Re: [Openstack] On Ubuntu 18.04, the package "gnocchi-api" install tries to remove both "keystone" and "openstack-dashboard"!

[Martinx - ジェームズ]

Worth to mention that I still have to test python-gnocchi, my confs are
wrong but, it is installed...   :-P

On Thu, 7 Jun 2018 at 15:57, Martinx - ジェームズ 
wrote:

> Corey,
>
>  I still don't get it... The package gnocchi-api is basically empty.
>
>  It have only 1 single text file:
> "/etc/apache2/sites-available/gnocchi-api.conf"!
>
>  The problem is that it depends on "python3-gnocchi" and
> "libapache2-mod-wsgi-py3", which brings the problem...
>
>  So, the workaround looks quite simple!
>
>  Instead of running "apt install gnocchi-api", I'm doing:
>
> ---
>  apt install python-gnocchi # which brings Gnocchi API
> (/usr/bin/gnocchi-api) for Python 2, then:
>
>  apt download gnocchi-api
>
>  # Extract the "/etc/apache2/sites-available/gnocchi-api.conf" file out
> from "gnocchi-api_4.2.4-0ubuntu1_all.deb" and manually copy to
> /etc/apache/sites-available".
> ---
>
>
>
> On 5 June 2018 at 10:30, Corey Bryant  wrote:
>
>> Hi Martin,
>>
>> Thanks for bringing this up. I replied in the bug, but pasting some of my
>> response below:
>>
>> gnocchi-api only supports py3 for now in bionic/queens and above.
>>
>> Most OpenStack packages in Ubuntu have not yet received py3 support. That
>> is a focus for the cosmic cycle. Where possible the plan is to install py2
>> by default and py3 as an alternative to prevent all-in-one issues like this.
>>
>> In other words, 'sudo apt install python3-keystone keystone' would
>> install python3 packages, and (assuming python3-keystone was not already
>> installed) 'sudo apt install keystone' would install python2 packages.
>>
>> That's not the case for gnocchi-api, obviously, as it only installs py3
>> packages which conflicts with other python packages if and only if you're
>> installing on the same machine as other py2 services. Most modern
>> production deployments isolate services in one way or another (e.g.
>> containers such as LXD or docker) so there's a clear work around.
>>
>> With that said, we can probably add py2 alternative packages to the
>> Depends for gnocchi-api allowing py2 installs, however I seem to remember
>> gnocchi-api was forced to move to py3 for a reason. I'm not remembering why
>> though atm.
>>
>> Thanks,
>>
>> Corey
>>
>> On Mon, Jun 4, 2018 at 10:31 PM, Martinx - ジェームズ <
>> thiagocmarti...@gmail.com> wrote:
>>
>>> Thank you!
>>>
>>> On 4 June 2018 at 21:25, pablo brunetti 
>>> wrote:
>>>
>>>> Hello,
>>>>
>>>>
>>>> This bug is confirmed in
>>>> https://bugs.launchpad.net/ubuntu/+source/gnocchi/+bug/1746992
>>>> Bug #1746992 “gnocchi-api in Bionic uses python3, and libapache2...” :
>>>> Bugs : gnocchi package : Ubuntu
>>>> <https://bugs.launchpad.net/ubuntu/+source/gnocchi/+bug/1746992>
>>>> bugs.launchpad.net
>>>> Since the two wsgi modules are conflicting, installing gnocchi-api
>>>> breaks other OpenStack services.
>>>>
>>>>
>>>> Pablo.
>>>>
>>>>
>>>>
>>>>
>>>> --
>>>> *De:* Martinx - ジェームズ 
>>>> *Enviado:* segunda-feira, 4 de junho de 2018 21:53
>>>> *Para:* openstack@lists.openstack.org; James Page
>>>> *Assunto:* [Openstack] On Ubuntu 18.04, the package "gnocchi-api"
>>>> install tries to remove both "keystone" and "openstack-dashboard"!
>>>>
>>>> Hello,
>>>>
>>>>  I'm trying to install Gnocchi on Ubuntu 18.04, but there is a problem,
>>>> the "gnocchi-api" package tries to uninstall Keystone and Horizon!
>>>>
>>>>  So, from what I'm seeing, it is impossible to build an OpenStack
>>>> Queens All-in-One on Ubuntu 18.04 with Gnocchi and Ceilometer...   :-(
>>>>
>>>>  The problem (line 29):
>>>>
>>>>  https://pastebin.com/sns8LG18
>>>> <https://pastebin.com/sns8LG18>
>>>> administrative@queens-1:~$ sudo apt install gnocchi-api Reading package
>>>> lists. - Pastebin.com <https://pastebin.com/sns8LG18>
>>>> pastebin.com
>>>>
>>>>
>>>>  Any workaround or bug reported on Launchpad about this?
>>>>
>>>>  Time for Canonical to provide OpenStack Snap packages!  lol
>>>>
>>>> Thanks!
>>>> Thiago
>>>>
>>>
>>>
>>> ___
>>> Mailing list:
>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>> Post to : openstack@lists.openstack.org
>>> Unsubscribe :
>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>>
>>>
>>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] On Ubuntu 18.04, the package "gnocchi-api" install tries to remove both "keystone" and "openstack-dashboard"!

[Martinx - ジェームズ]

Corey,

 I still don't get it... The package gnocchi-api is basically empty.

 It have only 1 single text file:
"/etc/apache2/sites-available/gnocchi-api.conf"!

 The problem is that it depends on "python3-gnocchi" and
"libapache2-mod-wsgi-py3", which brings the problem...

 So, the workaround looks quite simple!

 Instead of running "apt install gnocchi-api", I'm doing:

---
 apt install python-gnocchi # which brings Gnocchi API
(/usr/bin/gnocchi-api) for Python 2, then:

 apt download gnocchi-api

 # Extract the "/etc/apache2/sites-available/gnocchi-api.conf" file out
from "gnocchi-api_4.2.4-0ubuntu1_all.deb" and manually copy to
/etc/apache/sites-available".
---



On 5 June 2018 at 10:30, Corey Bryant  wrote:

> Hi Martin,
>
> Thanks for bringing this up. I replied in the bug, but pasting some of my
> response below:
>
> gnocchi-api only supports py3 for now in bionic/queens and above.
>
> Most OpenStack packages in Ubuntu have not yet received py3 support. That
> is a focus for the cosmic cycle. Where possible the plan is to install py2
> by default and py3 as an alternative to prevent all-in-one issues like this.
>
> In other words, 'sudo apt install python3-keystone keystone' would install
> python3 packages, and (assuming python3-keystone was not already installed)
> 'sudo apt install keystone' would install python2 packages.
>
> That's not the case for gnocchi-api, obviously, as it only installs py3
> packages which conflicts with other python packages if and only if you're
> installing on the same machine as other py2 services. Most modern
> production deployments isolate services in one way or another (e.g.
> containers such as LXD or docker) so there's a clear work around.
>
> With that said, we can probably add py2 alternative packages to the
> Depends for gnocchi-api allowing py2 installs, however I seem to remember
> gnocchi-api was forced to move to py3 for a reason. I'm not remembering why
> though atm.
>
> Thanks,
>
> Corey
>
> On Mon, Jun 4, 2018 at 10:31 PM, Martinx - ジェームズ <
> thiagocmarti...@gmail.com> wrote:
>
>> Thank you!
>>
>> On 4 June 2018 at 21:25, pablo brunetti 
>> wrote:
>>
>>> Hello,
>>>
>>>
>>> This bug is confirmed in
>>> https://bugs.launchpad.net/ubuntu/+source/gnocchi/+bug/1746992
>>> Bug #1746992 “gnocchi-api in Bionic uses python3, and libapache2...” :
>>> Bugs : gnocchi package : Ubuntu
>>> <https://bugs.launchpad.net/ubuntu/+source/gnocchi/+bug/1746992>
>>> bugs.launchpad.net
>>> Since the two wsgi modules are conflicting, installing gnocchi-api
>>> breaks other OpenStack services.
>>>
>>>
>>> Pablo.
>>>
>>>
>>>
>>>
>>> --
>>> *De:* Martinx - ジェームズ 
>>> *Enviado:* segunda-feira, 4 de junho de 2018 21:53
>>> *Para:* openstack@lists.openstack.org; James Page
>>> *Assunto:* [Openstack] On Ubuntu 18.04, the package "gnocchi-api"
>>> install tries to remove both "keystone" and "openstack-dashboard"!
>>>
>>> Hello,
>>>
>>>  I'm trying to install Gnocchi on Ubuntu 18.04, but there is a problem,
>>> the "gnocchi-api" package tries to uninstall Keystone and Horizon!
>>>
>>>  So, from what I'm seeing, it is impossible to build an OpenStack Queens
>>> All-in-One on Ubuntu 18.04 with Gnocchi and Ceilometer...   :-(
>>>
>>>  The problem (line 29):
>>>
>>>  https://pastebin.com/sns8LG18
>>> <https://pastebin.com/sns8LG18>
>>> administrative@queens-1:~$ sudo apt install gnocchi-api Reading package
>>> lists. - Pastebin.com <https://pastebin.com/sns8LG18>
>>> pastebin.com
>>>
>>>
>>>  Any workaround or bug reported on Launchpad about this?
>>>
>>>  Time for Canonical to provide OpenStack Snap packages!  lol
>>>
>>> Thanks!
>>> Thiago
>>>
>>
>>
>> ___
>> Mailing list:
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>> Post to : openstack@lists.openstack.org
>> Unsubscribe :
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>
>>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Geneve isn't working on Queens, Ubuntu 18.04 - Failed to bind port type geneve

[Martinx - ジェームズ]

Guys,

 Is there anyone running Queens on Ubuntu 18.04 with Geneve instead of
VXLAN (OpenvSwitch Agent, Neutron L3 Agents...)?

Thanks!
Thiago

On 4 June 2018 at 18:59, Martinx - ジェームズ  wrote:

> Hello!
>
>  I have OpenStack Queens up and running on Ubuntu 18.04 with VXLAN!
>
>  It is awesome!
>
>  However, I want to use it with Geneve, instead of VXLAN.
>
>  But, it doesn't work!
>
>  With VXLAN, I have the following ml2_conf.ini:
>
> ---
> [ml2]
> type_drivers = vxlan,flat,vlan
> tenant_network_types = vxlan,flat,vlan
> mechanism_drivers = openvswitch,l2population
> extension_drivers = port_security
> overlay_ip_version = 4
> ---
>
>  All good! I can launch instances and ping the Internet from within an
> instance.
>
>  So, when I try the Geneve instead, like this:
>
> ---
> [ml2]
> type_drivers = geneve,vxlan,flat,vlan
> tenant_network_types = geneve,vxlan,flat,vlan
> mechanism_drivers = openvswitch,l2population
> extension_drivers = port_security
> overlay_ip_version = 4
> ---
>
> NOTE: I can easily bring up a clean Ubuntu 18.04 server image, and install
> OpenStack from scratch, to do the each test independently. I have my own
> Ansible automation to deploy OpenStack Queens on Ubuntu 18.04. As I've
> said, it works fine with VXLAN but, not with Geneve (2 lines change from
> one deployment, to another).
>
>
> So, while trying to create an instance with Geneve, the following error
> appear on Neutron logs:
>
> ---
> 2018-06-04 22:55:40.677 11426 ERROR neutron.plugins.ml2.managers
> [req-c988c769-690e-45db-b810-3b35f0e8cba8 bda3aed0ccae46008f9928885280c085
> 46f4131518fd4b699ac80bc867fd1832 - default default] Failed to bind port
> 17e4dd94-7655-40f5-9077-b6e7c583c7eb on host queens-1 for vnic_type
> normal using segments [{'network_id': '6f805d1f-dfb6-42f6-846a-e39fac80ca8c',
> 'segmentation_id': 19, 'physical_network': None, 'id':
> '51a89942-5a91-47c6-872b-b33d2e6d418d', 'network_type': u'geneve'}]
> ---
>
>  Is it a bug?
>
>  I appreciate any help!
>
> Cheers!
> Thiago
>
>
>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] On Ubuntu 18.04, the package "gnocchi-api" install tries to remove both "keystone" and "openstack-dashboard"!

[Martinx - ジェームズ]

Thank you!

On 4 June 2018 at 21:25, pablo brunetti  wrote:

> Hello,
>
>
> This bug is confirmed in https://bugs.launchpad.net/
> ubuntu/+source/gnocchi/+bug/1746992
> Bug #1746992 “gnocchi-api in Bionic uses python3, and libapache2...” :
> Bugs : gnocchi package : Ubuntu
> <https://bugs.launchpad.net/ubuntu/+source/gnocchi/+bug/1746992>
> bugs.launchpad.net
> Since the two wsgi modules are conflicting, installing gnocchi-api breaks
> other OpenStack services.
>
>
> Pablo.
>
>
>
>
> --
> *De:* Martinx - ジェームズ 
> *Enviado:* segunda-feira, 4 de junho de 2018 21:53
> *Para:* openstack@lists.openstack.org; James Page
> *Assunto:* [Openstack] On Ubuntu 18.04, the package "gnocchi-api" install
> tries to remove both "keystone" and "openstack-dashboard"!
>
> Hello,
>
>  I'm trying to install Gnocchi on Ubuntu 18.04, but there is a problem,
> the "gnocchi-api" package tries to uninstall Keystone and Horizon!
>
>  So, from what I'm seeing, it is impossible to build an OpenStack Queens
> All-in-One on Ubuntu 18.04 with Gnocchi and Ceilometer...   :-(
>
>  The problem (line 29):
>
>  https://pastebin.com/sns8LG18
> <https://pastebin.com/sns8LG18>
> administrative@queens-1:~$ sudo apt install gnocchi-api Reading package
> lists. - Pastebin.com <https://pastebin.com/sns8LG18>
> pastebin.com
>
>
>  Any workaround or bug reported on Launchpad about this?
>
>  Time for Canonical to provide OpenStack Snap packages!  lol
>
> Thanks!
> Thiago
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] On Ubuntu 18.04, the package "gnocchi-api" install tries to remove both "keystone" and "openstack-dashboard"!

[Martinx - ジェームズ]

Hello,

 I'm trying to install Gnocchi on Ubuntu 18.04, but there is a problem, the
"gnocchi-api" package tries to uninstall Keystone and Horizon!

 So, from what I'm seeing, it is impossible to build an OpenStack Queens
All-in-One on Ubuntu 18.04 with Gnocchi and Ceilometer...   :-(

 The problem (line 29):

 https://pastebin.com/sns8LG18

 Any workaround or bug reported on Launchpad about this?

 Time for Canonical to provide OpenStack Snap packages!  lol

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Geneve isn't working on Queens, Ubuntu 18.04 - Failed to bind port type geneve

[Martinx - ジェームズ]

Hello!

 I have OpenStack Queens up and running on Ubuntu 18.04 with VXLAN!

 It is awesome!

 However, I want to use it with Geneve, instead of VXLAN.

 But, it doesn't work!

 With VXLAN, I have the following ml2_conf.ini:

---
[ml2]
type_drivers = vxlan,flat,vlan
tenant_network_types = vxlan,flat,vlan
mechanism_drivers = openvswitch,l2population
extension_drivers = port_security
overlay_ip_version = 4
---

 All good! I can launch instances and ping the Internet from within an
instance.

 So, when I try the Geneve instead, like this:

---
[ml2]
type_drivers = geneve,vxlan,flat,vlan
tenant_network_types = geneve,vxlan,flat,vlan
mechanism_drivers = openvswitch,l2population
extension_drivers = port_security
overlay_ip_version = 4
---

NOTE: I can easily bring up a clean Ubuntu 18.04 server image, and install
OpenStack from scratch, to do the each test independently. I have my own
Ansible automation to deploy OpenStack Queens on Ubuntu 18.04. As I've
said, it works fine with VXLAN but, not with Geneve (2 lines change from
one deployment, to another).


So, while trying to create an instance with Geneve, the following error
appear on Neutron logs:

---
2018-06-04 22:55:40.677 11426 ERROR neutron.plugins.ml2.managers
[req-c988c769-690e-45db-b810-3b35f0e8cba8 bda3aed0ccae46008f9928885280c085
46f4131518fd4b699ac80bc867fd1832 - default default] Failed to bind port
17e4dd94-7655-40f5-9077-b6e7c583c7eb on host queens-1 for vnic_type normal
using segments [{'network_id': '6f805d1f-dfb6-42f6-846a-e39fac80ca8c',
'segmentation_id': 19, 'physical_network': None, 'id':
'51a89942-5a91-47c6-872b-b33d2e6d418d', 'network_type': u'geneve'}]
---

 Is it a bug?

 I appreciate any help!

Cheers!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [openstack-dev] Ocata - Networking OVN - option "neutron_sync_mode = repair" looks broken

[Martinx - ジェームズ]

BUG reported: https://bugs.launchpad.net/neutron/+bug/1689880

On 10 May 2017 at 10:49, Martinx - ジェームズ <thiagocmarti...@gmail.com> wrote:

> Hello,
>
>  I'm playing with Networking OVN, and planning to deploy it into
> production in a couple weeks.
>
>  After deploying everything and starting using OVN, with Floating IPs,
> multiple compute nodes and everything else, I can say that it looks
> awesome! Way better than "neutron-*-agents".
>
>  However, I noted that after running:
>
> ---
>  systemctl restart neutron-server
> ---
>
>  Literally ALL my stacks, on all projects, becomes unreachable!!!
>
>  After double checking everything, I did one single change in my
> ml2_conf.ini, from:
>
> ---
> neutron_sync_mode = repair
> ---
>
>  To:
>
> ---
> # neutron_sync_mode = off
> ---
>
>  Then, problem solved!
>
>  Now, I can restart the neutron-server without any problems!
>
>  Deep inside me, I'm freaking out with this... Because this whole OVN
> solution looks very, very delicate. I mean, what to do if this happens
> again?
>
>  Ask ALL my customers to rebuild their stacks?!
>
>  How to REALLY repair OVN?
>
>  Now that the problem is solved, I'll keep trying to use and stress test
> it even more but, I'm losing confidence on Networking OVN on its current
> state.
>
>  I'm using Ubuntu 16.04 with Kernel 4.8 (HWE), plus Ocata from Cloud
> Archive.
>
> Cheers!
> Thiago
>
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

[openstack-dev] Ocata - Networking OVN - option "neutron_sync_mode = repair" looks broken

[Martinx - ジェームズ]

Hello,

 I'm playing with Networking OVN, and planning to deploy it into production
in a couple weeks.

 After deploying everything and starting using OVN, with Floating IPs,
multiple compute nodes and everything else, I can say that it looks
awesome! Way better than "neutron-*-agents".

 However, I noted that after running:

---
 systemctl restart neutron-server
---

 Literally ALL my stacks, on all projects, becomes unreachable!!!

 After double checking everything, I did one single change in my
ml2_conf.ini, from:

---
neutron_sync_mode = repair
---

 To:

---
# neutron_sync_mode = off
---

 Then, problem solved!

 Now, I can restart the neutron-server without any problems!

 Deep inside me, I'm freaking out with this... Because this whole OVN
solution looks very, very delicate. I mean, what to do if this happens
again?

 Ask ALL my customers to rebuild their stacks?!

 How to REALLY repair OVN?

 Now that the problem is solved, I'll keep trying to use and stress test it
even more but, I'm losing confidence on Networking OVN on its current state.

 I'm using Ubuntu 16.04 with Kernel 4.8 (HWE), plus Ocata from Cloud
Archive.

Cheers!
Thiago
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

[Openstack] Ubuntu - Ocata - Ceilometer / Gnocchi, docs doesn't looks consistent.

[Martinx - ジェームズ]

Hello!

 I'm trying to install Ceilometer (with Gnocchi) on Ubuntu with Ocata Cloud
Archive, however, the following documents doesn't match each other:

 The first one, to install Ceilometer at the controller, the
ceilometer.conf, points to Gnochi, it have the "[service_credentials]"
group, it does not have the "[keystone_authtoken]" group (which is only at
the gnocchi.conf) and it uses "project_domain_name" instead of old
"project_domain_id".

 Ceilometer Controller:


https://docs.openstack.org/project-install-guide/telemetry/ocata/install-base-ubuntu.html


 However, the next step, which is to configure the compute nodes, have a
very different ceilometer.conf!

 It have "auth_strategy = keystone" on its "[DEFAULT]" group. plus a
"[keystone_authtoken]" group, also, its "[service_credentials]" is
different! It uses old "project_domain_id" options, instead of new
"project_domain_name"...

 Ceilometer Compute:


https://docs.openstack.org/project-install-guide/telemetry/ocata/install-compute-ubuntu.html

 I'm confused... My ceilometer-agent-compute.log is printing many LibVirt
errors and there is no data being collected.


 Some of the errors:


 ceilometer-agent-compute.log:

 ceilometer.agent.manager [-] Skip pollster disk.usage, no resources found
this cycle
 .
 ceilometer-agent-compute: "ceilometer.agent.manager" "libvirtError"
"metadata not found"


 ceilometer-collector.log:

 ceilometer.agent.manager [-] Skip pollster disk.usage, no resources found
this cycle
 .
 ceilometer.dispatcher.gnocchi [-] Resource
079881dd-4363-40ec-bbee-6a00ea3816f3 does not exist (HTTP 404)



 Any tips?

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [openstack-dev] Ocata - Ubuntu 16.04 - OVN does not work with DPDK

[Martinx - ジェームズ]

On 11 April 2017 at 11:08, Russell Bryant <rbry...@redhat.com> wrote:

>
>
> On Mon, Apr 10, 2017 at 4:49 PM, Martinx - ジェームズ <
> thiagocmarti...@gmail.com> wrote:
>
>>
>>
>> On 8 April 2017 at 00:37, Martinx - ジェームズ <thiagocmarti...@gmail.com>
>> wrote:
>>
>>> Guys,
>>>
>>>  I manage to deploy Ocata on Ubuntu 16.04 with OVN for the first time
>>> ever, today!
>>>
>>>  It looks very, very good... OVN L3 Router is working, OVN DHCP
>>> working... bridge mappings "br-ex" on each compute node... All good!
>>>
>>>  Then, I've said: time for DPDK!
>>>
>>>  I manage to use OVS with DPDK easily on top of Ubuntu (plus Ocata Cloud
>>> Archive) with plain KVM, no OpenStack, so, I have experience about how to
>>> setup DPDK, OVS+DPDK, Libvirt vhostuser, KVM and etc...
>>>
>>>  After configuring DPDK on a compute node, I tried the following
>>> instructions:
>>>
>>>  https://docs.openstack.org/developer/networking-ovn/dpdk.html
>>>
>>>  It looks quite simple!
>>>
>>>  To make things even simpler, I have just 1 controller, and 1 compute
>>> node, to begin with, before enabling DPDK at the compute node and changing
>>> the "br-int" datapath, I deleted all OVN Routers and all Neutron Networks
>>> and Subnets, that was previously working with regular OVS (no DPDK).
>>>
>>>  Then, after enabling DPDK and updating the "br-int" and the "br-ex"
>>> interfaces, right after connecting the "OVN L3 Router" into the "ext-net /
>>> br-ex" network, the following errors appeared on OpenvSwitch logs of the
>>> related compute node (OpenFlow error):
>>>
>>>
>>>  * After connecting OVN L3 Router against the "ext-net / br-ex" Flat /
>>> VLAN Network:
>>>
>>>  ovs-vswitchd.log:
>>>
>>>  http://paste.openstack.org/show/605800/
>>>
>>>  ovn-controller.log:
>>>
>>>  http://paste.openstack.org/show/605801/
>>>
>>>
>>>  Also, after connecting the OVN L3 Router into the local (GENEVE)
>>> network, very similar error messages appeared on OpenvSwitch logs...
>>>
>>>
>>>  * After connecting OVN L3 Router on a "local" GENEVE Network:
>>>
>>>  ovs-vswitchd.log:
>>>
>>>  http://paste.openstack.org/show/605804/
>>>
>>>  ovn-controller.log:
>>>
>>>  http://paste.openstack.org/show/605805/
>>>
>>>
>>>  * Output of "ovs-vsctl show" at the single compute node, after plugging
>>> the OVN L3 Router against the two networks (external / GENEVE):
>>>
>>>  http://paste.openstack.org/show/605806/
>>>
>>>
>>>  Then, I tried to launch an Instance anyway and, for my surprise, the
>>> Instance was created! Using vhostuser OVS+DPDK socket!
>>>
>>>  Also, the Instance got its IP! Which is great!
>>>
>>>  However, the Instance can not ping its OVN L3 Router (its default
>>> gateway), with or without any kind of security groups applied, no deal...
>>> :-(
>>>
>>>  BTW, the Instance did not received the ARP stuff of the OVN L3 Router,
>>> I mean, for the instance, the gateway IP on "arp -an" shows "".
>>>
>>>
>>>  * The ovs-vswitchd.log after launching an Instance on top of
>>> OVN/OVS+DPDK:
>>>
>>>  http://paste.openstack.org/show/605807/
>>>
>>>  * The output of "ovs-vsctl show" after launching the above instance:
>>>
>>>  http://paste.openstack.org/show/605809/ - Line 33 is the dpdkvhostuser
>>>
>>>
>>>  Just to give another try, I started a second Instance, to see if the
>>> Instances can ping each other... Also did not worked, the Instances can not
>>> ping each other.
>>>
>>>
>>>  So, from what I'm seeing, OVN on top of DPDK does not work.
>>>
>>>  Any tips?
>>>
>>>
>>>  NOTE:
>>>
>>>  I tried to enable "hugepages" on my OpenStack's flavor, just in case...
>>> Then, I found another bug, it doesn't even boot the Instance:
>>>
>>>  https://bugs.launchpad.net/cloud-archive/+bug/1680956
>>>
>>>
>>>  For now, I'll deploy Ocata with regular OVN, no DPDK, but, my goal with
>>> this cloud is for high performance networks, s

Re: [openstack-dev] Ocata - Ubuntu 16.04 - OVN does not work with DPDK

[Martinx - ジェームズ]

On 11 April 2017 at 11:08, Russell Bryant <rbry...@redhat.com> wrote:

>
>
> On Mon, Apr 10, 2017 at 4:49 PM, Martinx - ジェームズ <
> thiagocmarti...@gmail.com> wrote:
>
>>
>>
>> On 8 April 2017 at 00:37, Martinx - ジェームズ <thiagocmarti...@gmail.com>
>> wrote:
>>
>>> Guys,
>>>
>>>  I manage to deploy Ocata on Ubuntu 16.04 with OVN for the first time
>>> ever, today!
>>>
>>>  It looks very, very good... OVN L3 Router is working, OVN DHCP
>>> working... bridge mappings "br-ex" on each compute node... All good!
>>>
>>>  Then, I've said: time for DPDK!
>>>
>>>  I manage to use OVS with DPDK easily on top of Ubuntu (plus Ocata Cloud
>>> Archive) with plain KVM, no OpenStack, so, I have experience about how to
>>> setup DPDK, OVS+DPDK, Libvirt vhostuser, KVM and etc...
>>>
>>>  After configuring DPDK on a compute node, I tried the following
>>> instructions:
>>>
>>>  https://docs.openstack.org/developer/networking-ovn/dpdk.html
>>>
>>>  It looks quite simple!
>>>
>>>  To make things even simpler, I have just 1 controller, and 1 compute
>>> node, to begin with, before enabling DPDK at the compute node and changing
>>> the "br-int" datapath, I deleted all OVN Routers and all Neutron Networks
>>> and Subnets, that was previously working with regular OVS (no DPDK).
>>>
>>>  Then, after enabling DPDK and updating the "br-int" and the "br-ex"
>>> interfaces, right after connecting the "OVN L3 Router" into the "ext-net /
>>> br-ex" network, the following errors appeared on OpenvSwitch logs of the
>>> related compute node (OpenFlow error):
>>>
>>>
>>>  * After connecting OVN L3 Router against the "ext-net / br-ex" Flat /
>>> VLAN Network:
>>>
>>>  ovs-vswitchd.log:
>>>
>>>  http://paste.openstack.org/show/605800/
>>>
>>>  ovn-controller.log:
>>>
>>>  http://paste.openstack.org/show/605801/
>>>
>>>
>>>  Also, after connecting the OVN L3 Router into the local (GENEVE)
>>> network, very similar error messages appeared on OpenvSwitch logs...
>>>
>>>
>>>  * After connecting OVN L3 Router on a "local" GENEVE Network:
>>>
>>>  ovs-vswitchd.log:
>>>
>>>  http://paste.openstack.org/show/605804/
>>>
>>>  ovn-controller.log:
>>>
>>>  http://paste.openstack.org/show/605805/
>>>
>>>
>>>  * Output of "ovs-vsctl show" at the single compute node, after plugging
>>> the OVN L3 Router against the two networks (external / GENEVE):
>>>
>>>  http://paste.openstack.org/show/605806/
>>>
>>>
>>>  Then, I tried to launch an Instance anyway and, for my surprise, the
>>> Instance was created! Using vhostuser OVS+DPDK socket!
>>>
>>>  Also, the Instance got its IP! Which is great!
>>>
>>>  However, the Instance can not ping its OVN L3 Router (its default
>>> gateway), with or without any kind of security groups applied, no deal...
>>> :-(
>>>
>>>  BTW, the Instance did not received the ARP stuff of the OVN L3 Router,
>>> I mean, for the instance, the gateway IP on "arp -an" shows "".
>>>
>>>
>>>  * The ovs-vswitchd.log after launching an Instance on top of
>>> OVN/OVS+DPDK:
>>>
>>>  http://paste.openstack.org/show/605807/
>>>
>>>  * The output of "ovs-vsctl show" after launching the above instance:
>>>
>>>  http://paste.openstack.org/show/605809/ - Line 33 is the dpdkvhostuser
>>>
>>>
>>>  Just to give another try, I started a second Instance, to see if the
>>> Instances can ping each other... Also did not worked, the Instances can not
>>> ping each other.
>>>
>>>
>>>  So, from what I'm seeing, OVN on top of DPDK does not work.
>>>
>>>  Any tips?
>>>
>>>
>>>  NOTE:
>>>
>>>  I tried to enable "hugepages" on my OpenStack's flavor, just in case...
>>> Then, I found another bug, it doesn't even boot the Instance:
>>>
>>>  https://bugs.launchpad.net/cloud-archive/+bug/1680956
>>>
>>>
>>>  For now, I'll deploy Ocata with regular OVN, no DPDK, but, my goal with
>>> this cloud is for high performance networks, s

