Hi Chris,
Unless I overlooked something, I don't see Cinder or Ceph versions posted.
Feel free to just post the codenames but give us some inkling.
On Thu, Aug 23, 2018 at 3:26 PM, Chris Martin wrote:
> I back up my volumes daily, using incremental backups to minimize
> network traffic and sto
compute-01 ~]# nova --version
> 9.1.2
>
> I am using Pike 16.0.15 (My deployment tool is openstack-ansible)
>
>
> What are my option here?
>
>
> On Wed, Jul 25, 2018 at 8:19 AM, David Medberry
> wrote:
> > It's not clear what version of Nova you are run
It's not clear what version of Nova you are running but perhaps it is badly
patched. The 16.x.x (Pike) release of Nova has no
"migrate_configure_max_speed" but as best I can tell you are running a
patched version of Nova Pike so it may be inconsistent.
This parameter was introduced on 2017-08-24:
Bernd has this right.
Host aggregates (sometimes called Haggs) is the right soluton to this
problem. You can setup a flavor to only run on a certain hagg.
This works well (in production, at scale).
On Sun, Jul 1, 2018 at 7:52 AM, Satish Patel wrote:
> Folks,
>
> Recently we build openstack for
Generally recommended to start with a cloud-image and these are available
for Centos and RHEL.
You can then customize (either via cloud-init which would be the cloud-way
or via making a new image from your existing customized instances.)
RHEL
https://access.redhat.com/downloads/content/69/ver=/rhe
Best practice is to use shared storage and then the "copy" is really only
the active memory. A few changes came about in about the newton? timeframe
that allows for some memory convergence.
Take a look at the nova release notes from that time forward and you should
see reference to the change(s).
Subscribe to this bug and click the "This affects me." link near the top.
https://bugs.launchpad.net/cloud-images/+bug/1585233
On Thu, Feb 8, 2018 at 1:53 PM, Volodymyr Litovka wrote:
> Hi colleagues,
>
> does anybody here know how to prepare Ubuntu Artful (17.10) image using
> diskimage-builde
Hi all,
Broad distribution to make sure folks are aware of the upcoming Ops Meetup
in Tokyo.
You can help "steer" this meetup by participating in the planning meetings
or more practically by editing this page (respectfully):
https://etherpad.openstack.org/p/TYO-ops-meetup-2018
Sign up for the me
Hi G.,
I don't have Ocata up anywhere, but as a "best practice", I generally do
the:
nova reset-state --active $UUID
followed immediately by
nova reboot --hard
to try and "restore/resurrect" errored instances. The reset-state --active
doesn't actually do anything to the instance, it just manipula
Best practice is to never migrate more than one at a time. Maybe that has
been encoded. I.e. no more than one off of a compute at a time and no more
than one onto a compute at a time.
On Sep 19, 2017 5:37 PM, "Steven D. Searles" wrote:
> Hello everyone and thanks in advance. I have Openstack Pi
Cirros is a MICRO distribution for verifying a cloud works, not for doing
performance analysis. There are no "packages" for it as it is not designed
to have functionality added to it. You can instead install a complete
version of Linux (say, Ubuntu) and then run performance analysis as you
would no
Does "virsh capabilities" return the same tree on both hosts?
On Wed, Jan 25, 2017 at 4:10 AM, fabrice grelaud <
fabrice.grel...@u-bordeaux.fr> wrote:
> Hi,
>
> i ‘ve got live migration issue in one direction but not in other.
> I deploy openstack with OSA, ubuntu trusty, stable/newton branch, 14
Others reported the same (some of my co-workers). Wasn't at all clear
why/how they got missed.
On Wed, Oct 19, 2016 at 1:39 PM, Jonnalagadda, Venkata <
venkata.jonnalaga...@intl.att.com> wrote:
> Hi,
>
>
>
> Myself contributed to multiple openstack projects during Mitaka & Newton
> but it seems m
cloud-init and cloud-config is how one can add a password to a cloud image
though it is NEVER recommended.
See:
http://cloudinit.readthedocs.io/en/latest/topics/examples.html
and read about users and groups and mkpasswd
cloud-init will read this at vm boot time and apply the cloud-config
(presu
depending on who created it and in what tenant/project, yes you should be
able to see it as that equivalent user in horizon.
If you made it public, any user should be able to see it.
On Mon, Aug 29, 2016 at 9:53 AM, Kevin Wilson wrote:
> Hi all,
> I use OpenStack (Kilo) on Ubuntu.
>
> When creat
Exactly. Since Xenial released alongside Xenial, UCA is not required. You
will find Newton packages in UCA (if not now in pre-release, at some point
closer to release.)
