| *** sweston has joined #openstack-neutron | 00:11 | |
| openstackgerrit | Ronak Shah proposed a change to openstack/neutron: Nuage Networks Plugin https://review.openstack.org/67272 | 00:19 |
|---|---|---|
| *** banix has joined #openstack-neutron | 00:20 | |
| *** matsuhashi has joined #openstack-neutron | 00:23 | |
| *** networkstatic has joined #openstack-neutron | 00:31 | |
| *** matsuhashi has quit IRC | 00:33 | |
| *** matsuhashi has joined #openstack-neutron | 00:34 | |
| *** Sukhdev has quit IRC | 00:35 | |
| *** matsuhashi has quit IRC | 00:38 | |
| *** xuhanp has joined #openstack-neutron | 00:39 | |
| *** yamahata has quit IRC | 00:41 | |
| *** alagalah has quit IRC | 00:42 | |
| *** yamahata has joined #openstack-neutron | 00:44 | |
| *** alagalah has joined #openstack-neutron | 00:50 | |
| *** sweston has quit IRC | 00:56 | |
| *** xuhanp has quit IRC | 00:56 | |
| *** rossigee has quit IRC | 00:57 | |
| *** rossigee has joined #openstack-neutron | 00:58 | |
| *** sweston has joined #openstack-neutron | 00:58 | |
| *** yamahata has quit IRC | 01:00 | |
| openstackgerrit | Ronak Shah proposed a change to openstack/neutron: Nuage Networks Plugin https://review.openstack.org/67272 | 01:01 |
| *** matsuhashi has joined #openstack-neutron | 01:04 | |
| *** WackoRobie has joined #openstack-neutron | 01:04 | |
| *** dkehn__ has joined #openstack-neutron | 01:04 | |
| *** clev has quit IRC | 01:04 | |
| *** clev has joined #openstack-neutron | 01:04 | |
| *** dkehn has quit IRC | 01:05 | |
| *** dkehn__ has quit IRC | 01:05 | |
| *** dkehn__ has joined #openstack-neutron | 01:05 | |
| *** dkehn_ has quit IRC | 01:06 | |
| *** dkehn has joined #openstack-neutron | 01:07 | |
| *** dkehn__ is now known as dkehn_ | 01:09 | |
| *** sweston has quit IRC | 01:14 | |
| *** oda-g has joined #openstack-neutron | 01:27 | |
| *** morganfainberg_Z is now known as morganfainberg | 01:28 | |
| *** alagalah_ has joined #openstack-neutron | 01:40 | |
| *** baoli has joined #openstack-neutron | 01:47 | |
| *** yamahata__ has quit IRC | 01:49 | |
| *** yamahata has joined #openstack-neutron | 01:51 | |
| *** yamahata_ has joined #openstack-neutron | 01:54 | |
| *** WackoRobie has quit IRC | 01:54 | |
| *** WackoRobie has joined #openstack-neutron | 01:55 | |
| *** Gamekiller77 has quit IRC | 01:55 | |
| *** ramishra has joined #openstack-neutron | 01:55 | |
| *** ramishra has quit IRC | 01:55 | |
| *** ramishra has joined #openstack-neutron | 01:56 | |
| *** alagalah has quit IRC | 01:58 | |
| *** WackoRobie has quit IRC | 02:00 | |
| *** WackoRob_ has joined #openstack-neutron | 02:00 | |
| *** yamahata has quit IRC | 02:01 | |
| *** xianghui has joined #openstack-neutron | 02:02 | |
| *** JPtheJP has quit IRC | 02:07 | |
| *** JPtheJP has joined #openstack-neutron | 02:07 | |
| *** Jianyong has joined #openstack-neutron | 02:14 | |
| *** liudong has joined #openstack-neutron | 02:16 | |
| *** matsuhashi has quit IRC | 02:18 | |
| *** matsuhashi has joined #openstack-neutron | 02:26 | |
| *** baoli has quit IRC | 02:27 | |
| *** baoli has joined #openstack-neutron | 02:37 | |
| *** baoli has quit IRC | 02:41 | |
| *** dave_tucker is now known as dave_tucker_zzz | 02:42 | |
| openstackgerrit | Kyle Mestery proposed a change to openstack/neutron: Add OpenDaylight ML2 MechanismDriver https://review.openstack.org/69775 | 02:46 |
| *** ianw has quit IRC | 02:55 | |
| *** pcm_ has quit IRC | 03:01 | |
| *** ianw has joined #openstack-neutron | 03:02 | |
| *** pcm_ has joined #openstack-neutron | 03:02 | |
| openstackgerrit | lifeless proposed a change to openstack/python-neutronclient: Don't reuse 'body' for response data https://review.openstack.org/69110 | 03:08 |
| *** RajeshMohan has quit IRC | 03:08 | |
| *** RajeshMohan has joined #openstack-neutron | 03:09 | |
| openstackgerrit | Yongli He proposed a change to openstack/neutron: Use six.StringIO/BytesIO instead of StringIO.StringIO https://review.openstack.org/73902 | 03:10 |
| *** baoli has joined #openstack-neutron | 03:12 | |
| *** networkstatic has quit IRC | 03:13 | |
| *** networkstatic has joined #openstack-neutron | 03:13 | |
| *** otherwiseguy has quit IRC | 03:17 | |
| *** sweston has joined #openstack-neutron | 03:22 | |
| *** otherwiseguy has joined #openstack-neutron | 03:29 | |
| *** otherwiseguy has quit IRC | 03:30 | |
| openstackgerrit | berlin proposed a change to openstack/neutron: Fix duplicate name of NVP LBaaS objs not allowed on vShield Edge https://review.openstack.org/73905 | 03:34 |
| *** dkehn__ has joined #openstack-neutron | 03:38 | |
| *** dkehn_ has quit IRC | 03:41 | |
| *** dkehn has quit IRC | 03:41 | |
| *** dkehn has joined #openstack-neutron | 03:42 | |
| *** matsuhashi has quit IRC | 03:48 | |
| *** baoli has quit IRC | 04:06 | |
| openstackgerrit | Hemanth Ravi proposed a change to openstack/neutron: One Convergence Neutron Plugin Implementation https://review.openstack.org/69246 | 04:06 |
| *** WackoRob_ has quit IRC | 04:06 | |
| *** WackoRobie has joined #openstack-neutron | 04:07 | |
| *** sweston has quit IRC | 04:08 | |
| openstackgerrit | shihanzhang proposed a change to openstack/neutron: Fix problem of deleting dhcp port https://review.openstack.org/73802 | 04:08 |
| *** banix has quit IRC | 04:09 | |
| *** liudong has quit IRC | 04:10 | |
| *** WackoRobie has quit IRC | 04:12 | |
| *** alagalah_ has quit IRC | 04:12 | |
| *** alagalah has joined #openstack-neutron | 04:12 | |
| *** morganfainberg is now known as morganfainberg_Z | 04:13 | |
| *** liudong has joined #openstack-neutron | 04:13 | |
| *** dkehn__ is now known as dkehn_ | 04:16 | |
| *** alagalah has quit IRC | 04:17 | |
| *** matsuhashi has joined #openstack-neutron | 04:32 | |
| *** amotoki has joined #openstack-neutron | 04:39 | |
| *** yamahata has joined #openstack-neutron | 04:42 | |
| *** liudong has quit IRC | 04:43 | |
| *** alagalah has joined #openstack-neutron | 04:43 | |
| openstackgerrit | Isaku Yamahata proposed a change to openstack/neutron: options: consolidate options definitions https://review.openstack.org/73520 | 04:51 |
| *** alagalah has quit IRC | 04:51 | |
| *** ramishra has quit IRC | 04:53 | |
| *** yfried has quit IRC | 04:55 | |
| *** chandan_kumar has joined #openstack-neutron | 05:02 | |
| openstackgerrit | stephen-ma proposed a change to openstack/neutron: Delete duplicate internal devices in router namespace https://review.openstack.org/57954 | 05:03 |
| *** ramishra has joined #openstack-neutron | 05:04 | |
| *** irenab has joined #openstack-neutron | 05:13 | |
| irenab | amotoki: hi, are you on IRC? | 05:15 |
| amotoki | irenab: hi | 05:16 |
| irenab | hi, may I ask question on the review comment you gave for https://review.openstack.org/#/c/53609/12? | 05:17 |
| amotoki | irenab: let me check a bit later. I am discussing with my collegue. | 05:18 |
| irenab | amotoki: sure, please let me know when | 05:19 |
| openstackgerrit | Isaku Yamahata proposed a change to openstack/neutron: l3-agent-consolidation: framework for consolidating l3 agents https://review.openstack.org/57627 | 05:23 |
| *** gdubreui has quit IRC | 05:27 | |
| *** RajeshMohan has quit IRC | 05:29 | |
| *** RajeshMohan has joined #openstack-neutron | 05:30 | |
| *** yfried has joined #openstack-neutron | 05:40 | |
| *** pcm_ has quit IRC | 05:45 | |
| *** rotbeard has joined #openstack-neutron | 06:02 | |
| *** ramishra has quit IRC | 06:06 | |
| *** saju_m has joined #openstack-neutron | 06:15 | |
| *** iwamoto has joined #openstack-neutron | 06:20 | |
| amotoki | irenab: thanks for waiting. | 06:23 |
| irenab | amotoki: Can you please take a look at the comment you posted? | 06:24 |
| irenab | I have a problem in reworking the code the way you suggested | 06:24 |
| amotoki | irenab: I checked the comments on patch set 12. | 06:25 |
| irenab | unit test fails, since in testing environment there is no zmq | 06:25 |
| amotoki | I see. Are you talking about Jenkins UT failure? | 06:26 |
| irenab | yes | 06:26 |
| irenab | there was previous comments regarding zmq, and following Mark's suggestion, I use try_import | 06:27 |
| openstackgerrit | Jenkins proposed a change to openstack/neutron: Imported Translations from Transifex https://review.openstack.org/73932 | 06:27 |
| irenab | in real deploymnet the external service that agent talks to requires zmq on its nstalation, so there is no way zmq is not installed | 06:27 |
| amotoki | The current code (patch set 14) looks good to me too. I agree with mark. | 06:28 |
| irenab | I am wondering if I can change the code back as it was prior to your comment or you have some other suggestion? | 06:28 |
| irenab | The current code does not pass unit tests, since zmq is on in testing required packages | 06:29 |
| amotoki | hmm.. does faking zmq work? | 06:29 |
| irenab | didn't find a way how to do it.... | 06:30 |
| amotoki | AFAIK, ryu unit test deos the similar. Let me check for a while. | 06:30 |
| irenab | will be great if you cangive me some reference to look at | 06:31 |
| amotoki | It is one example: https://review.openstack.org/#/c/71791/7/neutron/tests/unit/ofagent/fake_oflib.py (though it is under review) | 06:31 |
| amotoki | It replaces real ryu modules with mock. | 06:32 |
| openstackgerrit | Ann Kamyshnikova proposed a change to openstack/neutron: Sync models with migrations https://review.openstack.org/55411 | 06:34 |
| irenab | I looked at other examples with try_import, not sure its required to raise system exit, since it cannot be in real deployment | 06:35 |
| irenab | like hyperv agent: https://github.com/openstack/neutron/blob/master/neutron/plugins/hyperv/agent/utils.py | 06:36 |
| openstackgerrit | IWAMOTO Toshihiro proposed a change to openstack/neutron: LBaaS: implements LVS driver https://review.openstack.org/67405 | 06:36 |
| openstackgerrit | Yongli He proposed a change to openstack/python-neutronclient: Use six.moves cStringIO instead of cStringIO https://review.openstack.org/73937 | 06:39 |
| amotoki | irenab: SystemExit is not necessarily required. It depends on you, but it looks better to raise SystemExit. I think it is clearer to operators. | 06:40 |
