On 28.11.16 17:35, Nelson Lameiras wrote:
> Hello Shahar,
> Actually, our production servers use pool storage (+block devices), so I
> quickly tried to import from them and I have the same error : 0 disks
> "declared" on each VM.
> I tried the patch you advised last week, and eventhough it
I am getting theese events 2-3 times every minute, any ideas ?
Failed to delete snapshot '' for VM ‘khk9dsw16’
I have tried :
Set the host to maintenaince an activated again
Restarted ovirt-engine on Manager
SW : oVirt Engine Version: 188.8.131.52-1.el7.centos
That's a perfect explanation. Do we have a release date for 4.1? I'm
currently working with RHV in my production enviroment.
I may don't have the opportunity to wait for 4.1 to see the light as all my
customers VMs (im a web hoster) run on a non HA cluster as my provider does
"Derek Atkins" writes:
> All storage is NFS storage exported by the host itself. i didn't mention
> it because it shouldn't have mattered -- the HE export mounted but the
> others didn't.. Although it appears this may all have been due to the
> network issues
ovirt-shell will be deprecated and not supported or some functions on
ovirt-shell (or all package ovirt-engine-cli)?
We use ovirt-shell on client desktops who connected to SPICE consoles
and work (users provided by LDAP on ovirt-engine), like via RDP. For
this I wrote very fast-hack patch for
BTW, the ovirt-shell is something we deprecated. It is working on top of
the v3 api, which we plan to remove in 4.2.
So better not use it.
Alternatives are to use the SDKs according to your preference (we have
Java, python and Ruby SDKs) or the ansible modules mentioned above by Moran
the warning message means that Kdump integration is enabled for the host,
but the host doesn't contain proper kdump configuration. This can happen
due to following reasons:
1. Kdump integration was enabled, but host was not reinstalled after that
- every time you enable/disable
currently the only way how to test not-yet-merged modules is to download
the from github and save them manually into the existing directory with
oVirt modules which are part of Ansible 2.2 (this way you can use all new
modules even with 2.2).
Thanks a lot for any testing and feedback for
>> <http://lists.ovirt.org/mailman/listinfo/users> 
>> Users mailing list
Currently in case a host is not responding, the only way to make sure no
vms are running on it is to fence it, and we only support fencing through
In 4.1 we plan to have limited options for storage fencing. This will mean
that if a host is not responding then we
Thank you for your suggestions. Unfortunately I wasn't able to resize the
console either with the mouse method or with xrandr command.
When I resize the console window the image looks blurry similar to a
miss-sized image in a browser.
When I try the xrandr command. I receive the
Hi Charles, that worked, thanks!!! allthough im still wondering where this
change was listed at any point between 3.5 and 4.0.
Verstuurd vanaf mijn iPhone
> Op 28 nov. 2016 om 16:11 heeft Charles Kozler het
> volgende geschreven:
> What happens when you configure the
I’m having trouble with that plan and I’m not even trying to put the ovirtmgmt
bridge on an existing bond. I have a pre-existing gluster setup with gluster
running on a bonded interface. The ovirtmgmt should go on a different
interface, but even that fails on the bond with the “torn down
Actually, our production servers use pool storage (+block devices), so I
quickly tried to import from them and I have the same error : 0 disks
"declared" on each VM.
I tried the patch you advised last week, and eventhough it didn't work, clearly
the logs showed you are going in
Is there any way to have HA in a cluster which hosts have not got
configured the "Power Management"?
In my scenario I have 2 hosts (v4) on OVH and a hosted-engine to manage
them. Thing is that OVH does not allow direct communication with the hosts
IPMI interface. IPMI is only available in
I will follow them on Github.
Keep up the good work. Also if there is any possibility to start testing the
other modules(like managing networks) due to come out with ansible 2.3 please
let me know.
Jean Paul Gatt
From: Ondra Machacek
thank you very much for your feedback. I am sending few notes inline.
On 11/25/2016 04:27 PM, Jean Paul Gatt wrote:
I have been trying to use the following ansible modules (based on the
last devel commit in the ansible github repo for the extras modules
What happens when you configure the bond and then build the bridge manually
over the bond? oVirt installer should skip over it and not do anything.
Just make sure you have DEFROUTE set or routes configuration file as you
expect (this is what used to screw me up)
On Mon, Nov 28, 2016 at 10:06 AM,
Thanks for your responses but the ui is not an option for me as i am
dealing with loads of systems.
in 3.5 ovirt used to just accept the bridge as it was and incorporate
it, i am just wondering if i am facing a bug or a feature at the moment.
Charles Kozler schreef op 2016-11-28 15:48:
Thats what I used to do as well then on oVirt 4 it started screwing with
the the bond as well so I ended up just dumbing it down and figured using
the UI after the fact would be OK. I cant remember exactly what would
happen but it would be stupid little things like routing would break or
On Mon, Nov 28, 2016 at 3:42 PM, Charles Kozler
> What I've been doing since oVirt 4 is just configuring one NIC manually
> when I provision the server (eg: eth0, em1, etc) and then let oVirt do the
> bridge setup. Once the engine is up I login to the UI and I use it
What I've been doing since oVirt 4 is just configuring one NIC manually
when I provision the server (eg: eth0, em1, etc) and then let oVirt do the
bridge setup. Once the engine is up I login to the UI and I use it to bond
the NICs in whatever fashion I need (LACP or active-backup). Any time I
On Mon, Nov 28, 2016 at 12:24 PM, wrote:
> Hi All,
> In our ovirt 3.5 setup. i have always setup the ovirtmgmt bridge manually .
> The bridge consisted of 2 nic's
> I'd have /etc/vdsm/vdsm.conf list net_persist = ifcfg
> When i then deployed the host from the ovirt
In our ovirt 3.5 setup. i have always setup the ovirtmgmt bridge
The bridge consisted of 2 nic's
I'd have /etc/vdsm/vdsm.conf list net_persist = ifcfg
When i then deployed the host from the ovirt ui or api it would install
and would display the network setup correctly in
Le 26/11/2016 à 22:40, Chris Cowley a écrit :
A few reasons really. I fancied playing around with Click, so I made
this as I was also digging around the oVirt API. My workstation at work
is running Ubuntu, and I do not believe that ovirt-shell is packaged for it.
sudo apt-get install -y
On Fri, Nov 25, 2016 at 10:11 PM, Nathanaël Blanchet
> After registering a new host in an existing gluster cluster, it can't
> activate because it has no other gluster peer registered, so no existing
> declared gluster domain storage. It has no more
While I really like both ideas it is not a route I would PERSONALLY
To me it seems logical that there should be parallels between oVirt and
Openstack. After all, oVirt already integrates with Glance, Cinder and
Neutron. Thus it is not unreasonable to suggest that any sort of CLI
Mail list logo