After an apt-get upgrade on one of the object/account/container servers Ubuntu
decided to rename the devices.
I've been running with the following structure:
/srv/node/sda1
/srv/node/sdb1
/srv/node/sdc1
/srv/node/sdd1
sde and sdf are system disks in a software raid 1 array (md0). After rebootin
Pigging backing on the answer before me but yes, i had the same problem.
Following the setup of ssh keys for the user 'Nova' between nodes will
resolve the problem for you.
Kind Regards,
James Shaw
On 15 August 2013 02:10, Md. Maruful Hassan wrote:
> I have not done the live-migration setup
I was so convinced we resolved your issue Guilhermebut maybe it was someone else :)the error simply means the user "nova" from the first CN cannot connect as the user "nova" to the second CN it needs to send the image toHere are few checks : #1 - Exchange all nova's public keys between all compute
Perhaps you need to make your SCSI device names to be persistent across
reboots.
You may need to check on the udev rules
Best
Ashok
On Aug 15, 2013 1:39 PM, "Morten Møller Riis" wrote:
> After an apt-get upgrade on one of the object/account/container servers
> Ubuntu decided to rename the device
Deal all,
I followed this instruction -
https://github.com/mseknibilel/OpenStack-Grizzly-Install-Guide/blob/OVS_MultiNode/OpenStack_Grizzly_Install_Guide.rst
and finished Grizzly installation on ubuntu 12.04 (3 nodes), and followed
this instruction -
https://github.com/mseknibilel/OpenStack-Fo
Ok, let me write down what I've done:
#1 - Logged with nova's user at my first CN and generate a key with
ssh-keygen -t dsa at the /var/lib/nova/.ssh location;
#2 - Copied the id_dsa.pub from my first CN to my second CN at the
/var/lib/nova/.ssh location;
#3 - I'm sure I don't have te directive "A
You also need to change the MTU of either instance or of the network node.
Check the mailthread
https://lists.launchpad.net/openstack/msg24056.html for further
details.
-Regards
Rahul
On 8/15/13, Di DX Xu wrote:
>
>
> Deal all,
>
> I followed this instruction -
> https://github.com/mseknibile
Right, I've done all again, and now things starting change but, I'm getting
this now:
2013-08-15 11:18:05.705 ERROR nova.compute.manager
[req-dbbe3889-acd0-4c99-b22d-68c7005901a3 c402785616534f2096b34ce132b7d3f2
d532a4fc2e9e4b5f83b6dec7085237e5] [instance:
53f0a8ff-cd3b-4ddc-be9c-76655e8b8354] Une
Hi all,
What would be the good place in OpenStack to implement a feature that should
able to change a compute host configuration to make it suitable for a special
purpose VM? For example the special purpose VM needs huge page configured and
allocated on the host OS to run efficiently. What do y
Hi Josh,
On 14. 08. 13 09:47, Josh Durgin wrote:
> Yes, that's the problem. Could you report this as a bug against grizzly
> in launchpad? It would affect all backends if it's a regression in the
> scheduler.
Ok, here's the report.
https://bugs.launchpad.net/cinder/+bug/1212710
François
Guilherme Russi wrote 15.08.2013 11:21:
...
> Stderr: 'Bad owner or permissions on /var/lib/nova/.ssh/config\r\n'.
...
> My config file content:
>
> StrictHostKeyChecking no
>
> and its properties:
>
> -rw-rw 1 nova nova 25 Ago 15 10:39 config
>
> What Am I missing?
...
A .ssh/config fil
-BEGIN PGP MESSAGE-
Charset: ISO-8859-1
Version: GnuPG v1.4.12 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
hQEOAwfy5I47R+7fEAQA2yQ+dkcm0kUIksLwtsz9/cZpr1fDYaNvUTKOECQlNsN+
QFDBC4fNlEGjGDgNmmafGiYqNfPZvbkQN8XzJupGZSTF+rISOodq7vVxEmPuS75w
Z0Xirr8rHv8yTXM53j3k
And then I execute chmod 755 config but I got the first error again :(
2013-08-15 12:25:11.309 ERROR nova.compute.manager
[req-38d58bbf-370b-4b1d-9d41-b057641e8147 c402785616534f2096b34ce132b7d3f2
d532a4fc2e9e4b5f83b6dec7085237e5] [instance:
b8592ab7-39fe-49db-9bbe-196fa8fc60a1] Unexpected error w
Hello,
I am trying to implement the Mixed Flat and Private Network topology
with Neutron, where a VM gets 2 NICs, one to private network and other to
public shared network. My physical router is acting as a gateway for public
shared network (e.g. 192.168.4.1/24). When I boot up my instan
On 8/15/13 12:50 AM, Morten Møller Riis wrote:
After an apt-get upgrade on one of the object/account/container servers
Ubuntu decided to rename the devices.
I've been running with the following structure:
/srv/node/sda1
/srv/node/sdb1
/srv/node/sdc1
/srv/node/sdd1
sde and sdf are system disks
Hello Peter, thank you for your answer. Have you succeeded with live
migration? I'm getting this error now:
2013-08-15 14:09:34.969 ERROR nova.compute.manager
[req-34db0e55-2454-4ae9-a9e2-3d269e3786af c402785616534f2096b34ce132b7d3f2
d532a4fc2e9e4b5f83b6dec7085237e5] [instance:
95680109-8d83-4b54-
I can't figure out what options I need to use with "nova boot" in order to
boot from a bootable volume snapshot. Can somebody hit me with the clue
stick please? Thanks in advance.
--
\*..+.-
--Greg Chavez
+//..;};
___
Mailing list: http://lists.openst
Hi Greg -
This is not so well documented in Grizzly, but see folsom docs:
http://docs.openstack.org/folsom/openstack-compute/admin/content/boot-from-volume.html
to boot from a volume:
nova boot --flavor --block_device_mapping vda=:::0
test_from_vol
to boot from a snapshot of a bootable volume:
Thanks guys.
I guess my questions was mostly an effort to try an understand how the ring
reference affects local mounts.
If eg. two disks are mounted as:
/dev/sda1 as /srv/node/disk1
/dev/sdb1 as /srv/node/disk2
If these are then umount and mounted as:
/dev/sdb1 as /srv/node/disk1
/dev/sda1 a
Follow this link https://etherpad.openstack.org/grizzly-boot-from-volumes,
it describe the current UX for boot from volume.
I have done this successfully.
Additional, there is a BP to improve the UX.
https://blueprints.launchpad.net/nova/+spec/improve-boot-from-volume
On Fri, Aug 16, 2013 at 8:14
Hi, I'm setting up a grizzly installation
(trying to use the 'basic install guide' but with the network and controller
nodes combined.) I've got this issue:
# ip netns
qrouter-ce22fad3-41a1-4428-a782-26b04ca0b95c
ip netns exec qrouter-ce22fad3-41a1-4428-a782-26b04ca0b95c ip addr list
seting the
21 matches
Mail list logo