| *** rohit404 has quit IRC | 06:41 | |
| amotoki | irenab: i can take a look at the UT in your patch later today. It seems to requires some tries. | 06:41 |
| *** jaypipes has quit IRC | 06:42 | |
| *** afazekas_ has quit IRC | 06:42 | |
| *** jaypipes has joined #openstack-neutron | 06:42 | |
| irenab | amotoki: thanks, will try to look at the code you posted. But not sure its necessary to handle missing zmq due to depedency of additional service used by the agent | 06:43 |
| *** rohit404 has joined #openstack-neutron | 06:44 | |
| amotoki | irenab: your case looks more simple. Changing the test code to the following will address the unit test failures: | 07:00 |
| amotoki | with mock.patch.object(utils, 'zmq'): | 07:00 |
| amotoki | self.agent = eswitch_neutron_agent.MlnxEswitchNeutronAgent({}) | 07:00 |
| openstackgerrit | Gary Duan proposed a change to openstack/neutron: FWaaS integration with service type framewrok https://review.openstack.org/60699 | 07:01 |
| openstackgerrit | Xu Han Peng proposed a change to openstack/neutron: Change firewall to INACTIVE when admin state down https://review.openstack.org/73944 | 07:02 |
| amotoki | irenab: this is the diff: http://paste.openstack.org/show/66369/ In this case, it is okay if property _conn in utils.EswitchUtils, so mocking mlnx.agent.utils.zmq is enough. | 07:02 |
| irenab | amotoki: taking a look | 07:02 |
| irenab | amotoki: will try to run it now | 07:04 |
| amotoki | irenab: in my env, agent ut passed after the change. | 07:04 |
| *** oda-g has quit IRC | 07:07 | |
| *** ajo has joined #openstack-neutron | 07:08 | |
| irenab | amotoki: still waiting to tests results. Once completed sucessfully will upload the fix. Many thanks for your help | 07:14 |
| openstackgerrit | Lee Li proposed a change to openstack/neutron: Remove dependent module py3kcompat https://review.openstack.org/73766 | 07:16 |
| openstackgerrit | Yongli He proposed a change to openstack/python-neutronclient: Use six.moves cStringIO instead of cStringIO https://review.openstack.org/73937 | 07:16 |
| amotoki | irenab: do you know how to run specific unit tests? I just ran only mlnx unit tests. | 07:24 |
| irenab | amotoki: I am running it with testr run mlnx | 07:25 |
| irenab | any other way? | 07:25 |
| yfried | jianingy1afk: ping | 07:26 |
| amotoki | irenab: i ran "tox -e py27 neutron.tests.unit.mlnx.test_mlnx_neutron_agent". python module path can be used. | 07:27 |
| *** afazekas_ has joined #openstack-neutron | 07:27 | |
| yfried | jianingy1afk: is this patch https://review.openstack.org/#/c/60512/ still alive? is this bp - https://blueprints.launchpad.net/neutron/+spec/router-port-forwarding going to be ready for I3? | 07:28 |
| irenab | amotoki: it take a lot of time, not finished yet... | 07:29 |
| irenab | amotoki: finally over and passed! Thank you very much, Uploading the fix | 07:30 |
| amotoki | irenab: good to know that. you are welcome. | 07:31 |
| openstackgerrit | IWAMOTO Toshihiro proposed a change to openstack/neutron: LBaaS: implements LVS driver https://review.openstack.org/67405 | 07:32 |
| openstackgerrit | Berezovsky Irena proposed a change to openstack/neutron: Add update from agent to plugin on device up https://review.openstack.org/53609 | 07:32 |
| openstackgerrit | Ann Kamyshnikova proposed a change to openstack/neutron: Sync models with migrations https://review.openstack.org/55411 | 07:58 |
| *** bvandenh has joined #openstack-neutron | 08:03 | |
| *** slawek has joined #openstack-neutron | 08:04 | |
| *** mflobo has joined #openstack-neutron | 08:04 | |
| openstackgerrit | IWAMOTO Toshihiro proposed a change to openstack/neutron: LBaaS: add routed-service-insertion extension to the plugin https://review.openstack.org/73962 | 08:06 |
| slawek | hello | 08:07 |
| slawek | I have question about l3-agent: is it possible to have got multiple l3-agents, for example on every compute-node instead of one main on e.g. neutron server? | 08:08 |
| slawek | will it work like "vrouters" on each node? | 08:08 |
| *** rohit404 has quit IRC | 08:09 | |
| iwamoto | you might want to check 'distributed virtual router' | 08:10 |
| *** jlibosva has joined #openstack-neutron | 08:10 | |
| slawek | <iwamoto>: but where I should "check" it? | 08:12 |
| iwamoto | slawek: it's under development now, iirc. | 08:12 |
| slawek | I see only blueprints with such feature: https://blueprints.launchpad.net/neutron/+spec/neutron-ovs-dvr but it is probably not implemented yet | 08:12 |
| slawek | ok, is it this blueprint which I send above? | 08:13 |
| iwamoto | slawek: yes | 08:14 |
| openstackgerrit | Salvatore Orlando proposed a change to openstack/neutron: NSX plugin: Introduce operational status for security groups https://review.openstack.org/71693 | 08:14 |
| *** yamahata has quit IRC | 08:21 | |
| *** pasquier-s has joined #openstack-neutron | 08:22 | |
| *** alagalah has joined #openstack-neutron | 08:23 | |
| *** leseb has joined #openstack-neutron | 08:24 | |
| *** afazekas_ is now known as afazekas | 08:27 | |
| openstackgerrit | Vincent Untz proposed a change to openstack/neutron: Reset the policy after loading extensions https://review.openstack.org/73866 | 08:28 |
| *** alagalah has quit IRC | 08:28 | |
| *** saju_m has quit IRC | 08:31 | |
| *** jroovers has joined #openstack-neutron | 08:32 | |
| *** jroovers|afk has joined #openstack-neutron | 08:33 | |
| *** matrohon has joined #openstack-neutron | 08:34 | |
| *** salv-orlando_ has joined #openstack-neutron | 08:35 | |
| *** salv-orlando has quit IRC | 08:35 | |
| *** salv-orlando_ is now known as salv-orlando | 08:35 | |
| *** amuller has joined #openstack-neutron | 08:35 | |
| *** jroovers has quit IRC | 08:37 | |
| *** jgallard has joined #openstack-neutron | 08:40 | |
| *** iwamoto has quit IRC | 08:40 | |
| openstackgerrit | A change was merged to openstack/neutron: Remove DEBUG:....nsx_cluster:Attribute is empty or null https://review.openstack.org/73749 | 08:44 |
| *** amuller has quit IRC | 08:46 | |
| *** ygbo has joined #openstack-neutron | 08:53 | |
| *** amuller has joined #openstack-neutron | 08:55 | |
| *** saju_m has joined #openstack-neutron | 08:58 | |
| openstackgerrit | Marcos FermÃn Lobo proposed a change to openstack/python-neutronclient: Unexpected response in agent-list command https://review.openstack.org/70363 | 08:59 |
| *** jistr has joined #openstack-neutron | 09:01 | |
| *** Jianyong has quit IRC | 09:02 | |
| *** chandan_kumar has quit IRC | 09:10 | |
| *** ajo has quit IRC | 09:16 | |
| *** rohit404 has joined #openstack-neutron | 09:19 | |
| *** chandan_kumar has joined #openstack-neutron | 09:20 | |
| *** shausy has joined #openstack-neutron | 09:33 | |
| *** zenoway has joined #openstack-neutron | 09:33 | |
| openstackgerrit | berlin proposed a change to openstack/neutron: Fix duplicate name of NVP LBaaS objs not allowed on vShield Edge https://review.openstack.org/73905 | 09:47 |
| *** rossella-s has joined #openstack-neutron | 09:50 | |
| *** matsuhashi has quit IRC | 09:54 | |
| *** rossella-s has quit IRC | 09:54 | |
| openstackgerrit | Isaku Yamahata proposed a change to openstack/neutron: Implement service vm framework: REST API and DB model part https://review.openstack.org/56892 | 09:55 |
| openstackgerrit | Isaku Yamahata proposed a change to openstack/neutron: Implement service vm framework: drivers for load balander (WIP) https://review.openstack.org/72068 | 09:55 |
| *** networkstatic has quit IRC | 09:56 | |
| *** networkstatic has joined #openstack-neutron | 09:56 | |
| *** rossella-s has joined #openstack-neutron | 10:00 | |
| openstackgerrit | Trinath Somanchi proposed a change to openstack/neutron: implements:bp/fsl-sdn-os-mech-driver::fsl mech driver https://review.openstack.org/73985 | 10:00 |
| *** leseb has quit IRC | 10:01 | |
| *** leseb has joined #openstack-neutron | 10:02 | |
| *** matsuhashi has joined #openstack-neutron | 10:02 | |
| *** jp_at_hp has joined #openstack-neutron | 10:15 | |
| *** matsuhashi has quit IRC | 10:17 | |
| *** matsuhashi has joined #openstack-neutron | 10:17 | |
| openstackgerrit | Jun Jie Nan proposed a change to openstack/neutron: Check vxlan enablement via modinfo https://review.openstack.org/65902 | 10:18 |
| openstackgerrit | Tomoko Inoue proposed a change to openstack/neutron: Fix typo in service_drivers.ipsec. https://review.openstack.org/73848 | 10:20 |
| *** ygbo has quit IRC | 10:21 | |
| *** ygbo has joined #openstack-neutron | 10:21 | |
| *** matsuhashi has quit IRC | 10:22 | |
| *** ajo has joined #openstack-neutron | 10:27 | |
| *** ramishra has joined #openstack-neutron | 10:38 | |
| yfried | jianingy1afk: ping | 10:38 |
| openstackgerrit | Tomoko Inoue proposed a change to openstack/neutron: Fix typo in service_drivers.ipsec https://review.openstack.org/73848 | 10:39 |
| *** jp_at_hp has quit IRC | 10:39 | |
| openstackgerrit | Isaku Yamahata proposed a change to openstack/neutron: tests/service: consolidate setUp/tearDown logic https://review.openstack.org/70668 | 10:42 |
| openstackgerrit | Isaku Yamahata proposed a change to openstack/neutron: Implement service vm framework: REST API and DB model part https://review.openstack.org/56892 | 10:42 |
| openstackgerrit | Isaku Yamahata proposed a change to openstack/neutron: Implement service vm framework: drivers for load balander (WIP) https://review.openstack.org/72068 | 10:42 |
| *** vkozhukalov has joined #openstack-neutron | 10:46 | |
| *** matsuhashi has joined #openstack-neutron | 10:47 | |
| openstackgerrit | Trinath Somanchi proposed a change to openstack/neutron: implements:bp/fsl-sdn-os-mech-driver::fsl mech driver https://review.openstack.org/73985 | 10:51 |
| *** yamahata has joined #openstack-neutron | 10:51 | |
| openstackgerrit | Trinath Somanchi proposed a change to openstack/neutron: implements:bp/fsl-sdn-os-mech-driver::fsl mech driver https://review.openstack.org/73985 | 10:55 |
| *** networkstatic has quit IRC | 10:55 | |
| *** jroovers|afk is now known as jroovers | 10:58 | |
| *** yamahata has quit IRC | 11:00 | |
| *** xianghui has quit IRC | 11:00 | |
| shausy | I have multinode openstack installation up and running with instances launched | 11:01 |
