To test whether it's our new patch that causes the problem, I submitted a
dummy patch to trigger CI and the CI failed again. Checking the log of
nova scheduler, it's very strange that the scheduling starts with 0 host at
May 06 09:40:34.358585
Hi team, I would like to nominate Baisen Song (songbaisen) for tricircle
core reviewer. Baisen has actively joined the discussion of feature
development and has contributed important patches since Queens, like
resource deletion reliability and openstack-sdk new version adaption. I
really think his
I would like to announce my self nomination for the PTL candidacy in
Tricircle Rocky cycle. My name is Zhiyuan Cai, and my IRC handle is
zhiyuan. I am currently the PTL of Tricircle for Queens cycle and have been
actively participating in the development of this project since Mitaka
Tomorrow I will attend the bug smash in Wuhan so I will not be able to
chair the weekly meeting. We will resume the meeting next week.
OpenStack Development Mailing List (not for usage
For the first question, since api-paste.ini is a configuration file, you
can just add your new filter to the configuration file so your filter can
take effect. As Joe suggests, you can add embed the information in the
oslo.context object so plugin can get it from context which is passed to
> On Sep 30, 2017, at 7:35 AM, Vega Cai <luckyveg...@gmail.com> wrote:
> In Tricircle we use the "multinode" topology to setup a test environment
> with three regions, "CentralRegion" and "RegionOne" in one node, and
In Tricircle we use the "multinode" topology to setup a test environment
with three regions, "CentralRegion" and "RegionOne" in one node, and
"RegionTwo" in the other node. I notice that the job definition has been
After the discussion in the PTG on last Friday, we plan the following work
items for the Queens circle:
Integration with Nova cell V2
Distinguish request from user and local neutron server
Default security group update
Network deletion reliability
Just to remind that the PTG of Tricircle team will be held on Sep 15th,
from 9:00 am to 17:00 pm(UTC+8). We will discuss our topics in this
etherpad page and #openstack-tricircle channel, so you can join the
discussion via etherpad and IRC. Also, please feel free to add topic to
As discussed in our weekly meeting, since most of our developers are in
China, we will hold a small scale team gathering in China. Please refer to
this etherpad page for the date and place of the gathering. During the
gathering, we will discuss our topics in that etherpad page and
You can first check whether the nova-consoleauth and nova-novncproxy
services are correctly running. If you use DevStack, the window names for
these two services are n-cauth and n-novnc.
On Wed, 23 Aug 2017 at 23:55 wrote:
> Hello to the Tricircle
Also, you can upload a new image with OpenStack client, when horizon
service is not running:
Since Nova service is running on RegionOne and RegionTwo, the image should
be also uploaded to
The document structure of networking-sfc has been changed so the link in
the guide is not accurate. Since you have already deployed Tricircle and
would like to enable service function chaining feature, I think you can
refer to these guides[1, 2] to setup local Neutron server. The
I would like to announce my self nomination for the PTL candidacy in
Tricircle Queens cycle. My name is Zhiyuan Cai, and my IRC handle is
zhiyuan. As a core reviewer of Tricircle project, I have been actively
participating in the development of this project since Mitaka cycle.
According to the DevStack script, for RHEL, the Apache configuration files
are placed in /etc/httpd/conf.d
On Mon, 10 Jul 2017 at 22:14 wrote:
> Hello everybody,
> I posted before a problem related to installing the tricircle on a single
Welcome to join Tricircle!
Both Keystone and Tricircle services are running under Apache, so if
Keystone failed to start after enabling Tricircle, I guess there might be
some problems of Apache configuration. You can check what services are
enabled in Apache by listing this folder:
So cute XD
On Mon, 24 Apr 2017 at 16:11 Zhipeng Huang wrote:
> perfect XD
> On Mon, Apr 24, 2017 at 2:59 PM, joehuang wrote:
>> Hello, team,
>> What about the mascot of Tricircle project?
>> Your comments are welcome.
On Wed, 1 Mar 2017 at 11:41 Devale, Sindhu wrote:
> From: joehuang
> Reply-To: "OpenStack Development Mailing List (not for usage questions)" <
> Date: Tuesday, February 28, 2017 at
Wow, awesome guide!!
My suggestion will be no matter where we maintain the guide, it will be
better to not just update the original guide directly, to make readers easy
to follow the changes of logic.
