Issues with cluster shared storage adding disks?

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

Diavuno

Active Member
I'm playing with some equipment an ran into an interesting issue... something new I've never seen or heard of.

Two X8 systems (fat twin) each with:
LSI 8E HBA's flashed to IT mode,
2012R2 fully updated.
On My Domain
8088 into a DataOn 2.5"JBod.
both clustered with HyperV and storage roles installed.

Currently they have nothing on them do only 1GBE on a dumb switch that's shared to a separate DC

Seems like a good setup a shared pool of disk and some HA VM's right...?

Problems arise when I run cluster validation, few minor warnings (I'll post them in next post)

in particular the disks are visable from each node and seem to be OK, but when added to clustered storage they are marked "failed" but I know there is NO WAY they are all bad.

there are 4x PERFECT S3500's
4x 32GB Intel SLC's
4x 500GB sata
2x4TB


My end goal is to populate like so:
2x 2TB 4x4TB 3.5's per node
This will be for Boot and archive. (I might even put in 4x 8TB archive drives)
8088's to the JBOD for VM storage as well as my exchange and a few DB' used in house.
the Jbod will have :
4x s3500, I might up it to 8 units
Maybe? the 4x 32SLC's for hot writes.
then I'll fill the rest with the piles of 1TB spinners I have, eventually going to 4TB or SSD's as needed.
 
Last edited:

Diavuno

Active Member
Cluster Configuration

Name Result Description

Cluster Core Groups Pass

Cluster Network Information Pass

Cluster Resources Pass

Cluster Volumes Pass

Clustered Roles Pass

Quorum Configuration Warning

Resource Status Warning

Service Principal Name Pass

Volume Consistency Pass




Hyper-V Configuration

Name Result Description

Hyper-V VM Information Pass

Information About Servers Running Hyper-V Pass

Compatibility of Virtual Fibre Channel SANs for Hyper-V Pass

Hyper-V Integration Services Version Pass

Hyper-V Memory Resource Pool Compatibility Pass

Hyper-V Network Resource Pool And Virtual Switch Compatibility Pass

Hyper-V Processor Resource Pool Compatibility Pass

Hyper-V Role Installed Pass

Hyper-V Storage Resource Pool Compatibility Pass

Hyper-V VM Network Configuration Pass

Hyper-V VM Storage Configuration Pass

Matching Processor Manufacturers Pass




Inventory

Name Result Description

BIOS Information Pass

Environment Variables Pass

Fibre Channel Host Bus Adapters Pass

iSCSI Host Bus Adapters Pass

Memory Information Pass

Operating System Information Pass

Plug and Play Devices Pass

Running Processes Pass

SAS Host Bus Adapters Pass

Services Information Pass

Software Updates Pass

System Drivers Pass

System Information Pass

Unsigned Drivers Pass




Network

Name Result Description

Network Binding Order Pass

Cluster Network Configuration Pass

IP Configuration Pass

Multiple Subnet Properties Pass

Network Communication Warning

Windows Firewall Configuration Pass




Storage

Name Result Description

Disks Pass

Disks To Be d Pass

CSV Network Bindings Pass

CSV Settings Pass

Disk Access Latency Warning

Disk Arbitration Warning

Disk Failover Warning

File System Warning

Microsoft MPIO-based disks Warning

Multiple Arbitration Warning

SCSI device Vital Product Data (VPD) Warning

SCSI-3 Persistent Reservation Warning

Simultaneous Failover Warning

Storage Spaces Persistent Reservation Warning




System Configuration

Name Result Description

Active Directory Configuration Pass

All Drivers Signed Pass

Cluster Service and Driver Settings Pass

Memory Dump Settings Pass

Operating System Edition Pass

Operating System Installation Option Pass

Operating System Version Pass

Required Services Pass

Same Processor Architecture Pass

Service Pack Levels Pass

Software Update Levels Warning

System Drive Variable Pass




Overall Result

Testing has completed Passfully. The configuration appears to be suitable for clustering. However, you should review the report because it may contain warnings which you should address to attain the highest availability.
 
Last edited:

markpower28

Active Member
Apr 9, 2013
413
104
43
The issues for sata drive is they can not handle scsi reservation, means in a cluster environment they will always fight for the resource and cause the storage to go offline. It should work as a standby node. When the primary go offline, you can always mount the volume to the new node. You need to share the folder again. It should retain ntfs permissions.
 
  • Like
Reactions: Chuntzu

markpower28

Active Member
Apr 9, 2013
413
104
43
Any reason ha is required on the storage level?

If you want the ability to move vm between 2 nodes on hyper-v, maybe one node as storage and both can be hypervisors.