| shausy | but I have issue in ping /ssh to instance from outside | 11:01 |
| shausy | whereas i can able to ping the public IP from controller to instance | 11:02 |
| *** matrohon has quit IRC | 11:06 | |
| *** ramishra has quit IRC | 11:08 | |
| *** ramishra has joined #openstack-neutron | 11:09 | |
| *** djoreilly has joined #openstack-neutron | 11:10 | |
| *** matrohon has joined #openstack-neutron | 11:11 | |
| *** leseb has quit IRC | 11:11 | |
| *** b3nt_pin has joined #openstack-neutron | 11:12 | |
| *** leseb has joined #openstack-neutron | 11:12 | |
| *** b3nt_pin is now known as beagles | 11:12 | |
| *** leseb has quit IRC | 11:16 | |
| *** leseb has joined #openstack-neutron | 11:18 | |
| *** pcm_ has joined #openstack-neutron | 11:18 | |
| *** pcm_ has quit IRC | 11:20 | |
| *** pcm_ has joined #openstack-neutron | 11:21 | |
| *** akamyshnikova has quit IRC | 11:21 | |
| *** akamyshnikova has joined #openstack-neutron | 11:24 | |
| *** akamyshnikova has joined #openstack-neutron | 11:24 | |
| *** leseb has quit IRC | 11:28 | |
| *** leseb has joined #openstack-neutron | 11:29 | |
| *** leseb has quit IRC | 11:33 | |
| openstackgerrit | Oleg Bondarev proposed a change to openstack/neutron: Reschedule router if new external gateway is on other network https://review.openstack.org/52884 | 11:34 |
| *** jp_at_hp has joined #openstack-neutron | 11:37 | |
| *** matsuhashi has quit IRC | 11:37 | |
| ihrachys | mordred: ping | 11:38 |
| *** jp_at_hp has quit IRC | 11:38 | |
| *** jgallard has quit IRC | 11:38 | |
| openstackgerrit | Ann Kamyshnikova proposed a change to openstack/neutron: Sync models with migrations https://review.openstack.org/55411 | 11:42 |
| *** jp_at_hp has joined #openstack-neutron | 11:48 | |
| *** shausy has quit IRC | 11:48 | |
| *** shausy has joined #openstack-neutron | 11:48 | |
| *** zenoway has quit IRC | 11:53 | |
| *** alagalah has joined #openstack-neutron | 12:00 | |
| *** bvandenh has quit IRC | 12:01 | |
| *** alagalah has quit IRC | 12:04 | |
| *** yfried has quit IRC | 12:10 | |
| *** alexpilotti has quit IRC | 12:13 | |
| ihrachys | ok, anyone able to release neutronclient? I've asked for it for a week already here, with no result... and now we start to hit issues in latest stable release due to lack of new client: https://review.openstack.org/#/c/73904/ | 12:14 |
| ihrachys | I think we should release it asap, otherwise we leave neutron metadata agent broken for all users that are not running client from git master [aka everyone] | 12:15 |
| ihrachys | markmcclain said before that he's going to tag and release it, but it already took too much time, so if anyone wants to step in and unbreak neutron stable, please do so | 12:16 |
| amuller | ihrachys: Perhaps an upstream email about this would be more effective / gain more traction | 12:18 |
| ihrachys | amuller: actually there is already one, from Robert Collins | 12:19 |
| amuller | ihrachys: I see | 12:19 |
| ihrachys | but yeah, I've just added +1 and ASAP to the thread | 12:22 |
| *** rohit404 has quit IRC | 12:22 | |
| *** rohit404 has joined #openstack-neutron | 12:22 | |
| amuller | ihrachys: I haven't seen the thread but maybe you can explain what bugs are popping up, why they are urgent etc | 12:23 |
| *** baoli_ has joined #openstack-neutron | 12:23 | |
| *** yfried has joined #openstack-neutron | 12:26 | |
| ihrachys | amuller: neutron-metadata-agent assumes Client.get_auth_info() works ok if Client is instantiated with token= predefined, there was a bug with this scenario, it was fixed in master but was not released yet. Now neutron assumes it works ok and uses API, but since new client hasn't hit PyPI, and requirements.txt were not updated (for there is no new client), we get UnexpectedError raised in | 12:26 |
| ihrachys | metadata-agent, and metadata requests fail. | 12:26 |
| *** yfried has quit IRC | 12:26 | |
| amuller | ihrachys: I meant in the email thread =D | 12:26 |
| ihrachys | amuller: ah, ok :) | 12:28 |
| *** leseb has joined #openstack-neutron | 12:29 | |
| *** yfried1 has joined #openstack-neutron | 12:30 | |
| *** leseb has quit IRC | 12:31 | |
| *** leseb has joined #openstack-neutron | 12:31 | |
| yfried1 | jianingy1afk: ping | 12:32 |
| *** IanGovett has joined #openstack-neutron | 12:34 | |
| *** jgallard has joined #openstack-neutron | 12:34 | |
| ihrachys | amotoki: we should release new client asap. do you know how to achieve this? | 12:34 |
| amotoki | ihrachys: unfortunately I don't have a power to release the client. | 12:35 |
| amotoki | ihrachys: I think markmcclain will be online in a couple of hours. Otherwise we can request it in today's neutron team meeting. | 12:36 |
| pcm_ | anyone: I want to update the patch for a review someone else has out so that it is updated with latest. How do I do that? | 12:36 |
| pcm_ | (essentially rebase with master latest) | 12:37 |
| ihrachys | amotoki: ok, let's wait a bit more. | 12:38 |
| ihrachys | amotoki: but yes, the issue should be raised if not fixed | 12:39 |
| ihrachys | amotoki: I guess gongysh can also release new clients (since he's a Team Lead), right? | 12:40 |
| *** rohit404 has quit IRC | 12:41 | |
| amotoki | ihrachys: I am not sure. neutronclient is alwasy released by neutron ptl so far (danwent and mark). | 12:41 |
| amotoki | ihrachys: most important point is who can upload python package to PyPI. | 12:42 |
| ihrachys | ok, I see. then all we have is patience :) | 12:43 |
| *** leseb has quit IRC | 12:46 | |
| *** baoli_ has quit IRC | 12:47 | |
| *** leseb has joined #openstack-neutron | 12:47 | |
| *** leseb has quit IRC | 12:47 | |
| *** baoli has joined #openstack-neutron | 12:47 | |
| *** leseb has joined #openstack-neutron | 12:47 | |
| openstackgerrit | Oleg Bondarev proposed a change to openstack/neutron: Delete disassociated floating ips on external network deletion https://review.openstack.org/53364 | 12:51 |
| openstackgerrit | Oleg Bondarev proposed a change to openstack/neutron: LBaaS: check for associations before deleting health monitor https://review.openstack.org/53821 | 12:52 |
| pcm_ | amotoki: ping | 12:53 |
| amotoki | pcm_: pong | 12:53 |
| pcm_ | amotoki: do you know how to update someone else's review? | 12:53 |
| openstackgerrit | enikanorov proposed a change to openstack/neutron: Validate rule uuids provided for update_policy https://review.openstack.org/74023 | 12:54 |
| pcm_ | amotoki: Nachi has a review out and I need to push mine out dependent on his, but his is way out of date to master. | 12:54 |
| amotoki | pcm_: we can update other's review unless it is abandoned. | 12:55 |
| pcm_ | amotoki: haven't been able to get ahold of Nachi. Wanted to update his review with a rebase to master. | 12:55 |
| pcm_ | it's active | 12:55 |
| pcm_ | how do I go about that? | 12:55 |
| amotoki | pcm_: It is really same as usual. | 12:55 |
| pcm_ | Do I pull a new branch off master, pull in his patches, and then commit/review and uses the same chagne ID? | 12:56 |
| enikanorov_ | yes | 12:56 |
| pcm_ | cool. Just didn't want to muck it up. | 12:56 |
| enikanorov_ | in fact you just pull his change | 12:56 |
| enikanorov_ | rebase it and push back to gerrit | 12:56 |
| amotoki | the easiest way is to checkout his patch and then rebase it to the latest master. | 12:57 |
| pcm_ | thanks guys, I'll give it a whirl | 12:58 |
| pcm_ | amotoki: I used the "checkout" from github, pasted that into my workspace. It say's I'm in detached head state. Is that right? | 12:59 |
| openstackgerrit | Oleg Bondarev proposed a change to openstack/neutron: LBaaS: check for associations before deleting health monitor https://review.openstack.org/53821 | 13:00 |
| amotoki | pcm_: "checkout" I mean is the "checkout" command line of gerrit review. | 13:01 |
| *** markvoelker has joined #openstack-neutron | 13:01 | |
| amotoki | pcm_: you must see "download" link in the gerrit reivew. After that you are in detached head, | 13:02 |
| amotoki | pcm_: so you need to create a branch with the same name (bug/xxxx or bp/xxxx). | 13:03 |
| amotoki | it is recommended but i think we can update the patch even in a detached branch. | 13:03 |
| pcm_ | so I did the checkout (pasted the link from 'checkout' and am in detached head with branch refs/changes/27/41827/20 | 13:04 |
| *** shausy has quit IRC | 13:04 | |
| pcm_ | So do checkout -b pcm/bp/... | 13:05 |
| *** shausy has joined #openstack-neutron | 13:05 | |
| pcm_ | or do I have to match the same branch name as what Nachi had? | 13:05 |
| pcm_ | bp/vpn-service-types-integration | 13:05 |
| openstackgerrit | enikanorov proposed a change to openstack/neutron: LBaaS: Fix discrepancy between status and admin_state_up https://review.openstack.org/55032 | 13:05 |
| amotoki | pcm_: either can work.. I usually the same name. | 13:06 |
| amotoki | /usually/usually use/ | 13:06 |
| openstackgerrit | Ann Kamyshnikova proposed a change to openstack/neutron: Sync models with migrations https://review.openstack.org/55411 | 13:07 |
| pcm_ | amotoki: ok did that. Will rebase now. | 13:07 |
| *** yfried1 has quit IRC | 13:09 | |
| openstackgerrit | Evgeny Fedoruk proposed a change to openstack/neutron: New SSL extension https://review.openstack.org/74031 | 13:11 |
| *** rohit404 has joined #openstack-neutron | 13:13 | |
| *** dave_tucker_zzz is now known as dave_tucker | 13:19 | |
| *** bvandenh has joined #openstack-neutron | 13:19 | |
| *** alexpilotti has joined #openstack-neutron | 13:22 | |
| *** alagalah has joined #openstack-neutron | 13:23 | |
| *** yfried has joined #openstack-neutron | 13:25 | |
| openstackgerrit | A change was merged to openstack/neutron: Skip a test for nicira service plugin https://review.openstack.org/73832 | 13:28 |
| pcm_ | amotoki: enikanorov_: Thanks! Seems to be working. Will push up for review once I run tox | 13:39 |
| pcm_ | mestery: ping | 13:40 |
| openstackgerrit | enikanorov proposed a change to openstack/neutron: Validate rule uuids provided for update_policy https://review.openstack.org/74023 | 13:40 |
| *** salv-orlando has quit IRC | 13:43 | |
