[one-users] storage configured for nfs, using scp

Javier Fontan jfontan at gmail.com
Fri Sep 18 08:02:25 PDT 2009


Hello,

Deployment file is always copied using scp to the remote node. It is a
little file (no more that 5 Kb, usually a lot less) so it wont affect
a lot in the performance. It is done this way so it works for shared
and non-shared installations. The bulk of movements, that is, the
images, are moved using cp over nfs if you use nfs driver.

Bye

On Fri, Sep 18, 2009 at 4:51 PM, Michael W. Lucas
<mwlucas at blackhelicopters.org> wrote:
> Hi,
>
> I have a brand-new opennebula 1.2 cluster, one head node and two
> workers, running ubuntu 9.04 diskless.  NFS server is OpenSolaris.
>
> I want to use NFS as a common backing store.  My understanding is that
> this reduces the amount of time needed to copy images from one host to
> another.  When I deploy a machine, though, the log shows that the head
> is scping the image to the worker node.
>
> ...
> Thu Sep 17 14:41:53 2009 [LCM][I]: New VM state is BOOT
> Thu Sep 17 14:41:53 2009 [VMM][I]: Generating deployment file: /var/lib/one/4/deployment.0
> Thu Sep 17 14:41:53 2009 [VMM][I]: Command: scp /var/lib/one/4/deployment.0 clusternode1:/one_di
> r/one_images/4/images/deployment.0
> Thu Sep 17 14:41:54 2009 [VMM][I]: Copy success
> ...
>
> I've obviously made a bonehead mistake, any guidance would be
> appreciated.
>
> oned.conf is:
>
> HOST_MONITORING_INTERVAL = 10
> VM_POLLING_INTERVAL      = 10
> VM_DIR=/one_dir/one_images
> PORT=2633
> DEBUG_LEVEL=3
> NETWORK_SIZE = 254
>
> MAC_PREFIX   = "00:03"
>
> IM_MAD = [
>    name       = "im_kvm",
>    executable = "one_im_ssh",
>    arguments  = "im_kvm/im_kvm.conf",
>    default    = "im_kvm/im_kvm.conf" ]
>
> VM_MAD = [
>     name       = "vmm_kvm",
>     executable = "one_vmm_kvm",
>     default    = "vmm_kvm/vmm_kvm.conf",
>     type       = "kvm" ]
>
>  TM_MAD = [
>      name       = "tm_nfs",
>      executable = "one_tm",
>      arguments  = "tm_nfs/tm_nfs.conf",
>      default    = "tm_nfs/tm_nfs.conf" ]
>
> Each worker was added with the command
>
> onehost create <hostname> im_kvm vmm_kvm tm_nfs
>
> Any suggestions?
>
> Thanks,
> ==ml
>
> --
> Michael W. Lucas        mwlucas at BlackHelicopters.org
> http://www.MichaelWLucas.com/
> Latest book:  Cisco Routers for the Desperate, 2nd Edition
> http://www.CiscoRoutersForTheDesperate.com/
> _______________________________________________
> Users mailing list
> Users at lists.opennebula.org
> http://lists.opennebula.org/listinfo.cgi/users-opennebula.org
>



-- 
Javier Fontan, Grid & Virtualization Technology Engineer/Researcher
DSA Research Group: http://dsa-research.org
Globus GridWay Metascheduler: http://www.GridWay.org
OpenNebula Virtual Infrastructure Engine: http://www.OpenNebula.org


More information about the Users mailing list