[one-users] Installation From ISO images

[Viji V Nair]

Hi Ruben,

Thank you so much for your response

I have followed the steps and there are some issues. When I try to do a
"onevm create solaris10.one", both the images (sol.img and
osol-0906-x86.iso) are being copied to the shared storage as
/srv/cloud/images/<vmid>/disk.0 and /srv/cloud/images/<vmid>/disk.1. Then
the it tries to boot from the hard drive, I mean disk.0 (sol.img, hda). From
the VNC console I could see

    1. disk.0 is detected as ata0 - 2G in size (this is the image, sol.img,
created using dd)
    2. disk.1 is detected as ata1 - 677M (this is the iso image)
    3. a cdrom device

VNC is showing the boot error, tried to select F12 and selected cdrom, but
no luck.


# cat solaris10.one
NAME   = solaris10
CPU    = 0.2
MEMORY = 1280

DISK = [
 source   = "/srv/cloud/images/imgs/sol.img",
 target   = "hda",
 save     = "yes",
 clone    = "no" ]

DISK = [
 source   = "/srv/cloud/images/isos/osol-0906-x86.iso",
 type     = "cdrom",
 target   = "hdb",
 clone    = "no" ]

NIC = [ NETWORK="Pcloud Network01"]

GRAPHICS = [ type = "vnc", listen  = "node01.pcloud.com", port = "5902"]

REQUIREMENTS = "HOSTNAME = \"node01*\""


Thanks & Regards

Viji

On Fri, Jan 22, 2010 at 4:44 AM, Ruben S. Montero <rubensm at dacya.ucm.es>wrote:

> Hi,
>
> The procedure should be as follows, create a template that contains :
>
> 1- Define your main disk as usual, and tell OpenNebula to keep the changes
> DISK = [
>  source   = "/srv/cloud/images/master/fedora_disk.img",
>  target   = "hda",
>  save     = "yes"
>  clone    = "no" ]
> (this assumes that you have a plain file for fedora_disk.img (e.g.
> with a dd). You can also use separate partitions for swap and the OS
> installation...
>
> 2- Define an additional DISK for the ISO image with the installation CD
> DISK = [
>  source   = "/srv/cloud/images/isos/fedora_12.iso",
>  type     = "cdrom",
>  target   = "hdb",
>  clone    = "no" ]
>
> 3.- Add a VNC console so you can interact with the installer. And tell
> to listen in a cluster node hostname, by adding to the template:
> GRAPHICS = [ type = "vnc", listen  = "a_cluster_node_hostname", port =
> "5902"]
>
> 4.- Tell OpenNebula to place the VM in that clusternode, by adding to
> the template:
> REQUIREMENTS="HOSTNAME=a_cluster_node_hostname"
>
> Once you have the template ready....
>
> 1.- Create the VM with onevm create
> 2.- use you favorite VNC client to install the OS (you should be able
> to see the console at c<cluster_node_name>:5902) NOTE: You can always
> use 127.0.0.1 for the VNC server but then you have to start the VNC
> client from the cluster node, but we usually do not have an X server
> there...
> 3.- When you are done shutdown the VM. onevm shutdown
> 4.- The new installed image would be in
> $ONE_LOCATION/var/<VM_ID>/disk.0 Copy it to a safe location so you can
> use it later as a master image.
>
> Hope it helps
>
> Ruben
> On Thu, Jan 21, 2010 at 5:35 PM, Viji V Nair <viji at fedoraproject.org>
> wrote:
> > Hi,
> >
> > I have a two node cluser with OCFS2 running on OpenNebula 1.4 and Fedora
> 12
> > x86_64. Is it possible to start installation of vms directly from an iso
> > image other than running virt-install to create the image and then deploy
> > the same.
> >
> > Any help of the same greatly appreciated.
> >
> > Viji
> >
> > _______________________________________________
> > Users mailing list
> > Users at lists.opennebula.org
> > http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
> >
> >
>
>
>
> --
> Dr. Ruben Santiago Montero
> Associate Professor, Complutense University of Madrid
>
> URL:    http://dsa-research.org/doku.php?id=people:ruben
> Weblog: http://blog.dsa-research.org/?author=7
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opennebula.org/pipermail/users-opennebula.org/attachments/20100122/beb5124c/attachment.htm>