| *** yamahata has joined #openstack-neutron | 13:47 | |
| *** jecarey has quit IRC | 13:47 | |
| *** clev has quit IRC | 13:50 | |
| *** rohit404 has quit IRC | 13:52 | |
| openstackgerrit | Miguel Angel Ajo proposed a change to openstack/neutron: Provide agent status via status file (WIP) https://review.openstack.org/74045 | 14:00 |
| *** ramishra_ has joined #openstack-neutron | 14:01 | |
| *** jecarey has joined #openstack-neutron | 14:01 | |
| *** ramishra has quit IRC | 14:02 | |
| *** ramishra_ has quit IRC | 14:06 | |
| *** WackoRobie has joined #openstack-neutron | 14:06 | |
| *** alagalah has quit IRC | 14:08 | |
| *** julim has joined #openstack-neutron | 14:09 | |
| openstackgerrit | Will proposed a change to openstack/neutron: Add z/VM mechanism driver to support IBM z/VM agent https://review.openstack.org/73187 | 14:15 |
| *** ramishra has joined #openstack-neutron | 14:15 | |
| *** saju_m has quit IRC | 14:15 | |
| *** banix has joined #openstack-neutron | 14:16 | |
| *** irenab has quit IRC | 14:20 | |
| *** feleouet has joined #openstack-neutron | 14:21 | |
| *** leseb has quit IRC | 14:27 | |
| *** ramishra has quit IRC | 14:31 | |
| ajo | there is some way to indicate a patchset is "WIP" (work in progress) that avoids jenkins tests?, | 14:34 |
| ajo | I have published a very preliminary patchset minutes ago, but I realized that jenkins is testing.. | 14:34 |
| *** dguitarbite has joined #openstack-neutron | 14:38 | |
| *** alagalah has joined #openstack-neutron | 14:38 | |
| openstackgerrit | enikanorov proposed a change to openstack/neutron: Validate rule uuids provided for update_policy https://review.openstack.org/74023 | 14:38 |
| enikanorov_ | ajo: No | 14:39 |
| enikanorov_ | ajo: it will still run the tests | 14:39 |
| openstackgerrit | Cedric Brandily proposed a change to openstack/neutron: Allow to delegate some tunnel provider attributes choice to neutron https://review.openstack.org/74055 | 14:39 |
| ajo | enikanorov_, thanks, wouldn't it make sense? (for example, you want to share the code with others, to get feedback, but's in an early state) | 14:40 |
| ajo | but on the other hand, if there is compute power available, the tests are good, even if early | 14:40 |
| *** alagalah has quit IRC | 14:40 | |
| *** alagalah_ has joined #openstack-neutron | 14:40 | |
| openstackgerrit | Evgeny Fedoruk proposed a change to openstack/neutron: New SSL extension https://review.openstack.org/74031 | 14:40 |
| *** dkehn__ has joined #openstack-neutron | 14:41 | |
| *** yamahata has quit IRC | 14:43 | |
| *** dkehn has quit IRC | 14:43 | |
| *** yamahata has joined #openstack-neutron | 14:44 | |
| *** dkehn_ has quit IRC | 14:44 | |
| *** aymenfrikha has joined #openstack-neutron | 14:45 | |
| *** dkehn has joined #openstack-neutron | 14:46 | |
| *** afazekas has quit IRC | 14:46 | |
| *** clev has joined #openstack-neutron | 14:46 | |
| *** dkehn__ is now known as dkehn_ | 14:46 | |
| *** aymenfrikha has quit IRC | 14:50 | |
| *** peristeri has joined #openstack-neutron | 14:52 | |
| *** dguitarbite has quit IRC | 14:54 | |
| *** dguitarbite has joined #openstack-neutron | 14:59 | |
| *** rotbeard has quit IRC | 15:00 | |
| *** skraynev is now known as skraynev_afk | 15:02 | |
| *** julim_ has joined #openstack-neutron | 15:02 | |
| *** skraynev_afk is now known as skraynev_afk_goi | 15:03 | |
| *** skraynev_afk_goi is now known as skraynev_afk | 15:03 | |
| *** aymenfrikha has joined #openstack-neutron | 15:04 | |
| *** ygbo has quit IRC | 15:04 | |
| *** otherwiseguy has joined #openstack-neutron | 15:04 | |
| *** julim has quit IRC | 15:05 | |
| *** chandan_kumar has quit IRC | 15:07 | |
| *** yfried has quit IRC | 15:07 | |
| *** rustlebee is now known as russellb | 15:08 | |
| *** rwsu has joined #openstack-neutron | 15:09 | |
| *** matrohon_ has joined #openstack-neutron | 15:12 | |
| *** dguitarbite has quit IRC | 15:13 | |
| *** bada has quit IRC | 15:13 | |
| *** matrohon_ has quit IRC | 15:14 | |
| *** bada has joined #openstack-neutron | 15:15 | |
| openstackgerrit | Xiang Hui proposed a change to openstack/neutron: Add options for some commands executing run_vsctl https://review.openstack.org/58954 | 15:15 |
| *** ygbo has joined #openstack-neutron | 15:16 | |
| *** dguitarbite has joined #openstack-neutron | 15:18 | |
| *** bada_ has joined #openstack-neutron | 15:18 | |
| *** leseb has joined #openstack-neutron | 15:19 | |
| *** yfried has joined #openstack-neutron | 15:19 | |
| *** bada has quit IRC | 15:20 | |
| *** alagalah_ has quit IRC | 15:20 | |
| *** yfried has quit IRC | 15:22 | |
| *** yfried has joined #openstack-neutron | 15:23 | |
| *** dguitarbite has quit IRC | 15:24 | |
| *** ygbo has quit IRC | 15:26 | |
| *** rohit404 has joined #openstack-neutron | 15:26 | |
| *** shausy has quit IRC | 15:26 | |
| openstackgerrit | Evgeny Fedoruk proposed a change to openstack/neutron: New SSL extension https://review.openstack.org/74031 | 15:27 |
| *** dguitarbite has joined #openstack-neutron | 15:29 | |
| openstackgerrit | Xiang Hui proposed a change to openstack/neutron: Add options for some commands executing run_vsctl https://review.openstack.org/58954 | 15:30 |
| *** amuller has quit IRC | 15:30 | |
| *** enikanorov has joined #openstack-neutron | 15:31 | |
| *** ramishra has joined #openstack-neutron | 15:32 | |
| *** ramishra has quit IRC | 15:37 | |
| *** ygbo has joined #openstack-neutron | 15:38 | |
| *** rkukura has quit IRC | 15:38 | |
| *** cjellick has joined #openstack-neutron | 15:42 | |
| *** cjellick has quit IRC | 15:43 | |
| *** ramishra has joined #openstack-neutron | 15:44 | |
| *** cjellick has joined #openstack-neutron | 15:47 | |
| *** beagles is now known as beagles_brb | 15:50 | |
| openstackgerrit | Kyle Mestery proposed a change to openstack/neutron: Add additional unit tests for the ML2 plugin https://review.openstack.org/74079 | 15:52 |
| *** ramishra has quit IRC | 15:55 | |
| *** alagalah has joined #openstack-neutron | 16:00 | |
| *** salv-orlando has joined #openstack-neutron | 16:00 | |
| *** amotoki has quit IRC | 16:03 | |
| *** ttx has joined #openstack-neutron | 16:18 | |
| *** ttx has left #openstack-neutron | 16:18 | |
| openstackgerrit | Evgeny Fedoruk proposed a change to openstack/neutron: New SSL extension https://review.openstack.org/74031 | 16:20 |
| *** jgrimm has joined #openstack-neutron | 16:20 | |
| *** rohit404 has quit IRC | 16:21 | |
| *** rohit404 has joined #openstack-neutron | 16:21 | |
| *** bvandenh has quit IRC | 16:23 | |
| *** chandan_kumar has joined #openstack-neutron | 16:29 | |
| *** chandan_kumar has quit IRC | 16:29 | |
| *** chandan_kumar has joined #openstack-neutron | 16:30 | |
| *** skraynev_afk is now known as skraynev | 16:31 | |
| *** Gamekiller77 has joined #openstack-neutron | 16:32 | |
| *** yfried has quit IRC | 16:33 | |
| *** bvandenh has joined #openstack-neutron | 16:35 | |
| *** jistr has quit IRC | 16:35 | |
| *** rohit404_ has joined #openstack-neutron | 16:36 | |
| *** rohit404 has quit IRC | 16:36 | |
| *** rohit404_ is now known as rohit404 | 16:36 | |
| *** alexpilotti has quit IRC | 16:38 | |
| *** alexpilotti has joined #openstack-neutron | 16:38 | |
| *** cjellick1 has joined #openstack-neutron | 16:38 | |
| *** cjellick1 has quit IRC | 16:39 | |
| *** rkukura has joined #openstack-neutron | 16:41 | |
| *** cjellick has quit IRC | 16:42 | |
| pcm_ | enikanorov_: ping | 16:45 |
| *** chandan_kumar has quit IRC | 16:55 | |
| *** dave_tucker is now known as dave_tucker_zzz | 16:55 | |
| *** vkozhukalov has quit IRC | 17:00 | |
| *** jlibosva has quit IRC | 17:06 | |
| *** jaypipes has quit IRC | 17:07 | |
| openstackgerrit | Paul Michali proposed a change to openstack/neutron: VPNaaS integration with service type framework https://review.openstack.org/41827 | 17:08 |
| openstackgerrit | Cedric Brandily proposed a change to openstack/neutron: Allow to delegate some tunnel provider attributes choice to neutron https://review.openstack.org/74055 | 17:16 |
| *** beagles_brb is now known as beagles | 17:16 | |
| *** sn6i23a has joined #openstack-neutron | 17:17 | |
| *** clev has quit IRC | 17:18 | |
| *** clev has joined #openstack-neutron | 17:19 | |
| *** ianw has quit IRC | 17:23 | |
| *** ianw has joined #openstack-neutron | 17:24 | |
| *** clev has quit IRC | 17:25 | |
| *** bjornar has joined #openstack-neutron | 17:26 | |
| *** clev has joined #openstack-neutron | 17:29 | |
| *** jgallard has quit IRC | 17:33 | |
| openstackgerrit | Arvind Somya proposed a change to openstack/neutron: Cisco APIC ML2 mechanism driver, part 2 https://review.openstack.org/73372 | 17:34 |
| openstackgerrit | Arvind Somya proposed a change to openstack/neutron: Cisco APIC ML2 mechanism driver, part 1 https://review.openstack.org/73355 | 17:34 |
| *** ygbo has quit IRC | 17:35 | |
| *** SumitNaiksatam has quit IRC | 17:37 | |
| *** RajeshMohan has quit IRC | 17:37 | |
| *** RajeshMohan has joined #openstack-neutron | 17:39 | |
| *** marun has joined #openstack-neutron | 17:41 | |
| enikanorov_ | pcm_: pong | 17:46 |
| *** rossella-s has quit IRC | 17:46 | |
| *** leseb has quit IRC | 17:49 | |
| *** leseb has joined #openstack-neutron | 17:50 | |
| openstackgerrit | enikanorov proposed a change to openstack/neutron: Validate rule uuids provided for update_policy https://review.openstack.org/74023 | 17:50 |
| *** clev has quit IRC | 17:50 | |
| *** jaypipes has joined #openstack-neutron | 17:53 | |
| *** leseb has quit IRC | 17:54 | |
| *** SumitNaiksatam has joined #openstack-neutron | 17:58 | |
| *** mlavalle has joined #openstack-neutron | 17:59 | |
| *** cjellick has joined #openstack-neutron | 18:08 | |
| *** jp_at_hp has quit IRC | 18:08 | |
| *** alagalah has quit IRC | 18:08 | |
| *** alagalah has joined #openstack-neutron | 18:09 | |
| pcm_ | Had some database related Qs. We talked before about mapping UUIDs and I created a table to hold those mappings. I have code that queries another table and then either queries this table or inserts into it. All this is done in a single session. | 18:12 |
| pcm_ | enikanorov_: reping | 18:12 |
| pcm_ | enikanorov_: Sorry I missed your pong. | 18:13 |
