| *** Longgeek has quit IRC | 00:00 | |
| *** joshw42 has joined #openstack-neutron | 00:08 | |
| *** ecbanks has joined #openstack-neutron | 00:09 | |
| *** jamielennox is now known as jamielennox|away | 00:26 | |
| *** jamielennox|away is now known as jamielennox | 00:30 | |
| *** ferest has quit IRC | 00:39 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron: Updated from global requirements https://review.openstack.org/182746 | 00:43 |
|---|---|---|
| *** gildub has joined #openstack-neutron | 00:45 | |
| *** banix has quit IRC | 00:46 | |
| *** ijw_ has quit IRC | 00:50 | |
| *** ijw has joined #openstack-neutron | 00:50 | |
| openstackgerrit | Tomoaki Sato proposed openstack/neutron: Fix subnet updating failure on valid allocation pools https://review.openstack.org/196487 | 01:04 |
| *** park_heijlong has joined #openstack-neutron | 01:13 | |
| *** zhenguo has joined #openstack-neutron | 01:16 | |
| openstackgerrit | gongysh proposed openstack/neutron: Fixed L3 agent manual scheduling for HA routers https://review.openstack.org/143297 | 01:24 |
| *** banix has joined #openstack-neutron | 01:27 | |
| *** markvoelker has joined #openstack-neutron | 01:30 | |
| openstackgerrit | gongysh proposed openstack/neutron: populate port security default into network https://review.openstack.org/188209 | 01:32 |
| openstackgerrit | gongysh proposed openstack/neutron: Add extra subnet route to ha router https://review.openstack.org/159772 | 01:32 |
| *** iwamoto has joined #openstack-neutron | 01:34 | |
| *** markvoelker has quit IRC | 01:35 | |
| *** stanzgy has joined #openstack-neutron | 01:35 | |
| *** jasondotstar has joined #openstack-neutron | 01:40 | |
| *** kiran-r is now known as kiranr | 01:40 | |
| *** tflynn has joined #openstack-neutron | 01:43 | |
| *** tflynn_ has joined #openstack-neutron | 01:44 | |
| *** tflynn has quit IRC | 01:47 | |
| *** erkules_ has joined #openstack-neutron | 01:53 | |
| *** tflynn_ has quit IRC | 01:55 | |
| *** tflynn has joined #openstack-neutron | 01:55 | |
| *** erkules has quit IRC | 01:56 | |
| *** tflynn has quit IRC | 01:57 | |
| *** tflynn has joined #openstack-neutron | 01:58 | |
| *** Longgeek has joined #openstack-neutron | 01:59 | |
| *** Sukhdev has joined #openstack-neutron | 02:01 | |
| openstackgerrit | Hirofumi Ichihara proposed openstack/neutron: Prohibit updating gateway_ip of an external network https://review.openstack.org/194939 | 02:01 |
| openstackgerrit | shihanzhang proposed openstack/neutron: Add conntrack-tool to manage security groups https://review.openstack.org/147713 | 02:07 |
| *** yalie has joined #openstack-neutron | 02:10 | |
| *** yalie has left #openstack-neutron | 02:10 | |
| *** Longgeek has quit IRC | 02:11 | |
| *** oda-g has joined #openstack-neutron | 02:13 | |
| *** dboik has quit IRC | 02:16 | |
| *** yamamoto has joined #openstack-neutron | 02:18 | |
| *** surabhi has joined #openstack-neutron | 02:24 | |
| openstackgerrit | shihanzhang proposed openstack/neutron: Destroy ipset set when corresponding sg member is empty https://review.openstack.org/190991 | 02:37 |
| openstackgerrit | Shang Yong proposed openstack/neutron: Add parent_id to _item calling from _handle_action https://review.openstack.org/180834 | 02:42 |
| *** mwagner_afk has joined #openstack-neutron | 02:43 | |
| *** aranjan has joined #openstack-neutron | 02:49 | |
| *** ecbanks has quit IRC | 02:51 | |
| *** dave-mccowan has joined #openstack-neutron | 02:56 | |
| *** SumitNaiksatam has quit IRC | 03:01 | |
| *** SumitNaiksatam has joined #openstack-neutron | 03:01 | |
| *** tflynn has quit IRC | 03:01 | |
| openstackgerrit | gongysh proposed openstack/neutron: Fixed L3 agent manual scheduling for HA routers https://review.openstack.org/143297 | 03:05 |
| *** Longgeek has joined #openstack-neutron | 03:07 | |
| *** achanda has joined #openstack-neutron | 03:11 | |
| *** flwang1 has quit IRC | 03:11 | |
| *** jckasper has quit IRC | 03:13 | |
| *** tflynn has joined #openstack-neutron | 03:14 | |
| *** dave-mccowan has quit IRC | 03:15 | |
| *** achanda has quit IRC | 03:18 | |
| *** markvoelker has joined #openstack-neutron | 03:19 | |
| *** Longgeek has quit IRC | 03:20 | |
| *** flwang1 has joined #openstack-neutron | 03:20 | |
| *** markvoelker has quit IRC | 03:23 | |
| *** aranjan has quit IRC | 03:26 | |
| *** sbalukoff has quit IRC | 03:32 | |
| *** sdake has joined #openstack-neutron | 03:32 | |
| *** jasondotstar has quit IRC | 03:32 | |
| *** jasondotstar has joined #openstack-neutron | 03:33 | |
