upcoming version of windows server there would be storage spaced direct and it would take four servers to create a hyper-converged setup similar to eve:rail from vmware

Child of Wonder said:
In order to create a Scale Out File Server cluster (SOFS) your two file server nodes would need shared disk. Sounds like they each have their own so you won't be able to create a SOFS cluster.

However, you can still have the VMs failover by clustering the two Hyper-V hosts together and pointing them to each separate EasyStore NAS for shared storage. If a Hyper-V host fails, the VMs on it will automatically fail over to the other node.

Another option would be to not cluster anything together and run half your VMs on Hyper-V host 1 and NAS 1 and the other half on Hyper-V host 2 and NAS 2 and replicate each to the other. This would double your storage needs, but would provide the ability to failover should a Hyper-V host or storage node go offline. The failover would not be automatic, it would require you to log in and initiate the failover but you get the added benefit of storage protection as well as host failure protection. Hyper-V replication can sync as often as every 30 seconds so that's the most amount of data you'd lose.
Click to expand...
 
KISS

Olga-SAN said:
it's absolutely possible

you need some windows based virtual storage running on your sofs nodes and creating virtual luns with csv on top of them

starwind can do it and also datacore

http://www.slideshare.net/StarWindSAN/ms-scale-out-file-servers-without-sas-jbo-ds

however there's no point in basically wasting two servers

i'd build a 4 node hyper-v cluster and use two hypervisor nodes to run free starwind or datacore virtuaal san software

in upcoming version of windows server there would be storage spaced direct and it would take four servers to create a hyper-converged setup similar to eve:rail from vmware
Click to expand...
 
You must log in or register to reply here.
Back
Top