| pcm_ | enikanorov_: Have some questions for you regarding database (related) issue I see. | 18:15 |
| *** networkstatic has joined #openstack-neutron | 18:16 | |
| *** alagalah has quit IRC | 18:17 | |
| enikanorov_ | pcm_: what is it? | 18:17 |
| *** alagalah has joined #openstack-neutron | 18:17 | |
| pcm_ | enikanorov_: Don't know if you saw my text a few lines up? | 18:18 |
| * pcm_ I can retype | 18:18 | |
| enikanorov_ | yeah | 18:18 |
| enikanorov_ | i mean i can see it :) | 18:18 |
| pcm_ | enikanorov_: good. So I have it working and have UTs for some lower level functions and the top level function. | 18:19 |
| enikanorov_ | ok | 18:19 |
| pcm_ | enikanorov_: However, when I run the full tox suite or on the module, I get errors. | 18:19 |
| enikanorov_ | pls paste them to paste | 18:20 |
| pcm_ | enikanorov_: Do the tests run in parallel at the per test case level or class level? | 18:20 |
| pcm_ | enikanorov_: Sure I can paste (working on pushing for review...but the testing and some other issues are preventing me right now) | 18:21 |
| pcm_ | enikanorov_: http://paste.openstack.org/show/66573/ | 18:22 |
| enikanorov_ | looking | 18:23 |
| *** gongysh has joined #openstack-neutron | 18:23 | |
| enikanorov_ | oh, i ean can you paste those errors? | 18:23 |
| pcm_ | enikanorov_: sorry - it's huge... the tests for the mapping are in £TestCiscoIPsecDriverMapping | 18:24 |
| *** SumitNaiksatam has quit IRC | 18:24 | |
| pcm_ | enikanorov_: I was wondering if I needed to put some of the test cases in a transaction | 18:24 |
| enikanorov_ | ok, lets wait for jenkings to run uts | 18:25 |
| *** gongysh has quit IRC | 18:29 | |
| *** suresh12 has joined #openstack-neutron | 18:30 | |
| *** alagalah has quit IRC | 18:31 | |
| *** alagalah has joined #openstack-neutron | 18:31 | |
| *** markmcclain has joined #openstack-neutron | 18:33 | |
| *** markmcclain has quit IRC | 18:33 | |
| *** carl_baldwin has joined #openstack-neutron | 18:33 | |
| *** dguitarbite has quit IRC | 18:37 | |
| enikanorov_ | jaypipes: ping | 18:39 |
| jaypipes | enikanorov: pong | 18:39 |
| enikanorov_ | oh cool you're here | 18:39 |
| *** markmcclain has joined #openstack-neutron | 18:40 | |
| *** suresh12_ has joined #openstack-neutron | 18:41 | |
| *** suresh12 has quit IRC | 18:42 | |
| *** morganfainberg_Z is now known as morganfainberg | 18:42 | |
| enikanorov_ | markmcclain: Hi. Will you have time today to continue the discussion on lb instance? I've got wiki page & pictures ready | 18:43 |
| *** vkozhukalov has joined #openstack-neutron | 18:44 | |
| markmcclain | enikanorov_: in a bit.. say 1h15m from now? | 18:44 |
| enikanorov_ | markmcclain: good, works for me | 18:44 |
| *** carl_baldwin has quit IRC | 18:51 | |
| *** mlavalle has quit IRC | 18:53 | |
| *** mlavalle has joined #openstack-neutron | 18:53 | |
| *** openstackgerrit has quit IRC | 18:53 | |
| *** openstackgerrit has joined #openstack-neutron | 18:54 | |
| *** clev has joined #openstack-neutron | 18:54 | |
| *** dave_tucker_zzz is now known as dave_tucker | 18:54 | |
| *** bjornar has quit IRC | 18:56 | |
| *** bjornar has joined #openstack-neutron | 18:58 | |
| openstackgerrit | Miguel Angel Ajo proposed a change to openstack/neutron: Provide agent status via status file (WIP) https://review.openstack.org/74045 | 18:58 |
| *** Gamekiller77 has quit IRC | 18:59 | |
| *** Gamekiller77 has joined #openstack-neutron | 19:00 | |
| *** dave_tucker is now known as dave_tucker_zzz | 19:00 | |
| *** otherwiseguy has quit IRC | 19:03 | |
| *** jprovazn has joined #openstack-neutron | 19:04 | |
| *** dguitarbite has joined #openstack-neutron | 19:06 | |
| *** jog0 has left #openstack-neutron | 19:17 | |
| openstackgerrit | Sandhya Dasu proposed a change to openstack/neutron: Implements a ML2 mechanism driver for Cisco UCS Manager. https://review.openstack.org/74134 | 19:24 |
| *** mjbot has quit IRC | 19:25 | |
| *** mriedem has joined #openstack-neutron | 19:25 | |
| mriedem | salv-orlando: ping | 19:25 |
| mriedem | or anyone i guess :) | 19:26 |
| mriedem | newb question | 19:26 |
| mriedem | if i want to filter the response when calling neutronclient.list_floatingips I just pass in kwarg filters=['x','y','z']? | 19:27 |
| mriedem | something like that? | 19:27 |
| mriedem | say i want to filter list_floatingips by the 'id' field | 19:27 |
| mriedem | oops, fields=[], not filters | 19:27 |
| mriedem | nevermind, answered my own questoin | 19:30 |
| enikanorov_ | filter={'id': [ids]} | 19:34 |
| mriedem | yeah, i meant fields | 19:34 |
| enikanorov_ | fields parameter filters fields returned, not rows | 19:34 |
| enikanorov_ | ok | 19:34 |
| mriedem | i want all floating_ips but want to keep the response small, so only fields=['id'] | 19:35 |
| enikanorov_ | aha, fileds is what you need | 19:35 |
| mriedem | yup, thanks | 19:35 |
| mriedem | trying to resurrect this nova patch https://review.openstack.org/#/c/43822/ | 19:35 |
| mriedem | it's only been restored like 3 or 4 times now... | 19:35 |
| *** yfried has joined #openstack-neutron | 19:40 | |
| mtreinish | mriedem: wow >30k changes since then and it was from the summer :) | 19:40 |
| mriedem | mtreinish: what are you doing lurking in here? | 19:40 |
| mriedem | mtreinish: that patch should also fix a tempest fail with neutron too...that's why i care | 19:41 |
| *** armax has joined #openstack-neutron | 19:42 | |
| mtreinish | mriedem: don't we just skip those tests? | 19:42 |
| mriedem | mtreinish: yeah, because of the bug :) | 19:43 |
| *** Gamekiller77 has quit IRC | 19:43 | |
| mriedem | nova isn't getting quota information from neutron | 19:43 |
| *** zzelle has joined #openstack-neutron | 19:43 | |
| mriedem | and as long as v3 is expiremental this is something we want in nova to interact with neutron | 19:44 |
| mriedem | i should say, as long as nova v2 is not deprecated | 19:44 |
| mtreinish | yeah, it's one of several pairity issues | 19:44 |
| mtreinish | well v3 is going to have nova network stuff in it too | 19:44 |
| *** otherwiseguy has joined #openstack-neutron | 19:44 | |
| mtreinish | doesn't that mean we'll have to worry about this for v3 too? | 19:44 |
| mriedem | shiiii, maybe? | 19:44 |
| mriedem | i'd have to ask cyeoh | 19:45 |
| mriedem | the v3 api king | 19:45 |
| *** Gamekiller77 has joined #openstack-neutron | 19:45 | |
| *** suresh12_ has quit IRC | 19:53 | |
| openstackgerrit | Rajesh Mohan proposed a change to openstack/neutron: Firewall as a Service (FWaaS) Service Insertion https://review.openstack.org/62599 | 19:53 |
| *** suresh12 has joined #openstack-neutron | 19:53 | |
| *** alagalah has quit IRC | 19:57 | |
| *** alagalah has joined #openstack-neutron | 19:57 | |
| *** djoreilly has quit IRC | 19:58 | |
| *** suresh12 has quit IRC | 19:58 | |
| enikanorov_ | markmcclain: ping | 20:00 |
| markmcclain | enikanorov_: pong | 20:00 |
| enikanorov_ | markmcclain: so here are pictures i've prepared: https://wiki.openstack.org/wiki/Neutron/LBaaS/LoadbalancerInstance/Discussion | 20:01 |
| enikanorov_ | i'm mostly interested in discussing (2) and (3) | 20:01 |
| * markmcclain looking | 20:01 | |
| enikanorov_ | as far as i understood your proposal, it looks like (3) to me | 20:01 |
| enikanorov_ | please take a look | 20:01 |
| *** alagalah has quit IRC | 20:02 | |
| pcm_ | mestery: ping | 20:02 |
| mestery | pcm_: pong hi there | 20:02 |
| pcm_ | mestery: had a question on pushing for review. | 20:03 |
| mestery | pcm_: Shoot! | 20:03 |
| pcm_ | mestery: did a git review -d of another review, created a branch, patched in my changes and tested it out | 20:04 |
| pcm_ | mestery: did a git review, and it says that I have 2 outstanding commits (one is mine and one is the dependency). | 20:04 |
| mestery | OK | 20:04 |
| pcm_ | mestery: do I just say "YES" or did I miss some step? | 20:04 |
| *** suresh12 has joined #openstack-neutron | 20:05 | |
| mestery | I think it's fine to say yes. Worst case, it will push a new version of hte dependent patch, which is mostly harmless as the author will remain the same. | 20:05 |
| *** markwash has joined #openstack-neutron | 20:05 | |
| pcm_ | I just pushed the dependent commit earlier today (Nachi's review that I need). | 20:06 |
| markmcclain | enikanorov_: 3 is closer | 20:06 |
| *** dave_tucker_zzz is now known as dave_tucker | 20:06 | |
| pcm_ | Concerned as my next step is to do the device driver which will depend on this one. | 20:06 |
| enikanorov_ | markmcclain: so to understand your proposal better, could you provide the workflow for the following case: | 20:06 |
| enikanorov_ | 1st vip -> 1st pool, 2nd vip -> 1st pool and 2nd vip-> 2nd pool (via L7). | 20:06 |
| enikanorov_ | e.g. 2 vips, 2 pools, all objects are on the same logical config. | 20:07 |
| pcm_ | mestery: Just don't want to mess this up... | 20:07 |
| mestery | pcm_: I think you will be ok. I've done this before, and like I said, worst case is it pushes another rev out. | 20:07 |
| mestery | pcm_: You are doing great stuff here, no worries my friend. :) | 20:07 |
| pcm_ | mestery: OK here goes. | 20:07 |
| * pcm_ load explosions | 20:07 | |
| openstackgerrit | Paul Michali proposed a change to openstack/neutron: VPNaaS Service Driver for Cisco CSR https://review.openstack.org/74144 | 20:08 |
| pcm_ | mestery: Looks like it went out. | 20:08 |
| pcm_ | mestery: thanks | 20:09 |
| mestery | pcm_: Nice work! | 20:09 |
| markmcclain | enikanorov_: not sure I'm following 100% | 20:10 |
| markmcclain | only the 2nd case uses L7 and the first does not? | 20:10 |
| enikanorov_ | right, one vip handles only 1 pool, another vip handles 2 pools, one default and one via L7 rules | 20:10 |
| enikanorov_ | the question is how to tell neutron that all these objects should be on one logical instance | 20:11 |
| markmcclain | enikanorov_: the user does not care | 20:12 |
| enikanorov_ | the user may not care | 20:13 |
| enikanorov_ | but we care :) | 20:13 |
| ihrachys | markmcclain: hi. we got the first Critical bug reported due to missing new client release: https://bugs.launchpad.net/neutron/+bug/1280941 | 20:13 |