| *** sdake has quit IRC | 03:36 | |
| *** sdake has joined #openstack-neutron | 03:39 | |
| *** tflynn has quit IRC | 03:42 | |
| *** baoli has quit IRC | 03:46 | |
| openstackgerrit | meenakshi kaushik proposed openstack/neutron: Add documentation for Linux Bridge (previously missing) https://review.openstack.org/185421 | 03:46 |
| *** Jabadia has joined #openstack-neutron | 03:46 | |
| *** baoli has joined #openstack-neutron | 03:48 | |
| *** baoli has quit IRC | 03:48 | |
| *** baoli has joined #openstack-neutron | 03:48 | |
| *** tflynn has joined #openstack-neutron | 03:49 | |
| *** baoli has quit IRC | 03:49 | |
| *** coolsvap|away is now known as coolsvap | 03:52 | |
| *** armax has joined #openstack-neutron | 03:54 | |
| *** hanzhang has joined #openstack-neutron | 03:55 | |
| *** aranjan has joined #openstack-neutron | 03:58 | |
| *** sdake_ has joined #openstack-neutron | 03:58 | |
| *** Sukhdev has quit IRC | 04:01 | |
| *** sdake has quit IRC | 04:02 | |
| *** Marga_ has joined #openstack-neutron | 04:06 | |
| *** Marga_ has quit IRC | 04:07 | |
| *** Marga_ has joined #openstack-neutron | 04:07 | |
| openstackgerrit | Tomoaki Sato proposed openstack/neutron: Fix subnet updating failure on valid allocation pools https://review.openstack.org/196487 | 04:14 |
| *** Jabadia has quit IRC | 04:16 | |
| *** banix has quit IRC | 04:20 | |
| *** Marga_ has quit IRC | 04:20 | |
| openstackgerrit | Ryan Tidwell proposed openstack/neutron: [WIP] Add devref for BGP Speaker API extension https://review.openstack.org/196452 | 04:21 |
| *** Marga_ has joined #openstack-neutron | 04:21 | |
| *** kiran-r has joined #openstack-neutron | 04:23 | |
| *** Marga_ has quit IRC | 04:25 | |
| *** irenab has quit IRC | 04:27 | |
| *** kiranr has quit IRC | 04:30 | |
| *** zhhuabj_ has quit IRC | 04:35 | |
| *** rm_work|away is now known as rm_work | 04:36 | |
| *** yfried_ has quit IRC | 04:36 | |
| *** garyk has joined #openstack-neutron | 04:38 | |
| *** ekuris has joined #openstack-neutron | 04:39 | |
| *** agireud has joined #openstack-neutron | 04:40 | |
| *** yfried has joined #openstack-neutron | 04:43 | |
| *** dedery has joined #openstack-neutron | 04:44 | |
| *** yfried has quit IRC | 04:49 | |
| *** zhhuabj_ has joined #openstack-neutron | 04:51 | |
| *** gildub is now known as gildub_bbl | 04:52 | |
| *** Djinnn has joined #openstack-neutron | 04:56 | |
| *** doude has quit IRC | 04:57 | |
| *** red_trela has joined #openstack-neutron | 05:00 | |
| *** Djinnn has quit IRC | 05:00 | |
| *** patrickeast has quit IRC | 05:01 | |
| *** sudipto has joined #openstack-neutron | 05:01 | |
| *** sthillma has joined #openstack-neutron | 05:02 | |
| *** kiran-r has quit IRC | 05:03 | |
| *** markvoelker has joined #openstack-neutron | 05:08 | |
| *** dedery has quit IRC | 05:08 | |
| *** dedery has joined #openstack-neutron | 05:09 | |
| *** dedery_ has joined #openstack-neutron | 05:09 | |
| *** markvoelker has quit IRC | 05:12 | |
| *** dedery has quit IRC | 05:13 | |
| *** Longgeek has joined #openstack-neutron | 05:17 | |
| *** sthillma_ has joined #openstack-neutron | 05:17 | |
| openstackgerrit | YujiAzama proposed openstack/neutron-specs: [PATCH] Add common classifier resource https://review.openstack.org/190463 | 05:18 |
| openstackgerrit | YujiAzama proposed openstack/neutron-specs: [PATCH] Add common classifier resource https://review.openstack.org/190463 | 05:20 |
| *** sthillma has quit IRC | 05:20 | |
| *** sthillma_ is now known as sthillma | 05:20 | |
| *** shivrao has joined #openstack-neutron | 05:20 | |
| *** yuanying_ has joined #openstack-neutron | 05:22 | |
| *** yuanying has quit IRC | 05:25 | |
| *** Longgeek has quit IRC | 05:25 | |
| *** Longgeek has joined #openstack-neutron | 05:25 | |
| *** shivrao_ has joined #openstack-neutron | 05:27 | |
| *** shivrao has quit IRC | 05:28 | |
| *** shivrao_ is now known as shivrao | 05:28 | |
| *** gildub_bbl is now known as gildub | 05:28 | |
| gongysh_ | yamahata: AttributeError: 'module' object has no attribute '_orig_get_script_args' | 05:31 |
| *** yfried has joined #openstack-neutron | 05:32 | |
| gongysh_ | when I am stack.sh on devstack, the ryu complains. | 05:32 |
| *** Murali has joined #openstack-neutron | 05:33 | |
| yamamoto | gongysh_: do you mean me or yamahata? | 05:34 |
| *** Longgeek has quit IRC | 05:34 | |
| iwamoto | mmm, ryu | 05:34 |
| *** Longgeek has joined #openstack-neutron | 05:34 | |
| yamamoto | gongysh_: FYI, iwamoto is the best person to ask ryu questions | 05:34 |
