I've just converted my VMs from Workstation to ESXi5 and set the drives to be Thin during the conversion.
I have 3 VMs in total and I wanted to back them up after the conversion.
I went to the Datastore Browser to download them and looking at the first machine, the size is around the 2gb and the provisioned size is 20GB.
It seems to have taken a long time and it started making me worried that when I back it up it'll have to download the 20gb file size.
That's a big worry because if that's the case, my other two machines have disks of 100gb - 1000gb.
That would take forever and be really complicated to backup.
Is there also any free tools to use for backup that are better than just downloading the machine from the datastore browser?
when I click on the VMDK file in the datastore browser there's an option to inflate, what's that? Does that optimize the disk and reduce the disk size?
If so, I'm guessing it's best to backup first right?
Thank you in advance,
Inflate option is for converting thin disks to thick. It will not save the storage space. VMware has backup solution for vSphere environment name vSphere Data Recovery. It comes by default with Stadart, Enterprise and Enterprsie Plus license packages. You can find more details on VDR at the following link. http://www.vmware.com/support/vdr/doc/vdr_200_releasenotes.html#supported
seems like an interesting tool.
ovf is just for packaging? do you set the source and target and then set the target to the external source and by that downloading your backup?
what are the chances of an OVF going wrong and the backup not working?
is it more time consuming backing up and downloading using OVF or once you get the hang of it, it's as quick?
my main concern is the extra disk space on the vmdks that is not being used in the drive and therefore having a huge download. does it avoid that situation?
i'd probably be most interested to understand how to download my vm without having to download all the extra disk space that is not being used so that backing up doesn't take too long.
some of my vms for example are set to 20gb or 100gb but are in fact only using about 1-2gbs. therefore backup should be very quick, but at the moment i'd have to download hundreds of gbs.
that's probably the solution i'm looking for if there is one.
i also can't seem to find a way to split the disk so that it's breaks the download to multiple chunks, is there a way to do that?
do i set my drive to be 100gb instead of 1tb and then just keep extending and it'll create a new vmdk to follow it?
thank you for all the replies so far!
wow that's fantastic. a drive that claimed to be 8.5gb used and 100gb provisioned size was exported into a 4.5gb which is probably the part that was used.
how safe is it with restoration?
that's a perfect solution for me for backups if that's safe.
my other concern is mounting a drive into a linux vm and doing a quick format without allocating all the space.
how do i create a big drive such a 1tb drive but know that for backup, it may be preferable for the esxi to split the disk?
thanks again for all the replies so far!