| markmcclain | the vip defines the endpoint using the terminology from #3 the listener defines the relationship to the pool | 20:13 |
| markmcclain | ihrachys: sorry, I've have to rekey everything I use to sign releases | 20:14 |
| enikanorov_ | #3 solves the workflow more or less, but it still makes user to care about logical config | 20:14 |
| enikanorov_ | the root object is now VIP | 20:14 |
| enikanorov_ | and to create the pool, you need to pass vip_id | 20:14 |
| markmcclain | not seeing how the use cares about the logical config | 20:15 |
| markmcclain | I'd argue we could drop that semantic | 20:15 |
| markmcclain | and just create a pool | 20:15 |
| enikanorov_ | ok, so that will be pure-db object at that point | 20:15 |
| enikanorov_ | right? | 20:15 |
| markmcclain | yes | 20:15 |
| enikanorov_ | ok, then we assign it to a listener via l7 rules | 20:16 |
| enikanorov_ | and it get associated with existing lopgical config | 20:16 |
| enikanorov_ | ? | 20:16 |
| markmcclain | yes | 20:16 |
| markmcclain | so consider the case where vip1 and vip2 get assigned to two different backends | 20:17 |
| markmcclain | and but share the same pool | 20:17 |
| enikanorov_ | that means two backends will share the same pool configuration | 20:18 |
| markmcclain | the backend implementation has two choices… 1) reschedule of the vips onto other's instance or 2) duplicate the pools | 20:18 |
| *** carl_baldwin has joined #openstack-neutron | 20:18 | |
| ihrachys | markmcclain: any eta? :) I'm keen to update requirements.txt in global and neutron... ;) | 20:18 |
| enikanorov_ | in fact i'd say that the ability to get them deployed to single backedn is useful | 20:18 |
| markmcclain | ihrachys: as soon as I get my pub key published | 20:19 |
| *** carl_baldwin has quit IRC | 20:19 | |
| *** skraynev is now known as skraynev_afk | 20:19 | |
| markmcclain | enikanorov_: it is useful | 20:19 |
| markmcclain | but from a user perspective that is an implementation detail | 20:19 |
| enikanorov_ | markmcclain: yes, it should bnot be a requirement | 20:19 |
| ihrachys | markmcclain: is that on pypi side? | 20:19 |
| enikanorov_ | markmcclain: so this way the problems are addressed, but the biggest issue is that it's complete redesign of object model and API | 20:20 |
| markmcclain | ihrachys: I have push signed tags to kick off the publishing | 20:20 |
| enikanorov_ | otherwise I would be happy to go with it | 20:20 |
| *** dave_tucker is now known as dave_tucker_zzz | 20:20 | |
| markmcclain | ihrachys: to multiple places | 20:20 |
| markmcclain | enikanorov_: we are allowed to adapt the object model over time as the understand the problem space better | 20:21 |
| markmcclain | enikanorov_: remember the original object model was craved out over a week last Feb | 20:21 |
| enikanorov_ | markmcclain: well... but the public API breaks too | 20:22 |
| enikanorov_ | is it acceptable? | 20:22 |
| *** irenab has joined #openstack-neutron | 20:23 | |
| enikanorov_ | i remember tempest folks gave me no-no for changing the api... | 20:23 |
| markmcclain | I still don't see how the public API breaks if we make the appropriate mappings from new to old | 20:23 |
| markmcclain | within a major version (v2 in our case) we have to maintain compatibility | 20:23 |
| enikanorov_ | because the workflow for #3 is different then how we create configuration right now | 20:23 |
| markmcclain | it's different under the hood | 20:24 |
| markmcclain | but from the API, I don't see anything that would prevent us having a proper compatibility layer | 20:24 |
| enikanorov_ | ok, i'll think about it more. currently i'm not so sure we can maintain compatibility going with #3 | 20:25 |
| enikanorov_ | but at least it provides the same model as an 'instance' approach | 20:25 |
| enikanorov_ | which is good | 20:25 |
| *** suresh12 has quit IRC | 20:26 | |
| markmcclain | yeah even in #3 I still think we need to look at how we express the relationships of the pools | 20:27 |
| enikanorov_ | markmcclain: btw, looks like neither libra nor aws don't have API flexible enough to support multiple vips and multiple pools at once | 20:27 |
| markmcclain | I'm concerned that in some cases we'll have to know to go through l7 mappings and in others we have to detect to skip them | 20:27 |
| enikanorov_ | markmcclain: which cases ae you thinking about? | 20:28 |
| mriedem | arosen1: enikanorov: ping when you have a minute | 20:28 |
| markmcclain | enikanorov_: yeah supporting both increases complexity | 20:28 |
| *** suresh12 has joined #openstack-neutron | 20:29 | |
| enikanorov_ | markmcclain: ..and looks like both aws and libra has the 'loadbalancer' notion, but it maps to our 'pool' | 20:29 |
| enikanorov_ | so i really didn't have the concern of letting the user to care about logical config | 20:29 |
| enikanorov_ | because they do care already, working with the pool, rather then with instance/vip we're trying to introduce | 20:30 |
| enikanorov_ | so i'm just saying is this concern big enough to redefing the API and find a way to make it compatible | 20:31 |
| markmcclain | enikanorov_: they're probably only concerned with the pool now because that is the direction we steered them | 20:31 |
| *** rohit404_ has joined #openstack-neutron | 20:31 | |
| *** rohit404 has quit IRC | 20:31 | |
| *** rohit404_ is now known as rohit404 | 20:31 | |
| enikanorov_ | those who work with single pool will not change the workflow and will not notice the instance | 20:31 |
| enikanorov_ | and for those who need more than 1 pool instance_id is what attaching new object to existing conf | 20:32 |
| markmcclain | not sure I'm following the last statement | 20:33 |
| *** jroovers|afk has joined #openstack-neutron | 20:33 | |
| *** clev has quit IRC | 20:33 | |
| enikanorov_ | i mean that for the exampel above with two pools | 20:33 |
| enikanorov_ | you create the 1st one, the instance created for it automaticaly, and then you use instance_id to create 2nd pool | 20:34 |
| enikanorov_ | so they will be colocated on the backend | 20:34 |
| enikanorov_ | i'm not sure why do we need to isolate the user of such knowledge | 20:35 |
| enikanorov_ | (colocation) | 20:35 |
| *** jroovers has quit IRC | 20:35 | |
| enikanorov_ | user don't care about the backend, but they know how logical configuration envelops created objects | 20:37 |
| enikanorov_ | and, say, they can migrate configuration from flavor to flavor, router, agent, etc | 20:37 |
| ihrachys | markmcclain: ok, thanks, keep me informed :) btw there is also 'Urgentpython-neutronclientrelease needed' thread in openstack-dev. | 20:39 |
| *** itzikb has joined #openstack-neutron | 20:40 | |
| *** bada_ has quit IRC | 20:41 | |
| enikanorov_ | markmcclain: in #3 such logical instance is simply a VIP, and it is not what current VIP is. that's why i'm not sure how we can be bw compatible | 20:41 |
| *** jlibosva has joined #openstack-neutron | 20:42 | |
| *** jecarey has quit IRC | 20:43 | |
| *** jecarey has joined #openstack-neutron | 20:44 | |
| itzikb | Hi, Can someone please review this patch https://review.openstack.org/#/c/66545/ ? | 20:46 |
| markmcclain | enikanorov_: sorry | 20:47 |
| *** irenab has quit IRC | 20:47 | |
| enikanorov_ | markmcclain: ok, the meeting is soon. I'll think about how #3 could have compatible API with existing one. Currently it seems to me that it can not. I hope to continue discussion after I get more thoughts on that | 20:47 |
| markmcclain | enikanorov_: ok.. I'll read the scroll back and get caught up | 20:48 |
| *** jecarey has quit IRC | 20:49 | |
| jlibosva | markmcclain: hello | 20:50 |
| *** sweston has joined #openstack-neutron | 20:50 | |
| sweston | beagles: ping | 20:51 |
| beagles | sweston: pong | 20:52 |
| *** jecarey has joined #openstack-neutron | 20:52 | |
| *** gongysh has joined #openstack-neutron | 20:52 | |
| mriedem | arosen1: shouldn't 'fields' here be specying a list? https://github.com/openstack/nova/blob/master/nova/network/neutronv2/api.py#L818 | 20:52 |
| mriedem | *specifying | 20:52 |
| *** ajo has quit IRC | 20:52 | |
| sweston | beagles: hey there ;-). I finally had some time this weekend to review the NovaNeutronGap document ... it was a rough week | 20:53 |
| sweston | beagles: if you have time to chat about it soon, I am available | 20:53 |
| beagles | sweston: okay. I should have a bit of time right after the neutron IRC meeting | 20:54 |
| sweston | beagles: excellent! | 20:54 |
| beagles | 342904 | 20:54 |
| beagles | whoops | 20:55 |
| gongysh | mriedem: hi, L818 will be transported into networks.json?fields=xxx&router:external=true | 20:57 |
| mriedem | gongysh: but i thought fields was supposed to be a list? | 20:57 |
| gongysh | mriedem: if it is a list, it will be ?fields=xxx&fields=xxx2&router:external=true.. | 20:58 |
| mriedem | where does the client handle that? | 20:58 |
| openstackgerrit | Hareesh Puthalath proposed a change to openstack/neutron: Configuration agent for Cisco Service VMs https://review.openstack.org/74152 | 20:58 |
| gongysh | mriedem: the server will received it as a list anyway. | 20:58 |
| mriedem | gongysh: ah i see, safe_encode_dict in the client | 21:00 |
| mriedem | handles strings, lists and dicts in the kwargs passed in | 21:00 |
| *** emagana has joined #openstack-neutron | 21:00 | |
| *** amotoki has joined #openstack-neutron | 21:00 | |
| gongysh | mriedem: yes | 21:00 |
| *** SumitNaiksatam has joined #openstack-neutron | 21:01 | |
| *** JPtheJP has quit IRC | 21:03 | |
| *** nati_ueno has joined #openstack-neutron | 21:03 | |
| *** jecarey_ has joined #openstack-neutron | 21:06 | |
| *** nati_ueno has quit IRC | 21:06 | |
| *** nati_ueno has joined #openstack-neutron | 21:06 | |
| *** jecarey has quit IRC | 21:08 | |
| openstackgerrit | Paul Michali proposed a change to openstack/neutron: VPNaaS Device Driver for Cisco CSR https://review.openstack.org/74156 | 21:13 |
| *** ianw has quit IRC | 21:14 | |
| openstackgerrit | Vincent Untz proposed a change to openstack/neutron: Reset the policy after loading extensions https://review.openstack.org/73866 | 21:14 |
| *** ianw has joined #openstack-neutron | 21:15 | |