| gongysh_ | ok, thanks | 05:35 |
| *** yfried_ has joined #openstack-neutron | 05:35 | |
| *** Marga_ has joined #openstack-neutron | 05:35 | |
| gongysh_ | iwamoto: do u have any clues? thanks | 05:36 |
| *** spandhe has joined #openstack-neutron | 05:36 | |
| *** yfried has quit IRC | 05:36 | |
| yamamoto | gongysh_: ryu has some back magic regarding pbr. it might be broken depending on versions you are using. | 05:37 |
| yamamoto | black magic | 05:37 |
| gongysh_ | yamamoto: pbr 1.2.0 | 05:37 |
| iwamoto | is it happening from inside ryu/hooks.py? | 05:38 |
| gongysh_ | ryu: the trunk | 05:38 |
| yamamoto | iwamoto: it should be | 05:38 |
| gongysh_ | iwamoto: /opt/stack/ryu$ sudo python setup.py develop | 05:38 |
| gongysh_ | running develop | 05:38 |
| gongysh_ | running egg_info | 05:38 |
| gongysh_ | writing pbr to ryu.egg-info/pbr.json | 05:38 |
| gongysh_ | writing requirements to ryu.egg-info/requires.txt | 05:38 |
| gongysh_ | writing ryu.egg-info/PKG-INFO | 05:38 |
| gongysh_ | writing top-level names to ryu.egg-info/top_level.txt | 05:38 |
| gongysh_ | writing dependency_links to ryu.egg-info/dependency_links.txt | 05:38 |
| gongysh_ | writing entry points to ryu.egg-info/entry_points.txt | 05:38 |
| gongysh_ | [pbr] Processing SOURCES.txt | 05:38 |
| gongysh_ | [pbr] In git context, generating filelist from git | 05:38 |
| gongysh_ | warning: no files found matching 'AUTHORS' | 05:38 |
| gongysh_ | warning: no files found matching 'ChangeLog' | 05:38 |
| *** dedery has joined #openstack-neutron | 05:38 | |
| gongysh_ | warning: no previously-included files found matching '.gitreview' | 05:38 |
| gongysh_ | warning: no previously-included files matching '*.pyc' found anywhere in distribution | 05:38 |
| gongysh_ | reading manifest template 'MANIFEST.in' | 05:38 |
| gongysh_ | warning: no previously-included files matching '*' found under directory 'doc/build' | 05:39 |
| gongysh_ | warning: no previously-included files matching '*~' found anywhere in distribution | 05:39 |
| gongysh_ | warning: no previously-included files matching '*.pyc' found anywhere in distribution | 05:39 |
| gongysh_ | warning: no previously-included files matching '.gitignore' found anywhere in distribution | 05:39 |
| gongysh_ | writing manifest file 'ryu.egg-info/SOURCES.txt' | 05:39 |
| gongysh_ | running build_ext | 05:39 |
| gongysh_ | Creating /usr/local/lib/python2.7/dist-packages/ryu.egg-link (link to .) | 05:39 |
| gongysh_ | ryu 3.22 is already the active version in easy-install.pth | 05:39 |
| gongysh_ | Traceback (most recent call last): | 05:39 |
| gongysh_ | File "setup.py", line 29, in <module> | 05:39 |
| gongysh_ | pbr=True) | 05:39 |
| gongysh_ | File "/usr/lib/python2.7/distutils/core.py", line 151, in setup | 05:39 |
| gongysh_ | dist.run_commands() | 05:39 |
| gongysh_ | File "/usr/lib/python2.7/distutils/dist.py", line 953, in run_commands | 05:39 |
| gongysh_ | self.run_command(cmd) | 05:39 |
| gongysh_ | File "/usr/lib/python2.7/distutils/dist.py", line 972, in run_command | 05:39 |
| gongysh_ | cmd_obj.run() | 05:39 |
| gongysh_ | File "/usr/local/lib/python2.7/dist-packages/setuptools/command/develop.py", line 32, in run | 05:39 |
| gongysh_ | self.install_for_development() | 05:39 |
| gongysh_ | File "/usr/local/lib/python2.7/dist-packages/setuptools/command/develop.py", line 132, in install_for_development | 05:39 |
| gongysh_ | self.process_distribution(None, self.dist, not self.no_deps) | 05:39 |
| gongysh_ | File "/usr/local/lib/python2.7/dist-packages/setuptools/command/easy_install.py", line 68 | 05:39 |
| gongysh_ | self.install_egg_scripts(dist) | 05:39 |
| gongysh_ | File "/usr/local/lib/python2.7/dist-packages/setuptools/command/develop.py", line 160, in install_egg_scripts | 05:39 |
| gongysh_ | self.install_wrapper_scripts(dist) | 05:39 |
| gongysh_ | File "/usr/local/lib/python2.7/dist-packages/pbr/packaging.py", line 254, in install_wrapper_scripts | 05:39 |
| gongysh_ | for args in override_get_script_args(dist): | 05:40 |
| gongysh_ | File "/opt/stack/ryu/ryu/hooks.py", line 59, in my_get_script_args | 05:40 |
| gongysh_ | return _main_module()._orig_get_script_args(*args, **kwargs) | 05:40 |
| gongysh_ | AttributeError: 'module' object has no attribute '_orig_get_script_args' | 05:40 |
| *** Marga_ has quit IRC | 05:40 | |
| *** Marga_ has joined #openstack-neutron | 05:40 | |
| *** sridhar_ram has joined #openstack-neutron | 05:40 | |
