Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

I have recently taken over as an administrator for an IBM chassis and associated blades, along with accompanying IBM SAN.

One of the blades is running VMWare ESXi 5, and we don't need it anymore, so I'm going to wipe it and install Windows Server

The blade has no onboard storage, and is setup to boot from the SAN. According to the SAN software, it has two 120 gig partitions and a 30 gig partition carved out.

I've planned to just use the existing storage setup for the machine, so wiping and installing new. I loaded Windows Server 2012 R2 setup and when I get to the bit where I'm choosing where to install windows, I get the following oddness (image inline below):

This seems strange to me, but I'm new to this role. Is this normal behavior for a setup of this type? Why are some of them showing as Offline? Can I safely format the ones listed as Primary and just roll with the install process?

Thanks!

share|improve this question

3 Answers 3

up vote 6 down vote accepted

The most likely reason for seeing all the drives twice is that you have multiple paths to the SAN but haven't installed the appropriate MPIO driver for it. There should be a driver for the SAN to install so it recognizes the drives properly.

share|improve this answer
    
Oooh - good eye. That could easily be the issue. –  mfinni May 12 at 21:31

Check your SAN zones. They probably had the ESXi server with access to the LUNs that are shared with other ESXi servers, for failover capability. Remove this non-non-ESXi server from any zones that are shared with other current ESXi servers. If you format those disks with NTFS and they're hosting live VMFS volumes, you're going to be having a very bad day.

share|improve this answer
    
This is the only ESXi server in the environment. It used to host a machine that has long since been put into Hyper-V elsewhere. –  jeremy May 12 at 20:59
1  
OK - regardless, it's a good idea to check and edit your zones, particularly when you see volumes attached to the server that you didn't expect. –  mfinni May 12 at 21:08

As @Grant mentioned, your issue is multiple paths to the same LUNs. I found a guide for installing Windows on a LUN from a SAN that recommends removing all the redundant paths until you can get into the OS and install the multipath software for the storage.

Not sure what kind of disk it is, but you could do this two ways. The first and easiest would be to temporarily deactivate some zones on your switches. The second way would be to go to the storage and remove the LUN masking from the server. If you tell us what kind of SAN it is, we could guide you through that second option, but the first one is pretty general and easy to undo once you're ready. You need to identify the zones in use that allow your blade to communicate with the storage ports, and then simply remove all but one of them from the active zone set on one switch, and all of them on the other switch. When you're done, you can readd them and bring everything back the way it was.

share|improve this answer
    
Yes, you can get in an awful mess if you try and 'use' both paths concurrently. You will generally get away with it if you're careful to only use one path - e.g. not even mounting the 'alternate'. –  Sobrique May 15 at 14:29

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.