are budget is tight
With 24 * 256 GB SSD disks. .
In anyway. OpenNebula is a Management System which supports a wide range of
underlying Cluster / Virtual System implementations and does not care much
about the underlying hardware.
So its up to you to decide if you will use KVM,XEN,
Hi,
New Users are created by default in the same group as the user creating it,
except for the oneadmin group, that creates them in 'users'.
So, the easy way to do it is create the new users from another user that
belongs to the different group. Let's call this group 'newgroup', what you
need to
Hello Christopher,
actually to be honest there's something I didn't quite understand about
your use-case. What the manual procedure you followed to make the
data/config volume to be accessible from the VMs? How is it shared? Or is
it dynamically mounted by the VM, and if so, how?
regards,
Jaime
Hello Valentin,
Thanks for the tip, we will review it and include in the documentation.
Thanks!
Jaime
On Wed, Jul 25, 2012 at 1:36 PM, Valentin Bud valentin@gmail.comwrote:
Hello Javier,
Thank you for your answer. Indeed, changing /dev/kvm group to oneadmin
solved the problem. I can
Hello Lawrence,
could it be that tgtadm got stuck and didn't remove one of the images?
Could you try removing the image manually in tgtd? Off the top of my head,
I think the commands were: tdt-admin --delete TID and you can get the
TID from tgt-admin --dump (I may be wrong, though).
We've found
Hello Shivay,
could you please send us your oned.conf and the output of onedatastore
list -x
cheers,
Jaime
On Mon, Sep 10, 2012 at 8:29 AM, shivay veer sharma
shivay.v...@gmail.comwrote:
i am still getting the error
Mon Sep 10 11:53:32 2012 [DiM][I]: New VM state is ACTIVE.
Mon Sep 10
True, the datastore 0 doesn't need to be mounted on the frontend. Back in
the day, when we didn't have datastores, we used to document that you
needed to share the whole /var/lib/one directory to make things easier, I
guess the idea still lingers in my mind
--
Carlos Martín, MSc
Project Engineer
* there is the system volume (everyone uses it)
* configured into the system volume are GRUB changes that:
* set a cmdline variable (bubble=stateless)
* configured into the system volume are initramfs changes that:
* if bubble == stateless
* mounts the context disk in
I think that the export LANG=C thing in scripts_common.sh file
should make this work even if the user profile is not loaded.
If you are executing the commands manually then do as Marco says,
without the '-' parameter the environment will continue to be the same
as the original loged user.
On
I agree that the boilerplate in the scripts to prepare the environment
is messy. One solution could be loading a environment file as you say:
--8--
if [ -z ${ONE_LOCATION} ]; then
. /etc/one/env.sh
else
. $ONE_LOCATION/etc/env.sh
fi
--8--
It is much more concise and can be more
Hello
I'm having problem deploying VMs using context. According to VM's log, the
context block device gets generated but init.sh won't run. My goal is to
configure VM with static IP address (no DHCP server) and hostname on the new
VMs.
hu Sep 13 17:40:35 2012 [DiM][I]: New VM state is ACTIVE.
11 matches
Mail list logo