| yamamoto | it reminded me this http://sourceforge.net/p/ryu/mailman/message/34143855/ | 05:41 |
| *** dedery_ has quit IRC | 05:42 | |
| *** shausy has joined #openstack-neutron | 05:43 | |
| *** shivrao has quit IRC | 05:43 | |
| iwamoto | yamamoto: cannot this ryu/hooks.py black magic simply removed? | 05:45 |
| yamamoto | iwamoto: we put it for reasons. i forgot details but i think it was about making ryu-manager pick the right version of modules. | 05:47 |
| gongysh_ | yamanoto: I reinstalled my pbr to 0.11 | 05:47 |
| gongysh_ | yamamoto: it still does not work. gongysh@ubuntu1504-1:/opt/stack/ryu$ pip freeze | grep pbr | 05:47 |
| gongysh_ | pbr==0.11.0 | 05:47 |
| *** SridharG has joined #openstack-neutron | 05:47 | |
| yamamoto | iwamoto: you can consult git log. or fujita-san might remember | 05:48 |
| yamamoto | gongysh_: same error? | 05:49 |
| gongysh_ | yamamoto: yes the same error. | 05:49 |
| *** yog__ has joined #openstack-neutron | 05:50 | |
| gongysh_ | yamamoto: but I install ryu on fedora22, it works even with pbr 1.2.0 | 05:50 |
| yamamoto | gongysh_: i'm a bit puzzled as iirc ryu-ci uses devstack to install ofagent (and ryu) | 05:50 |
| iwamoto | gongysh_: on what os version are you seeing this problem? | 05:51 |
| gongysh_ | yamamoto: yes, I think it should be my ubuntu 15.04. | 05:52 |
| *** ihrachyshka has joined #openstack-neutron | 05:52 | |
| *** jasondotstar has quit IRC | 05:53 | |
| *** Murali has quit IRC | 05:54 | |
| iwamoto | gongysh_: I wonder if there's only single version of ryu on the system, you might revert ryu's d6a9890b0833044fef1b7e916cd355cec325303f and give a try. | 05:56 |
| *** vikram has joined #openstack-neutron | 05:57 | |
| *** vikram has quit IRC | 05:57 | |
| *** itzikb has joined #openstack-neutron | 05:57 | |
| *** gridinv_ has quit IRC | 05:58 | |
| *** agireud has quit IRC | 05:59 | |
| *** dedery_ has joined #openstack-neutron | 05:59 | |
| *** dedery has quit IRC | 06:00 | |
| *** tflynn has quit IRC | 06:04 | |
| gongysh_ | iwamoto: I have only one ryu. This VM is newly installed. | 06:04 |
| *** abregman has joined #openstack-neutron | 06:10 | |
| *** yog__ has quit IRC | 06:12 | |
| *** zhhuabj_ has quit IRC | 06:12 | |
| *** spandhe has quit IRC | 06:13 | |
| openstackgerrit | enikanorov proposed openstack/neutron: Add logging of agent heartbeats https://review.openstack.org/181132 | 06:15 |
| openstackgerrit | Sudipta Biswas proposed openstack/python-neutronclient: Determine ip version during subnet create. https://review.openstack.org/194078 | 06:16 |
| *** Marga_ has quit IRC | 06:17 | |
| *** Marga_ has joined #openstack-neutron | 06:17 | |
| openstackgerrit | enikanorov proposed openstack/neutron: Use reset_mock() method instead of accessing mock's attribute https://review.openstack.org/196551 | 06:21 |
| *** ijw has quit IRC | 06:21 | |
| *** sbalukoff has joined #openstack-neutron | 06:22 | |
| *** numan has joined #openstack-neutron | 06:23 | |
| *** moshele has joined #openstack-neutron | 06:23 | |
| moshele | ihrachyshka: hi | 06:23 |
| ihrachyshka | moshele, hi | 06:23 |
| moshele | ihrachyshka: the SR-IOV is broken after the Consolidation | 06:24 |
| ihrachyshka | moshele, bug/patch avail? | 06:24 |
| moshele | ihrachyshka: this is the fix https://review.openstack.org/#/c/196433/ | 06:24 |
| ihrachyshka | ack | 06:24 |
| *** trinaths has joined #openstack-neutron | 06:24 | |
| *** armax has quit IRC | 06:26 | |
| ihrachyshka | garyk, ^^ trivial one for sr-iov | 06:28 |
| gongysh_ | iwamoto, yamamoto, I redownloaded the ryu from git, and it works now. | 06:28 |
| *** zhhuabj_ has joined #openstack-neutron | 06:29 | |
| garyk | sriov is so 80's | 06:29 |
| moshele | ihrachyshka, garyk thanks | 06:29 |
| garyk | ihrachyshka: where is everyone meeting tomorrow. i will be able to attend the moring sessions | 06:30 |
| ihrachyshka | garyk, ;) a lot of cool stuff is 80s, like unix. | 06:30 |
| ihrachyshka | (well, it's 70s) | 06:30 |
| ihrachyshka | garyk, redhat office it is. ajo, more specific details? ^^ | 06:30 |
| garyk | ihrachyshka: i just need to know which floor :) - as far as i am concerned we can work in the coffee shop downstairs - they make a pretty good cup :) | 06:31 |
| *** watanabe_isao has joined #openstack-neutron | 06:32 | |
| yamamoto | gongysh_: with which pbr version? | 06:33 |
| *** scheuran has joined #openstack-neutron | 06:34 | |
| garyk | ajo: please let me know which room we will be sitting in | 06:34 |
| ajo | garyk, 5th floor :) | 06:35 |
| *** gampel has joined #openstack-neutron | 06:35 | |
