"Jake G." wrote on 13/08/2013 07:25:04 AM:
> On 2013/08/13, at 12:50, Robert Collins
wrote:
> >>
> >> I was wondering how to handle changing the compute_driver in
nova.conf? I
> >> currently have the default
> >>
> >> compute_driver = libvirt.LibvirtDriver
> >> libvirt_type=kvm
> >>
> >> I w
Steve,
You can look at cloud management platforms like Red Hat cloudforms or Scalr.
They both have the added benefit of adding a way to automate a lot of stuff for
you (which is the reason we are currently looking at them)
Cloudforms seems to have a bit more features but at this moment I could n
On 13/08/13 12:10, Jake G. wrote:
> I was trying to find information from other than official openstack
> documentation. Every time I have followed the openstack docs I have either
> rendered my whole openstack server unusable or can never get said feature to
> work. So I am a bit weary to foll
On 13/08/13 12:10, Jake G. wrote:
> I was trying to find information from other than official openstack
> documentation. Every time I have followed the openstack docs I have either
> rendered my whole openstack server unusable or can never get said feature to
> work. So I am a bit weary to follo
The last one about multi_host seems weird. The instances still need a
gateway in the end, so if you don't use multi_host, that gateway is a
single host and if you enable multi_host, the gateway is the host, that is
running the instances.
With ipv6 you would still not want a single host as gateway,
On 2013/08/13, at 12:50, Robert Collins wrote:
> On 13 August 2013 15:15, Jake G. wrote:
>>
>> I was wondering how to handle changing the compute_driver in nova.conf? I
>> currently have the default
>>
>> compute_driver = libvirt.LibvirtDriver
>> libvirt_type=kvm
>>
>> I want to be able to
On Tue, Aug 13, 2013 at 1:15 PM, Jake G. wrote:
>
> I was wondering how to handle changing the compute_driver in nova.conf? I
> currently have the default
>
> compute_driver = libvirt.LibvirtDriver
> libvirt_type=kvm
>
> I want to be able to add the driver for baremetal provisioning, but I am
> un
@Jason Ford, it turn out be the iptables issue. Thx a lot.
@Jiajun Liu, I am using the nova-network. So this BP is not helpful. But I
am expecting this feature.
What I done:
1. using virsh to disable the nwfilter for the two vms.
2. stop the iptables in the vms
3. configure the keepalived.
4. work
On 13 August 2013 15:15, Jake G. wrote:
>
> I was wondering how to handle changing the compute_driver in nova.conf? I
> currently have the default
>
> compute_driver = libvirt.LibvirtDriver
> libvirt_type=kvm
>
> I want to be able to add the driver for baremetal provisioning, but I am
> unclear on
I was wondering how to handle changing the compute_driver in nova.conf? I
currently have the default
compute_driver = libvirt.LibvirtDriver
libvirt_type=kvm
I want to be able to add the driver for baremetal provisioning, but I am
unclear on how to do this.
Can there be more than one compute
Excerpts from Jake G.'s message of 2013-08-12 19:10:42 -0700:
> I was trying to find information from other than official openstack
> documentation. Every time I have followed the openstack docs I have either
> rendered my whole openstack server unusable or can never get said feature to
> work.
Excerpts from Daniel Ellison's message of 2013-08-12 03:03:00 -0700:
> On Aug 12, 2013, at 5:17 AM, Jake G. wrote:
> >
> > How can I install/enable the baremetal provisioning feature for Openstack
> > Grizzly? Any good resources, links?
>
> http://lmgtfy.com/?q=openstack+grizzly+baremetal
-1
I was trying to find information from other than official openstack
documentation. Every time I have followed the openstack docs I have either
rendered my whole openstack server unusable or can never get said feature to
work. So I am a bit weary to follow, because I have no way to clone my curre
On 08/12/2013 09:09 AM, Jonathan Proulx wrote:
>
> This is particularly odd since glance was working fine yesterday and to
> my knowlege the only thing I did was turn on more compute nodes...
>
>
> Now when I try and launch an instance it goes almost immediately to
> error state with the fault:
Also,
There is no need for "multihost = true" when using with IPv6...
Why?
Because "Multihost = true" turns on a NAT table (MASQUERADE I think) within
each compute node... Effectively hiding its running Instances and acting as
its default gateway but... Let IPv6 do the job without NAT... =P
N
On 08/12/2013 03:24 PM, Jonathan Proulx wrote:
> Thanks Rob,
>
> That was it
Can you explain what you did to make it work?
The quantum-openvswitch-agent should take care of connecting the
integration bridge (br-int) with the physical network bridge (the one
listed after "trunk:" in bridge_mappin
Thanks Rob,
That was it
-Jon
On Sun, Aug 11, 2013 at 6:31 PM, Robert Collins
wrote:
> You need to connect the exterior network to the integration bridge
> yourself. This is in the deployer docs somewhere, I don't recall
> offhand - sorry.
>
> -Rob
>
> On 12 August 2013 06:26, Jonathan Proulx
This is particularly odd since glance was working fine yesterday and to my
knowlege the only thing I did was turn on more compute nodes...
Now when I try and launch an instance it goes almost immediately to error
state with the fault:
{u'message': u'ImageNotFound', u'code': 404, u'details': u'Im
On Aug 12, 2013, at 1:34 PM, Jake G. wrote:
>
> I have been messing around with a few different tools for deploying
> openstack, but most of them seem they are designed for test environments.
