Hi there,
I am having a problem with our OpenNebula installation.
After issuing the "onevm stop" command, the VM got into SAVE_STOP mode but
nothing happened afterwards. The vm.log shows
> Tue Jan 4 18:02:20 2011 [LCM][I]: New VM state is SAVE_STOP
and nothing more.
On the host, I can confirm t
Hi all
i built a cloud for testing purposes using the following
drivers:
- information driver: kvm
- virtualization driver: kvm
- transfer manager driver: lvm.
I want to use the OCCI API for the management and
i followed the guide from the Opennebula main site.
I have some questions about the sto
OpenNebula is not service aware. VMs are black boxes and has no way to
communicate with them or get information about internal metrics aside
from contextualization or external monitoring data (memory, cpu or
network consumption). Service management can be done with external
tools like Claudia [1] o
Check the log files at /var/log/libvirt/qemu/one-xx.log in both ends,
maybe it gives you more information.
On Tue, Jan 4, 2011 at 12:20 PM, wrote:
> Hello gentlement and Happy New Year.
>
> I am experiencing this strange behaviour:
>
> I set up the TLS certs that OpenNebula seems to require to
You can control the way the scheduler sorts resources using rank
parameter in the job template [1]. You can make the VM go to the host
with least free ram with
RANK="- FREEMEMORY"
or make it use the host with most free ram:
RANK="FREEMEMORY"
[1] http://www.opennebula.org/documentation:rel2.0:sc
OpenNebula cannot directly cope with that requirements but maybe an
external server that picks that jobs, encapsulates them inside a VM
(maybe using contextualization) and sends that to ONE can do the job.
I am just thinking loud as I don't know the complexity of your jobs or
their requirements.
M
We have an open issue [1] so we can implement network groups. That way
a network group can be specified and OpenNebula will select an free
lease from that pool.
Concerning the number of free leases in occi information I suppose it
can be implemented as an extension to the occi information provided
It seems that you have apparmor enabled and it does not let libvirt
write checkpoint files. From OpenNebula documentation:
--8<--
If apparmor is active (by default it is), you should add $ONE_LOCATION/var
to the end of /etc/apparmor.d/libvirt-qemu
owner /path-to-one-location/var/** rw,
--
Check after changing the configuration file that the log file reflects
the prot change, maybe it is failing to read that info and trying to
connect to the same port. Also, are you starting the occi server as
oneadmin user?
On Tue, Dec 28, 2010 at 10:37 AM, Aya Chaouachi wrote:
> i already chaged
Hello,
I also think that dhcp is the way to go with windows guests or images
that can not me modified to use contextualization data provided by
ONE. We have not included dhcp configuration scripts as we don't want
to marry with any specific external server and I suppose most of the
datacenters alr
10 matches
Mail list logo