| garyk | ack, thanks | 06:36 |
| ajo | garyk++ | 06:37 |
| ajo | thank you too :D | 06:37 |
| garyk | ajo: i have meetings in the afternoon so i will be there in the morning | 06:39 |
| gsagie | ohhh | 06:40 |
| *** mitchjameson has joined #openstack-neutron | 06:40 | |
| gongysh_ | yamamoto: I am trying to install dragonflow at https://github.com/openstack/dragonflow | 06:41 |
| gsagie | gongysh_ : what help do you need? | 06:41 |
| gongysh_ | yamamoto: it seems the its devstack will update the setup.py of ryu | 06:41 |
| gongysh_ | gsagie: :/opt/stack/ryu$ git status | 06:42 |
| gongysh_ | On branch master | 06:42 |
| gongysh_ | Your branch is up-to-date with 'origin/master'. | 06:42 |
| gongysh_ | Changes not staged for commit: | 06:42 |
| gongysh_ | (use "git add <file>..." to update what will be committed) | 06:42 |
| gongysh_ | (use "git checkout -- <file>..." to discard changes in working directory) | 06:42 |
| gongysh_ | modified: setup.py | 06:42 |
| gongysh_ | modified: tools/pip-requires | 06:42 |
| gongysh_ | modified: tools/test-requires | 06:42 |
| *** oda-g has quit IRC | 06:42 | |
| gongysh_ | the ryu's setup.py was updated by dragonflow's devstack script. | 06:43 |
| *** sthillma has quit IRC | 06:43 | |
| *** apuimedo has joined #openstack-neutron | 06:43 | |
| yamamoto | gongysh_: ah, i see. dragonflow's devstack script uses setup_develop. ryu is not compatible with it. (yet) | 06:44 |
| gsagie | If you use the master branch of dragonflow, we also clone from Ryu master git, which probably we will fix soo for a specific version | 06:46 |
| gongysh_ | yamamoto: so what should I do to setup dragonflow with ryu? | 06:46 |
| gsagie | gongysh_ : what is failing with this? | 06:47 |
| trinaths | Hi, Good Morning all | 06:47 |
| gsagie | you could just remove this line from the dragonflow devstack plugin and install ryu yourself | 06:47 |
| *** Jabadia has joined #openstack-neutron | 06:47 | |
| gongysh_ | gsagie: ok | 06:48 |
| gsagie | but i havent encounter any problems on my setup | 06:48 |
| gsagie | you can query me if you need any more assistance | 06:49 |
| *** apuimedo has quit IRC | 06:49 | |
| yamamoto | gsagie: i think dragonflow should not use setup_develop for ryu. iirc setup_develop is meant for openstack projects (ie. global-req compatible projects) and ryu isn't. | 06:49 |
| *** apuimedo has joined #openstack-neutron | 06:50 | |
| gsagie | yamamoto: ok thanks, will fix that | 06:50 |
| *** irenab has joined #openstack-neutron | 06:51 | |
| *** yfried_ is now known as yfried|afk | 06:52 | |
| yamamoto | gsagie: i think you can use pip_install instead | 06:53 |
| *** yuanying_ has quit IRC | 06:53 | |
| *** yfried|afk is now known as yfried_ | 06:53 | |
| *** nmagnezi has joined #openstack-neutron | 06:54 | |
| *** anshul has joined #openstack-neutron | 06:54 | |
| gongysh_ | gsagie: seems a clean cut. | 06:54 |
| *** yuanying_ has joined #openstack-neutron | 06:54 | |
| gsagie | yamamoto, gongysh_ : thanks, will send a patch now | 06:55 |
| gongysh_ | gsagie: I changed the dragonflow devstack and use pip_install it. | 06:55 |
| gsagie | gongysh_ : if its working, can you post a patch? ;) | 06:55 |
| *** jlanoux has joined #openstack-neutron | 06:55 | |
| openstackgerrit | enikanorov proposed openstack/neutron: Use reset_mock() method instead of accessing mock's attribute https://review.openstack.org/196551 | 06:56 |
| apuimedo | gsagie: Boker tov | 06:56 |
| gongysh_ | gsagie: ok, I will post a patch | 06:56 |
| gsagie | apuimedo : Hi | 06:56 |
| gsagie | didnt know you speak hebrew.. ;) | 06:56 |
| *** markvoelker has joined #openstack-neutron | 06:56 | |
| apuimedo | gsagie: a couple of words only | 06:57 |
| openstackgerrit | enikanorov proposed openstack/neutron: Don't access mock's attribute directly especially when it's needed https://review.openstack.org/196551 | 06:57 |
| openstackgerrit | enikanorov proposed openstack/neutron: Don't access mock's attribute directly especially when it's not needed https://review.openstack.org/196551 | 06:57 |
| gsagie | gongysh_: Thanks! and let me know if you need any more help to bring it up | 06:58 |
| *** safchain has joined #openstack-neutron | 06:59 | |
| gongysh_ | gsagie: sure | 06:59 |
| *** jpena has joined #openstack-neutron | 07:00 | |
| *** markvoelker has quit IRC | 07:01 | |
| *** anshul has quit IRC | 07:02 | |
| *** anshul has joined #openstack-neutron | 07:06 | |
| *** shwetaap has quit IRC | 07:06 | |
| *** spandhe has joined #openstack-neutron | 07:06 | |
