VMware Cloud Community
USGLOBAL
Contributor
Contributor
Jump to solution

ESXi 4.1 server with 2.7TB RAID only has a 750GB datastore

I have an ESXi 4.1 Build 260247 server. It has an Adaptec 5805Z controller with a hardware 2.7 TB RAID6. The ESXi is installed and running off a USB stick. The server runs great and have no issues with the performance but what I failed to notice (after creating and configuring 7 servers for an Exchange 2010 lab) is that the datatstore1 is only 740.75GB. I have looked everywhere in vSphere and on line on how to access the rest of this storage. Under configuration>hardware>storage it shows the "Local Adaptec Disk (mpx.vmhba2:C0:T0:L0)",  Type = Disk Transport Parallel SCSI, Capasity 2.73 TB.

How can I access the additional disk storage? I certainly don't want to have to reinstall the LAB if i don't have to. Attempts to increase it's size in the datastore properties has not been successful. No errors it just doesn't do anything of find anything...

VM Issue.jpg

VM Issue2.jpg

Reply
0 Kudos
1 Solution

Accepted Solutions
a_p_
Leadership
Leadership
Jump to solution

Creating logical volumes on the RAID is a function/feature of the RAID controller, respectively the RAID controller's software. Most RAID controllers allow splitting the RAID set (unless it is an x0 RAID level) into multiple logical volumes. You will have to verify this for your RAID controller.

Most likely you will have to migrate the VMs off the current 750GB datastore and delete the datastore. Then delete the current logical 2.75TB volume and create new logical volumes - max. 2TB - 512 Bytes - using the RAID controller's software or BIOS.

André

View solution in original post

Reply
0 Kudos
9 Replies
a_p_
Leadership
Leadership
Jump to solution

Due to the SCSI-2protocol, ESXi currently only supports LUNs up to a size of 2 TB minus 512 Bytes. What you need to do is to split the RAID into multiple logical volumes.

André

PS: see http://kb.vmware.com/kb/3371739

Reply
0 Kudos
a_p_
Leadership
Leadership
Jump to solution

Btw. depending on the virtual disk sizes you are planning to create, you may choose a larger block size (e.g. 8MB) when formatting the VMFS datastore.

see e.g. http://kb.vmware.com/kb/1003565

André

USGLOBAL
Contributor
Contributor
Jump to solution

What is the best path to do the logical drives? I would like to avoid reinstalling the 7 servers if at all possible. Can I use a partitioning software to resize the partition or is it feasable to copy the VMs off the RAID repartition it and copy them back? either way it looks like i have my work cut out for me... Any advice would be helpful.

Reply
0 Kudos
a_p_
Leadership
Leadership
Jump to solution

Creating logical volumes on the RAID is a function/feature of the RAID controller, respectively the RAID controller's software. Most RAID controllers allow splitting the RAID set (unless it is an x0 RAID level) into multiple logical volumes. You will have to verify this for your RAID controller.

Most likely you will have to migrate the VMs off the current 750GB datastore and delete the datastore. Then delete the current logical 2.75TB volume and create new logical volumes - max. 2TB - 512 Bytes - using the RAID controller's software or BIOS.

André

Reply
0 Kudos
USGLOBAL
Contributor
Contributor
Jump to solution

André,

Thank you for your help. I have reconfigured the array and have regained a mjority of my disk space. I lost some space due change from a RAID6 to a RAID10. So here is where I am at. I recreated the datastore and copied the VM machines (files) back to that store but the ESXi server doesn't recognize the machines (inventory shows Unknown inaccessible) and I cannot find an import option to put those VMs back into play. Is there something I am missing?

Thanks again for your help!

Rick

VM Issue3.jpg

Reply
0 Kudos
a_p_
Leadership
Leadership
Jump to solution

I lost some space due change from a RAID6 to a RAID10

Doesn't the RAID controller have an option to create multple logical volumes on the RAID6?

Anyway, regarding the inaccessible VMs. It look like you did not unregister the VM's prior to delete the datastore. What you should do now is to right click each of the VM's in the inventory and select "Remove from Inventory". Then open the datastore browser and right click each VM's vmx file and select "Add to Inventory".

André

sjumma
Contributor
Contributor
Jump to solution

I have something similar running 3 guest (exchange, FIleserver & domcain controller)

My exchange is cdrive is running out of space

1)windows 2003 64 bit R2
2) exchange 2007

for a guest

Exchange

c drive 35 g (full 33 gb needs space)
E drive 350 gb (5 gb full rest empty but already fully allocated )
vsphere client 4.1
vmware 4.1
On my host i have 1.6 TB allocated and 175 gb free but cannot allocated to one of the 3 host (exchange)
to extend c drive
image1.png
image1.png
image2.png
Reply
0 Kudos
a_p_
Leadership
Leadership
Jump to solution

Welcome to the Community,

if understand this correctly the issue is different. The original poster had an issue with the 2TB limit of a datastore.

Please consider reposting your question in the appropriate forum VMware vSphere Hypervisor and explain what exactly you are trying to expand. The VMFS datastore or the virtual disk for your Exchange server.

André

Reply
0 Kudos
USGLOBAL
Contributor
Contributor
Jump to solution

André,

Your advice was spot on! thank you for all your help!!

Reply
0 Kudos