On Wed, Aug 24, 2016 at 12:51 PM, Erdősi Péter wrote:
> 2016. 08. 24. 17:55 keltezéssel, Kiriakos Krastillis írta:
>
>> Is ther
nova service-disable $SHORTNAME nova-compute --reason
NO_MORE_SCHEDULING_HERE
will prevent new VMs from going on but doesn't do anything with existing.
On Tue, Aug 2, 2016 at 3:01 PM, Ken D'Ambrosio wrote:
> Hi, all. Trying to figure out how to disable a compute node from getting
> new VMs sch
No, I don't think there is a way to "add" a rule that isn't already in your
default settings to an instance so that it can reach the metadata server.
If users bypass the "default" (presuming you allow default to access the
metadata server), they simply won't have that access.
On Tue, Jul 12, 20
Asaka,
The openstack-dev mailing list may be a better place to ask but since
you asked here
Read this:
https://wiki.openstack.org/wiki/How_To_Contribute
with special attention to this:
https://wiki.openstack.org/wiki/How_To_Contribute#Bug_fixing
with these Low Hanging Fruit bugs:
https:/
1:16 AM, Jean-Pierre Ribeauville <
jpribeauvi...@axway.com> wrote:
> Hi,
>
>
>
> This parameter is set to true in nova.conf . :
>
>
>
> *enable_instance_password=true*
>
>
>
>
>
> J.P.
>
>
>
> *From:* medbe...@gmail.com [mailto:medbe...@gma
It all depends on how you have setup your cloud. Most/many OpenStack clouds
DO NOT inject into the instance and the password feature is null and void.
Look at:
enable_instance_password = *True* (BoolOpt) Enables returning of the
instance password by the relevant server API calls such as create, r
to the neutron router (but I suspect it doesn't really care if your
provider net is RFC 1918 or not.
On Sat, Jun 4, 2016 at 11:59 AM, David Medberry
wrote:
> Of course, there are almost always two routers on a subnet... (unless it
> is the end of the line). What makes you think
Of course, there are almost always two routers on a subnet... (unless it is
the end of the line). What makes you think the neutron router has to be at
.1? Mine are typically .1 on the VM side (but it is totally fine to use
something else) and not .1 on the provider side.
On Fri, Jun 3, 2016 at 12:
Which drivers (virtio) did you use in the Windows instances?
On Thu, May 12, 2016 at 9:24 AM, Chris Friesen
wrote:
> On 05/12/2016 08:52 AM, gulseren bulut wrote:
>
>> Hello,
>>
>> I done some tests for windows virtual server on KVM and Hyper-V.
>>
>> All resources of VMs is same.
>>
>> VM OS: W
Hi Jaison,
This is an issue that the Neutron team is aware. They will likely be
addressing this (at some point) but your understanding aligns with my own.
So, public IPV4 usage is a well known, well documented issue and DVR / HA
exacerbates it.
On Tue, Apr 26, 2016 at 1:33 AM, Jaison Peter wrote
There are "official" Centos 6 images here:
http://cloud.centos.org/centos/6/images/
On Sat, Mar 26, 2016 at 5:01 PM, Jagga Soorma wrote:
> Ahhh.. got further.. looks like eth0 was set to onboot=no :( rebooting
> with onboot=yes gets me a bit further but the curl is still failing..
> attached is
Cells v2 is in active development. If you can wait a bit longer and
upgrade/reinstall with Mitaka and v2 you will be in a better place.
Hopefully a nova core will weigh in to tell me I'm on or off track with
that comment. Kris Lindgren also keeps a pretty good eye on this and I've
cc'd him.
On Thu
http://www.slideshare.net/SajuMadhavan/open-stack-devstack-screen-commands-and-key-bindings
may be more current.
On Wed, Sep 2, 2015 at 6:57 PM, David Medberry
wrote:
> Be aware that there is a "screen" session running with each of the
> services. If you go to the appropria
Be aware that there is a "screen" session running with each of the
services. If you go to the appropriate screen in the screen session and
then "Ctrl-C" therein and hit up arrow return, you'll basically be
restarting just that service.
This may be out of date but give it a try:
http://www.rushiag
Seen similar issues.
There are some known glance issues and also so nova filter issues going on
by default.
I'd look at enabling "verbose=True" in Nova and seeing if you can get
closer to the truth of what's going on and then tail the nova logs. We've
put that down in our runbooks as a best practi
It's due out any time (this week).
On Tue, Aug 25, 2015 at 9:59 AM, Balaji Narayanan (பாலாஜி நாராயணன்) <
li...@balajin.net> wrote:
> Hi,
>
> Does anyone know when will the list of papers accepted for Tokyo be
> published? Given that it takes approx 5 weeks for visa, it would be
> great if this co
).
On Fri, Aug 14, 2015 at 11:45 AM, mad Engineer
wrote:
> so is editing cloudinit manually the only way to change hostname?.I wonder
> how would it get the instance name as hostname during build state.