| *** breitz1 has joined #openstack-neutron | 07:08 | |
| *** breitz has quit IRC | 07:08 | |
| *** ccard_ has quit IRC | 07:09 | |
| garyk | are people having difficulty accessing gerrit? | 07:10 |
| *** anilvenkata has joined #openstack-neutron | 07:11 | |
| *** ijw has joined #openstack-neutron | 07:12 | |
| *** kawa2014 has joined #openstack-neutron | 07:13 | |
| *** mitchjameson has quit IRC | 07:13 | |
| *** Longgeek has quit IRC | 07:13 | |
| *** Longgeek has joined #openstack-neutron | 07:13 | |
| *** gberginc has joined #openstack-neutron | 07:14 | |
| *** jlibosva has joined #openstack-neutron | 07:15 | |
| *** rotbeard has joined #openstack-neutron | 07:17 | |
| *** ronis has joined #openstack-neutron | 07:17 | |
| *** yog__ has joined #openstack-neutron | 07:17 | |
| *** amotoki has joined #openstack-neutron | 07:17 | |
| *** ijw has quit IRC | 07:17 | |
| hanzhang | now if we delete dhcp port in a network, the TAP device in network node will not be deleted, but its tag is 4095, who is know why? | 07:18 |
| *** ildikov has quit IRC | 07:20 | |
| openstackgerrit | Aman Kumar proposed openstack/neutron: ovs agent resync may miss port remove event https://review.openstack.org/165023 | 07:22 |
| *** aranjan has quit IRC | 07:22 | |
| *** sridhar_ram has quit IRC | 07:22 | |
| *** spandhe has quit IRC | 07:22 | |
| gongysh_ | gsagie: https://review.openstack.org/#/c/196561/ | 07:24 |
| gongysh_ | garyk: my access to gerrit is good. | 07:25 |
| *** nlahouti has joined #openstack-neutron | 07:25 | |
| garyk | gongysh: thanks | 07:26 |
| *** yamamoto has quit IRC | 07:27 | |
| *** nlahouti has quit IRC | 07:28 | |
| *** nlahouti has joined #openstack-neutron | 07:28 | |
| *** jlanoux has quit IRC | 07:29 | |
| *** yamamoto has joined #openstack-neutron | 07:31 | |
| *** jistr has joined #openstack-neutron | 07:32 | |
| scheuran | mestery: Kyle, could you please give your +1 on https://review.openstack.org/189644 during this day, that the infra team can continue to set up the repository? Thx. | 07:33 |
| *** norman has joined #openstack-neutron | 07:33 | |
| openstackgerrit | Ramanjaneya Reddy Palleti proposed openstack/python-neutronclient: [WIP]: Support CLI changes for QoS. https://review.openstack.org/189655 | 07:33 |
| openstackgerrit | Kevin Benton proposed openstack/neutron: Read vif port information in bulk https://review.openstack.org/186734 | 07:34 |
| openstackgerrit | Kevin Benton proposed openstack/neutron: Update DVR agent to use get_vifs_by_id https://review.openstack.org/196568 | 07:34 |
| *** vikram has joined #openstack-neutron | 07:36 | |
| *** devvesa has joined #openstack-neutron | 07:36 | |
| *** Marga_ has quit IRC | 07:37 | |
| norman | sbelous, ping | 07:39 |
| norman | SridharG, ping | 07:40 |
| *** kiran-r has joined #openstack-neutron | 07:42 | |
| *** neeti has joined #openstack-neutron | 07:44 | |
| *** Marga_ has joined #openstack-neutron | 07:46 | |
| *** saggi has quit IRC | 07:51 | |
| *** kawa2014 has quit IRC | 07:56 | |
| *** kawa2014 has joined #openstack-neutron | 07:56 | |
| *** lukasa has joined #openstack-neutron | 07:59 | |
| *** gildub has quit IRC | 08:01 | |
| *** jlanoux has joined #openstack-neutron | 08:01 | |
| *** mitchjameson has joined #openstack-neutron | 08:01 | |
| *** txomon|fon has joined #openstack-neutron | 08:03 | |
| *** ijw has joined #openstack-neutron | 08:04 | |
| *** mickeys has joined #openstack-neutron | 08:06 | |
| *** numan has quit IRC | 08:07 | |
| *** numan has joined #openstack-neutron | 08:08 | |
| *** ildikov has joined #openstack-neutron | 08:08 | |
| *** ijw has quit IRC | 08:09 | |
| *** Marga_ has quit IRC | 08:10 | |
| *** yuanying_ has quit IRC | 08:12 | |
| *** ygbo has joined #openstack-neutron | 08:13 | |
| *** nmagnezi has quit IRC | 08:14 | |
| *** mickeys has quit IRC | 08:16 | |
| *** dtynan has quit IRC | 08:21 | |
| *** erkules_ is now known as erkules | 08:21 | |
| *** erkules has joined #openstack-neutron | 08:21 | |
| *** apuimedo has quit IRC | 08:21 | |
| *** dtynan has joined #openstack-neutron | 08:21 | |
| *** hichihara has quit IRC | 08:22 | |
| *** Marga_ has joined #openstack-neutron | 08:22 | |
| *** aranjan has joined #openstack-neutron | 08:23 | |
| *** dguerri` is now known as dguerri | 08:23 | |
| *** mitchjameson has quit IRC | 08:26 | |
| *** nmagnezi has joined #openstack-neutron | 08:26 | |
| *** Marga_ has quit IRC | 08:27 | |
| openstackgerrit | Kevin Benton proposed openstack/neutron: Revert "Removed test_lib module" https://review.openstack.org/196580 | 08:27 |