Re: [openstack-dev] Ocata - Ubuntu 16.04 - OVN does not work with DPDK

[Martinx - ジェームズ]

On 8 April 2017 at 00:37, Martinx - ジェームズ <thiagocmarti...@gmail.com> wrote:

> Guys,
>
>  I manage to deploy Ocata on Ubuntu 16.04 with OVN for the first time
> ever, today!
>
>  It looks very, very good... OVN L3 Router is working, OVN DHCP working...
> bridge mappings "br-ex" on each compute node... All good!
>
>  Then, I've said: time for DPDK!
>
>  I manage to use OVS with DPDK easily on top of Ubuntu (plus Ocata Cloud
> Archive) with plain KVM, no OpenStack, so, I have experience about how to
> setup DPDK, OVS+DPDK, Libvirt vhostuser, KVM and etc...
>
>  After configuring DPDK on a compute node, I tried the following
> instructions:
>
>  https://docs.openstack.org/developer/networking-ovn/dpdk.html
>
>  It looks quite simple!
>
>  To make things even simpler, I have just 1 controller, and 1 compute
> node, to begin with, before enabling DPDK at the compute node and changing
> the "br-int" datapath, I deleted all OVN Routers and all Neutron Networks
> and Subnets, that was previously working with regular OVS (no DPDK).
>
>  Then, after enabling DPDK and updating the "br-int" and the "br-ex"
> interfaces, right after connecting the "OVN L3 Router" into the "ext-net /
> br-ex" network, the following errors appeared on OpenvSwitch logs of the
> related compute node (OpenFlow error):
>
>
>  * After connecting OVN L3 Router against the "ext-net / br-ex" Flat /
> VLAN Network:
>
>  ovs-vswitchd.log:
>
>  http://paste.openstack.org/show/605800/
>
>  ovn-controller.log:
>
>  http://paste.openstack.org/show/605801/
>
>
>  Also, after connecting the OVN L3 Router into the local (GENEVE) network,
> very similar error messages appeared on OpenvSwitch logs...
>
>
>  * After connecting OVN L3 Router on a "local" GENEVE Network:
>
>  ovs-vswitchd.log:
>
>  http://paste.openstack.org/show/605804/
>
>  ovn-controller.log:
>
>  http://paste.openstack.org/show/605805/
>
>
>  * Output of "ovs-vsctl show" at the single compute node, after plugging
> the OVN L3 Router against the two networks (external / GENEVE):
>
>  http://paste.openstack.org/show/605806/
>
>
>  Then, I tried to launch an Instance anyway and, for my surprise, the
> Instance was created! Using vhostuser OVS+DPDK socket!
>
>  Also, the Instance got its IP! Which is great!
>
>  However, the Instance can not ping its OVN L3 Router (its default
> gateway), with or without any kind of security groups applied, no deal...
> :-(
>
>  BTW, the Instance did not received the ARP stuff of the OVN L3 Router, I
> mean, for the instance, the gateway IP on "arp -an" shows "".
>
>
>  * The ovs-vswitchd.log after launching an Instance on top of OVN/OVS+DPDK:
>
>  http://paste.openstack.org/show/605807/
>
>  * The output of "ovs-vsctl show" after launching the above instance:
>
>  http://paste.openstack.org/show/605809/ - Line 33 is the dpdkvhostuser
>
>
>  Just to give another try, I started a second Instance, to see if the
> Instances can ping each other... Also did not worked, the Instances can not
> ping each other.
>
>
>  So, from what I'm seeing, OVN on top of DPDK does not work.
>
>  Any tips?
>
>
>  NOTE:
>
>  I tried to enable "hugepages" on my OpenStack's flavor, just in case...
> Then, I found another bug, it doesn't even boot the Instance:
>
>  https://bugs.launchpad.net/cloud-archive/+bug/1680956
>
>
>  For now, I'll deploy Ocata with regular OVN, no DPDK, but, my goal with
> this cloud is for high performance networks, so, I need DPDK, and I also
> need GENEVE and Provider Networks, everything on top of DPDK.
>
> ---
>  After researching more about this "high perf networks", I found this:
>
>  * DPDK-like performance in Linux kernel with XDP !
>
>  http://openvswitch.org/support/ovscon2016/7/0930-pettit.pdf
>
>  https://www.iovisor.org/technology/xdp
>  https://www.iovisor.org/technology/ebpf
>
>  https://qmonnet.github.io/whirl-offload/2016/09/01/dive-into-bpf/
>
>  But I have no idea about how to use OpenvSwitch with this thing, however,
> if I can achieve DPDK-Like performance, without DPDK, using just plain
> Linux, I'm a 100% sure that I'll prefer it!
>
>  I'm okay to give OpenvSwitch + DPDK another try, even knowing that it
> [OVS] STILL have serious problems (https://bugs.launchpad.net/
> ubuntu/+source/openvswitch/+bug/1577088)...
> ---
>
>  OpenStack on Ubuntu rocks!   :-D
>
> Thanks!
> Thiago
>

I just realized how cool IO Vis

[openstack-dev] Ocata - Ubuntu 16.04 - OVN does not work with DPDK

[Martinx - ジェームズ]

Guys,

 I manage to deploy Ocata on Ubuntu 16.04 with OVN for the first time ever,
today!

 It looks very, very good... OVN L3 Router is working, OVN DHCP working...
bridge mappings "br-ex" on each compute node... All good!

 Then, I've said: time for DPDK!

 I manage to use OVS with DPDK easily on top of Ubuntu (plus Ocata Cloud
Archive) with plain KVM, no OpenStack, so, I have experience about how to
setup DPDK, OVS+DPDK, Libvirt vhostuser, KVM and etc...

 After configuring DPDK on a compute node, I tried the following
instructions:

 https://docs.openstack.org/developer/networking-ovn/dpdk.html

 It looks quite simple!

 To make things even simpler, I have just 1 controller, and 1 compute node,
to begin with, before enabling DPDK at the compute node and changing the
"br-int" datapath, I deleted all OVN Routers and all Neutron Networks and
Subnets, that was previously working with regular OVS (no DPDK).

 Then, after enabling DPDK and updating the "br-int" and the "br-ex"
interfaces, right after connecting the "OVN L3 Router" into the "ext-net /
br-ex" network, the following errors appeared on OpenvSwitch logs of the
related compute node (OpenFlow error):


 * After connecting OVN L3 Router against the "ext-net / br-ex" Flat / VLAN
Network:

 ovs-vswitchd.log:

 http://paste.openstack.org/show/605800/

 ovn-controller.log:

 http://paste.openstack.org/show/605801/


 Also, after connecting the OVN L3 Router into the local (GENEVE) network,
very similar error messages appeared on OpenvSwitch logs...


 * After connecting OVN L3 Router on a "local" GENEVE Network:

 ovs-vswitchd.log:

 http://paste.openstack.org/show/605804/

 ovn-controller.log:

 http://paste.openstack.org/show/605805/


 * Output of "ovs-vsctl show" at the single compute node, after plugging
the OVN L3 Router against the two networks (external / GENEVE):

 http://paste.openstack.org/show/605806/


 Then, I tried to launch an Instance anyway and, for my surprise, the
Instance was created! Using vhostuser OVS+DPDK socket!

 Also, the Instance got its IP! Which is great!

 However, the Instance can not ping its OVN L3 Router (its default
gateway), with or without any kind of security groups applied, no deal...
:-(

 BTW, the Instance did not received the ARP stuff of the OVN L3 Router, I
mean, for the instance, the gateway IP on "arp -an" shows "".


 * The ovs-vswitchd.log after launching an Instance on top of OVN/OVS+DPDK:

 http://paste.openstack.org/show/605807/

 * The output of "ovs-vsctl show" after launching the above instance:

 http://paste.openstack.org/show/605809/ - Line 33 is the dpdkvhostuser


 Just to give another try, I started a second Instance, to see if the
Instances can ping each other... Also did not worked, the Instances can not
ping each other.


 So, from what I'm seeing, OVN on top of DPDK does not work.

 Any tips?


 NOTE:

 I tried to enable "hugepages" on my OpenStack's flavor, just in case...
Then, I found another bug, it doesn't even boot the Instance:

 https://bugs.launchpad.net/cloud-archive/+bug/1680956


 For now, I'll deploy Ocata with regular OVN, no DPDK, but, my goal with
this cloud is for high performance networks, so, I need DPDK, and I also
need GENEVE and Provider Networks, everything on top of DPDK.

---
 After researching more about this "high perf networks", I found this:

 * DPDK-like performance in Linux kernel with XDP !

 http://openvswitch.org/support/ovscon2016/7/0930-pettit.pdf

 https://www.iovisor.org/technology/xdp
 https://www.iovisor.org/technology/ebpf

 https://qmonnet.github.io/whirl-offload/2016/09/01/dive-into-bpf/

 But I have no idea about how to use OpenvSwitch with this thing, however,
if I can achieve DPDK-Like performance, without DPDK, using just plain
Linux, I'm a 100% sure that I'll prefer it!

 I'm okay to give OpenvSwitch + DPDK another try, even knowing that it
[OVS] STILL have serious problems (
https://bugs.launchpad.net/ubuntu/+source/openvswitch/+bug/1577088)...
---

 OpenStack on Ubuntu rocks!   :-D

Thanks!
Thiago
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [Openstack] Request for documentation about Ocata on Ubuntu with Networking OVN

[Martinx - ジェームズ]

I'm also seeing the following errors on ovs-vswitchd.log:

http://paste.openstack.org/show/605441/

On 4 April 2017 at 15:19, Martinx - ジェームズ <thiagocmarti...@gmail.com> wrote:

> There are more errors on Compute node as well, nova.log:
>
> http://paste.openstack.org/show/605439/
>
> On 4 April 2017 at 15:14, Martinx - ジェームズ <thiagocmarti...@gmail.com>
> wrote:
>
>> Hello!
>>
>>  Currently, I'm trying to deploy Ocata with Networking OVN, so far, no
>> success.
>>
>>  I can easily deploy Ocata in a multi-node environment (1 Controller, 1
>> Network and 2 Computes), using regular "neutron-openvswitch-agent" and its
>> "ovs-firewall-driver".
>>
>>  However, when I try the following instructions:
>>
>>  https://docs.openstack.org/developer/networking-ovn/install.html
>>
>>  It just doesn't work.
>>
>>  From what I'm seeing, on Ubuntu, the package names are different, like
>> "ovn-central" for the controllers and "ovn-host" for the compute nodes.
>>
>>  All steps on the above guide, works, no errors on logs. But, when I try
>> to launch an Instance, it fails, with errors on neutron-server.log:
>>
>>  http://paste.openstack.org/show/605438/
>>
>>  Looks like that Ubuntu have all the required packages but, just the
>> documentation that is not available.
>>
>>  I really want to move to OVN, can't deal with Network Node and its
>> namespaces anymore. Also, we need DPDK, and OVN have a better support for
>> tunneled networks on top of DPDK.
>>
>>  In my deployment, first, I deployed it with neutron-openvswitch-agents,
>> later, I deleted all instances, all stacks and all networks, changed it to
>> OVN, and tried again but, it failed.
>>
>>  I really appreciate any help in this direction...
>>
>> Thanks!
>> Thiago
>>
>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Request for documentation about Ocata on Ubuntu with Networking OVN

[Martinx - ジェームズ]

Hello!

 Currently, I'm trying to deploy Ocata with Networking OVN, so far, no
success.

 I can easily deploy Ocata in a multi-node environment (1 Controller, 1
Network and 2 Computes), using regular "neutron-openvswitch-agent" and its
"ovs-firewall-driver".

 However, when I try the following instructions:

 https://docs.openstack.org/developer/networking-ovn/install.html

 It just doesn't work.

 From what I'm seeing, on Ubuntu, the package names are different, like
"ovn-central" for the controllers and "ovn-host" for the compute nodes.

 All steps on the above guide, works, no errors on logs. But, when I try to
launch an Instance, it fails, with errors on neutron-server.log:

 http://paste.openstack.org/show/605438/

 Looks like that Ubuntu have all the required packages but, just the
documentation that is not available.

 I really want to move to OVN, can't deal with Network Node and its
namespaces anymore. Also, we need DPDK, and OVN have a better support for
tunneled networks on top of DPDK.

 In my deployment, first, I deployed it with neutron-openvswitch-agents,
later, I deleted all instances, all stacks and all networks, changed it to
OVN, and tried again but, it failed.

 I really appreciate any help in this direction...

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Request for documentation about Ocata on Ubuntu with Networking OVN

[Martinx - ジェームズ]

There are more errors on Compute node as well, nova.log:

http://paste.openstack.org/show/605439/

On 4 April 2017 at 15:14, Martinx - ジェームズ <thiagocmarti...@gmail.com> wrote:

> Hello!
>
>  Currently, I'm trying to deploy Ocata with Networking OVN, so far, no
> success.
>
>  I can easily deploy Ocata in a multi-node environment (1 Controller, 1
> Network and 2 Computes), using regular "neutron-openvswitch-agent" and its
> "ovs-firewall-driver".
>
>  However, when I try the following instructions:
>
>  https://docs.openstack.org/developer/networking-ovn/install.html
>
>  It just doesn't work.
>
>  From what I'm seeing, on Ubuntu, the package names are different, like
> "ovn-central" for the controllers and "ovn-host" for the compute nodes.
>
>  All steps on the above guide, works, no errors on logs. But, when I try
> to launch an Instance, it fails, with errors on neutron-server.log:
>
>  http://paste.openstack.org/show/605438/
>
>  Looks like that Ubuntu have all the required packages but, just the
> documentation that is not available.
>
>  I really want to move to OVN, can't deal with Network Node and its
> namespaces anymore. Also, we need DPDK, and OVN have a better support for
> tunneled networks on top of DPDK.
>
>  In my deployment, first, I deployed it with neutron-openvswitch-agents,
> later, I deleted all instances, all stacks and all networks, changed it to
> OVN, and tried again but, it failed.
>
>  I really appreciate any help in this direction...
>
> Thanks!
> Thiago
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Is there some Openstack version that supports Docker as a hypervisor driver?

[Martinx - ジェームズ]

Why not LXD instead of Docker?   ;-)

On 31 March 2017 at 08:17, Jorge Luiz Correa  wrote:

> Hi! I would like to know if Openstack supports Docker as a 'hypervisor'.
> Some time ago I started to test the nova-docker that was configured as a
> driver:
>
> compute_driver=novadocker.virt.docker.DockerDriver
>
> There is a documentation in Openstack Wiki that points to this scenario:
>
> https://wiki.openstack.org/wiki/Docker
>
> But the Git project says "This project is no longer maintained".
>
> There is also a Hypervisor Matrix support and Docker is marked as Class C
> (something like deprecated or not supported).
>
> But, I don't know rightly what times theses docs were written. Is the Wiki
> updated?
>
> I'm using Mitaka. Some other version has already a driver to be used like
> that, so we can configure one compute node to run docker virtual machines?
>
> Thanks!
>
> - JLC
>
> ___
> Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/
> openstack
> Post to : openstack@lists.openstack.org
> Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/
> openstack
>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Request for more documents: Ceilometer, Gnocchi, Aodh, Senlin and OVN.

[Martinx - ジェームズ]

Here is it, the "gnocchi.conf" example: https://paste.ubuntu.com/24098302/

Hope it helps!

On 2 March 2017 at 05:35, Mārtiņš Jakubovičs <martins-li...@hostnet.lv>
wrote:

> Hello Thiago,
>
> Could you, please, share gnocchi.conf file without sensitive data? Looks
> like I miss something, but can't find what exactly... :/
> Best regards,
> Martins
>
>
> On 2017.03.02. 03:09, Martinx - ジェームズ wrote:
>
> Hey Mārtiņš,
>
> I installed it from Ubuntu packages.
>
> I'm working right now, on a fully automated (and super simple) Ansible
> Ocata deployment, that includes Gnocchi.
>
> I think that it will be ready by Monday!
>
> If you have a fresh installed Ubuntu 16.04, you'll be able to give it a
> try... Or, just take a look at the automation source, it will be on Github.
>
> Cheers!
> Thiago
>
> On 1 March 2017 at 08:36, Mārtiņš Jakubovičs <martins-li...@hostnet.lv>
> wrote:
>
>> Hello Thiago,
>>
>> Did you installed gnocchi successfully?
>>
>> I also try to deploy gnocchi but stuck in step "gnocchi-upgrade", from
>> http://gnocchi.xyz/install.html documentation. Did you installed it from
>> pip or from OS packages?
>>
>> Thanks,
>>
>> Martins
>>
>> On 2017.02.16. 07:16, Martinx - ジェームズ wrote:
>>
>> Guys,
>>
>>  I'm trying very hard to put those pieces of software to work together,
>> however, it is not working as expected... I think that I am almost there!
>>
>>  Basically, I need a reliable AutoScaling solution.
>>
>>  Looks like that Senlin is the answer, I manage to make the "openstack
>> cluster ..." commands to work (Ocata on Ubuntu 16.04 via Cloud Archive), I
>> can run a few things here but, it is not complete... I can see errors on
>> Aodh, on Gnocchi...
>>
>>  Thing is, there is no clear documentation about how to install and
>> integrate those things on Ubuntu.
>>
>>  Also, I'm tired of L3 / DHCP agents and lots of Linux Namespaces, looks
>> like that OpenvSwitch and its OVN will replace those things but, again, I
>> can't find clear docs about that, especially when with OVS+DPDK...
>>
>>  Any plans for Ocata documentation?
>>
>>  BTW, OpenStack is getting better and better! This project is awesome!
>> :-D
>>
>> Thanks!
>> Thiago
>>
>>
>> ___
>> Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>> Post to : openstack@lists.openstack.org
>> Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>
>> ___ Mailing list:
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack Post to
>>  : openstack@lists.openstack.org Unsubscribe :
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Request for more documents: Ceilometer, Gnocchi, Aodh, Senlin and OVN.

[Martinx - ジェームズ]

Hey Mārtiņš,

I installed it from Ubuntu packages.

I'm working right now, on a fully automated (and super simple) Ansible
Ocata deployment, that includes Gnocchi.

I think that it will be ready by Monday!

If you have a fresh installed Ubuntu 16.04, you'll be able to give it a
try... Or, just take a look at the automation source, it will be on Github.

Cheers!
Thiago

On 1 March 2017 at 08:36, Mārtiņš Jakubovičs <martins-li...@hostnet.lv>
wrote:

> Hello Thiago,
>
> Did you installed gnocchi successfully?
>
> I also try to deploy gnocchi but stuck in step "gnocchi-upgrade", from
> http://gnocchi.xyz/install.html documentation. Did you installed it from
> pip or from OS packages?
>
> Thanks,
>
> Martins
>
> On 2017.02.16. 07:16, Martinx - ジェームズ wrote:
>
> Guys,
>
>  I'm trying very hard to put those pieces of software to work together,
> however, it is not working as expected... I think that I am almost there!
>
>  Basically, I need a reliable AutoScaling solution.
>
>  Looks like that Senlin is the answer, I manage to make the "openstack
> cluster ..." commands to work (Ocata on Ubuntu 16.04 via Cloud Archive), I
> can run a few things here but, it is not complete... I can see errors on
> Aodh, on Gnocchi...
>
>  Thing is, there is no clear documentation about how to install and
> integrate those things on Ubuntu.
>
>  Also, I'm tired of L3 / DHCP agents and lots of Linux Namespaces, looks
> like that OpenvSwitch and its OVN will replace those things but, again, I
> can't find clear docs about that, especially when with OVS+DPDK...
>
>  Any plans for Ocata documentation?
>
>  BTW, OpenStack is getting better and better! This project is awesome!
> :-D
>
> Thanks!
> Thiago
>
>
> ___
> Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> Post to : openstack@lists.openstack.org
> Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>
>
>
> ___
> Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/
> openstack
> Post to : openstack@lists.openstack.org
> Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/
> openstack
>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Ocata / Senlin / Ubuntu LTS: os.heat.stack profile create fails, SDKException.

[Martinx - ジェームズ]

Damn, just fixed it! Region name was wrong...   lol

Sorry about the buzz!

On 27 February 2017 at 16:04, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

> Bug reported (after talking about it on IRC / #senlin): https://bugs.
> launchpad.net/senlin/+bug/1668398
>
> On 23 February 2017 at 01:46, Martinx - ジェームズ <thiagocmarti...@gmail.com>
> wrote:
>
>> Hey guys,
>>
>>  My Senlin is not working as expected... I need some help.
>>
>>  I can create Senlin profiles of "os.nova.server" type, and then, create
>> clusters, policies, receivers and etc... Looks good.
>>
>>  However, when I try to do the same (profile create) but, using
>> "os.heat.stack", "openstack cluster profile create" fails.
>>
>>
>>  As follows:
>>
>>
>>  os.nova.server, good:
>>
>>  https://paste.ubuntu.com/24051251/
>>
>>
>>  os.heat.stack, fails:
>>
>>  https://paste.ubuntu.com/24051259/
>>
>>  "SDKException: Could not find requested endpoint in Service Catalog"
>>
>>  Output of command with "--debug":
>>
>>  https://paste.ubuntu.com/24051281/
>>
>>
>>  I can create a Senlin profile of type "os.nova.server" but not
>> "os.heat.stack"? Does it need a different endpoint?? I'm confused...
>>
>>  Heat is working, I can "openstack stack create ..." normally on this
>> box...
>>
>> Thanks,
>> Thiago
>>
>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Ocata / Senlin / Ubuntu LTS: os.heat.stack profile create fails, SDKException.

[Martinx - ジェームズ]

Bug reported (after talking about it on IRC / #senlin):
https://bugs.launchpad.net/senlin/+bug/1668398

On 23 February 2017 at 01:46, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

> Hey guys,
>
>  My Senlin is not working as expected... I need some help.
>
>  I can create Senlin profiles of "os.nova.server" type, and then, create
> clusters, policies, receivers and etc... Looks good.
>
>  However, when I try to do the same (profile create) but, using
> "os.heat.stack", "openstack cluster profile create" fails.
>
>
>  As follows:
>
>
>  os.nova.server, good:
>
>  https://paste.ubuntu.com/24051251/
>
>
>  os.heat.stack, fails:
>
>  https://paste.ubuntu.com/24051259/
>
>  "SDKException: Could not find requested endpoint in Service Catalog"
>
>  Output of command with "--debug":
>
>  https://paste.ubuntu.com/24051281/
>
>
>  I can create a Senlin profile of type "os.nova.server" but not
> "os.heat.stack"? Does it need a different endpoint?? I'm confused...
>
>  Heat is working, I can "openstack stack create ..." normally on this
> box...
>
> Thanks,
> Thiago
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] OpenStack Ocata for Ubuntu 16.04 LTS

[Martinx - ジェームズ]

This is awesome Corey! Really fast...

 Ubuntu is the leading operating system for both running your cloud, or for
running it at the cloud!

 However, OpenStack's future in both Debian and Ubuntu is in checkmate,
look:

 http://lists.openstack.org/pipermail/openstack-dev/2017-
February/112220.html

 I think that Mark Shuttleworth have to do something about this situation.

 What do you think?

Cheers!
Thiago

On 22 February 2017 at 22:09, Corey Bryant 
wrote:

> Hi All,
>
> The Ubuntu OpenStack team at Canonical is pleased to announce the general
> availability of OpenStack Ocata on Ubuntu 16.04 LTS via the Ubuntu Cloud
> Archive. Ocata is the latest release of OpenStack and the first in the new
> OpenStack release cadence, with a focus on stabilization of existing
> features along with usability improvements. Details of the Ocata release
> can be found at:  https://www.openstack.org/software/ocata
>
> To get access to the Ubuntu Ocata packages:
>
> Ubuntu 16.04 LTS
>
> 
>
> You can enable the Ubuntu Cloud Archive pocket for OpenStack Ocata on
> Ubuntu 16.04 installations by running the following commands:
>
> sudo add-apt-repository cloud-archive:ocata
>
> sudo apt update
>
> The Ubuntu Cloud Archive for Ocata includes updates for:
>
> aodh, barbican, ceilometer, cinder, congress, designate,
> designate-dashboard, dpdk (16.11), glance, gnocchi, heat, horizon, ironic,
> libvirt (2.5.0), keystone, magnum, manila, manila-ui, mistral, murano,
> murano-dashboard, networking-ovn, networking-sfc, neutron,
> neutron-dynamic-routing, neutron-fwaas, neutron-lbaas,
> neutron-lbaas-dashboard, nova, nova-lxd, openstack-trove, panko, qemu
> (2.8), sahara, sahara-dashboard, senlin, swift, trove-dashboard, watcher
> and zaqar
>
> For a full list of packages and versions, please refer to [0].
>
> Migration considerations
>
> 
>
> * API’s now running under apache2 with mod_wsgi
>
> In Ocata we’ve updated the following APIs to run under apache2 with
> mod_wsgi: aodh-api, barbican-api, ceilometer-api, cinder-api, and the
> nova-placement-api. Please keep this in mind as the packages will no longer
> install systemd unit files for these services, and will instead install
> apache2 with corresponding apache2 site configuration.
>
> * libvirt 2.5.0 changes
>
> In this release of libvirt, the libvirt-bin systemd service has been
> renamed to libvirtd, and the unix_sock_group has changed from libvirtd to
> libvirt.
>
> * openstack-dashboard changes
>
> In Ocata we’ve moved static assets from /usr/share/openstack-dashboard/static
> to /var/lib/openstack-dashboard/static.
>
> Known Issues
>
> ---
>
> nova console log is empty with libvirt 2.5.0:
> https://bugs.launchpad.net/bugs/1667033
>
> Branch Package Builds
>
> ---
>
> If you would like to try out the latest updates to branches, we deliver
> continuously integrated packages on each upstream commit via the following
> PPA’s:
>
>   sudo add-apt-repository ppa:openstack-ubuntu-testing/liberty
>
>   sudo add-apt-repository ppa:openstack-ubuntu-testing/mitaka
>
>   sudo add-apt-repository ppa:openstack-ubuntu-testing/newton
>
>   sudo add-apt-repository ppa:openstack-ubuntu-testing/ocata
>
> Reporting bugs
>
> -
>
> If you have any issues please report bugs using the 'ubuntu-bug' tool to
> ensure that bugs get logged in the right place in Launchpad:
>
> sudo ubuntu-bug nova-conductor
>
> Thanks to everyone who has contributed to OpenStack Ocata, both upstream
> and downstream.  And special thanks to the puppet modules team for their
> continued early testing of OpenStack development releases.
>
> Have fun and see you in Pike!
>
> Regards,
>
> Corey
>
> (on behalf of the Ubuntu OpenStack team)
>
> [0] http://reqorts.qa.ubuntu.com/reports/ubuntu-server/cloud-arc
> hive/ocata_versions.html
>
>
> ___
> Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstac
> k
> Post to : openstack@lists.openstack.org
> Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstac
> k
>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Ocata / Senlin / Ubuntu LTS: os.heat.stack profile create fails, SDKException.