| *** afazekas has joined #openstack-neutron | 21:21 | |
| *** suresh12 has quit IRC | 21:22 | |
| annegentle | emagana: I know you're in the neutron meeting now, but if you could look at https://review.openstack.org/#/c/72850/ for the Operations Guide it would be much appreciated! | 21:22 |
| emagana | annegentle: I am on it! | 21:23 |
| *** suresh12 has joined #openstack-neutron | 21:23 | |
| *** salv-orlando has quit IRC | 21:25 | |
| *** salv-orlando has joined #openstack-neutron | 21:25 | |
| *** JPtheJP has joined #openstack-neutron | 21:31 | |
| enikanorov_ | SumitNaiksatam: could you please take a look at https://review.openstack.org/#/c/74023/ a fix to a samll issue i've found today, thanks. | 21:35 |
| SumitNaiksatam | sure | 21:36 |
| *** dims has joined #openstack-neutron | 21:41 | |
| openstackgerrit | Miguel Angel Ajo proposed a change to openstack/neutron: Provide agent status via status file (WIP) https://review.openstack.org/74045 | 21:43 |
| *** nati_ueno has quit IRC | 21:49 | |
| *** nati_ueno has joined #openstack-neutron | 21:49 | |
| *** vkozhukalov has quit IRC | 21:53 | |
| *** alagalah has joined #openstack-neutron | 21:53 | |
| openstackgerrit | Miguel Angel Ajo proposed a change to openstack/neutron: Poll the agent child processes status (WIP) https://review.openstack.org/74179 | 21:54 |
| *** jprovazn has quit IRC | 21:55 | |
| *** devlaps1 has joined #openstack-neutron | 21:58 | |
| *** alagalah has quit IRC | 22:00 | |
| *** sputnik13 has joined #openstack-neutron | 22:00 | |
| *** WackoRobie has quit IRC | 22:01 | |
| *** WackoRobie has joined #openstack-neutron | 22:01 | |
| *** alagalah has joined #openstack-neutron | 22:02 | |
| *** nati_ueno has quit IRC | 22:04 | |
| pcm_ | nati_ueno: ping | 22:04 |
| *** dguitarbite_ has joined #openstack-neutron | 22:04 | |
| *** dguitarbite has quit IRC | 22:04 | |
| sweston | beagles: give me 5 minutes? | 22:05 |
| beagles | sweston, sure | 22:05 |
| *** WackoRobie has quit IRC | 22:06 | |
| sweston | beagles: thanks, trying to finish up a bug ;-) | 22:06 |
| *** ecbanks has quit IRC | 22:06 | |
| *** gongysh has quit IRC | 22:06 | |
| jlibosva | markmcclain: hi, do you have a minute? | 22:07 |
| markmcclain | jlibosva: I do | 22:07 |
| markmcclain | what's up? | 22:07 |
| jlibosva | markmcclain: wrt https://review.openstack.org/#/c/69433/3 and offline migration - I added condition for offline migration. Could you please take a look and consider your -2 or discuss other options how to avoid dropping non-existing table? | 22:08 |
| markmcclain | jlibosva: yes I'll take a 2nd look | 22:08 |
| jlibosva | markmcclain: thanks | 22:08 |
| *** pcm_ has left #openstack-neutron | 22:09 | |
| *** armax has left #openstack-neutron | 22:10 | |
| *** jecarey_ has quit IRC | 22:11 | |
| *** bjornar has quit IRC | 22:12 | |
| sweston | beagles: ok, I am ready when you are. Wanna discuss over PM? | 22:13 |
| beagles | sweston, probably best to discuss here in case others want to throw in | 22:14 |
| *** leseb has joined #openstack-neutron | 22:15 | |
| sweston | beagles: sure, that works for me. where would you like to start/ | 22:15 |
| sweston | ? | 22:15 |
| *** ecbanks has joined #openstack-neutron | 22:15 | |
| *** clev has joined #openstack-neutron | 22:16 | |
| *** markmcclain has quit IRC | 22:16 | |
| *** mriedem has left #openstack-neutron | 22:17 | |
| beagles | sweston, let's start with a working definition of parity: to date you could define parity as "whatever" is required to deprecate nova-network, eventually allowing it to be removed | 22:17 |
| beagles | sweston, there are multiple facets to that: functionality/features, stability/performance, quality of default implementation | 22:18 |
| beagles | these are very general categories that can get pretty big when you dig into them... | 22:19 |
| *** julim_ has quit IRC | 22:19 | |
| beagles | sweston: for Icehouse a lot of progress has been made in the 2nd and 3rd items and more to come | 22:19 |
| *** jlibosva has quit IRC | 22:19 | |
| sweston | beagles: so, then this is a question of documenting the differences, or coming up with a plan to address them, or both? | 22:19 |
| beagles | sweston: an implicit aspect of both of these is tempest coverage etc. ... these are all being handled | 22:20 |
| beagles | sweston, so the first item is left | 22:20 |
| beagles | sweston, yeah... so | 22:20 |
| beagles | sweston, really when it comes down to broad strokes, it is possible to configure neutron to behave very much like a nova deployment... let's ignore the steps for a moment, but the underlying "guts" are there | 22:21 |
| beagles | sweston, except... | 22:21 |
| *** hemanthravi has joined #openstack-neutron | 22:21 | |
| sweston | beagles: is there another document with nova parity identified issues? I have an open bug for one of them, I remember anita was going to put together a roadmap in December but I did not follow up on that yet. | 22:22 |
| beagles | sweston, when you compare the accessibility and ease-of-use of using floating IPs between the two, neutron is pretty hard | 22:22 |
| beagles | sweston: which bug are you referring to? | 22:22 |
| *** clev has quit IRC | 22:24 | |
| sweston | beagles: yes, but I do not think it will be difficult to change the default installation of devstack to create a provider network on the same range as the eth0 interface, for example | 22:24 |
| sweston | beagles: I assume there have already been discussions about that, though? ;-) | 22:24 |
| beagles | sweston, while you are digging that up, the other remaining hunk is multi-host | 22:24 |
| *** clev has joined #openstack-neutron | 22:24 | |
| beagles | sweston, it has more to do with how they are connected | 22:24 |
| beagles | sweston, e.g. in nova, the public interface may be a physical interface on the host and the floating ips are aliased to that | 22:25 |
| beagles | sweston, so add a floating ip and off you go | 22:25 |
| sweston | beagles: I've had this one for a while ... but no progress yet https://blueprints.launchpad.net/neutron/+spec/auto-associate-floating-ip | 22:26 |
| beagles | sweston, in neutron, you have to pick an approach of forwarding stuff into br-ex, set up some bridging so you can add a provider network interface to a physical interface, ya-da-ya-da | 22:26 |
| beagles | sweston, for folks who don't do networking it is a pretty difficult pill to swallow | 22:27 |
| *** banix has quit IRC | 22:28 | |
| beagles | sweston, it isn't the assignment of the floating IPs, but what you need to do with the system to get external access working. It seems like a trivial thing from a networking point of view, but if all you want is a private/developer/tiny-poc cloud it raises the bar | 22:28 |
| sweston | beagles: so why not make that the default installation? so i.e. the default network type would be a local provider network, bridged to the eth0 interface, using the ml2 openvswitch mechanism driver | 22:28 |
| *** rohit404 has quit IRC | 22:28 | |
| beagles | sweston, as soon as you bridge an eth, it loses its IP, etc. It requires some reconfig to get things working | 22:29 |
| sweston | beagles: nova network is fairly simple, i do not see why we cannot do the same for neutron. forgive me if I am asking "naive" questions | 22:29 |
| *** itzikb has quit IRC | 22:30 | |
| rms_13 | SumitNaiksatam: Thanks for reviewing and approving https://review.openstack.org/#/c/67272/. | 22:30 |
| beagles | sweston, for example.. all of my dev machines treat eth0 as a bridge interface and I configure a br0 as the "main interface" that gets my host's IP... I add whatever I want to that to get permit access between neutron L3 and L2 and my physical (or virtual) networks. | 22:30 |
| *** leseb has quit IRC | 22:31 | |
| beagles | sweston, basically, nova-network doesn't require bridging your public interface so it just kinda works | 22:31 |
| *** leseb has joined #openstack-neutron | 22:31 | |
| SumitNaiksatam | rms_13: hi, happy to help, it still needs one more core reviewer to approve it | 22:31 |
| beagles | sweston, the interesting dilemmas you run into when looking at how to mimic this behavior is that it obviates the ability to use some neutron features (not a huge deal for a transitional deployment), but also in some cases it seems like "hacking up" neutron or doing things that are against the spirit of neutron in order to get that type of thing | 22:33 |
| rms_13 | SumitNaiksatam: I understand. Probably markmcclain can take a look as he already did some initial digging | 22:33 |
| beagles | sweston, a similar set of issues arise when you start looking at multi-host. Nova-network multihost comes with some compromises that a similar implementation in neutron could avoid. There are discussions about creating similar functionality, however on occasion where similar compromises appear, they block further progress along those veins. | 22:36 |
| beagles | sweston, thinking of some examples... | 22:36 |
| beagles | sweston, as multi-host is basically a nova-network controller on each compute node that manages L3 and L2 for the instances on that node... | 22:37 |
| beagles | sweston, that implies multiple dhcp namespaces and dnsmasq instances... consuming an IP in that address space, | 22:38 |
| beagles | sweston, it also implies conditional or haphazard east-west routing... | 22:38 |
| zzelle | hi | 22:38 |
| beagles | sweston, in the former, nova actually has a configurable option to workaround that and the east-west routing thing is a problem in nova-network too | 22:39 |
| beagles | sweston, so one approach to the multi-host issue would be bite the bullet and accept those issues... | 22:39 |
| beagles | sweston, another approach would be to forget it and do something else. | 22:40 |
| zzelle | one question, does someone know why veth are used to connect physnets and br-int instead of ovs patchs when using VLAN + OVS | 22:40 |
| zzelle | ? | 22:40 |
| beagles | sweston, to date, I think we are doing the latter ;) | 22:40 |
| beagles | sweston, I'm thinking that multi-host is "too big a fish to swallow". There are guys working on DVR and it has the floor and may address some of the issues multi-host was meant to address. | 22:41 |