>
> Just wonder what everyone's opinion is on the best tool for deploying a
> production quality envir
On Aug 12, 2013, at 12:34 PM, Jake G. wrote:
> I have been messing around with a few different tools for deploying
> openstack, but most of them seem they are designed for test environments.
>
> Just wonder what everyone's opinion is on the best tool for deploying a
> production quality enviro
Hi all,
I have been messing around with a few different tools for deploying openstack,
but most of them seem they are designed for test environments.
Just wonder what everyone's opinion is on the best tool for deploying a
production quality environment? Or is it best to install each component
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
so I have a nice little openstack cluster running on local hardware.
I would like to be able to launch off to EC2 within the same dashboard
as launching to local hardware.
my google fu hasnt been able to find anything useful amongst all the
openstack/
On Aug 12, 2013, at 6:21 AM, Daniel Ellison wrote:
> I was trying out the new Horizon features, specifically the Hypervisors panel
> in the Admin dashboard. To my surprise it claimed I was using QEMU, not KVM!
> To verify, I did "nova hypervisor-show 1" and hypervisor_type is listed as
> QEMU.
Gents,
This isn't an indication of KVM or QEMU - see this bug:
https://bugs.launchpad.net/nova/+bug/1195361
Verify you're running KVM by checking libvirtd logs on your compute hosts.
Regards,
Kev
On 12 August 2013 13:50, Victor Ruiz wrote:
> I have the same problem in a Ubuntu server 12.04.2
I have the same problem in a Ubuntu server 12.04.2 LTS.
I have no idea of how to fix it.
2013/8/12 Daniel Ellison
> On Aug 12, 2013, at 8:02 AM, Kevin Jackson
> wrote:
> >
> > Hi Daniel,
> > in /etc/nova do you have a nova-compute.conf with libvirt_type=qemu in
> instead? If KVM wasn't workin
On Aug 12, 2013, at 8:02 AM, Kevin Jackson wrote:
>
> Hi Daniel,
> in /etc/nova do you have a nova-compute.conf with libvirt_type=qemu in
> instead? If KVM wasn't working, it wouldn't fall back to QEMU so it must be
> mentioned somewhere.
Good thought! Unfortunately /etc/nova/nova-compute.conf
Hi Daniel,
in /etc/nova do you have a nova-compute.conf with libvirt_type=qemu in
instead? If KVM wasn't working, it wouldn't fall back to QEMU so it must be
mentioned somewhere.
Kev
On 12 August 2013 12:35, Daniel Ellison wrote:
> On Aug 12, 2013, at 6:53 AM, Razique Mahroua
> wrote:
> >
> >
On Aug 12, 2013, at 6:53 AM, Razique Mahroua wrote:
>
> Can you try to reload the module and see if the issue persists?
After removing the kvm and kvm_intel modules, "modprobe kvm" and "modprobe
kvm_intel" give no errors
"lsmod | grep kvm" gives me
kvm_intel 137721 0
kvm 41555
Can you try to reload the module and see if the issue persists?
Razique Mahroua - Nuage & Corazique.mahr...@gmail.comTel : +33 9 72 37 94 15
Le 12 août 2013 à 12:21, Daniel Ellison a écrit :Hey all,I was trying out the new Horizon features, specifically the Hypervisors panel in
On Aug 12, 2013, at 6:20 AM, Jake G. wrote:
>
> Thanks! I know how to use google. haha thats a funny site.
> The only reason I ask here is because I am having trouble finding a good
> resource or documentation from google results.
Your question didn't specify what you had searched for previousl
Hey all,
I was trying out the new Horizon features, specifically the Hypervisors panel
in the Admin dashboard. To my surprise it claimed I was using QEMU, not KVM! To
verify, I did "nova hypervisor-show 1" and hypervisor_type is listed as QEMU.
Now I've been using this same server for a few yea
Thanks! I know how to use google. haha thats a funny site.
The only reason I ask here is because I am having trouble finding a good
resource or documentation from google results.
Not all good blogs show up on google btw.
Best regards,
Jake
From: Daniel Ellison
On Aug 12, 2013, at 5:17 AM, Jake G. wrote:
>
> How can I install/enable the baremetal provisioning feature for Openstack
> Grizzly? Any good resources, links?
http://lmgtfy.com/?q=openstack+grizzly+baremetal
___
Mailing list: http://lists.openstack.
Dear List.
some how i can not manage it to get the s3token working. Hope one can
give me i hint.
I am using grizzly swift with keystone. The s3 related stuff gets
called. Keystone does authenticate the request. But at the end swift
returns a HTTP 403.
Since the container is also world reada
Hi all,
How can I install/enable the baremetal provisioning feature for Openstack
Grizzly? Any good resources, links?
My host is a single node CentOS fully working Openstack server installed with
RDO packstack.
Thank you,
Jake
___
Mailing list: h
Ok i started from scratch and now everything is working just fine. Thank you
for all your help.
In the end I believe it was because I did not have haproxy installed.
From: Eugene Nikanorov
To: Jake G.
Cc: "openstack@lists.openstack.org"
Sent: Friday, Au
36 matches
Mail list logo