[Martinx - ジェームズ]

Hey guys,

 My Senlin is not working as expected... I need some help.

 I can create Senlin profiles of "os.nova.server" type, and then, create
clusters, policies, receivers and etc... Looks good.

 However, when I try to do the same (profile create) but, using
"os.heat.stack", "openstack cluster profile create" fails.


 As follows:


 os.nova.server, good:

 https://paste.ubuntu.com/24051251/


 os.heat.stack, fails:

 https://paste.ubuntu.com/24051259/

 "SDKException: Could not find requested endpoint in Service Catalog"

 Output of command with "--debug":

 https://paste.ubuntu.com/24051281/


 I can create a Senlin profile of type "os.nova.server" but not
"os.heat.stack"? Does it need a different endpoint?? I'm confused...

 Heat is working, I can "openstack stack create ..." normally on this box...

Thanks,
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Gnocchi (Pandas) error, ValueError: buffer source array is read-only...

[Martinx - ジェームズ]

Awesome! I've sent an e-mail to Ubuntu Devel and James Page about that...
 :-)

On 17 February 2017 at 16:32, Chuck Short <chuck.sh...@canonical.com> wrote:

> Hi
>
> Uploading a fix now.
>
> On Fri, Feb 17, 2017 at 3:55 PM, Martinx - ジェームズ <
> thiagocmarti...@gmail.com> wrote:
>
>> Thank you, that's precisely what happened! Filling a bug report ASAP...
>>
>> On 17 February 2017 at 13:48, Mehdi Abaakouk <sil...@sileht.net> wrote:
>>
>>> On Fri, Feb 17, 2017 at 01:14:11PM -0500, Martinx - ジェームズ wrote:
>>>
>>>> "ERROR gnocchi.storage._carbonara ValueError: buffer source array is
>>>> read-only"
>>>>
>>>>
>>> * Using Ubuntu 16.04 + Ocata Cloud Archive Staging PPA.
>>>>
>>>>
>>> It seems Ubuntu 16.04 ship pandas 0.17, while Gnocchi required 0.18.
>>>
>>> You should open a bug on ubuntu or cloud-archive packaging.
>>>
>>>
>>> --
>>> Mehdi Abaakouk
>>> mail: sil...@sileht.net
>>> irc: sileht
>>>
>>
>>
>> ___
>> Mailing list: http://lists.openstack.org/cgi
>> -bin/mailman/listinfo/openstack
>> Post to : openstack@lists.openstack.org
>> Unsubscribe : http://lists.openstack.org/cgi
>> -bin/mailman/listinfo/openstack
>>
>>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Where is Ceilometer endpoint? Is it with Gnocchi now!?

[Martinx - ジェームズ]

Bug reported: https://bugs.launchpad.net/aodh/+bug/1665784

On 17 February 2017 at 13:08, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

> I see... Maybe that page is right, and the Aodh one is wrong (outdated)...
>
> I tried to install Ceilometer-API, didn't worked as expected (dispacher is
> gnocchi), its log prints:
>
> ---
> WARNING ceilometer.api.app [-] Note: Ceilometer API is deprecated; use
> APIs from Aodh (alarms), Gnocchi (metrics) and/or Panko (events).
> ---
>
> So, thing is, how to config Aodh so it doesn't look for the
> Ceilometer/metering endpoint anymore...
>
> Cheers!
> Thiago
>
> On 17 February 2017 at 03:35, Tobias Urdin <tobias.ur...@crystone.com>
> wrote:
>
>> Hello,
>>
>> You are correct, please report a bug by clicking the "bug report" link on
>> that page.
>> https://docs.openstack.org/project-install-guide/telemetry/
>> draft/install-base-ubuntu.html
>>
>> Best regards
>>
>>
>> On 02/17/2017 06:14 AM, Martinx - ジェームズ wrote:
>>
>> Guys,
>>
>>  According to the new doc:
>>
>>  https://docs.openstack.org/project-install-guide/telemetry/
>> draft/install-base-ubuntu.html
>>
>>  There are no instructions to create the Ceilometer endpoint... Is this
>> correct?
>>
>>  If yes, how to configure Aodh when with a Ceilometer that have no
>> endpoint?
>>
>>  Right now, my aodh-evaluator.log prints this:
>>
>> ---
>> ERROR aodh.evaluator.threshold EndpointNotFound: internalURL endpoint for
>> metering service in ca-east-1 region not found
>> ---
>>
>>  Any tips?
>>
>> Thanks!
>> Thiago
>>
>>
>>
>> ___
>> Mailing list: http://lists.openstack.org/cgi
>> -bin/mailman/listinfo/openstack
>> Post to : openstack@lists.openstack.org
>> Unsubscribe : http://lists.openstack.org/cgi
>> -bin/mailman/listinfo/openstack
>>
>>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Gnocchi (Pandas) error, ValueError: buffer source array is read-only...

[Martinx - ジェームズ]

Thank you, that's precisely what happened! Filling a bug report ASAP...

On 17 February 2017 at 13:48, Mehdi Abaakouk <sil...@sileht.net> wrote:

> On Fri, Feb 17, 2017 at 01:14:11PM -0500, Martinx - ジェームズ wrote:
>
>> "ERROR gnocchi.storage._carbonara ValueError: buffer source array is
>> read-only"
>>
>>
> * Using Ubuntu 16.04 + Ocata Cloud Archive Staging PPA.
>>
>>
> It seems Ubuntu 16.04 ship pandas 0.17, while Gnocchi required 0.18.
>
> You should open a bug on ubuntu or cloud-archive packaging.
>
>
> --
> Mehdi Abaakouk
> mail: sil...@sileht.net
> irc: sileht
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Gnocchi (Pandas) error, ValueError: buffer source array is read-only...

[Martinx - ジェームズ]

Guys,

 I'm trying to install and configure Gnocchi but, the following error
appear the its Metricd log:


 https://paste.ubuntu.com/24014963/

 "ERROR gnocchi.storage._carbonara ValueError: buffer source array is
read-only"


 Something related to Python Pandas... What is this about and how to fix it?

 * Using Ubuntu 16.04 + Ocata Cloud Archive Staging PPA.

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Where is Ceilometer endpoint? Is it with Gnocchi now!?

[Martinx - ジェームズ]

I see... Maybe that page is right, and the Aodh one is wrong (outdated)...

I tried to install Ceilometer-API, didn't worked as expected (dispacher is
gnocchi), its log prints:

---
WARNING ceilometer.api.app [-] Note: Ceilometer API is deprecated; use APIs
from Aodh (alarms), Gnocchi (metrics) and/or Panko (events).
---

So, thing is, how to config Aodh so it doesn't look for the
Ceilometer/metering endpoint anymore...

Cheers!
Thiago

On 17 February 2017 at 03:35, Tobias Urdin <tobias.ur...@crystone.com>
wrote:

> Hello,
>
> You are correct, please report a bug by clicking the "bug report" link on
> that page.
> https://docs.openstack.org/project-install-guide/
> telemetry/draft/install-base-ubuntu.html
>
> Best regards
>
>
> On 02/17/2017 06:14 AM, Martinx - ジェームズ wrote:
>
> Guys,
>
>  According to the new doc:
>
>  https://docs.openstack.org/project-install-guide/
> telemetry/draft/install-base-ubuntu.html
>
>  There are no instructions to create the Ceilometer endpoint... Is this
> correct?
>
>  If yes, how to configure Aodh when with a Ceilometer that have no
> endpoint?
>
>  Right now, my aodh-evaluator.log prints this:
>
> ---
> ERROR aodh.evaluator.threshold EndpointNotFound: internalURL endpoint for
> metering service in ca-east-1 region not found
> ---
>
>  Any tips?
>
> Thanks!
> Thiago
>
>
>
> ___
> Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/
> openstack
> Post to : openstack@lists.openstack.org
> Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/
> openstack
>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Where is Ceilometer endpoint? Is it with Gnocchi now!?

[Martinx - ジェームズ]

Guys,

 According to the new doc:


https://docs.openstack.org/project-install-guide/telemetry/draft/install-base-ubuntu.html

 There are no instructions to create the Ceilometer endpoint... Is this
correct?

 If yes, how to configure Aodh when with a Ceilometer that have no endpoint?

 Right now, my aodh-evaluator.log prints this:

---
ERROR aodh.evaluator.threshold EndpointNotFound: internalURL endpoint for
metering service in ca-east-1 region not found
---

 Any tips?

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Request for more documents: Ceilometer, Gnocchi, Aodh, Senlin and OVN.

[Martinx - ジェームズ]

Another error on Senlin...


senling-engine.log:

https://paste.ubuntu.com/24005927/


While running:

openstack cluster profile create --spec-file heat_stack_nova_server.yaml
my_stack --debug

"--debug output":

https://paste.ubuntu.com/24005924/


;-(


On this very same server, the following commands works (from a Senlin
example):

---
openstack cluster profile create --spec-file cirros_basic.yaml mycirros

openstack cluster profile list
---


Something that works for "nova_server" profile, doesn't work for
"heat_stack" profile... I appreciate any help...

BTW, I can launch Heat stacks normally on this server, for example:
"openstack stack create nova_server_template.yaml", that is included inside
of "heat_stack_nova_server.yaml" (Senlin "heat_stack/nova_server" profile
example).

Thanks,
Thiago

On 16 February 2017 at 00:55, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

> BTW...
>
> So far, after reading the developer docs, I manage to launch "nova_server"
> from a Senlin Profile, looks awesome!
>
> Here are the errors that I'm seeing:
>
>
> * gnocchi-metricd.log:
>
>  https://paste.ubuntu.com/24005716/
>
>  On gnocchi-metricd.log, it prints an error related to "Pandas", some
> Python stuff... Not sure about how to fix it. Any clue?
>
>
> * aodh-evaluator.log:
>
>  https://paste.ubuntu.com/24005720/
>
>
>   About Aodh error, according to the following doc:
>
>  https://docs.openstack.org/project-install-guide/
> telemetry/draft/install-base-ubuntu.html
>
>  ...Ceilometer (when with Gnocchi?) doesn't have an endpoint for itself,
> the "metering" endpoint doesn't exist anymore, is this correct?
>
>
>  On Mitaka, Ceilometer endpoint is still there:
>
>  https://docs.openstack.org/mitaka/install-guide-ubuntu/
> ceilometer-install.html
>
>
>  However, on Newton / Ocata, it points to this new Ceilometer version doc,
> that apparently, doesn't have an endpoint on its own:
>
>  https://docs.openstack.org/project-install-guide/telemetry/
>
>  So, how to integrate Aodh with this new Ceilometer? And get rid of the
> "aodh-evaluator" error that I'm seeing...
>
> Cheers!
> Thiago
>
> On 16 February 2017 at 00:16, Martinx - ジェームズ <thiagocmarti...@gmail.com>
> wrote:
>
>> Guys,
>>
>>  I'm trying very hard to put those pieces of software to work together,
>> however, it is not working as expected... I think that I am almost there!
>>
>>  Basically, I need a reliable AutoScaling solution.
>>
>>  Looks like that Senlin is the answer, I manage to make the "openstack
>> cluster ..." commands to work (Ocata on Ubuntu 16.04 via Cloud Archive), I
>> can run a few things here but, it is not complete... I can see errors on
>> Aodh, on Gnocchi...
>>
>>  Thing is, there is no clear documentation about how to install and
>> integrate those things on Ubuntu.
>>
>>  Also, I'm tired of L3 / DHCP agents and lots of Linux Namespaces, looks
>> like that OpenvSwitch and its OVN will replace those things but, again, I
>> can't find clear docs about that, especially when with OVS+DPDK...
>>
>>  Any plans for Ocata documentation?
>>
>>  BTW, OpenStack is getting better and better! This project is awesome!
>> :-D
>>
>> Thanks!
>> Thiago
>>
>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Request for more documents: Ceilometer, Gnocchi, Aodh, Senlin and OVN.

[Martinx - ジェームズ]

BTW...

So far, after reading the developer docs, I manage to launch "nova_server"
from a Senlin Profile, looks awesome!

Here are the errors that I'm seeing:


* gnocchi-metricd.log:

 https://paste.ubuntu.com/24005716/

 On gnocchi-metricd.log, it prints an error related to "Pandas", some
Python stuff... Not sure about how to fix it. Any clue?


* aodh-evaluator.log:

 https://paste.ubuntu.com/24005720/


  About Aodh error, according to the following doc:


https://docs.openstack.org/project-install-guide/telemetry/draft/install-base-ubuntu.html

 ...Ceilometer (when with Gnocchi?) doesn't have an endpoint for itself,
the "metering" endpoint doesn't exist anymore, is this correct?


 On Mitaka, Ceilometer endpoint is still there:


https://docs.openstack.org/mitaka/install-guide-ubuntu/ceilometer-install.html


 However, on Newton / Ocata, it points to this new Ceilometer version doc,
that apparently, doesn't have an endpoint on its own:

 https://docs.openstack.org/project-install-guide/telemetry/

 So, how to integrate Aodh with this new Ceilometer? And get rid of the
"aodh-evaluator" error that I'm seeing...

Cheers!
Thiago

On 16 February 2017 at 00:16, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

> Guys,
>
>  I'm trying very hard to put those pieces of software to work together,
> however, it is not working as expected... I think that I am almost there!
>
>  Basically, I need a reliable AutoScaling solution.
>
>  Looks like that Senlin is the answer, I manage to make the "openstack
> cluster ..." commands to work (Ocata on Ubuntu 16.04 via Cloud Archive), I
> can run a few things here but, it is not complete... I can see errors on
> Aodh, on Gnocchi...
>
>  Thing is, there is no clear documentation about how to install and
> integrate those things on Ubuntu.
>
>  Also, I'm tired of L3 / DHCP agents and lots of Linux Namespaces, looks
> like that OpenvSwitch and its OVN will replace those things but, again, I
> can't find clear docs about that, especially when with OVS+DPDK...
>
>  Any plans for Ocata documentation?
>
>  BTW, OpenStack is getting better and better! This project is awesome!
> :-D
>
> Thanks!
> Thiago
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Request for more documents: Ceilometer, Gnocchi, Aodh, Senlin and OVN.

[Martinx - ジェームズ]

Guys,

 I'm trying very hard to put those pieces of software to work together,
however, it is not working as expected... I think that I am almost there!

 Basically, I need a reliable AutoScaling solution.

 Looks like that Senlin is the answer, I manage to make the "openstack
cluster ..." commands to work (Ocata on Ubuntu 16.04 via Cloud Archive), I
can run a few things here but, it is not complete... I can see errors on
Aodh, on Gnocchi...

 Thing is, there is no clear documentation about how to install and
integrate those things on Ubuntu.

 Also, I'm tired of L3 / DHCP agents and lots of Linux Namespaces, looks
like that OpenvSwitch and its OVN will replace those things but, again, I
can't find clear docs about that, especially when with OVS+DPDK...

 Any plans for Ocata documentation?

 BTW, OpenStack is getting better and better! This project is awesome!   :-D

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Newton, OpenvSwitch with DPDK, it is all or nothing! How to make it more modular?

[Martinx - ジェームズ]

Hey guys,

 I'm playing with OpenvSwitch with DPDK, in my KVM Ubuntu host, it is
awesome!

 My KVM guest, have 4 vNIC, it is a L2 Bridge (another DPDK App), the first
two (eth0 and eth1), are regular bridges (Libvirt Linux Bridges (or OVS
regular bridges)), and the latest two (eth2 and eth3), are type=vhostuser
with "/var/run/openvswitch/vhost-user-1" and "/var/run/openvswitch/vhost-
user-2".

 So, all good! I can ssh into my VM via its eth0 (public), my VM can talk
with other VMs via its eth1 (private), and my virtualized DPDK-App runs
fine on top of two VirtIO vhostuser NIC (eth2 and eth3).

 Now, time to do this via OpenStack! But, it doesn't work like this, let me
try to explain...

 On OpenStack, without OVS+DPDK, my KVM Guest also have 4 vNIC, first two
(eth0 and eth1), are VXLAN networks (using the OpenvSwitch Agent), and the
latest two (eth2 and eth3), are Provider Networks (also using the
OpenvSwitch Agent).

 So, here comes the problem with OVS and DPDK...

 I've configured Neutron OVS Agent like this:
http://docs.openstack.org/developer/neutron/devref/ovs_vhostuser.html

 My intention, is to use the OVS+DPDK _only_ for the Provider Networks,
but, when I started my Instance, all the 4 vNIC are being tied to the
OVS+DPDK bridges, which is clearly, not what I've configured on my Neutron
Agents.

 So, I'm thinking here, how can I do this?

 Regular "OVS Regular" bridges for VXLAN networks, and "OVS DPDPK" bridges
for Provider Networks?

 OR, how to use OVS+DPDK for VXLAN networks? If this is possible, does it
still supports the ovs-firewall-driver?

 Another workaround, that I'm thinking here, would be to launch 2 OVS
Agents on my Compute Nodes, one for OVS and VXLAN, and another only for
OVS+DPDK... Or, maybe, for VXLAN, I can go back to Linux Bridges...

 I've read that RedHat don't recommend DPDK for VXLAN networks but, I'm
using Newton on Ubuntu, so, things might be different, specially because
Ubuntu have native support for a very recent DPDK version, for free,
thoughts?

 I appreciate any comments!

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Mitaka - Why "OS::Neutron::FloatingIPAssociation" requires "fixed_ips" on "OS::Neutron::Port" ?

[Martinx - ジェームズ]

That's an amazing tip, thank you!

On 15 October 2016 at 07:27, Turbo Fredriksson <tu...@bayour.com> wrote:

> On Oct 14, 2016, at 11:13 PM, Martinx - ジェームズ wrote:
>
> > Looks like that it requires fixed_ips! Why is that?
>
> What version have you set for your stack?
>

I was still using 2013-05-23 specs, lol...


>
> Part of my stack looks like:
>
> - s n i p -
> heat_template_version: 2016-04-08
> description: Instance template
> [..]
> resources:
> [..]
>   port:
> type: OS::Neutron::Port
> properties:
>   name: { get_param: name }
>   network: { get_param: network }
>   security_groups: { get_param: secgrps }
> - s n i p -
>
> And that works just fine - no need for fixed_ips.
>

Awesome, it worked!


>
> >  name: {"Fn::Join": ["-", [{ get_param: "OS::stack_name" } ,
> > "instance-port-1"]]}
>
> Hint: Use list_join instead..
>

Good, working too...


>
>
> Oh, and just to 'blow your mind', this is one of my absolute
> "worst" :)
>
> - s n i p -
>   instance:
> [create the instance using OS::Nova::Server]
>
>   record:
> type: Turbo::Record
> properties:
>   dnsname: { list_join:
>['-', [
>  { get_param: name },
>  { str_split: ['.', { get_attr: [instance,
> first_address] }, 3] },
>] ]
>  }
>   dnsdomain: domain.tld.
>   data: { get_attr: [instance, first_address] }
>
>   reverse:
> type: Turbo::Record
> properties:
>   dnsname: { str_split: ['.', { get_attr: [instance, first_address] },
> 3] }
>   dnsdomain: { list_join:
>['.', [
>  { str_split: ['.', { get_attr: [instance,
> first_address] }, 2] },
>  { str_split: ['.', { get_attr: [instance,
> first_address] }, 1] },
>  { str_split: ['.', { get_attr: [instance,
> first_address] }, 0] },
>  'in-addr.arpa.'
>] ]
>  }
>   data: { list_join:
> ['.', [
>   { list_join:
> ['-', [
>   { get_param: name },
>   { str_split: ['.', { get_attr: [instance,
> first_address] }, 3] },
> ] ]
>   },
>   'domain.tld.'
> ] ]
> }
>   rec_type: PTR
> - s n i p -
>
> With 'Turbo::Record' defined as
>
> - s n i p -
> heat_template_version: 2016-04-08
> description: Designate domain record template
>
> parameters:
>   dnsdomain:
> type: string
> description: Domain name to attach dnsname to
> default: domain.tld
>
>   dnsname:
> type: string
> description: Host name to attach to domain
>
>   data:
> type: string
> description: IP address to attach to IP
>
>   rec_type:
> type: string
> description: DNS record type
> default: A
>
> resources:
>   record:
> type: OS::Designate::Record
> properties:
>   description: { list_join: [' - ', [ 'Floating IP DNS Record', {
> get_param: dnsname } ] ] }
>   name: { list_join: ['.', [ { get_param: dnsname }, { get_param:
> dnsdomain } ] ] }
>   domain: { get_param: dnsdomain }
>   data: { get_param: data }
>   type: { get_param: rec_type }
> - s n i p -
>
> I use it for creating a DNS record for the instance in Designate,
> because a auto scaling group doesn't have a "counter" ('%index%')
> like a resource group have..
>

Hmmm... This is very interesting... I was expecting to use the %index% with
AutoScaling but, now that you said it doesn't work, I'll definitely give
Designate a try.


> --
> Choose a job you love, and you will never have
> to work a day in your life.


Absolutely!

Cheers!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Mitaka, how to use Floating IPs under "OS::Nova::Server", without "OS::Neutron::Port"?

[Martinx - ジェームズ]

Guys,

 According to the following doc:


http://docs.openstack.org/developer/heat/template_guide/openstack.html#OS::Nova::Server-hot

 It is possible to associate (?) the Floating IP under "OS::Nova::Server"
but, it is not working as expected...

 The Heat template boots up but, the Floating IP is not associated with my
instance.

 Looks like that the following line:

https://github.com/sandvine-eng/svauto/blob/dev/misc/os-heat-templates/demo-simple-stack-3.yaml#L104

 Is being partially ignored, it is wired to the network but, no floating IP
associated.

Cheers!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] How to use "floating_ip" under "OS::Nova::Server", Mitaka ?

[Martinx - ジェームズ]

Hi Jose,

 That's awesome!

 Floating IPs without fixed_ips and no more
"OS::Neutron::FloatingIPAssociation"!

 Thank you!

Best,
Thiago

On 17 October 2016 at 05:42, Jose Manuel Ferrer Mosteiro <
jmferrer.paradigmatecnolog...@gmail.com> wrote:

> Hi
>
> Here is a template that creates a server and publish it:
>
> https://github.com/jmferrer/openstack-openshift-workshop/
> blob/master/openstack/nfs-server-heat-template.yaml
>
> Maybe it could help you.
>
> Have a nice day,
>
> Jose Manuel
>
>
>
>
> El 2016-10-14 00:01, Martinx - ジェームズ escribió:
>
>
>
> On 13 October 2016 at 17:26, Martinx - ジェームズ <thiagocmarti...@gmail.com>
> wrote:
>
>>
>>
>> On 13 October 2016 at 17:21, Martinx - ジェームズ <thiagocmarti...@gmail.com>
>> wrote:
>>
>>> Guys,
>>>
>>>  I'm trying to launch a stack via Heat, by using the following template
>>> (I'm copying just part of it here):
>>>
>>> ---
>>>   demosrv:
>>> type: OS::Nova::Server
>>> depends_on: ctrl_subnet
>>> properties:
>>>   name: { str_replace: { params: { $stack_name: { get_param:
>>> 'OS::stack_name' } }, template: '$stack_name-demosrv' } }
>>>   key_name: { get_param: 'ssh_key' }
>>>   image: { get_param: 'demosrv_image' }
>>>   flavor: "m1.small"
>>>   networks:
>>> - network: { get_resource: ctrl_net }
>>> ---
>>>
>>>  This works!
>>>
>>>  However, I want to associate a Floating IP to it, then, I added:
>>>
>>> ---
>>>  
>>> *  floating_ip_1:*
>>> *type: OS::Neutron::FloatingIP*
>>> *depends_on: router-0-port-1*
>>> *properties:*
>>> *  floating_network: { get_param: public_network }*
>>> 
>>>   demosrv:
>>> type: OS::Nova::Server
>>> depends_on: ctrl_subnet
>>> properties:
>>>   name: { str_replace: { params: { $stack_name: { get_param:
>>> 'OS::stack_name' } }, template: '$stack_name-demosrv' } }
>>>   key_name: { get_param: 'ssh_key' }
>>>   image: { get_param: 'demosrv_image' }
>>>   flavor: "m1.small"
>>>   networks:
>>> - network: { get_resource: ctrl_net }
>>> *- floating_ip: { get_resource: floating_ip_1 }*
>>> ---
>>>
>>>  Then, it does't work anymore!
>>>
>>>  heat stack-create fails with:
>>>
>>> -
>>> ERROR: One of the properties "network", "port" or "subnet" should be set
>>> for the specified network of server "demosrv".
>>> -
>>>
>>>  So, how to use it?
>>>
>>>  If I use "OS::Neutron::Port" and "OS::Neutron::FloatingIPAssociation",
>>> then, it works but, damn, I want a simpler Heat template, without fixed
>>> ips, without Port definitions, something like this:
>>>
>>> https://specs.openstack.org/openstack/heat-specs/specs/kilo/
>>> rich-network-prop.html
>>>
>>>  Is it supported on Mitaka? If yes, how to do it?
>>>
>>> Thanks!
>>> Thiago
>>>
>> Also, according to the following doc:
>>
>> http://docs.openstack.org/developer/heat/template_guide/open
>> stack.html#OS::Nova::Server-prop-networks
>>
>> The "networks" section supports "floating_ip", starting with Mitaka...
>>
> BTW, I'll be happy to first, get rid of "OS::Neutron::FloatingIPAssociation"
> AND the fixed_ips, then, later, if possobe, I'll get rid of
> "OS::Neutron::Port" as well...
>
> ___
> Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/
> openstack
> Post to : openstack@lists.openstack.org
> Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/
> openstack
>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Mitaka - Why "OS::Neutron::FloatingIPAssociation" requires "fixed_ips" on "OS::Neutron::Port" ?

[Martinx - ジェームズ]

Guys,

 I'm trying to launch a stack via Heat, without fixed IPs but, with
Floating IPs.

 Looks like that it requires fixed_ips! Why is that?

 For example, this:

---
  instance_ctrl_port:
type: OS::Neutron::Port
properties:
  name: {"Fn::Join": ["-", [{ get_param: "OS::stack_name" } ,
"instance-port-1"]]}
  network: { get_resource: control_net }
  fixed_ips:
- ip_address: 192.168.192.7
  security_groups:
- { get_resource: svpts_ctrl_sec }
---

 ...works but, if I try to remove:

---
  fixed_ips:
- ip_address: 192.168.192.7
---

 Then, "openstack stack create" fails!

 With the following error:

---
ERROR: The server could not comply with the request since it is either
malformed or otherwise incorrect.
---

 The workaround is to launch a stack without Floating IPs and then, via
Horizon, add the those manually, which is not a good idea... I'm planning
to add the Floating IP after launching the stack, via Ansible, or something
like that, as a second step but, I think that it would be nice to launch
the entire stack in one shot...

 I'm also trying to associate the Floating IP under "OS::Nova::Server" but,
no idea about its syntax, so, not working either (but for different reason).

 Have a nice weekend guys!

Cheers!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] How to use "floating_ip" under "OS::Nova::Server", Mitaka ?

[Martinx - ジェームズ]

On 13 October 2016 at 17:26, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

>
>
> On 13 October 2016 at 17:21, Martinx - ジェームズ <thiagocmarti...@gmail.com>
> wrote:
>
>> Guys,
>>
>>  I'm trying to launch a stack via Heat, by using the following template
>> (I'm copying just part of it here):
>>
>> ---
>>   demosrv:
>> type: OS::Nova::Server
>> depends_on: ctrl_subnet
>> properties:
>>   name: { str_replace: { params: { $stack_name: { get_param:
>> 'OS::stack_name' } }, template: '$stack_name-demosrv' } }
>>   key_name: { get_param: 'ssh_key' }
>>   image: { get_param: 'demosrv_image' }
>>   flavor: "m1.small"
>>   networks:
>> - network: { get_resource: ctrl_net }
>> ---
>>
>>  This works!
>>
>>  However, I want to associate a Floating IP to it, then, I added:
>>
>> ---
>>  
>> *  floating_ip_1:*
>> *type: OS::Neutron::FloatingIP*
>> *depends_on: router-0-port-1*
>> *properties:*
>> *  floating_network: { get_param: public_network }*
>> 
>>   demosrv:
>> type: OS::Nova::Server
>> depends_on: ctrl_subnet
>> properties:
>>   name: { str_replace: { params: { $stack_name: { get_param:
>> 'OS::stack_name' } }, template: '$stack_name-demosrv' } }
>>   key_name: { get_param: 'ssh_key' }
>>   image: { get_param: 'demosrv_image' }
>>   flavor: "m1.small"
>>   networks:
>> - network: { get_resource: ctrl_net }
>> *- floating_ip: { get_resource: floating_ip_1 }*
>> ---
>>
>>  Then, it does't work anymore!
>>
>>  heat stack-create fails with:
>>
>> -
>> ERROR: One of the properties "network", "port" or "subnet" should be set
>> for the specified network of server "demosrv".
>> -
>>
>>  So, how to use it?
>>
>>  If I use "OS::Neutron::Port" and "OS::Neutron::FloatingIPAssociation",
>> then, it works but, damn, I want a simpler Heat template, without fixed
>> ips, without Port definitions, something like this:
>>
>> https://specs.openstack.org/openstack/heat-specs/specs/kilo/
>> rich-network-prop.html
>>
>>  Is it supported on Mitaka? If yes, how to do it?
>>
>> Thanks!
>> Thiago
>>
>
> Also, according to the following doc:
>
> http://docs.openstack.org/developer/heat/template_guide/
> openstack.html#OS::Nova::Server-prop-networks
>
> The "networks" section supports "floating_ip", starting with Mitaka...
>

BTW, I'll be happy to first, get rid of
"OS::Neutron::FloatingIPAssociation" AND the fixed_ips, then, later, if
possobe, I'll get rid of "OS::Neutron::Port" as well...
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Race condition on a very simple Heat template - Mitaka

[Martinx - ジェームズ]

Guys,

 If I remove the following line out from my Heat template:


https://github.com/sandvine-eng/svauto/blob/dev/misc/os-heat-templates/demo-simple-stack-1.yaml#L80

---
depends_on: ctrl_subnet
---

 Then, the stack creation fails... But, why?