| sweston | beagles: ok I follow you ... but, the goal is to make this technology easily consumable, right? yes, there are some changes that have to happen with bridging networks and manipulating routing tables, but if it's an issue that is standing in the way of neutron adoption, I think it's a small sacrafice to make | 22:41 |
| beagles | sweston, I do not disagree... | 22:41 |
| beagles | sweston, so what would be cool if there were a minimal non-bizarro way to get this to work, even if it means saying "hey you can't do XYZ if you configure it like this" | 22:42 |
| beagles | sweston, e.g. pretty much either your L3 couldn't use namespaces and would have to play nice with the host, or there would need to be a "simple L3" router | 22:43 |
| beagles | sweston, some of the other folks may have some thoughts about whether that is a direction we want to go and/or how it might be done. | 22:44 |
| sweston | beagles: yes, I agree, it would be totally possible to create a poc with devstack and neutron that would bind everything together ... of course it would be limited functionality, but everything out of the box that is set up by default makes that compromise. | 22:45 |
| sweston | beagles: I think it would have to be a community decision, yes, and there would probably have to be frequent discussions about the best way to do it | 22:47 |
| sweston | beagles: it is always going to be a tug-of-war between functionality and simplicity | 22:47 |
| beagles | sweston, sure.. and to be fair, as soon as you start looking at "real world" nova-networking deployments, things get more complicated pretty quickly. It's the very simple cases where nova and neutron have a more noticeable gap | 22:47 |
| beagles | sweston, yup | 22:47 |
| *** oda-g has joined #openstack-neutron | 22:49 | |
| beagles | sweston, another mondo-question that comes to mind is whether these things are reasonable priorities at the moment | 22:49 |
| beagles | sweston, ie. is it more reasonable to focus on performance, stability and scalability, quality, etc. | 22:50 |
| sweston | beagles: so possibly the next step here, if everyone agrees that this is worth moving forward, is to set up a working group, with weekly meetings for direction and organization, leadership ... I think it is necessary in order to move neutron forward | 22:51 |
| beagles | sweston, bodies | 22:53 |
| beagles | sweston, yes | 22:53 |
| sweston | beagles: yes, that is a big question, it is a lot of work, and not as cool as LBaaS or DVR, but I think this effort is everyone's responsibility. this is a community and it is important to realize that if nova network never gets deprecated, then we will never have any of the cool things happening in the subgroups right now | 22:54 |
| * beagles chuckles | 22:54 | |
| * sweston smiles | 22:54 | |
| beagles | sweston, I'm chuckling because it sounds pretty familiar | 22:54 |
| sweston | beagles: I think we all know it's true, though ;-) | 22:55 |
| *** dave_tucker_zzz is now known as dave_tucker | 22:55 | |
| beagles | sweston, well, that actually begs the question: does nova-network *not* being deprecate obviate the existence of all of the aforementioned cool things | 22:56 |
| sweston | beagles: how many thousands of times have development projects gotten stomped on and put out because all of the developers wanted to work on what was fun instead of having to drive the basic functionality | 22:57 |
| *** RajeshMohan has quit IRC | 22:58 | |
| *** gdubreui has joined #openstack-neutron | 22:58 | |
| beagles | sweston, :) | 22:59 |
| sweston | beagles: would anything that is going on in the subgroups really be possible without ml2? nova-network is not modular, it is stable yes, but fragile, and hard to extend. | 22:59 |
| *** RajeshMohan has joined #openstack-neutron | 22:59 | |
| *** JPtheJP has quit IRC | 23:00 | |
| sweston | beagles: but I can tell that I am preaching to the choir here, so what in your opinion is the best way to proceed? | 23:00 |
| sweston | beagles: ;-) | 23:00 |
| beagles | sweston, oh my | 23:00 |
| beagles | sweston, well... | 23:03 |
| *** pcm_ has joined #openstack-neutron | 23:03 | |
| *** alagalah_ has joined #openstack-neutron | 23:04 | |
| beagles | sweston, I think to start it needs to be established whether there is a true business case behind parity and the notion of re-casting the simplified - or in the case of multi-host- functionality using the neutron component | 23:07 |
| *** alagalah has quit IRC | 23:07 | |
| sweston | beagles: so, I think that everyone realizes what a good thing we have going on with Neutron, and that we want to do everything we can to favor the adoption and success of the project, right? I know that many people are working very hard on contributing, I certainly am, and I would hate to see a lot of good work go to waste because there was a lack of effort to make the code usable. | 23:08 |
| beagles | sweston: let me state it another way, does neutron need to close those gap items in order to continue to exist and evolve | 23:10 |
| * beagles stipulates he is punching out of his weight class and wonders if the cores are watching :) | 23:11 | |
| sweston | beagles: I have been given the opportunity to begin working on developing this software full time, as of the beginning of the year, and now that I can be working with the community (as my job) to make this effort successful, and I am willing to help out in whatever way I can. | 23:12 |
| *** ecbanks has quit IRC | 23:12 | |
| *** amotoki has quit IRC | 23:12 | |
| sweston | me is chuckling. | 23:13 |
| * sweston is chuckling | 23:13 | |
| sweston | beagles: well, I think anyone is free to join in here, but yes, I do believe the gaps need to be closed in order to make it an easier transition. | 23:14 |
| beagles | sweston, the reason I'm playing devil's advocate because it is not clear to me that it is commonly viewed this way. | 23:16 |
| *** yamahata has quit IRC | 23:17 | |
| beagles | sweston, but that's enough of that. | 23:17 |
| beagles | sweston, there are two challenges moving forward that: one is the gap that's already been discussed and the other will be keeping up with the changes in the Openstack ecosystem. | 23:19 |
| sweston | beagles: yes ;-). I do think, though that it is reasonable to address the issues of nova-network out of the box Neutron parity, at least to some extent, in order to favor the adoption of the technology. I know that I do not want to be the only one working on parity, though, I believe that it should be a community effort ... although I am willing to at least help facilitate some amount of compromise here. | 23:21 |
| beagles | sweston, e.g. changes in how nova and neutron interact - these may be significant moving forward. In some cases it may make things a lot easier/smoother in others it may "break" things or at the very least obsolete efforts that might be underway | 23:21 |
| beagles | sweston, understood | 23:21 |
| *** otherwiseguy has quit IRC | 23:22 | |
| sweston | beagles: understood, as well, ;-) | 23:22 |
| openstackgerrit | Akihiro Motoki proposed a change to openstack/neutron: NEC plugin: PFC packet fitler support https://review.openstack.org/74192 | 23:23 |
| *** peristeri has quit IRC | 23:25 | |
| *** jgrimm has quit IRC | 23:26 | |
| sweston | beagles: so, what are a few things I can get started on to help? let's pick some tasks, in your opinion, that will not hinder the current functionality and restore some of the community goodness in making Neutron more "palatable" | 23:26 |
| openstackgerrit | Ryu Ishimoto proposed a change to openstack/neutron: Implement MidoNet plugin for Icehouse https://review.openstack.org/74193 | 23:26 |
| *** aymenfrikha has quit IRC | 23:28 | |
| openstackgerrit | Bob Melander proposed a change to openstack/neutron: Implements Neutron routing API using Cisco appliances. https://review.openstack.org/74194 | 23:30 |
| *** alexpilotti has quit IRC | 23:30 | |
| beagles | sweston, to be honest, I think we should focus on fixing the bugs that affect common usability. If we are looking at functionality that doesn't exist yet while the key functionality that does exist, it doesn't matter if we close the remaining gap. It's gotta be ... smooooooth ... | 23:31 |
| beagles | sweston, there is an image problem now as well, so scrutiny is always going to be very high | 23:32 |
| beagles | sweston, I guess in a vague way, I feel that dilution of the resource pool doesn't get us there. If we can make neutron be the best neutron it can be, then the last bits either will be acceptable or easier to resolve | 23:34 |
| beagles | sweston, strong foundations and all of that. | 23:34 |
| beagles | sweston, and ... | 23:34 |
| beagles | sweston, unfortunately I have to run.. it's 2004h here and we haven't had supper yet | 23:35 |
| beagles | I missed some words there somewhere ... | 23:35 |
| beagles | ... looking at functionality that doesn't exist yet while the key functionality that does exist doesn't work very well... | 23:36 |
| *** leseb has quit IRC | 23:36 | |
| beagles | sweston, anyways.. the natives are now hollering. Catch up tomorrow? | 23:37 |
| sweston | beagles: Yes, you bring some good issues and problems to light. Perhaps one of the best strengths of working as a community is that there are enough bright minds involved to make the vast majority happy with some common ground. Many other open source projects have been successful in this area, there is no reason, in my opinion, why we cannot strive for the same "togetherness". :-). | 23:39 |
| *** oda-g_ has joined #openstack-neutron | 23:40 | |
| sweston | beagles: Yes, let's catch up tomorrow ... I have been working all night on the lbaas ovs hybrid interface driver, and I could use a break as well. Have a good evening ;-) | 23:41 |
| *** oda-g has quit IRC | 23:42 | |
| *** IanGovett has left #openstack-neutron | 23:43 | |
| *** mwagner_lap has quit IRC | 23:51 | |
| *** mlavalle has quit IRC | 23:55 | |
| *** arosen has joined #openstack-neutron | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!