| *** aranjan has quit IRC | 08:28 | |
| *** ronis has quit IRC | 08:29 | |
| openstackgerrit | Kevin Benton proposed openstack/neutron: Read vif port information in bulk https://review.openstack.org/186734 | 08:33 |
| openstackgerrit | Kevin Benton proposed openstack/neutron: Update DVR agent to use get_vifs_by_id https://review.openstack.org/196568 | 08:33 |
| *** Marga_ has joined #openstack-neutron | 08:35 | |
| *** sudipto has quit IRC | 08:36 | |
| *** sudipto has joined #openstack-neutron | 08:38 | |
| *** Longgeek has quit IRC | 08:38 | |
| * pavel_bondar is looking for send core reviewer for https://review.openstack.org/#/c/190623/ | 08:40 | |
| * pavel_bondar is looking for *second* core reviewer for https://review.openstack.org/#/c/190623/ | 08:40 | |
| *** watanabe_isao has quit IRC | 08:41 | |
| *** beagles has quit IRC | 08:44 | |
| *** Marga_ has quit IRC | 08:45 | |
| *** Marga_ has joined #openstack-neutron | 08:45 | |
| *** Marga_ has quit IRC | 08:45 | |
| *** markvoelker has joined #openstack-neutron | 08:45 | |
| Ramanjaneya | hi, | 08:46 |
| *** lukasa_work has joined #openstack-neutron | 08:46 | |
| *** Marga_ has joined #openstack-neutron | 08:46 | |
| Ramanjaneya | how to change branch for posted changes.. | 08:46 |
| Ramanjaneya | how to change branch name for posted changes.. | 08:46 |
| gsagie | git review -r <destination-branch> | 08:46 |
| sbelous | norman: pong | 08:47 |
| sbelous | norman: hi | 08:47 |
| *** kbyrne has quit IRC | 08:47 | |
| *** lukasa has quit IRC | 08:49 | |
| *** trinaths has quit IRC | 08:49 | |
| *** lukasa_work has quit IRC | 08:49 | |
| *** markvoelker has quit IRC | 08:50 | |
| *** dguitarbite has joined #openstack-neutron | 08:53 | |
| *** kbyrne has joined #openstack-neutron | 08:53 | |
| *** kir_ has joined #openstack-neutron | 08:54 | |
| *** afazekas_ has joined #openstack-neutron | 08:56 | |
| openstackgerrit | Masaki Matsushita proposed openstack/python-neutronclient: Make security-group-rule-create accept protocol "None" https://review.openstack.org/196582 | 08:56 |
| norman | sbelous, hi | 08:57 |
| openstackgerrit | Masaki Matsushita proposed openstack/python-neutronclient: Make security-group-rule-create accept protocol "None" https://review.openstack.org/196582 | 08:57 |
| norman | sbelous, I thought you are knowing well about the HA router, and I have some question about it | 08:57 |
| norman | sbelous, when working with the legacy_router, I can know that it will eventually call the IPDevice with 'ip addr' to add the floating ip into the router namespace when I associated floating ip to VM | 08:58 |
| *** ronis has joined #openstack-neutron | 08:59 | |
| norman | but with HA router , I only saw that it add the vip list , but not found where had add it into the namespaces' interface in the linux level | 09:00 |
| openstackgerrit | Ramanjaneya Reddy Palleti proposed openstack/neutron: [WIP] Support Basic QoS CRUD operation https://review.openstack.org/195050 | 09:02 |
| openstackgerrit | Ramanjaneya Reddy Palleti proposed openstack/neutron: [WIP] Support Basic QoS CRUD operation https://review.openstack.org/195050 | 09:04 |
| norman | sbelous, code path as below ,in ha_router.py, add_floating_ip --> _add_vip --> keepalived instance.add_vip --> which only in the internal list | 09:08 |
| *** doude has joined #openstack-neutron | 09:08 | |
| *** eezhova has quit IRC | 09:08 | |
| SridharG | norman: pong | 09:09 |
| norman | SridharG, Hi , also the same question about the how the HA router add the floating ip, :P | 09:10 |
| SridharG | norman: is it about the same query you are asking sbelous? | 09:10 |
| norman | SridharG, yes | 09:10 |
| *** shausy has quit IRC | 09:11 | |
| *** yamamoto has quit IRC | 09:11 | |
| SridharG | norman: ok. As you figured it out, the floating ip is also treated as a VIP (i.e., neutron simply configures the floating ip in the keepalived.conf file and leaves it to keepalived daemon to configure it on the gateway-interface when in Master state). | 09:11 |
| sbelous | norman: sorry, I'm not very well versed in ha routers :( | 09:11 |
| *** shausy has joined #openstack-neutron | 09:12 | |
| *** Longgeek has joined #openstack-neutron | 09:12 | |
| SridharG | norman: if keepalived daemon moves to backup state, it removes the ip from the qateway-iface (i.e., qg-xxx). At any point of time, only the Master HA Router would own the IP. | 09:12 |
| SridharG | norman: ok. If you have any specific question, I'll be happy to help. | 09:12 |