 I didn't used this option with Kilo / IceHouse / Juno, why this line is
required now?

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] How to use "floating_ip" under "OS::Nova::Server", Mitaka ?

[Martinx - ジェームズ]

On 13 October 2016 at 17:21, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

> Guys,
>
>  I'm trying to launch a stack via Heat, by using the following template
> (I'm copying just part of it here):
>
> ---
>   demosrv:
> type: OS::Nova::Server
> depends_on: ctrl_subnet
> properties:
>   name: { str_replace: { params: { $stack_name: { get_param:
> 'OS::stack_name' } }, template: '$stack_name-demosrv' } }
>   key_name: { get_param: 'ssh_key' }
>   image: { get_param: 'demosrv_image' }
>   flavor: "m1.small"
>   networks:
> - network: { get_resource: ctrl_net }
> ---
>
>  This works!
>
>  However, I want to associate a Floating IP to it, then, I added:
>
> ---
>  
> *  floating_ip_1:*
> *type: OS::Neutron::FloatingIP*
> *depends_on: router-0-port-1*
> *properties:*
> *  floating_network: { get_param: public_network }*
> 
>   demosrv:
> type: OS::Nova::Server
> depends_on: ctrl_subnet
> properties:
>   name: { str_replace: { params: { $stack_name: { get_param:
> 'OS::stack_name' } }, template: '$stack_name-demosrv' } }
>   key_name: { get_param: 'ssh_key' }
>   image: { get_param: 'demosrv_image' }
>   flavor: "m1.small"
>   networks:
> - network: { get_resource: ctrl_net }
> *- floating_ip: { get_resource: floating_ip_1 }*
> ---
>
>  Then, it does't work anymore!
>
>  heat stack-create fails with:
>
> -
> ERROR: One of the properties "network", "port" or "subnet" should be set
> for the specified network of server "demosrv".
> -
>
>  So, how to use it?
>
>  If I use "OS::Neutron::Port" and "OS::Neutron::FloatingIPAssociation",
> then, it works but, damn, I want a simpler Heat template, without fixed
> ips, without Port definitions, something like this:
>
> https://specs.openstack.org/openstack/heat-specs/specs/
> kilo/rich-network-prop.html
>
>  Is it supported on Mitaka? If yes, how to do it?
>
> Thanks!
> Thiago
>

Also, according to the following doc:

http://docs.openstack.org/developer/heat/template_guide/openstack.html#OS::Nova::Server-prop-networks

The "networks" section supports "floating_ip", starting with Mitaka...
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] How to use "floating_ip" under "OS::Nova::Server", Mitaka ?

[Martinx - ジェームズ]

Guys,

 I'm trying to launch a stack via Heat, by using the following template
(I'm copying just part of it here):

---
  demosrv:
type: OS::Nova::Server
depends_on: ctrl_subnet
properties:
  name: { str_replace: { params: { $stack_name: { get_param:
'OS::stack_name' } }, template: '$stack_name-demosrv' } }
  key_name: { get_param: 'ssh_key' }
  image: { get_param: 'demosrv_image' }
  flavor: "m1.small"
  networks:
- network: { get_resource: ctrl_net }
---

 This works!

 However, I want to associate a Floating IP to it, then, I added:

---
 
*  floating_ip_1:*
*type: OS::Neutron::FloatingIP*
*depends_on: router-0-port-1*
*properties:*
*  floating_network: { get_param: public_network }*

  demosrv:
type: OS::Nova::Server
depends_on: ctrl_subnet
properties:
  name: { str_replace: { params: { $stack_name: { get_param:
'OS::stack_name' } }, template: '$stack_name-demosrv' } }
  key_name: { get_param: 'ssh_key' }
  image: { get_param: 'demosrv_image' }
  flavor: "m1.small"
  networks:
- network: { get_resource: ctrl_net }
*- floating_ip: { get_resource: floating_ip_1 }*
---

 Then, it does't work anymore!

 heat stack-create fails with:

-
ERROR: One of the properties "network", "port" or "subnet" should be set
for the specified network of server "demosrv".
-

 So, how to use it?

 If I use "OS::Neutron::Port" and "OS::Neutron::FloatingIPAssociation",
then, it works but, damn, I want a simpler Heat template, without fixed
ips, without Port definitions, something like this:

https://specs.openstack.org/openstack/heat-specs/specs/kilo/rich-network-prop.html

 Is it supported on Mitaka? If yes, how to do it?

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] [openstack-dev] OpenStack Newton B3 for Ubuntu 16.04 LTS and Ubuntu 16.10

[Martinx - ジェームズ]

On 8 September 2016 at 14:40, Corey Bryant 
wrote:

> Hi All,
>
> The Ubuntu OpenStack team is pleased to announce the general availability
> of OpenStack Newton B3 milestone in Ubuntu 16.10 and for Ubuntu 16.04 LTS
> via the Ubuntu Cloud Archive.
>
> Ubuntu 16.04 LTS
> 
> You can enable the Ubuntu Cloud Archive pocket for OpenStack Newton on
> Ubuntu 16.04 installations by running the following commands:
>
> sudo add-apt-repository cloud-archive:newton
> sudo apt update
>
> The Ubuntu Cloud Archive for Newton includes updates for Aodh, Barbican,
> Ceilometer, Cinder, Designate, Glance, Heat, Horizon, Ironic (6.1.0),
> Keystone, Manila, Networking-OVN, Neutron, Neutron-FWaaS, Neutron-LBaaS,
> Neutron-VPNaaS, Nova, and Trove.
>
> You can see the full list of packages and versions at [0].
>
> Ubuntu 16.10
> --
> No extra steps required; just start installing OpenStack!
>
> Branch Package Builds
> ---
> If you want to try out the latest master branch updates, or updates to
> stable branches, we are delivering continuously integrated packages on each
> upstream commit in the following PPA’s:
>
>sudo add-apt-repository ppa:openstack-ubuntu-testing/liberty
>sudo add-apt-repository ppa:openstack-ubuntu-testing/mitaka
>sudo add-apt-repository ppa:openstack-ubuntu-testing/newton
>
> bear in mind these are built per-commitish (30 min checks for new commits
> at the moment) so ymmv from time-to-time.
>
> Reporting bugs
> 
> Any issues please report bugs using the 'ubuntu-bug' tool:
>
>   sudo ubuntu-bug nova-conductor
>
> this will ensure that bugs get logged in the right place in Launchpad.
>
> Thanks and have fun!
>
> Regards,
>
> Corey
> (on behalf of the Ubuntu OpenStack team)
>
> [0] http://reqorts.qa.ubuntu.com/reports/ubuntu-server/cloud-arc
> hive/newton_versions.html
>
>
That's awesome!

I think that we also need new OpenvSwitch (2.6) and new DPDK-16.07 inside
of UCA (Xenial) too...

I can't find those packages neither here:

https://launchpad.net/~openstack-ubuntu-testing/+
archive/ubuntu/newton?field.series_filter=xenial

Nor here:

https://launchpad.net/~ubuntu-cloud-archive/+archive/ubuntu/newton-staging

I just sent an e-mail to Christian Ehrhardt about this...

Cheers!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [openstack-dev] [Openstack] OpenStack Newton B3 for Ubuntu 16.04 LTS and Ubuntu 16.10

[Martinx - ジェームズ]

On 8 September 2016 at 14:40, Corey Bryant 
wrote:

> Hi All,
>
> The Ubuntu OpenStack team is pleased to announce the general availability
> of OpenStack Newton B3 milestone in Ubuntu 16.10 and for Ubuntu 16.04 LTS
> via the Ubuntu Cloud Archive.
>
> Ubuntu 16.04 LTS
> 
> You can enable the Ubuntu Cloud Archive pocket for OpenStack Newton on
> Ubuntu 16.04 installations by running the following commands:
>
> sudo add-apt-repository cloud-archive:newton
> sudo apt update
>
> The Ubuntu Cloud Archive for Newton includes updates for Aodh, Barbican,
> Ceilometer, Cinder, Designate, Glance, Heat, Horizon, Ironic (6.1.0),
> Keystone, Manila, Networking-OVN, Neutron, Neutron-FWaaS, Neutron-LBaaS,
> Neutron-VPNaaS, Nova, and Trove.
>
> You can see the full list of packages and versions at [0].
>
> Ubuntu 16.10
> --
> No extra steps required; just start installing OpenStack!
>
> Branch Package Builds
> ---
> If you want to try out the latest master branch updates, or updates to
> stable branches, we are delivering continuously integrated packages on each
> upstream commit in the following PPA’s:
>
>sudo add-apt-repository ppa:openstack-ubuntu-testing/liberty
>sudo add-apt-repository ppa:openstack-ubuntu-testing/mitaka
>sudo add-apt-repository ppa:openstack-ubuntu-testing/newton
>
> bear in mind these are built per-commitish (30 min checks for new commits
> at the moment) so ymmv from time-to-time.
>
> Reporting bugs
> 
> Any issues please report bugs using the 'ubuntu-bug' tool:
>
>   sudo ubuntu-bug nova-conductor
>
> this will ensure that bugs get logged in the right place in Launchpad.
>
> Thanks and have fun!
>
> Regards,
>
> Corey
> (on behalf of the Ubuntu OpenStack team)
>
> [0] http://reqorts.qa.ubuntu.com/reports/ubuntu-server/cloud-arc
> hive/newton_versions.html
>
>
That's awesome!

I think that we also need new OpenvSwitch (2.6) and new DPDK-16.07 inside
of UCA (Xenial) too...

I can't find those packages neither here:

https://launchpad.net/~openstack-ubuntu-testing/+
archive/ubuntu/newton?field.series_filter=xenial

Nor here:

https://launchpad.net/~ubuntu-cloud-archive/+archive/ubuntu/newton-staging

I just sent an e-mail to Christian Ehrhardt about this...

Cheers!
Thiago
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [Openstack] Ubuntu Cloud Image - Forbidden access! Glance fails with error 500.

[Martinx - ジェームズ]

But I need to rely on the upstream URLs for two reasons:

1- During Glance provision, I can't download the images, the images MUST be
downloaded by Glance itself, by demand (that's why I always use --location
and that's why I'm still using Glance v1;

2- By relying on a remote URL, I don't need to re-add the images every
single time that upstream updates its image, Glance will always download
the latest directly from upstream.

BTW, I've sent those messages to both lists (Ubuntu / OpenStack) because
this interests Ubuntu and since Glance is failing with Error 500, OpenStack
guys might be interested as well.

Cheers!
Thiago

On 29 August 2016 at 05:37, <stuart.mcla...@hp.com> wrote:

>
> To prevent this kind of thing recurring you can upload the image bytes
> into Glance rather than relying on the third party url always being
> available, eg:
>
> curl http://uec-images.ubuntu.com/releases/16.04/release/ubuntu-1
> 6.04-server-cloudimg-amd64-disk1.img | glance image-create --name "Ubuntu
> 16.04.1 LTS - Xenial Xerus - 64-bit - Cloud Based Ima
> ge" --is-public true --container-format bare --disk-format qcow2
>
>
> On Sun, 28 Aug 2016, Kaustubh Kelkar wrote:
>
> Broken link?
>>
>> https://cloud-images.ubuntu.com/xenial/
>>
>> -Kaustubh
>>
>> From: Martinx - ジェームズ
>> Sent: Saturday, August 27, 23:06
>> Subject: [Openstack] Ubuntu Cloud Image - Forbidden access! Glance
>> failswith error 500.
>> To: ubuntu-server, Ubuntu user technical support, not for general
>> discussions, openstack@lists.openstack.org
>>
>> Guys,
>>
>> It is impossible to download Ubuntu Cloud Image right now:
>>
>> http://uec-images.ubuntu.com/releases/16.04/release/ubuntu-1
>> 6.04-server-cloudimg-amd64-disk1.img
>>
>> Returns: Forbidden!
>>
>> 
>>
>> wget http://uec-images.ubuntu.com/releases/16.04/release/ubuntu-1
>> 6.04-server-cloudimg-amd64-disk1.img
>>
>> --2016-08-28 02:50:36--  http://uec-images.ubuntu.com/r
>> eleases/16.04/release/ubuntu-16.04-server-cloudimg-amd64-disk1.img
>>
>> Resolving uec-images.ubuntu.com<http://uec-images.ubuntu.com> (
>> uec-images.ubuntu.com<http://uec-images.ubuntu.com>)... 91.189.88.140
>>
>> Connecting to uec-images.ubuntu.com<http://uec-images.ubuntu.com> (
>> uec-images.ubuntu.com<http://uec-images.ubuntu.com>)|91.189.88.140|:80...
>> connected.
>>
>> HTTP request sent, awaiting response... 403 Forbidden
>>
>> 2016-08-28 02:50:36 ERROR 403: Forbidden.
>>
>> 
>>
>> This broke my OpenStack deployment, because Glance tries to download it
>> and then it fails (error 500 on Glance).
>>
>> ---
>>
>> http://paste.openstack.org/show/564302/
>>
>> ---
>>
>> Here is how I'm adding Ubuntu images to my OpenStack Mitaka Cloud:
>>
>> ---
>>
>> glance image-create --location http://uec-images.ubuntu.com/r
>> eleases/16.04/release/ubuntu-16.04-server-cloudimg-amd64-disk1.img
>> --name "Ubuntu 16.04.1 LTS - Xenial Xerus - 64-bit - Cloud Based Image"
>> --is-public true --container-format bare --disk-format qcow2
>>
>> ---
>>
>> Cheers!
>>
>> Thiago
>>
>>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Ubuntu Cloud Image - Forbidden access! Glance fails with error 500.

[Martinx - ジェームズ]

Guys,

 It is impossible to download Ubuntu Cloud Image right now:

http://uec-images.ubuntu.com/releases/16.04/release/ubuntu-16.04-server-cloudimg-amd64-disk1.img

Returns: Forbidden!


wget
http://uec-images.ubuntu.com/releases/16.04/release/ubuntu-16.04-server-cloudimg-amd64-disk1.img
--2016-08-28 02:50:36--
http://uec-images.ubuntu.com/releases/16.04/release/ubuntu-16.04-server-cloudimg-amd64-disk1.img
Resolving uec-images.ubuntu.com (uec-images.ubuntu.com)... 91.189.88.140
Connecting to uec-images.ubuntu.com
(uec-images.ubuntu.com)|91.189.88.140|:80...
connected.
HTTP request sent, awaiting response... 403 Forbidden
2016-08-28 02:50:36 ERROR 403: Forbidden.


 This broke my OpenStack deployment, because Glance tries to download it
and then it fails (error 500 on Glance).

---
http://paste.openstack.org/show/564302/
---

 Here is how I'm adding Ubuntu images to my OpenStack Mitaka Cloud:

---
glance image-create --location
http://uec-images.ubuntu.com/releases/16.04/release/ubuntu-16.04-server-cloudimg-amd64-disk1.img
--name "Ubuntu 16.04.1 LTS - Xenial Xerus - 64-bit - Cloud Based Image"
--is-public true --container-format bare --disk-format qcow2
---

Cheers!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] [neutron] - vlan-aware-vms

[Martinx - ジェームズ]

On 2 August 2016 at 17:49, Armando M. <arma...@gmail.com> wrote:

>
>
> On 29 July 2016 at 12:59, Martinx - ジェームズ <thiagocmarti...@gmail.com>
> wrote:
>
>> Quick question:
>>
>> Can I start testing Newton VLAN Aware VMs now (Beta 2)?
>>
>> Thanks,
>> Thiago
>>
>>
> If you're paying close attention the LinuxBridge version is almost
> functional, and the OVS one is coming along. I'd advise to wait a tad
> longer. I am trying to keep [1] up to date, so you might want to check that
> out before pulling down  the code.
>
> [1] https://blueprints.launchpad.net/neutron/+spec/vlan-aware-vms
>

Whee! That's great news!

I had to give up on Mitaka for a big project (almost 300 Instances) just
because it did not support VLAN Aware VMs.

We deployed Ubuntu 16.04 with OpenvSwitch and DPDK at the hosts... And we
used shell scripts to manage everything, super creepy...

I can't wait to give it a try!

Thank you,
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Federated users login into Horizon but, an error appear on apache logs: "Unable to retrieve project list".

[Martinx - ジェームズ]

Guys,

 I trying to configure OpenStack Federation and, right after logging into
Horizon with a Federated user, the following error appear on Apache /
Keystone logs:

---
Unable to retrieve project list.
Traceback (most recent call last):
  File "/usr/lib/python2.7/dist-packages/openstack_auth/user.py", line 313,
in authorized_tenants
is_federated=self.is_federated)
  File "/usr/lib/python2.7/dist-packages/openstack_auth/utils.py", line
307, in get_project_list
projects = client.federation.projects.list()
  File
"/usr/lib/python2.7/dist-packages/keystoneclient/v3/contrib/federation/base.py",
line 34, in list
tenant_list = self._list(url, self.object_type)
  File "/usr/lib/python2.7/dist-packages/keystoneclient/base.py", line 124,
in _list
resp, body = self.client.get(url, **kwargs)
  File "/usr/lib/python2.7/dist-packages/keystoneauth1/adapter.py", line
173, in get
return self.request(url, 'GET', **kwargs)
  File "/usr/lib/python2.7/dist-packages/keystoneauth1/adapter.py", line
331, in request
resp = super(LegacyJsonAdapter, self).request(*args, **kwargs)
  File "/usr/lib/python2.7/dist-packages/keystoneauth1/adapter.py", line
98, in request
return self.session.request(url, method, **kwargs)
  File "/usr/lib/python2.7/dist-packages/positional/__init__.py", line 94,
in inner
return func(*args, **kwargs)
  File "/usr/lib/python2.7/dist-packages/keystoneauth1/session.py", line
467, in request
raise exceptions.from_response(resp, method, url)
BadRequest: Bad Request (HTTP 400)
---

 What this means?

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Federated users can use Horizon, but can not use `openstack` CLI, error "Could not find user".

[Martinx - ジェームズ]

Hey guys,

 I'm facing a hard time here to configure OpenStack Federation...

 So far, I can login into Horizon using my Windows AD credentials but, I
can not use command line interface, the `openstack` command, with Federated
users.

 Here is the error:

---
ubuntu@controller-1:~$ source ~/tmartins-openrc.conf



Please enter your OpenStack Password:

ubuntu@controller-1:~$ openstack token issue



The request you have made requires authentication. (HTTP 401) (Request-ID:
req-15d779ba-211e-4bc9-b7b6-a3fc887c7f92)
---

 On Apache log, I see the following error ("Could not find user"):

---
http://paste.openstack.org/show/545677/
---

 Note that the "~/tmartins-openrc.conf" file was downloaded from Horizon
itself, after logging in with that very same user.

 It worth to mention that both "admin" and "demo" local users still works
even after enabling Federation, on both Horizon and `openstack` CLI.

 I really appreciate any help!

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] [neutron] - vlan-aware-vms

[Martinx - ジェームズ]

Quick question:

Can I start testing Newton VLAN Aware VMs now (Beta 2)?

Thanks,
Thiago

On 22 July 2016 at 04:45, Kevin Benton  wrote:

> Since they are essentially regular ports in the neutron data model, the
> regular rules for attaching to networks would apply. So you can should be
> able to create a sub-port on another network if that network is shared with
> you (either globally shared or via RBAC).
>
> On Wed, Jul 13, 2016 at 8:55 AM, Farhad Sunavala  wrote:
>
>>
>> Below is the latest spec for vlan-aware-vms
>>
>>
>> https://specs.openstack.org/openstack/neutron-specs/specs/newton/vlan-aware-vms.html
>> 
>>
>>
>>
>> I have a quick question on the above. (multi-tenancy).
>>
>> Assume the case of nested containers in a VM.
>>
>> Yes, the containers can be in different networks of the same tenant and
>> the above blue-print will handle the case very well.
>> How does it work when the containers are in different networks in
>> different tenants ?
>>
>> The trick is to create neutron ports (for the subports) and then link
>> them to the trunk port using
>>
>> neutron trunk-subport-add TRUNK \
>>PORT[,SEGMENTATION-TYPE,SEGMENTATION-ID] \
>>[PORT,...]
>>
>>
>> In the above command all the neutron ports (trunk  ports and subports)
>> must be in the same tenant.
>> As far as I know, a tenant will not see neutron ports from another
>> tenant.Or will this command allow
>> neutron ports from different tenants to be attached ?
>>
>> Solution1:
>>
>>
>> C1(ten1)   C2(ten2)
>> |   |
>> 
>> OVS bridge inside VM
>> 
>> |
>> | Trunk port
>> |
>> 
>> br-trunk (vlan-aware-vms spec)
>> 
>>
>> E.g.  VM "X" consists of containers C1 in Tenant 1 with portID = C1
>> (network dn1)
>> container C2 in Tenant 2 with portID = C2 (network dn2)
>> The trunk port of VM "X" is in tenant 100 with portID = T1 (network
>> dt)
>>
>> Will the above command allow a neutron trunk to have neutron sub-ports in
>> different tenants ?
>>
>> neutron trunk-subport-add T1 \
>>A  vlan 1 \
>>B vlan 2
>>
>>
>> Solution2:
>> Have a separate trunk port for each tenant connected to the vM
>>
>> C1(Ten1)C2(Ten2)
>> ||
>> ||
>> ---
>> OVS bridge inside VM
>> 
>> |  |
>> |Trunk(Ten1)  | (Trunk(Ten2)
>> |  |
>> -
>> br-trunk (vlan-aware-vms spec)
>> ---
>>
>> If the approach is solution2, then the issue is that Nova will not
>> allow a neutron port to be attached to a VM (if the neutron port
>> belongs to another tenant).
>>
>>
>> Any pointers will be highly appreciated.
>>
>> thanks,
>> Farhad.
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
> ___
> Mailing list:
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> Post to : openstack@lists.openstack.org
> Unsubscribe :
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Reaching VXLAN tenant networks from outside (without floating IPs)

[Martinx - ジェームズ]

 You can use IPv6...

On 29 June 2016 at 15:17, Gustavo Randich  wrote:

> Hi operators...
>
> Transitioning from nova-network to Neutron (Mitaka), one of the key issues
> we are facing is how to reach VMs in VXLAN tenant networks without using
> precious floating IPs.
>
> Things that are outside Neutron in our case are:
>
> - in-house made application orchestrator: needs SSH access to instances to
> perform various tasks (start / shutdown apps, configure filesystems, etc.)
>
> - various centralized and external monitoring/metrics pollers: need SNMP /
> SSH access to gather status and trends
>
> - internal customers: need SSH access to instance from non-openstack VPN
> service
>
> - ideally, non-VXLAN aware traffic balancer appliances
>
>
>
> We have considered these approaches:
>
> - putting some of the external components inside a Network Node: inviable
> because components need access to multiple Neutron deployments
>
>
> - Neutron's VPNaaS: cannot figure how to configure a client-to-site VPN
> topology
>
> - integrate hardware switches capable of VXLAN VTEP: for us in this stage,
> it is complex and expensive
>
>
> - other?
>
>
> Thank you in advance,
> Gustavo
>
>
> ___
> Mailing list:
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> Post to : openstack@lists.openstack.org
> Unsubscribe :
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Mitaka installation process

[Martinx - ジェームズ]

On 14 June 2016 at 16:20, Nasir Mahmood  wrote:

> What you want to achieve?
>
> Without packstack, try using the docs.openstacks.org for your
> environment.  these guides are awesome source to bootstrap  the raw
> openstack and configure it as per your requirements.  Then expand
> horizontally or vertically.  Start configuring and share your questions and
> ideas on the relevant mailing lists.
>
> My first POC  deployment was on  3 vmware VMs but it was the best way to
> learn ins and outs , learned a lot  and then I moved along with multi
> purpose , diversified  physical architecture  comprising of different
> nodes for different needs.
>
> Do it before you ask.
>
> Regards,
>
> NM
> On Jun 15, 2016 12:32 AM, "Venkatesh Kotipalli" <
> openstackvenkat...@gmail.com> wrote:
>
> Hi Folks,
>
> I am new to  openstack. Please guide me which way to configure the
> openstack without packstack.
>
> And i want to install the *MITAKA* version of openstack. Please share the
> links.
>
> If any one succeeded installed the mitaka version without packstack
> software, please help me.
>
> Regards,
> Venkatesh.k
>
>
Hi!

I have an Ansible playbook that deploys Mitaka on top of Ubuntu 16.04.

The good thing about it is that it is SUPER simple, it only tries to
automate the instructions from here:

docs.openstacks.org

Here is it:

https://github.com/tmartinx/svauto

You can look at the startup script here:

https://github.com/tmartinx/svauto/blob/dev/scripts/os-install.sh

Or you can execute it directly via "curl | bash" on top of a fresh
installed Ubuntu 16.04, like this:

bash <(curl -s
https://raw.githubusercontent.com/tmartinx/svauto/dev/scripts/os-install.sh)

By default, it deploys OpenStack "all-in-one".

Cheers!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [openstack-dev] [Openstack] OpenStack Newton B1 for Ubuntu 16.04 LTS and Ubuntu 16.10

[Martinx - ジェームズ]

I can't wait to try "VLAN Aware VMs"... But it is not there yet... Maybe on
Newton B2...

https://blueprints.launchpad.net/neutron/+spec/vlan-aware-vms

This is *very* important for NFV Instances on OpenStack... Many telcos
needs this...

