• 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.

Migration To ESXI

A

andymiller6891

n00b
Joined
Dec 27, 2013
Messages
38
currently I run hyper v 2012, have been on windows servers since I saw the need for one.

due to constantly changing hardware ive decided that switching to esxi should make hardware swap out more simple.

but I have a few questions.

firstly I have an adaptec 6805, 6x 2tb sammys running raid 5, 3 virtual drives.

im guessing ill have to move it all for a reformat, which will be a struggle without dropping the parity drive.

so what is the best way to approach building the array?? whats is the best way to build and manage the data stores?? how to I go about replacing a dead drive, adding another drive or as i have future plans to swap out to 8x 3tb wd reds.

data is my main concern as i have stuff that is personal, photos, home movies etc. i have backups of course but i don't want to be redoing them once a week due to unreliability.

esxi will be held on a 40gb corsair ssd i have spare, vm's will be held on a 2x 500gb mirror.

i have a vm for a dc/dns/dhcp, a vm for cctv, a vm for pf sense and lastly a vm for media streaming.

hardware is currently i5 2500k, 16gb ram, onboard gb lan, dual port intel server lan, onboard port is wan in, others are teamed into lacp currently.
 
Is the data and the OS on separate drives? Why do you need to rebuild the array? Are you using Storage Spaces or something?

I don't see why you can't just do V2V migrations to ESXi formats on the drives as they are now and then install ESXi on the OS drives then import those V2V disks?

Your explanation of your situation is terrible, not a very clear picture of what you're trying to accomplish and how you currently have it set up.
 
ok ill try my best to explain current setup.

I have windows hyper-v and 4 vm's stored on a 500gb mirrored drives, 40gb os, 40gb per vm.

on the raid I have 6x 2tb disks, 3 virtual drives. 4tb for vm1, 4tb for vm2 and 1tb for vm3.

windows sees them as 3 separate hard disks not 3 partitions.

I considered pass through but didn't know if I could pass the drives back through to the vm's they need to be connected to or would I have to pass the whole card to 1 vm.

I don't really know how datastores etc work on esxi.

my aim is to have a virtual host I can pull a mainboard from at any time, with windows hyper-v I cannot do this, it requires host reload, vm imports, lots of downtime.
 
my main concern is the ability to control the raid arrays through esxi, incase I need to yank a drive, rebuild, expand etc, can I install the management utility to 1 vm and will it pick up the controller through esxi??
 
Also when you say drives do you mean volumes? Do you mean virtual disks created in Storage Spaces?

I still can't make sense of your wording. You have a 6x 2TB RAID5....I got that much...the rest makes no sense at all. I can't tell what is physical and what is logical in your explanation.

"on a 500gb mirrored drives" Mirrored how? Where did the 500Gb come from? Did it come from the pool of 6x 2TB RAID5 or are these separate drives?

on the raid I have 6x 2tb disks, 3 virtual drives. 4tb for vm1, 4tb for vm2 and 1tb for vm3

Does this mean that you've created 3 volumes from the pool? If so are you using passthrough to give them to the VMs or just using them as three different volumes in Windows.

Sounds like a way over complicated setup.
 
what i mean is on the raid controller utility it gives you the option to make an array and split it into logical drives, which the os reads as physical disks not partitions, i have all 6 disks configured into 1 pool and split to 3 logicals, my vm hosts reads 3 separate hard disks, which i take offine and add as a physical drive to each required vm.

not storage spaces or any windows software implemented raid config, all done through the hardware controller.

the 500gb mirror is currently done from the onboard software raid, being a mirror it should transfer from mainboard to mainboard aslong as they have onboard raid 1, which most do, will just be re-mirrored as required or possibly managed by esxi, making hardware swap out easier.

what i hoped to be able to do is to install something to esxi which will allow me to manage the raid controller from one of my vm's.

the vm's i hope to convert from hyperv to esxi and just import into the host.

hopefully that makes more sense.
 
Last edited:
Yeah that makes a lot more sense. If the OS is separate then you shouldn't have to do anything with the current RAID, just leave it alone. But again, managing the RAID is going to be specific to your raid controller. You'll want to Google your RAID controller + EXSi + Management

I'm not sure how you would migrate those hosts though...you'll have to do some kind of V2V from Hyper-V to VMWare. That's probably going to require at least 50% free space and somewhere to store them. I've never tried to do a V2V with VMs that have pass through HDDs...

One option would be to convert from passthrough to VHDX and then V2V that.

http://www.petri.co.il/convert-hyper-v-pass-through-disk-to-a-vhdx.htm
 
I found one article that says "Neither VMWare or Hyper-V adds any hypervisor information on a passed through disk, that I'm aware of."

That would suggest that you MAY be able to just install ESXi, create a VM for each machine with the same settings and attach the disk. Not sure how the guest would handle that though as they have different drivers and whatnot...
 
i would prefer to keep the logical drives on the raid card as pass through drives to the vm's as opposed to virtual storage, preference being to keep the drives in ntfs so they can be mounted to any windows os in the event i need to backup or move around.

all i really need to worry about is converting my 2 main 2008r2 vm's to esxi to avoid having to reload my domain server and mess about pulling all my machines in and out of domains and new group policies. also have a lot of mac management done to control scheduling of kids internet access.
 
andymiller6891 said:
i would prefer to keep the logical drives on the raid card as pass through drives to the vm's as opposed to virtual storage, preference being to keep the drives in ntfs so they can be mounted to any windows os in the event i need to backup or move around.

all i really need to worry about is converting my 2 main 2008r2 vm's to esxi to avoid having to reload my domain server and mess about pulling all my machines in and out of domains and new group policies. also have a lot of mac management done to control scheduling of kids internet access.
Click to expand...
Wouldnt your hardware need to be capable of doing passthrough?
 
my mistake, it is a k, didn't think they would put it in the none k and remove it from the k, what a waste of potential.
 
figured that.

have loaded esxi onto a 40gb ssd, have it running, plan to create rdm links to the physical disks.

problem is I need to ssh to create them, I have tried 2 different windows pc's and 2 different versions of putty, both will allow me to connect to the host but say access denied after I type my password using keyboard authentication, cannot get around it.
 
You must log in or register to reply here.
Back
Top