So, I'm new to the VMware world and have a few questions regarding having a hot spare setup in case something crashes.
There are two main things I want to know...
1.
We have a Vmware box currently set up with 2 VM's on it(I didn't set it
up). We would like to have a spare server ready in case that crashes,
but without spending too much money. Would it be better to set up ESXi
on the spare and, in case of failure, use the images from the Main box
to get them back up temporarily, or to have a spare server and just
take the drives from the downed server and drop them in the
spare(assuming it wasn't a drive failure). Would either of those work?
2.
Is it possible to take snapshots of non-VM machine and then use those
images in VMware to get it back up in a virtual enviroment until the
physical server is repaired? Just curious.
Any tips or advice would be greatly appreciated.
1.
We have a Vmware box currently set up with 2 VM's on it(I didn't set it
up). We would like to have a spare server ready in case that crashes,
but without spending too much money. Would it be better to set up ESXi
on the spare and, in case of failure, use the images from the Main box
to get them back up temporarily, or to have a spare server and just
take the drives from the downed server and drop them in the
spare(assuming it wasn't a drive failure). Would either of those work?
Your best bet would be to have ESXi setup on the other server and have images of the VMs running on you existing server that are taken periodically. I assume from the tone of email you do not have an overabundance of funds to work with, there fore shared storage, clustering and Snapshot based backups would be out.
If you count on being able to recover the VMs from the drives on the existing system you might be in for a surprise.
2.
Is it possible to take snapshots of non-VM machine and then use those
images in VMware to get it back up in a virtual enviroment until the
physical server is repaired? Just curious.
Short answer No.
Long answer you can do a couple of things that may not work. You may be able to occasionally ghost the physical systems and then you may be able to recover the system as a VM from the ghost image. You might be able to restore a backup image to a VM and then repair it to a working state.
There are a number of options available and i am sure some other folks will pile on shortly
1.
We have a Vmware box currently set up with 2 VM's on it(I didn't set it
up). We would like to have a spare server ready in case that crashes,
but without spending too much money. Would it be better to set up ESXi
on the spare and, in case of failure, use the images from the Main box
to get them back up temporarily, or to have a spare server and just
take the drives from the downed server and drop them in the
spare(assuming it wasn't a drive failure). Would either of those work?
Your best bet would be to have ESXi setup on the other server and have images of the VMs running on you existing server that are taken periodically. I assume from the tone of email you do not have an overabundance of funds to work with, there fore shared storage, clustering and Snapshot based backups would be out.
If you count on being able to recover the VMs from the drives on the existing system you might be in for a surprise.
2.
Is it possible to take snapshots of non-VM machine and then use those
images in VMware to get it back up in a virtual enviroment until the
physical server is repaired? Just curious.
Short answer No.
Long answer you can do a couple of things that may not work. You may be able to occasionally ghost the physical systems and then you may be able to recover the system as a VM from the ghost image. You might be able to restore a backup image to a VM and then repair it to a working state.
There are a number of options available and i am sure some other folks will pile on shortly
Thanks for the advice. We are somewhat limited on cashflow, but not entirely. We don't want to spend a lot of money on an expensive high availability situation, just something that can temporarily get up without too much delay until the main server is back.
We had a scare this weekend where somehow the power AND a UPS went out, resulting in ESX not shutting down properly and the linux kernel not booting back up properly. We had to reinstall the base system, but all the images were fine. We're trying to take precaution for next time.
I'll look into some of the options you suggested. Thanks!
I would suggest setting this up so that you can back up another physical server that will go live as soon as the primary fails.
That would give you the most accurate data of using ESXi would allow you to keep an image of the drive very up to date and allow
for almost immediate ability to be up again.
For your second question just replacing the drives may not work so well you may have issues and problems with
the differences between servers and the configurations in both. Also the number 1 failed parts tend to be fans or drives so I wouldn't rely on
using the drives to get you back up fast and reliably.