Cheers!
Thiago

On 14 June 2016 at 07:25, Jeffrey Zhang  wrote:

> Very Cool! Wait this for long.
> we(kolla project) will enable the ubuntu binary deploy gate in the CI.
>
> On Mon, Jun 13, 2016 at 8:54 PM, Corey Bryant 
> wrote:
>
>> Hi All,
>>
>> The Ubuntu OpenStack team is pleased to announce the general availability
>> of the OpenStack Newton B1 milestone in Ubuntu 16.10 and for Ubuntu 16.04
>> LTS via the Ubuntu Cloud Archive.
>>
>> Ubuntu 16.04 LTS
>> 
>>
>> You can enable the Ubuntu Cloud Archive pocket for OpenStack Newton on
>> Ubuntu 16.04 installations by running the following commands:
>>
>> echo "deb http://ubuntu-cloud.archive.canonical.com/ubuntu
>> xenial-updates/newton main" | sudo tee
>> /etc/apt/sources.list.d/newton-uca.list
>> sudo apt-get install -y ubuntu-cloud-keyring
>> sudo apt-get update
>>
>> The Ubuntu Cloud Archive for Newton includes updates for Cinder,
>> Designate, Glance, Heat, Horizon, Keystone, Manila, Neutron, Neutron-FWaaS,
>> Neutron-LBaaS, Neutron-VPNaaS, Nova, and Swift (2.8.0).
>>
>> You can check out the full list of packages and versions at [0].
>>
>> Ubuntu 16.10
>> --
>>
>> No extra steps required; just start installing OpenStack!
>>
>> Branch Package Builds
>> ---
>>
>> We’ve resurrected the branch package builds of OpenStack projects that we
>> had in place a while back - if you want to try out the latest master branch
>> updates, or updates to stable branches, the following PPA’s are now
>> up-to-date and maintained:
>>
>>sudo add-apt-repository ppa:openstack-ubuntu-testing/liberty
>>sudo add-apt-repository ppa:openstack-ubuntu-testing/mitaka
>>sudo add-apt-repository ppa:openstack-ubuntu-testing/newton
>>
>> bear in mind these are built per-commit-ish (30 min checks for new
>> commits at the moment) so ymmv from time-to-time.
>>
>> Reporting bugs
>> -
>>
>> Any issues please report bugs using the 'ubuntu-bug' tool:
>>
>>   sudo ubuntu-bug nova-conductor
>>
>> this will ensure that bugs get logged in the right place in Launchpad.
>>
>> Thanks and have fun!
>>
>> Regards,
>> Corey
>> (on behalf of the Ubuntu OpenStack team)
>>
>> [0]
>> http://reqorts.qa.ubuntu.com/reports/ubuntu-server/cloud-archive/newton_versions.html
>>
>>
>> __
>> OpenStack Development Mailing List (not for usage questions)
>> Unsubscribe:
>> openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>>
>>
>
>
> --
> Regards,
> Jeffrey Zhang
> Blog: http://xcodest.me
>
> __
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>
>
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [Openstack] [openstack-dev] OpenStack Newton B1 for Ubuntu 16.04 LTS and Ubuntu 16.10

[Martinx - ジェームズ]

On 13 June 2016 at 08:54, Corey Bryant  wrote:

> Hi All,
>
> The Ubuntu OpenStack team is pleased to announce the general availability
> of the OpenStack Newton B1 milestone in Ubuntu 16.10 and for Ubuntu 16.04
> LTS via the Ubuntu Cloud Archive.
>
> Ubuntu 16.04 LTS
> 
>
> You can enable the Ubuntu Cloud Archive pocket for OpenStack Newton on
> Ubuntu 16.04 installations by running the following commands:
>
> echo "deb http://ubuntu-cloud.archive.canonical.com/ubuntu
> xenial-updates/newton main" | sudo tee
> /etc/apt/sources.list.d/newton-uca.list
> sudo apt-get install -y ubuntu-cloud-keyring
> sudo apt-get update
>
> The Ubuntu Cloud Archive for Newton includes updates for Cinder,
> Designate, Glance, Heat, Horizon, Keystone, Manila, Neutron, Neutron-FWaaS,
> Neutron-LBaaS, Neutron-VPNaaS, Nova, and Swift (2.8.0).
>
> You can check out the full list of packages and versions at [0].
>
> Ubuntu 16.10
> --
>
> No extra steps required; just start installing OpenStack!
>
> Branch Package Builds
> ---
>
> We’ve resurrected the branch package builds of OpenStack projects that we
> had in place a while back - if you want to try out the latest master branch
> updates, or updates to stable branches, the following PPA’s are now
> up-to-date and maintained:
>
>sudo add-apt-repository ppa:openstack-ubuntu-testing/liberty
>sudo add-apt-repository ppa:openstack-ubuntu-testing/mitaka
>sudo add-apt-repository ppa:openstack-ubuntu-testing/newton
>
> bear in mind these are built per-commit-ish (30 min checks for new commits
> at the moment) so ymmv from time-to-time.
>
> Reporting bugs
> -
>
> Any issues please report bugs using the 'ubuntu-bug' tool:
>
>   sudo ubuntu-bug nova-conductor
>
> this will ensure that bugs get logged in the right place in Launchpad.
>
> Thanks and have fun!
>
> Regards,
> Corey
> (on behalf of the Ubuntu OpenStack team)
>
> [0]
> http://reqorts.qa.ubuntu.com/reports/ubuntu-server/cloud-archive/newton_versions.html
>
>
This is awesome!   :-D
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [openstack-dev] [Monasca] influxDB clustering and HA will be "commercial option".

[Martinx - ジェームズ]

On 30 May 2016 at 11:59, Jaesuk Ahn  wrote:

> Hi, Monasca developers and users,
>
>
> https://influxdata.com/blog/update-on-influxdb-clustering-high-availability-and-monetization/
> "For our current and future customers, we’ll be offering clustering and
> high availability through Influx Cloud, our managed hosting offering, and
> Influx Enterprise, our on-premise offering, in the coming months.”
>
>
> It seems like “clustering” and “high availablity” of influxDB will be
> available only in commercial version.
> Monasca is currently leveraging influxDB as a metrics and alarm database.
> Beside vertical, influxDB is currently only an open source option to use.
>
> With this update stating “influxDB open source sw version will not have
> clustering / ha feature”,
> I would like to know if there has been any discussion among monasca
> community to add more database backend rather than influxDB, especially
> OpenTSDB.
>
>
> Thank you.
>
>
>
>
>
> --
> Jaesuk Ahn, Ph.D.
> Software Defined Infra Tech. Lab.
> SKT
>


What about Prometheus?

https://prometheus.io/

https://prometheus.io/docs/introduction/comparison/

Cheers!
Thiago
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [Neutron][L2GW] Mitaka release of L2 Gateway now available

[Martinx - ジェームズ]

On 11 May 2016 at 16:05, Sukhdev Kapur  wrote:

>
> Folks,
>
> I am happy to announce that Mitaka release for L2 Gateway is released and
> now available at https://pypi.python.org/pypi/networking-l2gw.
>
> You can install it by using "pip install networking-l2gw"
>
> This release has several enhancements and fixes for issues discovered in
> liberty release.
>
> Thanks
> Sukhdev Kapur
>
>
Sounds very interesting!

Currently, I have a DPDK App that is a L2 Bridge, however, when I bridge
two Neutron Networks together (under the same L2 broadcast domain),
OpenStack itself is "not aware" of this!

Basically, OpenStack doesn't "knows" that a "regular Instance" is a L2
Bridge, which make things very weird for NFV applications like mine.

So, my question is: can this "L2 Gateway" help my setup? I mean, can I use
"L2 Gateway" to tell: "Hey, OpenStack, those two Networks X & Y are in
fact, just one. Is this possible?

Cheers!
Thiago
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

[openstack-dev] Mitaka - OVS Firewall Driver, not working = OVSFWPortNotFound

[Martinx - ジェームズ]

Guys,

 I'm trying to enable OVS Firewall Driver in my Cloud Env but, it is not
working...

 I'm trying to replace the following line (openvswitch_agent.ini config
across the cloud):

---
 firewall_driver =
neutron.agent.linux.iptables_firewall.OVSHybridIptablesFirewallDriver
---

 By this:

---
 firewall_driver = openvswitch
---

 But it does not work in a multi-node env (all-in-one looks good).

 I'm seeing the following error on Neutron OpenvSwitch Agents, while trying
to launch Instances:

 "OVSFWPortNotFound" Port "is not managed by this agent"

 Full log putput:

 http://paste.openstack.org/show/495252/

 The result is that instances are inaccessible from its Floating IP.
Rolling back to OVSHybrid, all goo again.

 What am I missing?

Thanks!
Thiago
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [Openstack] Mitaka - OVS Firewall Driver, not working = OVSFWPortNotFound

[Martinx - ジェームズ]

On 25 April 2016 at 01:51, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

> Guys,
>
>  I'm trying to replace this (openvswitch_agent.ini across the cloud):
>
> ---
>  firewall_driver =
> neutron.agent.linux.iptables_firewall.OVSHybridIptablesFirewallDriver
> ---
>
>  By this:
>
> ---
>  firewall_driver = openvswitch
> ---
>
>  But it does not work.
>
>  I'm seeing the following error on Neutron OpenvSwitch Agents, while
> trying to launch Instances:
>
>  "OVSFWPortNotFound" Port "is not managed by this agent"
>
>  Full log putput:
>
>  http://paste.openstack.org/show/495252/
>
>  The result is that instances are inaccessible from its Floating IP.
> Rolling back to OVSHybrid, all goo again.
>
>  What am I missing?
>
> Thanks!
> Thiago
>

Guys? How to enable OVS Firewall Driver?
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Mitaka - OVS Firewall Driver, not working = OVSFWPortNotFound

[Martinx - ジェームズ]

Guys,

 I'm trying to replace this (openvswitch_agent.ini across the cloud):

---
 firewall_driver =
neutron.agent.linux.iptables_firewall.OVSHybridIptablesFirewallDriver
---

 By this:

---
 firewall_driver = openvswitch
---

 But it does not work.

 I'm seeing the following error on Neutron OpenvSwitch Agents, while trying
to launch Instances:

 "OVSFWPortNotFound" Port "is not managed by this agent"

 Full log putput:

 http://paste.openstack.org/show/495252/

 The result is that instances are inaccessible from its Floating IP.
Rolling back to OVSHybrid, all goo again.

 What am I missing?

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Mitaka, does it supports "VLAN aware VMs"?

[Martinx - ジェームズ]

On 22 April 2016 at 18:33, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

> Guys,
>
>  I need to create Instances on Mitaka, that needs to see the VLAN tags...
>
>  Does Mitaka supports that?
>
>  If not, what is the state of this? If yes, where are the docs?   =)
>
> Thanks!
> Thiago
>

Mmm... Looks like that Mitaka doesn't support this...   :-(

https://blueprints.launchpad.net/neutron/+spec/vlan-aware-vms

I would like to apply any patches manually! If possible... Any tips?

Cheers!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Mitaka, does it supports "VLAN aware VMs"?

[Martinx - ジェームズ]

Guys,

 I need to create Instances on Mitaka, that needs to see the VLAN tags...

 Does Mitaka supports that?

 If not, what is the state of this? If yes, where are the docs?   =)

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] OpenStack Mitaka for Ubuntu 14.04 LTS and Ubuntu 16.04 LTS

[Martinx - ジェームズ]

Mitaka on Xenial looks good now! YAY!!   :-P

On 21 April 2016 at 16:24, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

> This is awesome!
>
> However, I am facing a hard time to make it work with OpenvSwitch and
> multi-node environment... All works in an "All in One" fashion...
>
> I'll keep researching and testing it, until it works.
>
> Nice job guys, congrats!
>
> Cheers!
> Thiago
>
> On 21 April 2016 at 16:08, Corey Bryant <corey.bry...@canonical.com>
> wrote:
>
>> Hi All,
>>
>> The Ubuntu OpenStack Engineering team is pleased to announce the general
>> availability of OpenStack Mitaka in Ubuntu 16.04 LTS and for Ubuntu 14.04
>> LTS via the Ubuntu Cloud Archive.
>>
>> Ubuntu 14.04 LTS
>> 
>>
>> You can enable the Ubuntu Cloud Archive for OpenStack Mitaka on Ubuntu
>> 14.04 installations by running the following commands (make sure you have
>> an up-to-date software-properties-common package first):
>>
>>   sudo add-apt-repository cloud-archive:mitaka
>>   sudo apt-get update
>>
>> The Ubuntu Cloud Archive for Mitaka includes updates for Nova, Glance,
>> Keystone, Neutron, Neutron-FWaaS, Neutron-LBaaS, Neutron-VPNaaS, Cinder,
>> Swift, Horizon, Ceilometer, Aodh, Heat, Designate, Barbican, Manila,
>> Mistral, Murano, Ironic, Trove, Sahara, Senlin, and Zaqar; Ceph (10.1.2),
>> Nova-LXD (13.0.0), RabbitMQ (3.5.7), QEMU (1.2.5), libvirt (1.3.1),
>> OpenvSwitch (2.5.0) and DPDK (2.2.0) backports from Ubuntu 16.04
>> development have also been provided.
>>
>> You can see the full list of packages and versions at [0].
>>
>> Ubuntu 16.04 LTS
>> 
>>
>> No extra steps are required; just start installing OpenStack!
>>
>> Release notes for Ubuntu 16.04 LTS can be found at [1].
>>
>> Reporting bugs
>> -
>>
>> If you have any issues, please report bugs using the 'ubuntu-bug' tool:
>>
>>   sudo ubuntu-bug nova-conductor
>>
>> This will ensure that bugs get logged in the right place in Launchpad.
>>
>> Also feel free to poke anyone on our team in #ubuntu-server (coreycb,
>> ddellav, beisner, jamespage) if you have any issues.
>>
>> Finally, thank you to the entire OpenStack community and everyone who's
>> made OpenStack Mitaka a reality!
>>
>> [0]
>> http://reqorts.qa.ubuntu.com/reports/ubuntu-server/cloud-archive/mitaka_versions.html
>> [1] https://wiki.ubuntu.com/XenialXerus/ReleaseNotes
>>
>> --
>> Regards,
>> Corey Bryant
>> Ubuntu Core Developer
>>
>>
>> ___
>> Mailing list:
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>> Post to : openstack@lists.openstack.org
>> Unsubscribe :
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>
>>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Mitaka with OpenvSwitch, no VXLAN tunnels between compute and network nodes

[Martinx - ジェームズ]

On 21 April 2016 at 16:05, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

>
>
> On 21 April 2016 at 15:54, Martinx - ジェームズ <thiagocmarti...@gmail.com>
> wrote:
>
>>
>> On 21 April 2016 at 15:52, Martinx - ジェームズ <thiagocmarti...@gmail.com>
>> wrote:
>>
>>> Guys,
>>>
>>>  I'm trying to deploy Mitaka on Xenial, using OpenvSwitch.
>>>
>>>  I am using the following documents:
>>>
>>>  http://docs.openstack.org/mitaka/install-guide-ubuntu
>>>
>>>
>>> http://docs.openstack.org/mitaka/networking-guide/scenario-classic-ovs.html
>>>
>>>
>>>  * The topology is like this:
>>>
>>>  1- Controller + Network Node;
>>>
>>>  2- Compute Node.
>>>
>>>  NOTE: Both have two NICs, em1 for management and services, and p1p1 for
>>> VXLAN tunnels.
>>>
>>>
>>>  However, I see no "vxlan peers" on the output of "ovs-vsctl show"...
>>>
>>>  So, Instances doesn't get IP from DHCP server, and if I try to
>>> configure static IPs, Instances can only talk with each other (that are
>>> running on the same, and uniq, compute node). Also, instances can't ping
>>> Neutron L2 Namespace router.
>>>
>>>  What I am missing?
>>>
>>> Thanks!
>>> Thiago
>>>
>>
>> BTW, I am using an Ansibe automation tool that I created for this. If I
>> use the same roles, to deploy an "All in One" OpenStack box, with
>> OpenvSwitch, of course, it works!
>>
>> Only when I split the Compute Node, out from the "Controller+Network"
>> node, that it doesn't work anymore...
>>
>
> Also, I can create the Network and its Subnet, the Router attached to both
> "External Net" and with a internal interface...
>
> All network agents are up and running...
>
> But, VXLAN tunnels, no connectivity
>

It is working now!
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Xenial / Mitaka = Instance failed network setup / nova.compute.manager Unauthorized

[Martinx - ジェームズ]

I just manage to make OpenStack Mitaka to work with both Linux Bridges, and
OpenvSwitch...

Everything is work on both "All in One" and multi-node environments.

Very soon, I'll post instructions about how to use the Ansible automation
that I am developing to play with this...

Then, you guys will be able to deploy it, on a spare box (or VMs), and
then, compare with failing deployemnts...

Cheers!

On 21 April 2016 at 16:44, Paras pradhan <pradhanpa...@gmail.com> wrote:

> The neutron section was missing from nova.conf and now the instances work
> but having issues with metadata server. instances boot but no network
> access.
> Thanks
> Paras.
>
> On Thu, Apr 21, 2016 at 9:58 AM, Martinx - ジェームズ <
> thiagocmarti...@gmail.com> wrote:
>
>> My nova.conf is this one:
>>
>>
>> https://github.com/tmartinx/svauto/blob/dev/ansible/roles/os_nova_cmpt/templates/mitaka/nova.conf
>>
>> However, I'm facing connectivity problems when with OpenvSwitch
>> deployments, investigating it now...
>>
>> On 21 April 2016 at 11:04, Paras pradhan <pradhanpa...@gmail.com> wrote:
>>
>>> Yes . I did. here is my nova.conf. Can you share yours?
>>> http://paste.openstack.org/show/494988/
>>>
>>> On Thu, Apr 21, 2016 at 7:36 AM, Eugen Block <ebl...@nde.ag> wrote:
>>>
>>>> Okay, did you restart the respective services on all nodes after the
>>>> changes in your config files? If the same error still occurs then you might
>>>> haven't found all occurrences of the option auth_plugin, did you replace it
>>>> in all config files? I'm just guessing here...
>>>>
>>>>
>>>>
>>>> Zitat von Paras pradhan <pradhanpa...@gmail.com>:
>>>>
>>>> No I still have that error. Other than that I don't see any other
>>>>> errors.
>>>>>
>>>>> On Wed, Apr 20, 2016 at 9:22 AM, Eugen Block <ebl...@nde.ag> wrote:
>>>>>
>>>>> So I guess the mentioned error should be resolved? Does it work now?
>>>>>>
>>>>>>
>>>>>>
>>>>>> Zitat von Paras pradhan <pradhanpa...@gmail.com>:
>>>>>>
>>>>>> Yes I I have it set up in nova.conf and neutron.conf
>>>>>>
>>>>>>>
>>>>>>> On Wed, Apr 20, 2016 at 9:11 AM, Eugen Block <ebl...@nde.ag> wrote:
>>>>>>>
>>>>>>> And did you change it to auth_type instead of auth_plugin? Also you
>>>>>>> should
>>>>>>>
>>>>>>>> make sure that this option is in the correct section of your config
>>>>>>>> file,
>>>>>>>> for example
>>>>>>>>
>>>>>>>> [keystone_authtoken]
>>>>>>>> ...
>>>>>>>> auth_type = password
>>>>>>>>
>>>>>>>> or
>>>>>>>>
>>>>>>>> [neutron]
>>>>>>>> ...
>>>>>>>> auth_type = password
>>>>>>>>
>>>>>>>>
>>>>>>>> Regards,
>>>>>>>> Eugen
>>>>>>>>
>>>>>>>>
>>>>>>>> Zitat von Paras pradhan <pradhanpa...@gmail.com>:
>>>>>>>>
>>>>>>>> Hi Eugen,
>>>>>>>>
>>>>>>>>
>>>>>>>>> Thanks. Log says its an error. Here is the full log.
>>>>>>>>> http://pastebin.com/K1f4pJhB
>>>>>>>>>
>>>>>>>>> -Paras.
>>>>>>>>>
>>>>>>>>> On Tue, Apr 19, 2016 at 2:05 AM, Eugen Block <ebl...@nde.ag>
>>>>>>>>> wrote:
>>>>>>>>>
>>>>>>>>> Hi Paras,
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>> the option auth_plugin is deprecated (from nova.conf):
>>>>>>>>>>
>>>>>>>>>> ---cut here---
>>>>>>>>>> # Authentication type to load (unknown value)
>>>>>>>>>> # Deprecated group/name - [DEFAULT]/auth_plugin
>>>>>>>>>> auth_type = password
>>>>>>>>>> ---cut here---
>>&g

Re: [Openstack] OpenStack Mitaka for Ubuntu 14.04 LTS and Ubuntu 16.04 LTS

[Martinx - ジェームズ]

This is awesome!

However, I am facing a hard time to make it work with OpenvSwitch and
multi-node environment... All works in an "All in One" fashion...

I'll keep researching and testing it, until it works.

Nice job guys, congrats!

Cheers!
Thiago

On 21 April 2016 at 16:08, Corey Bryant  wrote:

> Hi All,
>
> The Ubuntu OpenStack Engineering team is pleased to announce the general
> availability of OpenStack Mitaka in Ubuntu 16.04 LTS and for Ubuntu 14.04
> LTS via the Ubuntu Cloud Archive.
>
> Ubuntu 14.04 LTS
> 
>
> You can enable the Ubuntu Cloud Archive for OpenStack Mitaka on Ubuntu
> 14.04 installations by running the following commands (make sure you have
> an up-to-date software-properties-common package first):
>
>   sudo add-apt-repository cloud-archive:mitaka
>   sudo apt-get update
>
> The Ubuntu Cloud Archive for Mitaka includes updates for Nova, Glance,
> Keystone, Neutron, Neutron-FWaaS, Neutron-LBaaS, Neutron-VPNaaS, Cinder,
> Swift, Horizon, Ceilometer, Aodh, Heat, Designate, Barbican, Manila,
> Mistral, Murano, Ironic, Trove, Sahara, Senlin, and Zaqar; Ceph (10.1.2),
> Nova-LXD (13.0.0), RabbitMQ (3.5.7), QEMU (1.2.5), libvirt (1.3.1),
> OpenvSwitch (2.5.0) and DPDK (2.2.0) backports from Ubuntu 16.04
> development have also been provided.
>
> You can see the full list of packages and versions at [0].
>
> Ubuntu 16.04 LTS
> 
>
> No extra steps are required; just start installing OpenStack!
>
> Release notes for Ubuntu 16.04 LTS can be found at [1].
>
> Reporting bugs
> -
>
> If you have any issues, please report bugs using the 'ubuntu-bug' tool:
>
>   sudo ubuntu-bug nova-conductor
>
> This will ensure that bugs get logged in the right place in Launchpad.
>
> Also feel free to poke anyone on our team in #ubuntu-server (coreycb,
> ddellav, beisner, jamespage) if you have any issues.
>
> Finally, thank you to the entire OpenStack community and everyone who's
> made OpenStack Mitaka a reality!
>
> [0]
> http://reqorts.qa.ubuntu.com/reports/ubuntu-server/cloud-archive/mitaka_versions.html
> [1] https://wiki.ubuntu.com/XenialXerus/ReleaseNotes
>
> --
> Regards,
> Corey Bryant
> Ubuntu Core Developer
>
>
> ___
> Mailing list:
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> Post to : openstack@lists.openstack.org
> Unsubscribe :
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Mitaka with OpenvSwitch, no VXLAN tunnels between compute and network nodes

[Martinx - ジェームズ]

On 21 April 2016 at 15:54, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

>
> On 21 April 2016 at 15:52, Martinx - ジェームズ <thiagocmarti...@gmail.com>
> wrote:
>
>> Guys,
>>
>>  I'm trying to deploy Mitaka on Xenial, using OpenvSwitch.
>>
>>  I am using the following documents:
>>
>>  http://docs.openstack.org/mitaka/install-guide-ubuntu
>>
>>
>> http://docs.openstack.org/mitaka/networking-guide/scenario-classic-ovs.html
>>
>>
>>  * The topology is like this:
>>
>>  1- Controller + Network Node;
>>
>>  2- Compute Node.
>>
>>  NOTE: Both have two NICs, em1 for management and services, and p1p1 for
>> VXLAN tunnels.
>>
>>
>>  However, I see no "vxlan peers" on the output of "ovs-vsctl show"...
>>
>>  So, Instances doesn't get IP from DHCP server, and if I try to configure
>> static IPs, Instances can only talk with each other (that are running on
>> the same, and uniq, compute node). Also, instances can't ping Neutron L2
>> Namespace router.
>>
>>  What I am missing?
>>
>> Thanks!
>> Thiago
>>
>
> BTW, I am using an Ansibe automation tool that I created for this. If I
> use the same roles, to deploy an "All in One" OpenStack box, with
> OpenvSwitch, of course, it works!
>
> Only when I split the Compute Node, out from the "Controller+Network"
> node, that it doesn't work anymore...
>

Also, I can create the Network and its Subnet, the Router attached to both
"External Net" and with a internal interface...

All network agents are up and running...

But, VXLAN tunnels, no connectivity
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Mitaka with OpenvSwitch, no VXLAN tunnels between compute and network nodes

[Martinx - ジェームズ]

On 21 April 2016 at 15:52, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

> Guys,
>
>  I'm trying to deploy Mitaka on Xenial, using OpenvSwitch.
>
>  I am using the following documents:
>
>  http://docs.openstack.org/mitaka/install-guide-ubuntu
>
>
> http://docs.openstack.org/mitaka/networking-guide/scenario-classic-ovs.html
>
>
>  * The topology is like this:
>
>  1- Controller + Network Node;
>
>  2- Compute Node.
>
>  NOTE: Both have two NICs, em1 for management and services, and p1p1 for
> VXLAN tunnels.
>
>
>  However, I see no "vxlan peers" on the output of "ovs-vsctl show"...
>
>  So, Instances doesn't get IP from DHCP server, and if I try to configure
> static IPs, Instances can only talk with each other (that are running on
> the same, and uniq, compute node). Also, instances can't ping Neutron L2
> Namespace router.
>
>  What I am missing?
>
> Thanks!
> Thiago
>

BTW, I am using an Ansibe automation tool that I created for this. If I use
the same roles, to deploy an "All in One" OpenStack box, with OpenvSwitch,
of course, it works!

Only when I split the Compute Node, out from the "Controller+Network" node,
that it doesn't work anymore...
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Xenial / Mitaka = Instance failed network setup / nova.compute.manager Unauthorized

[Martinx - ジェームズ]

My nova.conf is this one:

https://github.com/tmartinx/svauto/blob/dev/ansible/roles/os_nova_cmpt/templates/mitaka/nova.conf

However, I'm facing connectivity problems when with OpenvSwitch
deployments, investigating it now...

On 21 April 2016 at 11:04, Paras pradhan <pradhanpa...@gmail.com> wrote:

> Yes . I did. here is my nova.conf. Can you share yours?
> http://paste.openstack.org/show/494988/
>
> On Thu, Apr 21, 2016 at 7:36 AM, Eugen Block <ebl...@nde.ag> wrote:
>
>> Okay, did you restart the respective services on all nodes after the
>> changes in your config files? If the same error still occurs then you might
>> haven't found all occurrences of the option auth_plugin, did you replace it
>> in all config files? I'm just guessing here...
>>
>>
>>
>> Zitat von Paras pradhan <pradhanpa...@gmail.com>:
>>
>> No I still have that error. Other than that I don't see any other errors.
>>>
>>> On Wed, Apr 20, 2016 at 9:22 AM, Eugen Block <ebl...@nde.ag> wrote:
>>>
>>> So I guess the mentioned error should be resolved? Does it work now?
>>>>
>>>>
>>>>
>>>> Zitat von Paras pradhan <pradhanpa...@gmail.com>:
>>>>
>>>> Yes I I have it set up in nova.conf and neutron.conf
>>>>
>>>>>
>>>>> On Wed, Apr 20, 2016 at 9:11 AM, Eugen Block <ebl...@nde.ag> wrote:
>>>>>
>>>>> And did you change it to auth_type instead of auth_plugin? Also you
>>>>> should
>>>>>
>>>>>> make sure that this option is in the correct section of your config
>>>>>> file,
>>>>>> for example
>>>>>>
>>>>>> [keystone_authtoken]
>>>>>> ...
>>>>>> auth_type = password
>>>>>>
>>>>>> or
>>>>>>
>>>>>> [neutron]
>>>>>> ...
>>>>>> auth_type = password
>>>>>>
>>>>>>
>>>>>> Regards,
>>>>>> Eugen
>>>>>>
>>>>>>
>>>>>> Zitat von Paras pradhan <pradhanpa...@gmail.com>:
>>>>>>
>>>>>> Hi Eugen,
>>>>>>
>>>>>>
>>>>>>> Thanks. Log says its an error. Here is the full log.
>>>>>>> http://pastebin.com/K1f4pJhB
>>>>>>>
>>>>>>> -Paras.
>>>>>>>
>>>>>>> On Tue, Apr 19, 2016 at 2:05 AM, Eugen Block <ebl...@nde.ag> wrote:
>>>>>>>
>>>>>>> Hi Paras,
>>>>>>>
>>>>>>>
>>>>>>>> the option auth_plugin is deprecated (from nova.conf):
>>>>>>>>
>>>>>>>> ---cut here---
>>>>>>>> # Authentication type to load (unknown value)
>>>>>>>> # Deprecated group/name - [DEFAULT]/auth_plugin
>>>>>>>> auth_type = password
>>>>>>>> ---cut here---
>>>>>>>>
>>>>>>>> But as far as I can tell, you should only get a warning, not an
>>>>>>>> error,
>>>>>>>> I've seen some of these warnings in my logs, but it works (I work
>>>>>>>> with
>>>>>>>> openSUSE). To get Mitaka working at all I simply tried to set the
>>>>>>>> same
>>>>>>>> options as in my working Liberty configs, and then I searched for
>>>>>>>> deprecation warnings and additional options mentioned in the Mitaka
>>>>>>>> guide.
>>>>>>>>
>>>>>>>> Hope this helps!
>>>>>>>>
>>>>>>>> Regards,
>>>>>>>> Eugen
>>>>>>>>
>>>>>>>>
>>>>>>>> Zitat von Paras pradhan <pradhanpa...@gmail.com>:
>>>>>>>>
>>>>>>>>
>>>>>>>> Can somebody share the nova.conf and neutron.conf from working
>>>>>>>> mitaka?
>>>>>>>> I
>>>>>>>> am
>>>>>>>>
>>>>>>>> also following the same guide and ran into a problem.
>>>>>>>>
>>>>>>>>>
>>>>>>>>> 2016-04-18 16:51:07.