On Tue, 21 Feb 2017 at 11:00 joehuang wrote:
In Nova Ocata release, deploying placement API is a must. The nova-compute
service will fail to start if [placement] section is not configured in
DevStack script is updated to generate Apache configuration file for
placement API and it works fine in single region scenario.
I am heading for Atlanta PTG as the representative of Tricircle project, if
you are interested in Tricircle project and will also attend Atlanta PTG,
we can meet and discuss some topics there.
It's cool to store tunnel termination endpoint information in the port
model. Will this information support updating so external devices or
controllers can inject the tunnel termination endpoints via the API? This
is very useful to support VxLAN type external network, which there are
During the development of Tricircle, we find that one feature whose spec
has already been accepted is very useful, that is to enhance floating IP
router lookup to consider extra routes. With this feature, floating IP
is allowed to be associated to internal addresses that are
Hello, patch for installation guide update of multi-node deployment has
just been submitted. Here is the link:
On Wed, 12 Oct 2016 at 10:17 joehuang wrote:
> Hello, Team,
> Understand this period is keeping us quite
for usage questions)
> > Subject: RE: [openstack-dev] [tricircle][trio2o] trio2o git repo ready
> > tricircle splitting
> > Thank you for your comment, Zhiyuan.
> > For pod management, because these two projects need to run
> independently, I
+1 for the proposal. What about the codes for Pod operation? It seems that
both Tricircle and Trio2o need these codes. We submit patches to these two
joehuang 于2016年9月18日周日 下午4:17写道：
> hello, team,
> Trio2o git repository is ready now:
Oops, sorry for "triangel". Then what about "trio2o", openstack-to-openstack
Yipei Niu 于2016年9月5日周一 上午11:22写道：
> Trifennel +1
> Best regards,
> OpenStack Development Mailing List
+1 for Triangel
On 2 September 2016 at 17:34, joehuang wrote:
> After the discussion in the #openstack-tricircle channel, 3 candidates
> available now, please vote the name for the new sub-project for api-gateway
> 1. Triangel
> The Triangel are dolls
You can also refer to my network type driver implementation:
Type driver is registered in setup.cfg and loaded by code.
On 27 June 2016 at 21:44, Yipei Niu wrote:
> Dear all,
> Recently, I learn to name a
Thanks for attending the work group meeting. Our discussion log can be
On 24 May 2016 at 09:51, Yipei Niu wrote:
> Hi, all,
> Since I have
Sorry that it's a bit late to launch our cross-pod L2 networking work group
since I was attending OSCON last week.
What about having our first work group meeting on Wednesday 10am(Beijing
time UTC+8), in #openstack-tricircle channel?
Topics I think we can discuss:
+1, Shinobu has given me many suggestions in my patches.
On 4 May 2016 at 07:57, Zhipeng Huang wrote:
> +1, still remember Shinobu getting started from basic concept of
> tricircle, and now becoming a maester of multisite openstack :)
> On Tue, May 3, 2016 at 9:03 PM,
Just would like to share a way to test Tricircle north-south L3 networking
without requiring the third interface.
In the Tricircle readme, it is said that you need to add an interface in
your host to br-ext bridge. One interface to access the host, one interface
for east-west networking
Hi, I have submitted the second patch for asynchronous job management,
please help to review. Here is the link:
The first patch has been merged. Link for the first patch:
-29 15:41:04.122 TRACE neutron.api.v2.resource
> ^[[01;35m^[[00m File
> "/usr/local/lib/python2.7/dist-packages/neutronclient/v2_0/client.py", line
> 69, in exception_handler_v20
> ^[[01;31m2016-03-29 15:41:04.122 TRACE neutron.api.v2.resource
le with two node
>> Sorry for interruption.
>> >> I followed the README.md in github, every thing goes well until I
>> >> boot virtual machines with the following command
>> Point here is that even Yipei was
>> > # Use Neutron instead of nova-network
>> > disable_service n-net
>> > enable_service q-svc
>> > enable_service q-svc1
>> > enable_service q-dhcp
>> > enable_service q-a
Check "network_api_class" and "use_neutron" options in your nova.conf. It
seems that your nova API is not configured to use neutron.
On 28 March 2016 at 13:25, Yipei Niu wrote:
> Hi, all,
> After I execute the command "nova boot --flavor 1 --image
Added a work item in phase 4 in our todo list:
On 26 March 2016 at 17:31, joehuang wrote:
> Hi, Shinobu,
> This BP is a good entrance for Tricircle source code, this BP listed the
> all basic patches building Tricircle from
On 22 March 2016 at 12:09, Shinobu Kinjo <shinobu...@gmail.com> wrote:
> Thank you for your comment (inline for my message).