>
> On Fri, Aug 14, 2015 at 11:06 PM, David Medberry
> wrote:
>
>> Thi
c
> Second compute : 899f90c8-e8e8-47b7-8d25-3e14fc2f270a
>
> but, # virsh secret-get-value uuid gave me same results on my two nodes
>
>
> Regards,
>
>
>
> 2015-08-14 3:22 GMT+02:00 David Medberry :
>
>> Yep, exactly what Mike Smith says.
>>
>> sud
Yep, exactly what Mike Smith says.
sudo virsh secret-list # should list one or more on nova-compute nodes
using ceph
sudo virsh secret-get-value $UUID #from the previous should have a value
that looks something like:
AQDoABCDEFGHIJKLMNOPQRSTUVWXYZoA/A
That value come from a ceph keyring.
On Thu
The column type for the foreign keys don't match EXACTLY. (Strings). They
have to be exact in value and the collation order must be the same.
Two issues: a character set (UTFetc) and the collation order must be
IDENTICAL.
It will use the database default if you don't SET it in the new column..
This is just different projects. The vote numbers won't be public for a day
or two, if ever. The votes don't fully determine what gets selected. The
Track Chairpeople make those final calls.
On Wed, Jul 29, 2015 at 12:44 PM, Perry, Sean wrote:
> Where are the actual vote numbers? How did you acq
Nice writeup maish! very nice.
On Wed, Jul 29, 2015 at 10:27 AM, Maish Saidel-Keesing
wrote:
> Some of my thoughts on the Voting process.
>
>
> http://technodrone.blogspot.com/2015/07/openstack-summit-voting-by-numbers.html
>
> Guess which category has the most number of submissions??
> ;)
>
> -
hot?
>
> Thank you all,
> Giuseppe
>
> 2015-07-06 18:40 GMT+02:00 David Medberry :
>
>> Giuseppe,
>>
>> A great deal of Disaster Recovery planning relies on you / your
>> organization to define what they mean by DR planning. Yes, Nova snaps
>&
Giuseppe,
A great deal of Disaster Recovery planning relies on you / your
organization to define what they mean by DR planning. Yes, Nova snaps
(instance snaps) and Cinder snaps (volume snaps) can be used in DR.
Additionally, depending on your architecture you can utilize Swift (if it
is geo distr
-tunnels/
On Sat, Jun 20, 2015 at 5:59 AM, David Medberry
wrote:
> https://wiki.opendaylight.org/view/OpenStack_and_OpenDaylight
>
> On Fri, Jun 19, 2015 at 10:53 PM, chinasubbareddy m <
> subbareddy.openst...@gmail.com> wrote:
>
>> Hi,
>>
>> could you ple
https://wiki.opendaylight.org/view/OpenStack_and_OpenDaylight
On Fri, Jun 19, 2015 at 10:53 PM, chinasubbareddy m <
subbareddy.openst...@gmail.com> wrote:
> Hi,
>
> could you please guide me good blogs for integrating opeack and
> opendaylight ?
>
>
> Regards,
> subbareddy.
>
> _
We have run with (really) Liberty keystone, Liberty horizon and everything
else Juno.
I suspect you can do this mixed thing as long as you pay careful attention.
Likewise, when we upgrade our TWO REGIONAL DATACENTERS we always do one and
then the next a couple days later, so all of our upgrades h
I think the best practices are to use the operating system's cloud images
(and not bake them yourself.)
Ie:
http://cloud-images.ubuntu.com/
http://www.cloudbase.it/ws2012r2/
https://fedoraproject.org/wiki/Cloud_images
etc
Converting a running instance into an image for glance is a worst practice
This page, presumably correct, shows one more Grizzly release in March:
https://wiki.openstack.org/wiki/StableBranchRelease
Planned stable/grizzly releases
2013.1.5 Mar 20 2014
On Tue, Dec 17, 2013 at 8:13 AM, Drew Fisher wrote:
> Are there plans for additional Grizzly releases, or is Hava
Anne, it does appear to have been "stuck" for some time.
-dave
On Mon, Jul 29, 2013 at 4:38 PM, Anne Gentle wrote:
>
>
>
> On Mon, Jul 29, 2013 at 11:48 AM, Jarret Raim
> wrote:
>
>> Two quick questions.
>>
>>
>> When you sign up to present presentations at the summit, you enter your
>> name
If you want to be a speaker at the conference, go here:
http://www.openstack.org/summit/openstack-summit-hong-kong-2013/become-a-speaker/
The Icehouse design summit doesn't appear to be up yet (it still refers to
the Havana Design Summit which was last April.)
On Mon, Jul 29, 2013 at 10:48 AM,
46 matches
Mail list logo