Re: [Openstack] Xenial / Mitaka = Instance failed network setup / nova.compute.manager Unauthorized

[Martinx - ジェームズ]

FIXED!! I knew it was a typo somewhere!   LOL

https://github.com/tmartinx/svauto/commit/40ce6566cd0e6435cf75bb34116b6c3bacbeaf02

Thank you guys!

Sorry about the buzz on TWO mail lists...

At least, now we know that Nova silent fail on its start up, if somethings
aren't configured according... And there is no verification steps to test
the communication between Nova and Neutron.

 Mitaka is working now on Xenial! YAY!!

I'm about to commit changes to enable OpenvSwitch with DPDK and multi-node
deployments, fully automated!

Cheers!
Thiago

On 17 April 2016 at 21:26, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

> On 17 April 2016 at 17:39, Martinx - ジェームズ <thiagocmarti...@gmail.com>
> wrote:
>
>> Guys,
>>
>>  I am trying to deploy Mitaka, on top of Ubuntu 16.04, by using the
>> following document:
>>
>>  http://docs.openstack.org/mitaka/install-guide-ubuntu
>>
>>  Yes, I know, the above document is for installing Mitaka on top of
>> Ubuntu 14.04 but, from what I understand, the only difference is that on
>> Xenial, there is no need to add the Ubuntu Mitaka Cloud Archive, since
>> Mitaka is the default on Xenial, so, I can follow that document, right?
>>  =)
>>
>>  At first, OpenStack installation goes okay, without any errors, all
>> services comes up, etc... However, it is not possible to launch an Instance.
>>
>>  *** Errors on launching the Instance:
>>
>>  - Right after launching it:
>>
>>  https://paste.ubuntu.com/15902503/
>>
>>  - Spawning it, after Glance finishes the download, similar error a
>> second time:
>>
>>  https://paste.ubuntu.com/15902556/
>>
>>  What am I missing?
>>
>>  Apparently, Nova is not authorized to talk with Neutron but, I am
>> following the docs (maybe it is just a typo somewhere?)...
>>
>>  Also, I have an Ansible automation to deploy it, so, it is less error
>> prone. And however can help me, will see how I am deploying it.
>>
>>  I see no error on Admin Dashboard, all services are up.
>>
>>  NOTE: My Ansible playbooks is in a sense, "
>> docs.openstack.org/mitaka/install-guide-ubuntu fully automated", it is
>> very close to it, step-by-step.
>>
>>  How can I debug this? I mean, how can I try to do, what Nova is doing
>> (its connection with Neutron), to make sure that the settings are in place
>> correctly?
>>
>>  Here is how I am installing Mitaka on Xenial:
>>
>> ---
>>  1- Install Ubuntu 16.04 server 64-bit on bare-metal;
>>
>>* Configure your /etc/hostname and /etc/hosts.
>>
>>DETAILS:
>> https://github.com/tmartinx/svauto/blob/dev/README.OpenStack.md
>>
>>
>>  2- Clone the automation:
>>
>> cd ~
>> git clone https://github.com/tmartinx/svauto.git
>>
>>
>>  3- Run the automation to install OpenStack (all-in-one)
>>
>> cd ~/svauto
>>
>>./os-deploy.sh --br-mode=LBR --use-dummies --base-os=ubuntu16
>> --base-os-upgrade=yes --openstack-release=mitaka --openstack-installation
>> --dry-run
>>
>>ansible-playbook -c local site-openstack.yml --extra-vars
>> "openstack_installation=yes"
>> ---
>>
>>  NOTE: If you don't use "--dry-run" option, Ansible will be executed
>> automatically by "os-deploy.sh".
>>
>>  I am sharing the Ansible playbooks, because it will be easier to see
>> what I am doing.
>>
>>  About the relevant configuration blocks, I believe that I have it
>> properly configured (I followed Mitaka docs), like this:
>>
>>  * neutron.conf:
>> -
>>
>> https://github.com/tmartinx/svauto/blob/dev/ansible/roles/os_neutron_aio/templates/mitaka/neutron.conf
>> -
>>
>>  * nova.conf:
>> -
>>
>> https://github.com/tmartinx/svauto/blob/dev/ansible/roles/os_nova_aio/templates/mitaka/nova.conf
>> -
>>
>>  I already installed OpenStack many, many, many times, since Havana
>> release, I'm confident that I am doing it right but, of course, maybe I did
>> something wrong this time...   =P
>>
>>  I appreciate any help!
>>
>> Thanks!
>> Thiago
>>
>
> Hey guys,
>
>  I am reviewing everything, to make sure that my automation matches
> docs.openstack.org/mitaka/install-guide-ubuntu/ in details.
>
>  Two things...
>
> ---
>  1- All the verification steps (from official doc) passed
>
>
> http://docs.openstack.org/mitaka/install-guide-ubuntu/keystone-verify.html
> - passed
>
>
> http://docs.openstack.org/mitaka/install-gui

[Openstack] nova.virt.osinfo - Cannot load Libosinfo: (No module named Libosinfo)

[Martinx - ジェームズ]

Guys,

 What this message means?

 I'm seeing it on Mitaka / Xenial...

 What it [Libosinfo] brings to the table and how can I enabled it?

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Xenial / Mitaka = Instance failed network setup / nova.compute.manager Unauthorized

[Martinx - ジェームズ]

On 17 April 2016 at 17:39, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

> Guys,
>
>  I am trying to deploy Mitaka, on top of Ubuntu 16.04, by using the
> following document:
>
>  http://docs.openstack.org/mitaka/install-guide-ubuntu
>
>  Yes, I know, the above document is for installing Mitaka on top of Ubuntu
> 14.04 but, from what I understand, the only difference is that on Xenial,
> there is no need to add the Ubuntu Mitaka Cloud Archive, since Mitaka is
> the default on Xenial, so, I can follow that document, right?=)
>
>  At first, OpenStack installation goes okay, without any errors, all
> services comes up, etc... However, it is not possible to launch an Instance.
>
>  *** Errors on launching the Instance:
>
>  - Right after launching it:
>
>  https://paste.ubuntu.com/15902503/
>
>  - Spawning it, after Glance finishes the download, similar error a second
> time:
>
>  https://paste.ubuntu.com/15902556/
>
>  What am I missing?
>
>  Apparently, Nova is not authorized to talk with Neutron but, I am
> following the docs (maybe it is just a typo somewhere?)...
>
>  Also, I have an Ansible automation to deploy it, so, it is less error
> prone. And however can help me, will see how I am deploying it.
>
>  I see no error on Admin Dashboard, all services are up.
>
>  NOTE: My Ansible playbooks is in a sense, "
> docs.openstack.org/mitaka/install-guide-ubuntu fully automated", it is
> very close to it, step-by-step.
>
>  How can I debug this? I mean, how can I try to do, what Nova is doing
> (its connection with Neutron), to make sure that the settings are in place
> correctly?
>
>  Here is how I am installing Mitaka on Xenial:
>
> ---
>  1- Install Ubuntu 16.04 server 64-bit on bare-metal;
>
>* Configure your /etc/hostname and /etc/hosts.
>
>DETAILS:
> https://github.com/tmartinx/svauto/blob/dev/README.OpenStack.md
>
>
>  2- Clone the automation:
>
> cd ~
> git clone https://github.com/tmartinx/svauto.git
>
>
>  3- Run the automation to install OpenStack (all-in-one)
>
> cd ~/svauto
>
>./os-deploy.sh --br-mode=LBR --use-dummies --base-os=ubuntu16
> --base-os-upgrade=yes --openstack-release=mitaka --openstack-installation
> --dry-run
>
>ansible-playbook -c local site-openstack.yml --extra-vars
> "openstack_installation=yes"
> ---
>
>  NOTE: If you don't use "--dry-run" option, Ansible will be executed
> automatically by "os-deploy.sh".
>
>  I am sharing the Ansible playbooks, because it will be easier to see what
> I am doing.
>
>  About the relevant configuration blocks, I believe that I have it
> properly configured (I followed Mitaka docs), like this:
>
>  * neutron.conf:
> -
>
> https://github.com/tmartinx/svauto/blob/dev/ansible/roles/os_neutron_aio/templates/mitaka/neutron.conf
> -
>
>  * nova.conf:
> -
>
> https://github.com/tmartinx/svauto/blob/dev/ansible/roles/os_nova_aio/templates/mitaka/nova.conf
> -
>
>  I already installed OpenStack many, many, many times, since Havana
> release, I'm confident that I am doing it right but, of course, maybe I did
> something wrong this time...   =P
>
>  I appreciate any help!
>
> Thanks!
> Thiago
>

Hey guys,

 I am reviewing everything, to make sure that my automation matches
docs.openstack.org/mitaka/install-guide-ubuntu/ in details.

 Two things...

---
 1- All the verification steps (from official doc) passed

  http://docs.openstack.org/mitaka/install-guide-ubuntu/keystone-verify.html
- passed

  http://docs.openstack.org/mitaka/install-guide-ubuntu/keystone-openrc.html
- passed

  http://docs.openstack.org/mitaka/install-guide-ubuntu/glance-verify.html
- passed

  http://docs.openstack.org/mitaka/install-guide-ubuntu/nova-verify.html -
passed

  http://docs.openstack.org/mitaka/install-guide-ubuntu/neutron-verify.html
- passed


http://docs.openstack.org/mitaka/install-guide-ubuntu/neutron-verify-option2.html
- passed

  http://docs.openstack.org/mitaka/install-guide-ubuntu/horizon-verify.html
- passed (mine doesn't have /horizon)

  http://docs.openstack.org/mitaka/install-guide-ubuntu/cinder-verify.html
- passed but, I'm not using any cinder-volume yet


http://docs.openstack.org/mitaka/install-guide-ubuntu/launch-instance-selfservice.html
- "6. Launch the instance" - FAIL

  NOTE: I can create Networks, Subnets, Routers... I can see the Linux
Namespaces at the Neutron L3 Router (All-in-One)


 2- This problem happen no matter if deploying it with Linux Bridges
(--br-mode=LBR) or with OpenvSwitch (--br-mode=OVS)
---

 So, I think that OpenStack should provide more verification steps,
specially to test the communication / authentication between its differ

[Openstack] Xenial / Mitaka = Instance failed network setup / nova.compute.manager Unauthorized

[Martinx - ジェームズ]

Guys,

 I am trying to deploy Mitaka, on top of Ubuntu 16.04, by using the
following document:

 http://docs.openstack.org/mitaka/install-guide-ubuntu

 Yes, I know, the above document is for installing Mitaka on top of Ubuntu
14.04 but, from what I understand, the only difference is that on Xenial,
there is no need to add the Ubuntu Mitaka Cloud Archive, since Mitaka is
the default on Xenial, so, I can follow that document, right?=)

 At first, OpenStack installation goes okay, without any errors, all
services comes up, etc... However, it is not possible to launch an Instance.

 *** Errors on launching the Instance:

 - Right after launching it:

 https://paste.ubuntu.com/15902503/

 - Spawning it, after Glance finishes the download, similar error a second
time:

 https://paste.ubuntu.com/15902556/

 What am I missing?

 Apparently, Nova is not authorized to talk with Neutron but, I am
following the docs (maybe it is just a typo somewhere?)...

 Also, I have an Ansible automation to deploy it, so, it is less error
prone. And however can help me, will see how I am deploying it.

 I see no error on Admin Dashboard, all services are up.

 NOTE: My Ansible playbooks is in a sense, "
docs.openstack.org/mitaka/install-guide-ubuntu fully automated", it is very
close to it, step-by-step.

 How can I debug this? I mean, how can I try to do, what Nova is doing (its
connection with Neutron), to make sure that the settings are in place
correctly?

 Here is how I am installing Mitaka on Xenial:

---
 1- Install Ubuntu 16.04 server 64-bit on bare-metal;

   * Configure your /etc/hostname and /etc/hosts.

   DETAILS: https://github.com/tmartinx/svauto/blob/dev/README.OpenStack.md


 2- Clone the automation:

cd ~
git clone https://github.com/tmartinx/svauto.git


 3- Run the automation to install OpenStack (all-in-one)

cd ~/svauto

   ./os-deploy.sh --br-mode=LBR --use-dummies --base-os=ubuntu16
--base-os-upgrade=yes --openstack-release=mitaka --openstack-installation
--dry-run

   ansible-playbook -c local site-openstack.yml --extra-vars
"openstack_installation=yes"
---

 NOTE: If you don't use "--dry-run" option, Ansible will be executed
automatically by "os-deploy.sh".

 I am sharing the Ansible playbooks, because it will be easier to see what
I am doing.

 About the relevant configuration blocks, I believe that I have it properly
configured (I followed Mitaka docs), like this:

 * neutron.conf:
-
https://github.com/tmartinx/svauto/blob/dev/ansible/roles/os_neutron_aio/templates/mitaka/neutron.conf
-

 * nova.conf:
-
https://github.com/tmartinx/svauto/blob/dev/ansible/roles/os_nova_aio/templates/mitaka/nova.conf
-

 I already installed OpenStack many, many, many times, since Havana
release, I'm confident that I am doing it right but, of course, maybe I did
something wrong this time...   =P

 I appreciate any help!

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] [OpenStack] How to deploy Liberty Neutron with linuxbridge on physical machine having single NIC

[Martinx - ジェームズ]

On 14 April 2016 at 05:17, liyulei  wrote:

> Hello,
>
>
>
> Is it possible that deploying Liberty Neutron with linuxbridge on physical
> machine having single NIC ?
>
> If it is, how should do ?
>
>
>
> Thanks
>
>
>
> Li yulei
>

Quite simple... At a glance:

1- Install Ubuntu 14.04;

2- Prepare it;

3- Install OpenStack Liberty with Linux Bridges:

bash <(curl -s
https://raw.githubusercontent.com/sandvine/os-ansible-deployment-lite/liberty/misc/os-install.sh
)


Full instructions:

https://github.com/sandvine/os-ansible-deployment-lite/tree/liberty


Basically, this automation automates docs.openstack.org...

Hope it helps!

Cheers!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] [Nova] [DPDK] [Liberty] Issues in booting VMs via Nova onto DPDK-enabled OVS

[Martinx - ジェームズ]

On 28 March 2016 at 05:21, N Vivekanandan 
wrote:

> Hi Openstack Team,
>
>
>
> We were trying to boot VMs via Openstack Nova (off Liberty release) into
> DPDK-enabled OVS instances.
>
> Our OVS is version 2.5  + DPDK 2.2 + Qemu 2.3 + Libvirt 1.2.16.
>
>
>
> We followed the guidelines here for installation of OVS Datapath and
> enable them for DPDK:
>
> https://github.com/openvswitch/ovs/blob/master/INSTALL.DPDK.md
>
>
>
> Howeve,r we were not able to get the VMs booted successfully via Nova.
>   When we attempted to boot the VMs , they went into error state.
>
> The vhostdpdkuser ports got created in the OVS-DPDK instance, but Libvirt
> Driver invoked by Nova, could not successfully attach this port to the VM.
>
>
>
> *We got these set of errors in libvirtd.log whenever we booted the VMs:*
>
> 2016-03-26 13:57:20.535+: 7354: error : virSecurityDriverLookup:80 :
> unsupported configuration: Security driver apparmor not enabled
>
> 2016-03-26 13:57:51.007+: 7339: error : virNetSocketReadWire:1520 :
> End of file while reading data: Input/output error
>
> 2016-03-26 13:58:36.293+: 7341: error : qemuMonitorOpenUnix:358 :
> failed to connect to monitor socket: No such process
>
> 2016-03-26 13:58:36.293+: 7341: error : qemuProcessWaitForMonitor:2113
> : internal error: process exited while connecting to monitor: 
> 2016-03-*26T13:58:36.235874Z
> qemu-system-x86_64: -chardev
> socket,id=charnet0,path=/var/lib/libvirt/qemu/vhu189ade8e-3b: Failed to
> connect socket: Permission denied*
>
>
>
>
>
> *On the QEMU log:*
>
> 2016-03-26 13:58:35.479+: starting up libvirt version: 1.2.16,
> package: 1.2.16-2ubuntu11.15.10.4~cloud0, qemu version: 2.3.0 (Debian
> 1:2.3+dfsg-5ubuntu9.2~cloud0)
>
> LC_ALL=C PATH=/usr/local/sbin:/usr/local/bin:/usr/bin:/usr/sbin:/sbin:/bin
> QEMU_AUDIO_DRV=none /usr/bin/kvm-spice -name instance-000c -S -machine
> pc-i440fx-vivid,accel=kvm,usb=off -m 512 -realtime mlock=off -smp
> 1,sockets=1,cores=1,threads=1 -uuid dbe34ace-960e-4915-884a-5a30b8ec896d
> -smbios type=1,manufacturer=OpenStack Foundation,product=OpenStack
> Nova,version=12.0.3,serial=----002590883f20,uuid=dbe34ace-960e-4915-884a-5a30b8ec896d,family=Virtual
> Machine -no-user-config -nodefaults -chardev
> socket,id=charmonitor,path=/var/lib/libvirt/qemu/instance-000c.monitor,server,nowait
> -mon chardev=charmonitor,id=monitor,mode=control -rtc
> base=utc,driftfix=slew -global kvm-pit.lost_tick_policy=discard -no-hpet
> -no-shutdown -boot strict=on -kernel
> /opt/stack/data/nova/instances/dbe34ace-960e-4915-884a-5a30b8ec896d/kernel
> -initrd
> /opt/stack/data/nova/instances/dbe34ace-960e-4915-884a-5a30b8ec896d/ramdisk
> -append root=/dev/vda console=tty0 console=ttyS0 -device
> piix3-usb-uhci,id=usb,bus=pci.0,addr=0x1.0x2 -drive
> file=/opt/stack/data/nova/instances/dbe34ace-960e-4915-884a-5a30b8ec896d/disk,if=none,id=drive-virtio-disk0,format=qcow2,cache=none
> -device
> virtio-blk-pci,scsi=off,bus=pci.0,addr=0x4,drive=drive-virtio-disk0,id=virtio-disk0,bootindex=1
> -drive
> file=/opt/stack/data/nova/instances/dbe34ace-960e-4915-884a-5a30b8ec896d/disk.config,if=none,id=drive-ide0-1-1,readonly=on,format=raw,cache=none
> -device ide-cd,bus=ide.1,unit=1,drive=drive-ide0-1-1,id=ide0-1-1 -chardev
> socket,id=charnet0,path=/var/lib/libvirt/qemu/vhu189ade8e-3b -netdev
> type=vhost-user,id=hostnet0,chardev=charnet0 -device
> virtio-net-pci,netdev=hostnet0,id=net0,mac=fa:16:3e:bd:cb:d6,bus=pci.0,addr=0x3
> -chardev
> file,id=charserial0,path=/opt/stack/data/nova/instances/dbe34ace-960e-4915-884a-5a30b8ec896d/console.log
> -device isa-serial,chardev=charserial0,id=serial0 -chardev
> pty,id=charserial1 -device isa-serial,chardev=charserial1,id=serial1 -vnc
> 10.183.254.64:0 -k en-us -device cirrus-vga,id=video0,bus=pci.0,addr=0x2
> -device virtio-balloon-pci,id=balloon0,bus=pci.0,addr=0x5 -msg timestamp=on
>
> *2016-03-26T13:58:36.235874Z qemu-system-x86_64: -chardev
> socket,id=charnet0,path=/var/lib/libvirt/qemu/vhu189ade8e-3b: Failed to
> connect socket: Permission denied*
>
>
>
> We verified in our setup that we have given user/group as ‘root/root’ in
> etc/libvirt/qemu.conf and also
>
> affirmed that our ovs-vswitchd runs as ‘root’ with DPDK enabled.
>
>
>
> The ‘vhuxxx’ ports are appearing in /var/lib/libvirt/qemu which is the
> vsock_host_dir folder that we have configured
>
> for ovs-vswitchd as well as in Nova.
>
> srwxr-xr-x 1 root root0 Mar 26 20:12 vhu98c8f690-41
>
>
>
> But QEMU is not able to complete attaching the VM to the vhostdpdkuser
> port.
>
> We included "/var/lib/libvirt/qemu" which is the vhost_sock_dir into
> cgroup_device_acl as well.
>
>
>
> And we also tried to put ‘security_driver’ to ‘None’ in qemu.conf to see
> if that helps (just to remove any bearing on SELinux (or) apparmor), but
> that
>
> didn’t help either.
>
>
>
> Highly Appreciate any tips.
>
>
>
> --
>
> Thanks,
>
>
>
> Vivek
>

Hello,

Maybe you're facing this 

Re: [Openstack] How to achieve OpenStack compute nodes "HA" ?

[Martinx - ジェームズ]

Guys,

 What about Virtual Machine Asynchronous replication under OpenStack?

 Something like:

 http://wiki.xen.org/wiki/Remus

 Or:

 http://wiki.qemu.org/Features/FaultTolerance

 This way, your can take the power cable off the physical compute node,
that all of your instances will be up and running on second node. No
downtime, not even TCP connection will be lost, like SSH or Windows RDP
sessions!

 Can we make OpenStack aware of Remus? And deal with this? I mean, I think
that OpenStack must create two instances, and put it in sync, when you
launch one instance that must be under H.A.. Looks like "quantum
entanglement" for Virtual Machines...   :-P

 I know that there are something similar to Remux (Xen) for QEmu, called
Kemari but, I don't know if it is under development...

 http://www.linux-kvm.org/images/0/0d/0.5.kemari-kvm-forum-2010.pdf

 Remus Xen:

 https://www.youtube.com/watch?v=2jV4lOgFJMY

 Kemari QEMu:

 https://www.youtube.com/watch?v=YrVNZG77PhY

Cheers!
Thiago

On 19 March 2016 at 00:09, Shinobu Kinjo  wrote:

> Thank you for the pointer.
> This project is a bit old, isn't it? Is it still on going ... no??
>
> Have a nice weekend too.
>
> Cheers,
> S
>
> On Sat, Mar 19, 2016 at 11:30 AM, CHOW Anthony
>  wrote:
> >
> https://blueprints.launchpad.net/nova/+spec/dynamic-consolidation-of-virtual-machines
> >
> > Seems like this is a feature under Nova.
> >
> > Nowadays with the Big Tent approach, we cannot rely on GitHub and
> StackForge to see if it is a full OpenStack "project".
> >
> > Have a nice weekend,
> >
> > Anthony.
> >
> > -Original Message-
> > From: Shinobu Kinjo [mailto:shinobu...@gmail.com]
> > Sent: Friday, March 18, 2016 6:54 PM
> > To: Jean-Pierre Ribeauville
> > Cc: openstack@lists.openstack.org
> > Subject: Re: [Openstack] How to achieve OpenStack compute nodes "HA" ?
> >
> > What is *this* project?
> >
> > Cheers,
> > S
> >
> > On Fri, Mar 18, 2016 at 10:28 PM, Jean-Pierre Ribeauville <
> jpribeauvi...@axway.com> wrote:
> >> Hi,
> >>
> >> Good point.
> >>
> >> I was only aware of  OpenStack Neat project .
> >>
> >> Are these talks related to this project ?
> >>
> >> BTW, is there a chance that RedHat OpenStack Platform  offers this
> feature ?
> >>
> >> Thx for help.
> >>
> >> Regards,
> >>
> >>
> >> J.P.
> >>
> >> -Original Message-
> >> From: Danny Abukalam [mailto:danny.abuka...@codethink.co.uk]
> >> Sent: jeudi 17 mars 2016 12:56
> >> To: Jean-Pierre Ribeauville
> >> Cc: openstack@lists.openstack.org
> >> Subject: Re: [Openstack] How to achieve OpenStack compute nodes "HA" ?
> >>
> >> Jean-Pierre,
> >>
> >> In case you're not already aware of it, just a quick pointer to a talk
> that will cover this exact topic at the summit:
> >>
> >> https://www.openstack.org/summit/austin-2016/summit-schedule/events/73
> >> 27
> >>
> >> Thanks,
> >>
> >> Danny
> >>
> >> On 16/03/16 16:13, Jean-Pierre Ribeauville wrote:
> >>> Hi,
> >>>
> >>> I'm quite a newbie  in Openstack HA stuff.
> >>>
> >>> My concern is the following :
> >>>
> >>> By using a two compute nodes infrastructure ( with shared iSCSi
> >>> storage) , I want to build an Openstack  environment  which offers
> >>> automatic Guest migration from a compute node to the second one.
> >>>
> >>> I presume that these two compute nodes must be part of a "cluster".
> >>>
> >>> I'm a little bit lost in all docs I 've found related to Openstack HA
> stuff.
> >>>
> >>> As I understood HA for two controller nodes , I don't see clearly the
> >>> equivalent for the compute nodes.
> >>>
> >>> Any hint  to clarify the infrastructure and softwares pieces  ( in
> >>> addition to  all 'classical"  OpenStack components)  I need ?
> >>>
> >>> Thanks  for help.
> >>>
> >>> Regards,
> >>>
> >>> Jean-Pierre RIBEAUVILLE
> >>>
> >>> +33 1 4717 2049
> >>>
> >>> axway_logo_tagline_87px
> >>>
> >>>
> >>>
> >>> ___
> >>> Mailing list:
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> >>> Post to : openstack@lists.openstack.org
> >>> Unsubscribe :
> >>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> >>>
> >>
> >> ___
> >> Mailing list:
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> >> Post to : openstack@lists.openstack.org
> >> Unsubscribe :
> >> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> >
> >
> >
> > --
> > Email:
> > shin...@linux.com
> > GitHub:
> > shinobu-x
> > Blog:
> > Life with Distributed Computational System based on OpenSource
> >
> > ___
> > Mailing list:
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> > Post to : openstack@lists.openstack.org
> > Unsubscribe :
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> >
> > ___
> > Mailing list:
> 

[openstack-dev] Mitaka, Xenial, OVS Firewall Driver, DPDK, VXLAN and Provider Networks

[Martinx - ジェームズ]

Hey guys!

 Next Ubuntu and Mitaka are promising something ultra mega cool!

 Look at this!

---
root@mitaka-1:~# apt install neutron-openvswitch-agent
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following additional packages will be installed:
  dpdk libdpdk0 openvswitch-common openvswitch-switch
---

 Xenial will brings DPDK-2.2 fully supported for 5 years!

 However, I am curious about the following scenarios:

 Will be possible to use, at the same time (same Network and Compute nodes
/ Host Aggregate):

 1- Regular OVS bridges without DPDK for VXLAN Networks, with
OVS-Firewall-Driver and;

 2- OVS powered by DPDK for Provider Networks only ( without any firewall,
current case anyway, due to https://bugs.launchpad.net/neutron/+bug/1531205
).

?

 I have NFV Instances that are also, DPDK L2 Bridges running on KVM Guest /
VirtIO, that are physically wired using Provider Networks (flat and vlans).

 So, for the Instance's vNICs (eth1 and eth2) that are used as a L2 bridge,
I don't want any kind of ovs-firewall (I'm not affected by LP #1531205 on
this case) and I want OVS+DPDK under it but, for SSH into the Instance to
manage it (via its eth0), it is still using regular VXLAN with Security
Groups - OVS-Firewall from now on (no need for DPDK under eth0 / VXLAN).

 I'm curious about this specially because the OVS Ubuntu package, makes use
of Debian's Alternatives subsystem, and we need to choose one OVS
(default), or another (with DPDK), via "update-alternatives", so, will be
possible to select OVS with DPDK but, use regular bridges with it as well
(for VXLAN networks)?

 If yes, how to create a VXLAN network with regular OVS and another
FLAT/VLAN network with OVS+DPDK ?

 Thanks in advance!

Best,
Thiago
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

[Openstack] Mitaka, Xenial, OVS Firewall Driver and DPDK

[Martinx - ジェームズ]

Hey guys!

 Next Ubuntu and Mitaka are promising something ultra mega cool!

 Look at this!

---
root@mitaka-1:~# apt install neutron-openvswitch-agent
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following additional packages will be installed:
  dpdk libdpdk0 openvswitch-common openvswitch-switch
---

 Xenial will brings DPDK-2.2 fully supported for 5 years!

 However, I am curious about the following scenarios:

 Will be possible to use, at the same time (same Network and Compute nodes
/ Host Aggregate):

 1- Regular OVS bridges without DPDK for VXLAN Networks, with
OVS-Firewall-Driver and;

 2- OVS powered by DPDK for Provider Networks only ( without any firewall,
current case anyway, due to https://bugs.launchpad.net/neutron/+bug/1531205
).

?

 I have NFV Instances that are also, DPDK L2 Bridges running on KVM Guest /
VirtIO, that are physically wired using Provider Networks (flat and vlans).

