Hi, i have a VMware image with two partition (2 .vmdk files), i started this machine with a .vmx file. It's possibile import this VM to proxmox?
I have read thid guide http://pve.proxmox.com/wiki/Migration_of_servers_to_Proxmox_VE, but i don't know which method to follow (clonezilla or import vmdk directly)
hi,
stop the vmware machine, then convert (directly through network or a local copy) the vmdk (the "full" one) to whatever format you need on proxmox ve.
remember to use mergeide.reg in vmware before, if converting a windows machine.
Marco
I have to use clonezilla?
all you need is to have vmdk on the proxmox node (or wherever qemu-img is)
so, if you can access the vmware vmdk in other ways (eg: network) from your proxmox node, no, you don't need to use clonezilla
just copy the vmdk to your proxmox node, in any way. then, use qemu-img to convert them to other formats usable by proxmox (btw proxmox can also use vmdk copied files directly, although is not the best format usable)
Marco
I have tried to copy and convert in .qcow2 with qemu-image a .vmdk file, but i have 2 file .vmdk, it's possible to have one sigle kvm with 2 vmdk files?
hi, sure.
the img file is the RAW file proxmox expects, when vm is configured to use it.
so, in the 100.conf file you will need a line like (eg)
ide0: pve_nfs_storage:100/vm-100-disk-1.RAW
you can use that filename (like created by pve) or another, at your taste.
Marco
if i create image with clonezilla, i can restore this image
...afaik, an "image" is a single file...
what you did exactly with clonezilla, and what you got from it?
Marco
-rw------- 1 stefano stefano 55771453 nov 20 12:46 sda1.ext4-ptcl-img.gz.aa
-rw------- 1 stefano stefano 1048064 nov 20 12:46 sda-hidden-data-after-mbr
-rw------- 1 stefano stefano 1048064 nov 20 12:46 sdb-hidden-data-after-mbr
Hi Marco,of course.
in proxmox vm (kvm) you can add several disks, in three formats, but you choice depends also on what type of "pve storage" you choose to host them on (eg: storage could be nfs, or iscsi/lvm, local and so on)
see http://pve.proxmox.com/wiki/Storage_Model
if your storage is lvm/iscsi, you can only choose raw, since lvm can host only filesystems, not files
if your storage can host files, you must use a a file, so you can choose between qcow2, vmdk, raw (each one have pros/cons)
for each disk you also have to choose a bus/deice type: vitrio, ide, scsi, sata (each one have pros/cons)
see http://pve.proxmox.com/wiki/Installation#Hard_disk
when you create a new vm (kvm) you *must* create at least an empty disk following infos above. You can add more, of course.
Once the vm is "saved" you will find all its info, including all hardware, in its .conf file, under /etc/pve/qemu-server/
like /etc/pve/qemu-server/100.conf, if your vm as VMID = "100"
if you look in the file, you will see that for each disk "created" there's a line like
ide0: pve_lvm_storage:vm-100-disk-1,size=32G
this means i'm using a 32GB disk on lvm storage "pve_lvm_storage", and disk data is on lv name "vm-100-disk-1", using bus "ide", and ide0 means is the first of the 4 (in this case, since is ide) channels allowed on this bus type
if you use a file, instead it will be like
virtio0: pve_nfs_storage:100/vm-100-disk-1.qcow,size=32G
or
virtio0: pve_nfs_storage:100/vm-100-disk-1.vmdk,size=32G
or
virtio0: pve_nfs_storage:100/vm-100-disk-1.RAW,size=32G
so if you have 2 .vmdk files, create the vm with a 1GB vmdk disk, then add another empty 1GB one, so your 100.conf will have lines like
ide0: pve_nfs_storage:100/vm-100-disk-1.vmdk,size=1G
ide1: pve_nfs_storage:100/vm-100-disk-2.vmdk,size=1G
To use your .vmdk you have to replace the above 1GB files with yours, renaming yours or adapting the .conf file to your names, if you get what I mean: you can't do this through the web gui, sadly, you must use the pve command line on a node.
Then, just check that boot order is right, so that your vm will boot from the right .vmdk
It should work, if you have windows on them you should have followed all notes about windows vms and mergeide
Marco
We use essential cookies to make this site work, and optional cookies to enhance your experience.