> On Tue, Mar 22, 2016 at 11:53 AM, Vega Cai <luckyveg...@gmail.com> wrote:
> > Let me try to explain some.
> > On 22 Marc
Let me try to explain some.
On 22 March 2016 at 10:09, Shinobu Kinjo wrote:
> On Tue, Mar 22, 2016 at 10:22 AM, joehuang wrote:
> > Hello, Shinobu,
> > Yes, as what you described here, the "initialize" in "core.py" is used
> for unit/function test
What I would like to discuss has been included :), the Reliable async, job.
On 1 March 2016 at 17:19, joehuang wrote:
> If you have any topic to discuss, please reply in the M-L.
> IRC meeting:
I suppose you are using "internal network" in virtual box to connect your
virtual machines, and I find a link that seems to related to your problem.
On 1 March 2016 at 11:14, Vega Cai <luckyveg...@gma
>> So it can be fixed by adding one more configurable global variable to
>> separate KeyStone region name and other services(Nova,Cinder,Neutron)
>> region name.
>> Best Regards
>> Chaoyi Huang ( Joe Huang )
On 29 February 2016 at 16:12, Vega Cai <luckyveg...@gmail.com> wrote:
> Hi Pengfei,
> It's expected that there's no port for 188.8.131.52 since what we attach to
> router are 184.108.40.206 and 220.127.116.11.
> Try to ping 10.0.2.1 in vm1 and
It's expected that there's no port for 18.104.22.168 since what we attach to
router are 22.214.171.124 and 126.96.36.199.
Try to ping 10.0.2.1 in vm1 and to ping 10.0.1.1 in vm2 to see if you can
get reply. There may be security group rules that prevent vm to be accessed.
The check if all the
One reason for that error is that the API service is down. You can run
"rejoin-stack.sh" under your DevStack folder to enter the "screen" console
of DevStack, to check if services are running well. If you are not familiar
with "screen", which is a window manager for Linux, you can do a
I did a test about the performance of port query in Tricircle yesterday.
The result is attached.
Three observations in the test result:
(1) Neutron client costs much more time than curl, the reason may be
neutron client needs to apply for a new token in each run.
(2) Eventlet doesn't
When implementing L3 north-south networking functionality, I meet the DHCP
port problem again.
First let me briefly explain the DHCP port problem. In Tricircle, we have a
Neutron server using Tricircle plugin in top pod to take control all the
Neutron servers in bottom pods. The strategy
Could you check the version of your flake8? It's located in .tox/pep8/bin.
D102 is one of the pep8 error, and I find it's not ignored in Tricircle
tox.ini, so I guess that flake8 on Jenkins doesn't enable this error by
default but in your flake8, this error is enabled.
I check the
> your configuration file in Node1 and Node2 ( local.conf in the devstack
> folder )
> Best Regards
> Chaoyi Huang ( Joe Huang )
> *From:* Vega Cai [mailto:luckyveg...@gmail.com]
> *Sent:* Tuesday, January 26, 2016 10:03 AM
> *To:* OpenStack De
Strange... One suggestion to dig into this problem is that you add a
"--debug" parameter to the command like:
"openstack --debug project show admin -f value -c id"
and run DevStack again. With "--debug" parameter, OpenStack client will
print out raw HTTP request and response so may bring you
As we discussed before, DAL needs to provide API to access resources in the
top and bottom OpenStack, so I implemented a client wrapper which has API
list_servers(self, cxt, site, fileters)
cxt is the context object storing authorization information, site tells DAL
Agree that we use polling mode to synchronize resource status, but it's a
bit weird to deploy a localized service for just polling. Maybe we can
define a status synchronizatioin task to handle such work so it can be done
by cascade service itself. If multiple cascade service is deployed, to
In my current implementation, there are a core module and a models module.
Core module handles all the database stuff, start a session, issue sql
operation, then end a session. Models module invokes methods in core module
to access database, as showed below:
, but how to
decide when to commit the data, end a session?
Chaoyi Huang ( Joe Huang )
*From:* Vega Cai [mailto:luckyveg...@gmail.com]
*Sent:* Thursday, July 30, 2015 3:03 PM
*Subject:* [openstack-dev] [tricircle] DAL implementation
Mail list logo