 So, for the Instance's vNICs (eth1 and eth2) that are used as a L2 bridge,
I don't want any kind of ovs-firewall (I'm not affected by LP #1531205 on
this case) and I want OVS+DPDK under it but, for SSH into the Instance to
manage it (via its eth0), it is still using regular VXLAN with Security
Groups - OVS-Firewall from now on (no need for DPDK under eth0 / VXLAN).

 I'm curious about this specially because the OVS Ubuntu package, makes use
of Debian's Alternatives subsystem, and we need to choose one OVS
(default), or another (with DPDK), via "update-alternatives", so, will be
possible to select OVS with DPDK but, use regular bridges with it as well
(for VXLAN networks)?

 If yes, how to create a VXLAN network with regular OVS and another
FLAT/VLAN network with OVS+DPDK ?

 Thanks in advance!

Best,
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Protest against Glance V2 usability, it is worse than V1! Bring back "--location to image-create"! Or am I missing something here?

[Martinx - ジェームズ]

BTW, you're right, OpenStack developers do a GREAT job, it is impressive.

I am just frustrated with new behavior of Glance V2... I still don't like
it, I prefer 1 command, instead of 3.

Again, sorry for choosing bad works, I love OpenStack, this community
always help me and I'm working with it for about 4~5 years, only good
things happening in my life because of this project!

^_^

On 20 February 2016 at 00:59, Martinx - ジェームズ <thiagocmarti...@gmail.com>
wrote:

> That would be great!
>
> Sorry for being rude, I'm gonna try to auto-generate the ID from now on...
>
> Thanks!
>
> On 20 February 2016 at 00:53, Mohammed Naser <mna...@vexxhost.com> wrote:
>
>> Hello,
>>
>> I think you’re being a bit extreme to just use the terms you’ve used
>> to describe the API.  A lot of people worked on this to make it ideal,
>> and while it’s made life a bit harder, it’s actually made the system
>> more robust.  You can fallback to the V1 api if you really can’t
>> retrieve the ID.
>>
>> http://www.gossamer-threads.com/lists/openstack/dev/50406
>>
>> This can be solved by one simple thing:
>>
>> glance image-create .. | fgrep ‘| id’ | awk ‘{ print $4 }'
>>
>>
>> If it’s really an issue, you can actually create a image with your own
>> UUID.
>>
>> Thanks,
>> Mohammed
>>
>> On Fri, Feb 19, 2016 at 9:17 PM, Martinx - ジェームズ
>> <thiagocmarti...@gmail.com> wrote:
>> > Guys,
>> >
>> >  I want to formalize my protest against Glance V2 usability.
>> >
>> >  With Glance V1, you can straight away, add an image based on its
>> remote URL
>> > location, then, Glance will download the image ONLY when someone launch
>> it.
>> >
>> >  This is very cool and I can add, easily and automatically by Ansible,
>> for
>> > example, dozens of images at once, right after deploying OpenStack.
>> >
>> >  However, this is much worse now, when with Glance V2, as follows:
>> >
>> > -
>> > *** Glance V1 procedure:
>> >
>> >  1) Add an image and its remote location at once:
>> >
>> >  glance image-create --location
>> >
>> http://cdimage.debian.org/cdimage/openstack/8.3.0/debian-8.3.0-openstack-amd64.qcow2
>> > --is-public true --name "Debian 8.3.0 - Jessie - 64-bit - Cloud Based
>> Image"
>> > --container-format bare --disk-format qcow2
>> >
>> >  See it? Glance V1 is awesome!
>> >
>> > --
>> >
>> > *** Glance V2 procedure:
>> >
>> >  1) Add an Image WITHOUT a location:
>> >
>> >  glance image-create --visibility public --name \"Debian 8.3.0 - Jessie
>> -
>> > 64-bit - Cloud Based Image\" --container-format bare --disk-format qcow2
>> >
>> >  2) Look (or parse somehow) the stdout, copy to clipboad the Image ID.
>> >
>> >  3) Add the location of the image:
>> >
>> >  glance location-add --url
>> >
>> http://cdimage.debian.org/cdimage/openstack/8.3.0/debian-8.3.0-openstack-amd64.qcow2
>> > $PASTE_IMAGE_ID
>> >
>> >  See it? Glance V2 sucks.
>> > -
>> >
>> >  It is more complicated now, specially to automate this with Ansible or
>> > other tool! Because there is a need to parser the stdout of step "1" and
>> > store the Image ID somewhere (like a variable), and then, add the remote
>> > location URL for the image.
>> >
>> >  Please guys, do something about this!
>> >
>> >  NOTE: I will be happy to be wrong here, let me know if I am missing
>> > something.
>> >
>> >  I believe that, if the "glance location-add" starts to accept "image
>> name"
>> > (instead of only ID), it will make the automation easier, because the
>> name
>> > is predictable, and the ID is random. Or, if someone tells me how to
>> > auto-generate random IDs for OpenStack, so it will be predictable (I'll
>> be
>> > able to use auto-genarated ID with "image-create", I believe), then, it
>> will
>> > also make the automation easier, I'll be okay with that too!
>> >
>> >  Thanks in advance!
>> >
>> > Regards,
>> > Thiago
>> >
>> > ___
>> > Mailing list:
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>> > Post to : openstack@lists.openstack.org
>> > Unsubscribe :
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>> >
>>
>
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Protest against Glance V2 usability, it is worse than V1! Bring back "--location to image-create"! Or am I missing something here?

[Martinx - ジェームズ]

That would be great!

Sorry for being rude, I'm gonna try to auto-generate the ID from now on...

Thanks!

On 20 February 2016 at 00:53, Mohammed Naser <mna...@vexxhost.com> wrote:

> Hello,
>
> I think you’re being a bit extreme to just use the terms you’ve used
> to describe the API.  A lot of people worked on this to make it ideal,
> and while it’s made life a bit harder, it’s actually made the system
> more robust.  You can fallback to the V1 api if you really can’t
> retrieve the ID.
>
> http://www.gossamer-threads.com/lists/openstack/dev/50406
>
> This can be solved by one simple thing:
>
> glance image-create .. | fgrep ‘| id’ | awk ‘{ print $4 }'
>
>
> If it’s really an issue, you can actually create a image with your own
> UUID.
>
> Thanks,
> Mohammed
>
> On Fri, Feb 19, 2016 at 9:17 PM, Martinx - ジェームズ
> <thiagocmarti...@gmail.com> wrote:
> > Guys,
> >
> >  I want to formalize my protest against Glance V2 usability.
> >
> >  With Glance V1, you can straight away, add an image based on its remote
> URL
> > location, then, Glance will download the image ONLY when someone launch
> it.
> >
> >  This is very cool and I can add, easily and automatically by Ansible,
> for
> > example, dozens of images at once, right after deploying OpenStack.
> >
> >  However, this is much worse now, when with Glance V2, as follows:
> >
> > -
> > *** Glance V1 procedure:
> >
> >  1) Add an image and its remote location at once:
> >
> >  glance image-create --location
> >
> http://cdimage.debian.org/cdimage/openstack/8.3.0/debian-8.3.0-openstack-amd64.qcow2
> > --is-public true --name "Debian 8.3.0 - Jessie - 64-bit - Cloud Based
> Image"
> > --container-format bare --disk-format qcow2
> >
> >  See it? Glance V1 is awesome!
> >
> > --
> >
> > *** Glance V2 procedure:
> >
> >  1) Add an Image WITHOUT a location:
> >
> >  glance image-create --visibility public --name \"Debian 8.3.0 - Jessie -
> > 64-bit - Cloud Based Image\" --container-format bare --disk-format qcow2
> >
> >  2) Look (or parse somehow) the stdout, copy to clipboad the Image ID.
> >
> >  3) Add the location of the image:
> >
> >  glance location-add --url
> >
> http://cdimage.debian.org/cdimage/openstack/8.3.0/debian-8.3.0-openstack-amd64.qcow2
> > $PASTE_IMAGE_ID
> >
> >  See it? Glance V2 sucks.
> > -
> >
> >  It is more complicated now, specially to automate this with Ansible or
> > other tool! Because there is a need to parser the stdout of step "1" and
> > store the Image ID somewhere (like a variable), and then, add the remote
> > location URL for the image.
> >
> >  Please guys, do something about this!
> >
> >  NOTE: I will be happy to be wrong here, let me know if I am missing
> > something.
> >
> >  I believe that, if the "glance location-add" starts to accept "image
> name"
> > (instead of only ID), it will make the automation easier, because the
> name
> > is predictable, and the ID is random. Or, if someone tells me how to
> > auto-generate random IDs for OpenStack, so it will be predictable (I'll
> be
> > able to use auto-genarated ID with "image-create", I believe), then, it
> will
> > also make the automation easier, I'll be okay with that too!
> >
> >  Thanks in advance!
> >
> > Regards,
> > Thiago
> >
> > ___
> > Mailing list:
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> > Post to : openstack@lists.openstack.org
> > Unsubscribe :
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> >
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Protest against Glance V2 usability, it is worse than V1! Bring back "--location to image-create"! Or am I missing something here?

[Martinx - ジェームズ]

Guys,

 I want to formalize my protest against Glance V2 usability.

 With Glance V1, you can straight away, add an image based on its remote
URL location, then, Glance will download the image ONLY when someone launch
it.

 This is very cool and I can add, easily and automatically by Ansible, for
example, dozens of images at once, right after deploying OpenStack.

 However, this is much worse now, when with Glance V2, as follows:

-
*** Glance V1 procedure:

 1) Add an image and its remote location at once:

 glance image-create --location
http://cdimage.debian.org/cdimage/openstack/8.3.0/debian-8.3.0-openstack-amd64.qcow2
--is-public true --name "Debian 8.3.0 - Jessie - 64-bit - Cloud Based
Image" --container-format bare --disk-format qcow2

 See it? Glance V1 is awesome!

--

*** Glance V2 procedure:

 1) Add an Image WITHOUT a location:

 glance image-create --visibility public --name \"Debian 8.3.0 - Jessie -
64-bit - Cloud Based Image\" --container-format bare --disk-format qcow2

 2) Look (or parse somehow) the stdout, copy to clipboad the Image ID.

 3) Add the location of the image:

 glance location-add --url
http://cdimage.debian.org/cdimage/openstack/8.3.0/debian-8.3.0-openstack-amd64.qcow2
$PASTE_IMAGE_ID

 See it? Glance V2 sucks.
-

 It is more complicated now, specially to automate this with Ansible or
other tool! Because there is a need to parser the stdout of step "1" and
store the Image ID somewhere (like a variable), and then, add the remote
location URL for the image.

 Please guys, do something about this!

 NOTE: I will be happy to be wrong here, let me know if I am missing
something.

 I believe that, if the "glance location-add" starts to accept "image name"
(instead of only ID), it will make the automation easier, because the name
is predictable, and the ID is random. Or, if someone tells me how to
auto-generate random IDs for OpenStack, so it will be predictable (I'll be
able to use auto-genarated ID with "image-create", I believe), then, it
will also make the automation easier, I'll be okay with that too!

 Thanks in advance!

Regards,
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Howto remove "/horizon" from URL?

[Martinx - ジェームズ]

YAY!! It worked! Thank you guys!   :-D

On 8 February 2016 at 05:51, Erdősi Péter  wrote:

> Hy there!
> 2016. 02. 07. 22:33 keltezéssel, Mohammed Naser írta:
>
>> python manage.py collectstatic
>> python manage.py compress
>>
> Thanks mate! :)
> (maybe my question will be a bit off for the thread, but I think, they
> related)
>
> So, I did not run this now, but i will... can you maybe tell me, which
> directory i should be when I run that 2 commands?
> The google says, the py file will be there:
> /usr/share/openstack-dashboard/manage.py
> Is that okay, if I just run in a /tmp for example?
>
> Thanks,
>  Peter
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [openstack-dev] Questions regarding image "location" and glanceclient behaviour ...

[Martinx - ジェームズ]

>
> > But this procedure will force me to download all images in advance,
> which I
> > can not do.
> >
> > I NEED the previous behavior, where Glance download the images by itself,
> > on demand.
> >
> > How to do this with V2 ?
>
> You can use glance image-create without passing it any image data. You
> should then be able to use location-add and I believe Glance will download
> the image itself in that case.
>
> Cheers,
> --
> Ian Cordasco
>

That is good news! I am going to try it right now!

Thank you!

Cheers!
Thiago
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] Questions regarding image "location" and glanceclient behaviour ...

[Martinx - ジェームズ]

But this procedure will force me to download all images in advance, which I
can not do.

I NEED the previous behavior, where Glance download the images by itself,
on demand.

How to do this with V2 ?

On 18 February 2016 at 01:47, Fei Long Wang <feil...@catalyst.net.nz> wrote:

> Glance v2 doesn't support 'location' anymore, now there are multi
> locations for image in V2. You can use 'glance location-add' after create
> the the image by 'glance image-create --file xxx'
>
>
> On 18/02/16 15:51, Martinx - ジェームズ wrote:
>
> Hey guys, any news about this?
>
> I want to use Glance v2 but, without --location that points to a URL and,
> for me, without it, it is impossible to use it (v2).
>
> So, any plans to bring back --location, I want to use v2 like this:
>
> --
> glance image-create --location
> http://uec-images.ubuntu.com/releases/14.04.3/release/ubuntu-14.04-server-cloudimg-amd64-disk1.img
> --name "Ubuntu 14.04.3 LTS - Trusty Tahr - 64-bit - Cloud Based Image"
> --visibility public --container-format bare --disk-format qcow2
> --
>
> But, "--location" isn't recognized anymore!
>
> Thanks!
> Thiago
>
>
> On 22 January 2014 at 14:30, Mark Washenberger <
> mark.washenber...@markwash.net> wrote:
>
>>
>>
>>
>> On Wed, Jan 22, 2014 at 1:05 AM, Public Mail < <kpublicm...@gmail.com>
>> kpublicm...@gmail.com> wrote:
>>
>>> Hi All,
>>>
>>> I have two questions ...
>>>
>>> 1) Glance v1 APIs can take a --location argument when creating an
>>> image
>>>but v2 APIs can't - bug or feature? (Details below)
>>>
>>
>> I'd call that a missing feature. I think we probably need a glance
>> image-location-add command somewhere in the client. But fair warning, this
>> is typically a role-restricted operation.
>>
>>
>>>
>>> 2) How should glanceclient (v2 commands) handle reserved attributes?
>>> a) status quo: (Apparently) let the user set them but the server
>>>will return "attribute is reserved" error.  Pros: No missing
>>>functionality, no damage done.  Cons: Bad usability.
>>> b) hard-code list of reserved attributes in client and don't
>>> expose
>>>them to the user.
>>> Pros: quick to implement.
>>> Cons: Need to track reserved attributes in server
>>> implementation.
>>> c) get-reserved words from schema downloaded from server (and
>>> don't
>>>expose them to the user).
>>> Pros: Don't need to track server implmentation.
>>> Cons: Complex - reserved words can vary from command to
>>> command.
>>>
>>>   I personally favor (b) on the grounds that a client implementation
>>>   needs to closely understand server behaviour anyway so the sync-ing
>>>   of reserved attributes shouldn't be a big problem (*provided* the
>>>   list of reserved attributes is made available in the reference
>>>   documentation which doesn't seem to be the case currently).
>>>
>>
>>
>> We are in a bit of a bind with schemas--what's needed is schema resources
>> to represent each request and response, not just each resource. Because,
>> obviously, the things you can PATCH and POST are necessarily different than
>> the things you can GET in any service api. However, it is not clear to me
>> how we get from one schema per resource to one schema per request and
>> response in a backwards compatible way. So b) might be the only way to go.
>>
>>
>>
>>>
>>> So what does everybody think?
>>>
>>> 
>>> When using glance client's v1 interface I can image-create an image and
>>> specify the image file's location via the --location parameter.
>>> Alternatively I can image-create an empty image and then image-update the
>>> image's location to some url.
>>>
>>> However, when using the client's v2 commands I can neither image-create
>>> the
>>> file using the --location parameter, nor image-update the file later.
>>>
>>> When using image-create with --location, the client gives the following
>>> error (printed by warlock):
>>>
>>>   Unable to set 'locations' to '[u' <http://192.168.1.111/foo/bar%27>
>>> http://192.168.1.111/foo/bar']'
>>>
>>> This is because the schema dictates that the location should be an 

Re: [openstack-dev] Questions regarding image "location" and glanceclient behaviour ...

[Martinx - ジェームズ]

Hey guys, any news about this?

I want to use Glance v2 but, without --location that points to a URL and,
for me, without it, it is impossible to use it (v2).

So, any plans to bring back --location, I want to use v2 like this:

--
glance image-create --location
http://uec-images.ubuntu.com/releases/14.04.3/release/ubuntu-14.04-server-cloudimg-amd64-disk1.img
--name "Ubuntu 14.04.3 LTS - Trusty Tahr - 64-bit - Cloud Based Image"
--visibility public --container-format bare --disk-format qcow2
--

But, "--location" isn't recognized anymore!

Thanks!
Thiago


On 22 January 2014 at 14:30, Mark Washenberger <
mark.washenber...@markwash.net> wrote:

>
>
>
> On Wed, Jan 22, 2014 at 1:05 AM, Public Mail 
> wrote:
>
>> Hi All,
>>
>> I have two questions ...
>>
>> 1) Glance v1 APIs can take a --location argument when creating an
>> image
>>but v2 APIs can't - bug or feature? (Details below)
>>
>
> I'd call that a missing feature. I think we probably need a glance
> image-location-add command somewhere in the client. But fair warning, this
> is typically a role-restricted operation.
>
>
>>
>> 2) How should glanceclient (v2 commands) handle reserved attributes?
>> a) status quo: (Apparently) let the user set them but the server
>>will return "attribute is reserved" error.  Pros: No missing
>>functionality, no damage done.  Cons: Bad usability.
>> b) hard-code list of reserved attributes in client and don't
>> expose
>>them to the user.
>> Pros: quick to implement.
>> Cons: Need to track reserved attributes in server
>> implementation.
>> c) get-reserved words from schema downloaded from server (and
>> don't
>>expose them to the user).
>> Pros: Don't need to track server implmentation.
>> Cons: Complex - reserved words can vary from command to
>> command.
>>
>>   I personally favor (b) on the grounds that a client implementation
>>   needs to closely understand server behaviour anyway so the sync-ing
>>   of reserved attributes shouldn't be a big problem (*provided* the
>>   list of reserved attributes is made available in the reference
>>   documentation which doesn't seem to be the case currently).
>>
>
>
> We are in a bit of a bind with schemas--what's needed is schema resources
> to represent each request and response, not just each resource. Because,
> obviously, the things you can PATCH and POST are necessarily different than
> the things you can GET in any service api. However, it is not clear to me
> how we get from one schema per resource to one schema per request and
> response in a backwards compatible way. So b) might be the only way to go.
>
>
>
>>
>> So what does everybody think?
>>
>> 
>> When using glance client's v1 interface I can image-create an image and
>> specify the image file's location via the --location parameter.
>> Alternatively I can image-create an empty image and then image-update the
>> image's location to some url.
>>
>> However, when using the client's v2 commands I can neither image-create
>> the
>> file using the --location parameter, nor image-update the file later.
>>
>> When using image-create with --location, the client gives the following
>> error (printed by warlock):
>>
>>   Unable to set 'locations' to '[u'http://192.168.1.111/foo/bar']'
>>
>> This is because the schema dictates that the location should be an object
>> of the form [{"url": "string", "metadata": object}, ...] but there is no
>> way to specify such an object from the command line - I cannot specify a
>> string like '{"url": "192.168.1.111/foo/bar", "metadata": {}}' for there
>> is
>> no conversion from command line strings to python dicts nor is there any
>> conversion from a simple URL string to a suitable location object.
>>
>> If I modify glanceclient.v2.images.Controller.create to convert the
>> locations parameter from a URL string to the desired object then the
>> request goes through to the glance server where it fails with a 403 error
>> (Attribute 'locations' is reserved).
>>
>> So is this discrepancy between V1 & V2 deliberate (a feature :)) or is it
>> a
>> bug?
>> 
>>
>> ___
>> OpenStack-dev mailing list
>> OpenStack-dev@lists.openstack.org
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>>
>
>
> ___
> OpenStack-dev mailing list
> OpenStack-dev@lists.openstack.org
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>
>
__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [Openstack] Virtual Firewall Appliance

[Martinx - ジェームズ]

I don't think that you'll be able to do that in IceHouse, neither on Juno.

Only Kilo and Liberty have a native function to disable the port_security
per port. Without it, OpenStack Neutron (and also Nova Network, I guess)
will not allow the firewall Instance to work correctly. It will not see any
packets that are not destined to it and also, it will not be able to
forward packets, because the Neutron (and Nova Network), will drop the
packets soon as it leaves the firewall Instance.

I'm not aware of a solution nice for IceHouse...

On 16 February 2016 at 06:26, Georgios Dimitrakakis 
wrote:

> Mark and Martinx thank you both for your suggestions.
>
> I had tried to build PFSense in the past but without success.
>
> Indeed my goal is to run the virtual firewall as an instance since I am on
> an older OpenStack version (IceHouse) with nova-networking and therefore I
> cannot have control over the outgoing connections.
>
> Regards,
>
> G.
>
>
> For running it as an Instance?
>>
>> You can try:
>>
>> - PFSense;
>>
>> - Zentyal;
>>
>> However, youll need to make use of the Neutron feature called
>> "port_security_enabled = false" for the vNIC attached to the
>> "internal" subnet (behind the firewall).
>>
>> Just a curiosity, why dont you use the Neutron native firewall that
>> resides on each L3 Router?
>>
>> On 15 February 2016 at 15:56, Georgios Dimitrakakis  wrote:
>>
>> Hi!
>>>
>>> Can anyone suggest me of a virtual firewall appliance which is
>>> compatible with OpenStack?
>>>
>>> Best regards,
>>>
>>> G.
>>>
>>> ___
>>> Mailing list:
>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack [1]
>>> Post to : openstack@lists.openstack.org [2]
>>> Unsubscribe :
>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack [3]
>>>
>>
>>
>>
>> Links:
>> --
>> [1] http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>> [2] mailto:openstack@lists.openstack.org
>> [3] http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>> [4] mailto:gior...@acmac.uoc.gr
>>
>
> ___
> Mailing list:
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> Post to : openstack@lists.openstack.org
> Unsubscribe :
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] Virtual Firewall Appliance

[Martinx - ジェームズ]

For running it as an Instance?

You can try:

- PFSense;

- Zentyal;

However, you'll need to make use of the Neutron feature called
"port_security_enabled = false" for the vNIC attached to the "internal"
subnet (behind the firewall).

Just a curiosity, why don't you use the Neutron native firewall that
resides on each L3 Router?

On 15 February 2016 at 15:56, Georgios Dimitrakakis 
wrote:

> Hi!
>
> Can anyone suggest me of a virtual firewall appliance which is compatible
> with OpenStack?
>
> Best regards,
>
> G.
>
>
> ___
> Mailing list:
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
> Post to : openstack@lists.openstack.org
> Unsubscribe :
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] Howto remove "/horizon" from URL?

[Martinx - ジェームズ]

Guys,

 I'm trying to remove the "/horizon" from my Ubuntu deployment but, it does
not work, it loses the style.

 I'm editing the file:

 "/etc/apache2/conf-enabled/openstack-dashboard.conf"

 and changing:

 "WSGIScriptAlias /horizon
/usr/share/openstack-dashboard/openstack_dashboard/wsgi/django.wsgi"

 to:

 "WSGIScriptAlias /
/usr/share/openstack-dashboard/openstack_dashboard/wsgi/django.wsgi"

 and the file:

 "/etc/openstack-dashboard/local_settings.py"

 And configuring:

 WEBROOT='/'

 But, it does not work! Login screen loses it style (CSS, I think)...

 I managed to do this on Juno but, I am unable to this on Kilo, Liberty
neither or Mitaka.

 What am I missing?

 Also, I see the "WEBROOT" entry is duplicated on Ubuntu deployments! Is
this expected?

Thanks!
Thiago
___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [openstack-dev] Any chances to backport port_security_enabled support into Heat for Kilo 2015.1.3?

[Martinx - ジェームズ]

On 31 January 2016 at 18:23, Steve Baker <sba...@redhat.com> wrote:
> On 29/01/16 09:45, Martinx - ジェームズ wrote:
>>
>> Guys,
>>
>>   This is important and Kilo is missing it:
>>
>>   https://review.openstack.org/#/c/179989/
>>
>>   Is it possible to backport it to Kilo 2015.1.3?
>>
>>   Currently, I am manually patching Kilo / Heat by using the following
>> diff:
>>
>>
>> https://review.openstack.org/gitweb?p=openstack%2Fheat.git;a=commitdiff;h=811c8714aa2442e68980561d3e8dd435378f164c
>>
>>   But it is a pain to maintain...
>>
>>
> Rather than carrying a backport you can always modify your templates to set
> port_security_enabled via the value_specs property:
>
>   value_specs: {port_security_enabled: false}

WoW! That is cool! I'm gonna try it now! Thank you!

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] Any chances to backport port_security_enabled support into Heat for Kilo 2015.1.3?

[Martinx - ジェームズ]

Oh, that's okay... Thanks guys!:)

On 29 January 2016 at 06:39, Sergey Kraynev  wrote:
> Sean, thank you for the spotting.
>
> Martinx, According to the information mentioned by Sean, we unfortunately
> can not do it :(
>
> On 29 January 2016 at 10:45, Sean M. Collins  wrote:
>>
>> Kilo is in the "security supported" stage of the lifecycle. So no,
>> that's not going to get backported.
>>
>> http://docs.openstack.org/releases/index.html
>> --
>> Sean M. Collins
>>
>> __
>> OpenStack Development Mailing List (not for usage questions)
>> Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>
>
>
>
> --
> Regards,
> Sergey.
>
> __
> OpenStack Development Mailing List (not for usage questions)
> Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev
>

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

[openstack-dev] Any chances to backport port_security_enabled support into Heat for Kilo 2015.1.3?

[Martinx - ジェームズ]

Guys,

 This is important and Kilo is missing it:

 https://review.openstack.org/#/c/179989/

 Is it possible to backport it to Kilo 2015.1.3?

 Currently, I am manually patching Kilo / Heat by using the following diff:

 
https://review.openstack.org/gitweb?p=openstack%2Fheat.git;a=commitdiff;h=811c8714aa2442e68980561d3e8dd435378f164c

 But it is a pain to maintain...

Thanks!
Thiago

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] The command "neutron-db-manage" of 8.0.0~b1 fails

[Martinx - ジェームズ]

On 4 January 2016 at 12:20, Ihar Hrachyshka <ihrac...@redhat.com> wrote:
> Mike Bayer <mba...@redhat.com> wrote:
>
>>
>>
>> On 01/04/2016 06:59 AM, Ihar Hrachyshka wrote:
>>>
>>> Martinx - ジェームズ <thiagocmarti...@gmail.com> wrote:
>>>
>>>> Guys,
>>>>
>>>>  I'm trying to experiment Mitaka on Ubuntu Xenial, which already have
>>>> beta version on its repositories, however, "neutron-db-manage" fails.
>>>>
>>>>  Here is the output of it:
>>>>
>>>>  http://paste.openstack.org/show/482920/
>>>>
>>>>  Any clue?
>>>>
>>>>  I'm using the Kilo instructions as a start point, of course, I'm
>>>> using new neutron.conf and new ml2_conf.ini as well.
>>>>
>>>> Thanks in advance!
>>>> Thiago
>>>
>>>
>>> I believe it was fixed in:
>>>
>>> https://review.openstack.org/#/c/253150/2/neutron/db/migration/alembic_migrations/versions/mitaka/contract/8a6d8bdae39_migrate_neutron_resources_table.py
>>
>>
>> doh and I'm the one who fixed it!
>
>
> Busy man you are.
>
>
> Ihar

LOL

AWESOME!

Thank you guys!

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [openstack-dev] [networking-ovs-dpdk] availability

[Martinx - ジェームズ]

On 3 January 2016 at 23:41, Palanisamy, Prabhakaran (Contractor)
 wrote:
> Hi,
>
>
>
> Is networking-ovs-dpdk package available non devstack installation ??
>
>
>
> Thanks,
>
> PP

Hi,

Don't know if it will help you but, Ubuntu Xenial (development
branch), have interesting packages:

root@xenial-1:~# apt-cache search ovs dpdk
openvswitch-switch-dpdk - DPDK enabled Open vSwitch switch implementation
python-networking-ovs-dpdk - OpenStack virtual network service - Open
vSwitch DPDK ML2 mechanism driver

Also, Xenial will support DPDK 2.2 natively, for at least, 5 years.

I'm trying to experiment Mitaka beta release on Xenial these days...

Best,
Thiago

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

[openstack-dev] The command "neutron-db-manage" of 8.0.0~b1 fails

[Martinx - ジェームズ]

Guys,

 I'm trying to experiment Mitaka on Ubuntu Xenial, which already have
beta version on its repositories, however, "neutron-db-manage" fails.

 Here is the output of it:

 http://paste.openstack.org/show/482920/

 Any clue?

 I'm using the Kilo instructions as a start point, of course, I'm
using new neutron.conf and new ml2_conf.ini as well.

Thanks in advance!
Thiago

__
OpenStack Development Mailing List (not for usage questions)
Unsubscribe: openstack-dev-requ...@lists.openstack.org?subject:unsubscribe
http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

Re: [Openstack] CPU Allocation Ratio / OverCommit, not working, need help...

[Martinx - ジェームズ]

Hey man!

Thanks for your reply... It was really helpful!

It is working now! It was disk_allocation_ratio problem...  lol

Thanks again! Debug is always good...   =)

Best,
Thiago

On 22 October 2015 at 08:40, Vladimir Eremin <vere...@mirantis.com> wrote:
> Hi Martinx,
>
> You could check what’s going wrong if you set debug to True in your nova.con 
> on controller, then restart nova-scheduler. Scheduler’s log contains lot of 
> debug, so you could easily figure out the problem.
>
> I bet it because of RAM by the way.
>
> --
> With best regards,
> Vladimir Eremin,
> Fuel Deployment Engineer,
> Mirantis, Inc.
>
>
>
>> On 22 Oct 2015, at 02:33, Martinx - ジェームズ <thiagocmarti...@gmail.com> wrote:
>>
>> Guys,
>>
>> I deployed Liberty on Trusty, on a 2Core physical box, with HT
>> enabled, so, I can see 4 Cores.
>>
>> However, I'm trying to launch 3 Instances (m1.medium, 2 vCPUs each),
>> totalizing 6 vCPUs but, it doesn't work.
>>
>> I have enough RAM and enough disk...
>>
>> Also, I tried to increase the cpu_allocation_ratio to 100.0, didn't
>> worked either...
>>
>> My nova.conf contains:
>>
>> https://github.com/sandvine/os-ansible-deployment-lite/blob/liberty/ansible/roles/nova_aio/templates/nova.conf
>>
>> What am I missing? Isn't it 16:1 by default?
>>
>> Thanks!
>> Thiago
>>
>> ___
>> Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>> Post to : openstack@lists.openstack.org
>> Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] [Liberty][Trusty] AppArmor interferes with instance launching...

[Martinx - ジェームズ]

I don't think that this problem is related to AppArmor.

It is happening again but, it is random, on a host without AppArmor installed...

The logs shows:


nova-compute.log:

http://paste.openstack.org/show/477033/


nova-conductor.log:

http://paste.openstack.org/show/477036/

libvirtd.log:

---
2015-10-21 15:07:14.578+: 2805: error : virNetDevTapCreate:290 :
Unable to create tap device tapab04671d-75: Device or resource busy
---

I'm trying to launch the following Stack:

https://github.com/sandvine/os-ansible-deployment-lite/blob/liberty/misc/os-heat-templates/sandvine-stack-0.1-cs-1.yaml

It failed (CREATE_FAILED), then, I tried it again, and it works
(CREATE_COMPLETE)! :-/

This never happened with Kilo (very same Heat Template)...

It seems that, the first time that of launch, after uploading a qcow2,
it fails, then, the second time, it works...

Am I missing something here?

Thanks!
Thiago

On 20 October 2015 at 10:04, Mohammed Naser <mna...@vexxhost.com> wrote:
> What do you see in /var/log/syslog regarding apparmor?
>
> On Mon, Oct 19, 2015 at 10:12 PM, Martinx - ジェームズ
> <thiagocmarti...@gmail.com> wrote:
>> Guys,
>>
>> Finally my Liberty setup is working!
>>
>> But, I removed the apparmor and libapparmor-perl packages out from
>> Ubuntu, otherwise, it is not possible to launch any instance, the
>> following error(s) appear:
>>
>> libvirtd.log:
>> ---
>> 2015-10-20 00:58:35.393+: 12689: error : virNetDevTapCreate:290 :
>> Unable to create tap device tap3805bb26-ef: Device or resource busy
>> ---
>>
>> nova-compute.log:
>>
>> ---
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager
>> [req-3a40fb40-ab82-4446-9e2d-641e451f3c85
>> b0e9378fbc1a48328e29f28e9109c25a d896204e6b6040da88039585bedb0824 - -
>> -] [instance: e305b8bf-2d74-4819-bc92-3eb1bd59db50] Instance failed to
>> spawn
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-2d74-4819-bc92-3eb1bd59db50] Traceback (most recent call
>> last):
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-2d74-4819-bc92-3eb1bd59db50]   File
>> "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 2155,
>> in _build_resources
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-2d74-4819-bc92-3eb1bd59db50] yield resources
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-2d74-4819-bc92-3eb1bd59db50]   File
>> "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 2009,
>> in _build_and_run_instance
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-2d74-4819-bc92-3eb1bd59db50]
>> block_device_info=block_device_info)
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-2d74-4819-bc92-3eb1bd59db50]   File
>> "/usr/lib/python2.7/dist-packages/nova/virt/libvirt/driver.py", line
>> 2444, in spawn
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-2d74-4819-bc92-3eb1bd59db50]
>> block_device_info=block_device_info)
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-2d74-4819-bc92-3eb1bd59db50]   File
>> "/usr/lib/python2.7/dist-packages/nova/virt/libvirt/driver.py", line
>> 4519, in _create_domain_and_network
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-2d74-4819-bc92-3eb1bd59db50] xml, pause=pause,
>> power_on=power_on)
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-2d74-4819-bc92-3eb1bd59db50]   File
>> "/usr/lib/python2.7/dist-packages/nova/virt/libvirt/driver.py", line
>> 4449, in _create_domain
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-2d74-4819-bc92-3eb1bd59db50] guest.launch(pause=pause)
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-2d74-4819-bc92-3eb1bd59db50]   File
>> "/usr/lib/python2.7/dist-packages/nova/virt/libvirt/guest.py", line
>> 141, in launch
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-2d74-4819-bc92-3eb1bd59db50] self._encoded_xml,
>> errors='ignore')
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-2d74-4819-bc92-3eb1bd59db50]   File
>> "/usr/lib/python2.7/dist-packages/oslo_utils/excutils.py", line 195,
>> in __exit__
>> 2015-10-19 20:58:35.476 13685 ERROR nova.compute.manager [instance:
>> e305b8bf-

[Openstack] CPU Allocation Ratio / OverCommit, not working, need help...

[Martinx - ジェームズ]

Guys,

I deployed Liberty on Trusty, on a 2Core physical box, with HT
enabled, so, I can see 4 Cores.

However, I'm trying to launch 3 Instances (m1.medium, 2 vCPUs each),
totalizing 6 vCPUs but, it doesn't work.

I have enough RAM and enough disk...

Also, I tried to increase the cpu_allocation_ratio to 100.0, didn't
worked either...

My nova.conf contains:

https://github.com/sandvine/os-ansible-deployment-lite/blob/liberty/ansible/roles/nova_aio/templates/nova.conf

What am I missing? Isn't it 16:1 by default?

Thanks!
Thiago

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] New Glance is worse than previous version... Specially API 2... Why?!

[Martinx - ジェームズ]

Am I experimenting a problem that was introduced by this fix:

https://review.openstack.org/#/c/136520/

???

On 20 October 2015 at 04:13, Martinx - ジェームズ <thiagocmarti...@gmail.com> wrote:
> Guys,
>
>  I have an Ansible automation to deploy OpenStack Kilo, it works like
> a charm, specially the following playbook:
>
>
> https://github.com/sandvine/os-ansible-deployment-lite/blob/kilo/ansible/roles/glance/tasks/default-images.yml
>
>
>  In one shot, I can add remote images (Glance downloads on demand) and
> update its properties immediatly... But, new Glance lost all
> functionality that make this to work.
>
>  To begin with, OS_IMAGE_API version 2, doesn't have "--location"
> anymore... Why not? It is VERY useful. Please, bring it back. Or at
> least, document how to do the same when using API 2...
>
>  Also, new (API 2) "glance image-update" REQUIRES the Image ID, this
> sucks because it is much easier to use the Image Name instead...
>
>  So, I updated my Ansible playbook to deploy Liberty... But Glance
> doesn't work the way it should work...
>
>  For example, here...:
>
>
>  
> https://github.com/sandvine/os-ansible-deployment-lite/blob/liberty/ansible/roles/glance/tasks/default-images.yml
>
>
> ...I had to disable all "glance image-update" calls, because it simple
> doesn't work anymore, even using OS_IMAGE_API=1 !!!
>
>
>  Look what happens now:
>
> 
> changed: [localhost] => (item=glance image-create --location
> http://download.cirros-cloud.net/0.3.4/cirros-0.3.4-i386-disk.img
> --name "CirrOS 0.3.4 - Minimalist - 32-bit - Cloud Based Image"
> --is-public true --container-format bare --disk-format qcow2)
> 
> failed: [localhost] => (item=glance image-update --property
> hw_scsi_model=virtio-scsi --property hw_disk_bus=scsi "CirrOS 0.3.4 -
> Minimalist - 32-bit - Cloud Based Image") => {"changed": true, "cmd":
> ["glance", "image-update", "--property", "hw_scsi_model=virtio-scsi",
> "--property", "hw_disk_bus=scsi", "CirrOS 0.3.4 - Minimalist - 32-bit
> - Cloud Based Image"], "delta": "0:00:00.898007", "end": "2015-10-20
> 01:10:03.837781", "item": "glance image-update --property
> hw_scsi_model=virtio-scsi --property hw_disk_bus=scsi \"CirrOS 0.3.4 -
> Minimalist - 32-bit - Cloud Based Image\"", "rc": 1, "start":
> "2015-10-20 01:10:02.939774", "warnings": []}
> stderr: Unable to upload image data to an image which is active.
> 
>
> It is even more weird, because those two commands, that doesn't work
> when with Ansible, works manually... For example:
>
> ---
> #! /bin/bash
> source ~/admin-openrc.sh
> glance image-create --location
> http://uec-images.ubuntu.com/releases/14.04.3/release/ubuntu-14.04-server-cloudimg-amd64-disk1.img
> --is-public true --disk-format qcow2 --container-format bare --name
> "Ubuntu 14.04.3 LTS - Trusty Tahr - 64-bit - Cloud Based Image"
> glance image-update --property hw_scsi_model=virtio-scsi --property
> hw_disk_bus=scsi "Ubuntu 14.04.3 LTS - Trusty Tahr - 64-bit - Cloud
> Based Image"
> ---
>
> Works! But only outside of Ansible...
>
> I triple checked everything (credentials that Ansible uses
> (ansible/group_vars/all), with the one that I source under Bash)...
> Maybe I am missing some typo again...=)
>
> I appreciate any help!
>
> Thanks in advance!
>
> Best,
> Thiago

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

[Openstack] New Glance is worse than previous version... Specially API 2... Why?!

[Martinx - ジェームズ]

Guys,

 I have an Ansible automation to deploy OpenStack Kilo, it works like
a charm, specially the following playbook:


https://github.com/sandvine/os-ansible-deployment-lite/blob/kilo/ansible/roles/glance/tasks/default-images.yml


 In one shot, I can add remote images (Glance downloads on demand) and
update its properties immediatly... But, new Glance lost all
functionality that make this to work.

 To begin with, OS_IMAGE_API version 2, doesn't have "--location"
anymore... Why not? It is VERY useful. Please, bring it back. Or at
least, document how to do the same when using API 2...

 Also, new (API 2) "glance image-update" REQUIRES the Image ID, this
sucks because it is much easier to use the Image Name instead...

 So, I updated my Ansible playbook to deploy Liberty... But Glance
doesn't work the way it should work...

 For example, here...:


 
https://github.com/sandvine/os-ansible-deployment-lite/blob/liberty/ansible/roles/glance/tasks/default-images.yml


...I had to disable all "glance image-update" calls, because it simple
doesn't work anymore, even using OS_IMAGE_API=1 !!!


 Look what happens now:


changed: [localhost] => (item=glance image-create --location
http://download.cirros-cloud.net/0.3.4/cirros-0.3.4-i386-disk.img
--name "CirrOS 0.3.4 - Minimalist - 32-bit - Cloud Based Image"
--is-public true --container-format bare --disk-format qcow2)

failed: [localhost] => (item=glance image-update --property
hw_scsi_model=virtio-scsi --property hw_disk_bus=scsi "CirrOS 0.3.4 -
Minimalist - 32-bit - Cloud Based Image") => {"changed": true, "cmd":
["glance", "image-update", "--property", "hw_scsi_model=virtio-scsi",
"--property", "hw_disk_bus=scsi", "CirrOS 0.3.4 - Minimalist - 32-bit
- Cloud Based Image"], "delta": "0:00:00.898007", "end": "2015-10-20
01:10:03.837781", "item": "glance image-update --property
hw_scsi_model=virtio-scsi --property hw_disk_bus=scsi \"CirrOS 0.3.4 -
Minimalist - 32-bit - Cloud Based Image\"", "rc": 1, "start":
"2015-10-20 01:10:02.939774", "warnings": []}
stderr: Unable to upload image data to an image which is active.


It is even more weird, because those two commands, that doesn't work
when with Ansible, works manually... For example:

---
#! /bin/bash
source ~/admin-openrc.sh
glance image-create --location
http://uec-images.ubuntu.com/releases/14.04.3/release/ubuntu-14.04-server-cloudimg-amd64-disk1.img
--is-public true --disk-format qcow2 --container-format bare --name
"Ubuntu 14.04.3 LTS - Trusty Tahr - 64-bit - Cloud Based Image"
glance image-update --property hw_scsi_model=virtio-scsi --property
hw_disk_bus=scsi "Ubuntu 14.04.3 LTS - Trusty Tahr - 64-bit - Cloud
Based Image"
---

Works! But only outside of Ansible...

I triple checked everything (credentials that Ansible uses
(ansible/group_vars/all), with the one that I source under Bash)...
Maybe I am missing some typo again...=)

I appreciate any help!

Thanks in advance!

Best,
Thiago

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] New Glance is worse than previous version... Specially API 2... Why?!

[Martinx - ジェームズ]

Thanks man! I'll take a look into this...  :-D

On 20 October 2015 at 09:58, Mohammed Naser <mna...@vexxhost.com> wrote:
> It actually is possible (and even possible to have multiple locations
> with the v2 API).
>
> http://www.sebastien-han.fr/blog/2015/05/13/openstack-glance-use-multiple-location-for-an-image/
>
> Regarding image-update with name, this is something that happens at
> the client side, I suspect you're not getting the exact correct image
> name.
>
> "Unable to upload image data to an image which is active."
>
> We do something like this...
>
> $ glance image-create --architecture x86_64 --protected True --name
> "Ubuntu 12.04.5 LTS" --visibility public --tags ubuntu precise latest
> --os-version 12.04.5 --disk-format qcow2 --os-distro ubuntu
> --container-format bare --property libvirt_use_agent=True
> 
> $ glance location-add --url http:// 
>
> Note that I strongly suggest you do not add/remove locations once an
> image is used, because you might have cache issues (image would be
> cached on a compute host, so deploying after updating location might
> not result in a re-download).
>
> On Tue, Oct 20, 2015 at 2:24 AM, Martinx - ジェームズ
> <thiagocmarti...@gmail.com> wrote:
>> Am I experimenting a problem that was introduced by this fix:
>>
>> https://review.openstack.org/#/c/136520/
>>
>> ???
>>
>> On 20 October 2015 at 04:13, Martinx - ジェームズ <thiagocmarti...@gmail.com> 
>> wrote:
>>> Guys,
>>>
>>>  I have an Ansible automation to deploy OpenStack Kilo, it works like
>>> a charm, specially the following playbook:
>>>
>>>
>>> https://github.com/sandvine/os-ansible-deployment-lite/blob/kilo/ansible/roles/glance/tasks/default-images.yml
>>>
>>>
>>>  In one shot, I can add remote images (Glance downloads on demand) and
>>> update its properties immediatly... But, new Glance lost all
>>> functionality that make this to work.
>>>
>>>  To begin with, OS_IMAGE_API version 2, doesn't have "--location"
>>> anymore... Why not? It is VERY useful. Please, bring it back. Or at
>>> least, document how to do the same when using API 2...
>>>
>>>  Also, new (API 2) "glance image-update" REQUIRES the Image ID, this
>>> sucks because it is much easier to use the Image Name instead...
>>>
>>>  So, I updated my Ansible playbook to deploy Liberty... But Glance
>>> doesn't work the way it should work...
>>>
>>>  For example, here...:
>>>
>>>
>>>  
>>> https://github.com/sandvine/os-ansible-deployment-lite/blob/liberty/ansible/roles/glance/tasks/default-images.yml
>>>
>>>
>>> ...I had to disable all "glance image-update" calls, because it simple
>>> doesn't work anymore, even using OS_IMAGE_API=1 !!!
>>>
>>>
>>>  Look what happens now:
>>>
>>> 
>>> changed: [localhost] => (item=glance image-create --location
>>> http://download.cirros-cloud.net/0.3.4/cirros-0.3.4-i386-disk.img
>>> --name "CirrOS 0.3.4 - Minimalist - 32-bit - Cloud Based Image"
>>> --is-public true --container-format bare --disk-format qcow2)
>>> 
>>> failed: [localhost] => (item=glance image-update --property
>>> hw_scsi_model=virtio-scsi --property hw_disk_bus=scsi "CirrOS 0.3.4 -
>>> Minimalist - 32-bit - Cloud Based Image") => {"changed": true, "cmd":
>>> ["glance", "image-update", "--property", "hw_scsi_model=virtio-scsi",
>>> "--property", "hw_disk_bus=scsi", "CirrOS 0.3.4 - Minimalist - 32-bit
>>> - Cloud Based Image"], "delta": "0:00:00.898007", "end": "2015-10-20
>>> 01:10:03.837781", "item": "glance image-update --property
>>> hw_scsi_model=virtio-scsi --property hw_disk_bus=scsi \"CirrOS 0.3.4 -
>>> Minimalist - 32-bit - Cloud Based Image\"", "rc": 1, "start":
>>> "2015-10-20 01:10:02.939774", "warnings": []}
>>> stderr: Unable to upload image data to an image which is active.
>>> 
>>>
>>> It is even more weird, because those two commands, that doesn't work
>>> when with Ansible, works manually... For example:
>>>
>>> ---
>>> #! /bin/bash
>>> source ~/admin-openrc.sh
>>> glance image-create --location
>>> http://uec-images.ubuntu.com

Re: [Openstack] OpenStack "Liberty" is officially released !

[Martinx - ジェームズ]

WOooOW!! OpenStack Liberty on Ubuntu Trusty already includes DPDK,
plus OpenvSwitch with DPDK enabled!!

This is freaking AWESOME!!

Thank you Canonical!!

And special thanks to Stefan Bader and to James Page, two amazing guys
that worked to package those projects for Ubuntu!

This is very cool! Ubuntu rocks!!

Exciting times...:-)

On 18 October 2015 at 02:02, Martinx - ジェームズ <thiagocmarti...@gmail.com> wrote:
> I'm installing Liberty on Trusty right now. Using Ubuntu 14.04 packages...
>
>
> https://wiki.ubuntu.com/ServerTeam/CloudArchive
>
>
> NOTE: Liberty cloud archive only supported on Trusty
>
> sudo add-apt-repository cloud-archive:liberty
>
>
> You can also follow this:
>
> http://docs.openstack.org/liberty/install-guide-ubuntu
>
>
> That's it...
>
>
> Also, I have an OpenStack environment that I installed using Ubuntu
> 12.04 with Havana years ago, and now, this very same system is on
> Ubuntu 14.04 with Kilo, upgrading using an Ubuntu tool called
> "do-release-upgrade" plus new Cloud Archives, good lucky trying
> something like this on CentOS... You can not even upgrade from CentOS
> 5 to 6... Comedy...  lol
>
> Cheers!
>
> On 16 October 2015 at 16:03, Matt Kassawara <mkassaw...@gmail.com> wrote:
>> Ubuntu has not published Liberty release packages to the cloud archive
>> repository (for 14.04) and RDO has not provided a permanent location for any
>> Liberty packages yet.
>>
>> On Fri, Oct 16, 2015 at 12:04 PM, Martinx - ジェームズ
>> <thiagocmarti...@gmail.com> wrote:
>>>
>>> Amazing times!!! Thanks you guys!
>>>
>>> It is already included in Ubuntu 15.10 and for Ubuntu 14.04 via Ubuntu
>>> Cloud Archive... Perfect! Super easy to upgrade...
>>>
>>> Congratulations!
>>>
>>> On 15 October 2015 at 13:43, Thierry Carrez <thie...@openstack.org> wrote:
>>> > Hello everyone,
>>> >
>>> > I'm very pleased to announce the final releases of OpenStack Liberty,
>>> > which conclude the 6-month Liberty development cycle.
>>> >
>>> > This is the first of our "Big Tent" releases, which means that we have a
>>> > lot more components around. You can find the complete list of
>>> > already-released Liberty versions at:
>>> >
>>> > http://docs.openstack.org/releases/releases/liberty.html
>>> >
>>> > The Liberty Release Notes wikipage contains an overview of the key
>>> > features, as well as upgrade notes and current lists of known issues.
>>> > You can access them at:
>>> >
>>> > https://wiki.openstack.org/wiki/ReleaseNotes/Liberty
>>> >
>>> > The OpenStack release management team directly handled a number of
>>> > services. You can find their source tarballs, together with complete
>>> > lists of features and bugfixes for each project, at the following links:
>>> >
>>> > nova:   https://launchpad.net/nova/liberty/12.0.0
>>> > swift:  https://launchpad.net/swift/liberty/2.5.0
>>> > glance: https://launchpad.net/glance/liberty/11.0.0
>>> > neutron:https://launchpad.net/neutron/liberty/7.0.0
>>> > cinder: https://launchpad.net/cinder/liberty/7.0.0
>>> > keystone:   https://launchpad.net/keystone/liberty/8.0.0
>>> > horizon:https://launchpad.net/horizon/liberty/8.0.0
>>> > ceilometer: https://launchpad.net/ceilometer/liberty/5.0.0
>>> > heat:   https://launchpad.net/heat/liberty/5.0.0
>>> > trove:  https://launchpad.net/trove/liberty/4.0.0
>>> > sahara: https://launchpad.net/sahara/liberty/3.0.0
>>> > ironic: https://launchpad.net/ironic/liberty/4.2.0
>>> > designate:  https://launchpad.net/designate/liberty/1.0.0
>>> > zaqar:  https://launchpad.net/zaqar/liberty/1.0.0
>>> > manila: https://launchpad.net/manila/liberty/1.0.0
>>> > barbican:   https://launchpad.net/barbican/liberty/1.0.0
>>> >
>>> > Thanks again to all the individuals who contributed to this development
>>> > cycle and helped in making this release a success !
>>> >
>>> > Our next development cycle, Mitaka, is already started. We'll all gather
>>> > in Tokyo in 10 days at the Mitaka Design Summit to brainstorm and plan
>>> > this next cycle.
>>> >
>>> > See you there !
>>> >
>>> > --
>>> > Thierry Carrez (ttx)
>>> >
>>> >
>>> > ___
>>> > Mailing list:
>>> > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>> > Post to : openstack@lists.openstack.org
>>> > Unsubscribe :
>>> > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>>
>>> ___
>>> Mailing list:
>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>> Post to : openstack@lists.openstack.org
>>> Unsubscribe :
>>> http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
>>
>>

___
Mailing list: http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack
Post to : openstack@lists.openstack.org
Unsubscribe : http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack

Re: [Openstack] [Liberty] What happened with "glance image-create --location http://..." ?

[Martinx - ジェームズ]

Hey guys,

The recommended workaround worked! On Liberty, I'm using it like this now:

---
glance --os-image-api-version 1 image-create --location
http://uec-images.ubuntu.com/releases/14.04.3/release/ubuntu-14.04-server-cloudimg-amd64-disk1.img
--is-public true --disk-format qcow2 --container-format bare --name
"Ubuntu 14.04.3 LTS - Trusty Tahr - 64-bit - Cloud Based Image"

glance --os-image-api-version 1 image-update --property
hw_scsi_model=virtio-scsi --property hw_disk_bus=scsi "Ubuntu 14.04.3
LTS - Trusty Tahr - 64-bit - Cloud Based Image"
---

Nevertheless, how can we achieve the same results, but using new OS
Image API version 2?

I mean, with "--location" from API 1, we can add lots of images in
advance and then, Glance will only download those images, on demand
(i.e., on first launch). So, how to do the same but, using API 2? This
is a very important feature...

Also, it seems that "image-update" is different on v2 as well...

Thanks!
Thiago


On 19 October 2015 at 01:06, Morales, Victor <victor.mora...@intel.com> wrote:
> Agree, it has been implemented that argument for V2(new default version)
> for python glance client
> (https://github.com/openstack/python-glanceclient/blob/master/glanceclient/
> v2/shell.py#L46-L59). A workaround could be use the OS_IMAGE_API_VERSION=1
> to use the --location argument of the
> client(https://github.com/openstack/python-glanceclient/blob/master/glancec
> lient/v1/shell.py#L195-L200)
>
> Regards/Saludos
> Victor Morales
>
> On 10/18/15, 6:31 PM, "James Denton" <james.den...@rackspace.com> wrote:
>
>>Hi Thiago,
>>
>>I'm not sure, but this may be a change from v1 API to v2 API. Here's a
>>bug I found a few months ago that may be related:
>>
>>https://bugs.launchpad.net/python-glanceclient/+bug/1399778
>>
>>James
>>
>>From: Martinx - ジェームズ <thiagocmarti...@gmail.com>
>>Sent: Sunday, October 18, 2015 2:28 AM
>>To: openstack@lists.openstack.org
>>Subject: [Openstack] [Liberty] What happened with "glance image-create
>>--location http://...; ?
>>
>>Hey guys,
>>
>>I'm trying Liberty (on Trusty) for the first time now... I'm facing
>>one problem that I think it might be easy to obtain help...
>>
>>To begin with:
>>
>>source admin-openrc.sh
>>glance image-list
>>
>>...works...
>>
>>But, I'm trying to add an image to it and it fails, like this:
>>
>>---
>>myuser@liberty-1:~$ glance image-create --location
>>http://uec-images.ubuntu.com/releases/14.04.3/release/ubuntu-14.04-server-
>>cloudimg-amd64-disk1.img
>>--visibility public --disk-format qcow2 --container-format bare --name
>>"Ubuntu 14.04.3 LTS - Trusty Tahr - 64-bit - Cloud Based Image"
>>usage: glance [--version] [-d] [-v] [--get-schema] [--timeout TIMEOUT]
>>  [--no-ssl-compression] [-f] [--os-image-url OS_IMAGE_URL]
>>  [--os-image-api-version OS_IMAGE_API_VERSION]
>>  [--profile HMAC_KEY] [-k] [--os-cert OS_CERT]
>>  [--cert-file OS_CERT] [--os-key OS_KEY] [--key-file OS_KEY]
>>  [--os-cacert ] [--ca-file OS_CACERT]
>>  [--os-username OS_USERNAME] [--os-user-id OS_USER_ID]
>>  [--os-user-domain-id OS_USER_DOMAIN_ID]
>>  [--os-user-domain-name OS_USER_DOMAIN_NAME]
>>  [--os-project-id OS_PROJECT_ID]
>>  [--os-project-name OS_PROJECT_NAME]
>>  [--os-project-domain-id OS_PROJECT_DOMAIN_ID]
>>  [--os-project-domain-name OS_PROJECT_DOMAIN_NAME]
>>  [--os-password OS_PASSWORD] [--os-tenant-id OS_TENANT_ID]
>>  [--os-tenant-name OS_TENANT_NAME] [--os-auth-url
>>OS_AUTH_URL]
>>  [--os-region-name OS_REGION_NAME]
>>  [--os-auth-token OS_AUTH_TOKEN]
>>  [--os-service-type OS_SERVICE_TYPE]
>>  [--os-endpoint-type OS_ENDPOINT_TYPE]
>>   ...
>>glance: error: unrecognized arguments: --location
>>http://uec-images.ubuntu.com/releases/14.04.3/release/ubuntu-14.04-server-
>>cloudimg-amd64-disk1.img
>>---
>>
>>What happens to "--location" option?
>>
>>I'm using a very similar line to add images to Kilo, only difference
>>is that on Kilo, I'm using "--is-public true", instead of
>>"--visibility public" (Liberty)...
>>
>>If I download the file, and use "--file", instead of "--location" as
>>before, then it works... But I prefer to add download the images o