| SridharG | norman: In the mean time, I'll share few links from Assaf which are very good. | 09:13 |
| norman | SridharG, yes, actually the question is that HA router did not send out the GARP when the floating ip is associated to an VM | 09:13 |
| norman | so I want to find where (in the code) it was added to the interface of namespace ,and then I want to call the arping like in the legacy_router | 09:14 |
| SridharG | norman: Ok. For an HA Router we do not send out GARP from Neutron. Keepalived after configuring the IP address to the interface, would automatically send out the GARPs/Neighbor Solicitations. | 09:15 |
| *** park_heijlong has quit IRC | 09:15 | |
| norman | SridharG, but seens that keepalived did not send out the GARPs ,and only when I manually send it out , the arp cache will be updated in another host on which I ever ping the floating ip | 09:16 |
| norman | SridharG, is it configurable? | 09:16 |
| *** sudipto has quit IRC | 09:16 | |
| SridharG | norman: there is some configuration flag in keepalived which will periodically send out GARPs, but currently we are not using it. However, if keepalived is not sending the GARPs then its an issue with keepalived version you are using. I've tested it extensively in my setup and did not find any issue related to GARPs with keepalived. | 09:17 |
| SridharG | norman: may i know the version of keepalived you are using? | 09:17 |
| *** ronis_ has joined #openstack-neutron | 09:17 | |
| norman | SridharG, Ok I will check the keepalived version, and I am working on the Centos | 09:18 |
| norman | SridharG, it's keepalived-1.2.13-1 | 09:19 |
| SridharG | numan: AFAIK it should keepalived should be sending out GARPs after transitioning to Master State and assigning the IPaddresses (including the floatingip) on the interfaces. | 09:20 |
| SridharG | oops its norman.. | 09:20 |
| SridharG | norman: in the meantime, please take a look at the following links - https://review.openstack.org/#/c/150918/ | 09:21 |
| SridharG | norman: http://assafmuller.com/2014/08/16/layer-3-high-availability/ | 09:21 |
| *** ronis has quit IRC | 09:22 | |
| norman | SridharG, hi thanks , I 'd already go through the blog of assafmuller, it was really good | 09:22 |
| *** ijw has joined #openstack-neutron | 09:22 | |
| norman | SridharG, with my testcases, I will assign the floating ip to one VM, and diassociated/associated to another VM in another router | 09:23 |
| SridharG | norman: We have an open bug in Neutron related to GARP which talks about the keepalived configuration flags - https://bugs.launchpad.net/neutron/+bug/1453855 | 09:23 |
| openstack | Launchpad bug 1453855 in neutron "HA routers may fail to send out GARPs when node boots" [Medium,New] - Assigned to Nir Magnezi (nmagnezi) | 09:23 |
| norman | and then the MAC will changed | 09:23 |
| *** aranjan has joined #openstack-neutron | 09:23 | |
| norman | without the GARP, the floating ip will be not accessbile | 09:23 |
| *** ronis__ has joined #openstack-neutron | 09:25 | |
| *** ijw has quit IRC | 09:27 | |
| *** ronis_ has quit IRC | 09:27 | |
| *** aranjan has quit IRC | 09:28 | |
| norman | SridharG, I have not run into this issue since I did not reboot my nodes ever, it was more general in my case that when the floating ip was reused ,then you may run into with it ,as least in my test env | 09:28 |
| *** breitz1 has quit IRC | 09:28 | |
| *** yamamoto has joined #openstack-neutron | 09:29 | |
| *** fawadkhaliq has joined #openstack-neutron | 09:30 | |
| *** ronis__ has quit IRC | 09:30 | |
| *** breitz has joined #openstack-neutron | 09:30 | |
| *** ronis has joined #openstack-neutron | 09:31 | |
| SridharG | norman: I'll check this use-case locally, but personally i feel MAC should not change when we move the floatingip from VM1 to VM2. | 09:33 |
| SridharG | norman: as the floatingip is configured on the gateway-interface in the neutron router. | 09:34 |
| *** Marga_ has quit IRC | 09:34 | |
| *** ronis_ has joined #openstack-neutron | 09:34 | |
| SridharG | norman: I'll be back in sometime, have to step out for a meeting. | 09:35 |
| *** ronis has quit IRC | 09:36 | |
| norman | SridharG, Ok, thanks. if the VM1 and VM2 attached to different router ,then MAC will changed | 09:36 |
| *** Jabadia has quit IRC | 09:37 | |
| *** Jabadia has joined #openstack-neutron | 09:38 | |
| *** ronis_ has quit IRC | 09:40 | |
| *** mohankumar has joined #openstack-neutron | 09:40 | |
| *** Jabadia has quit IRC | 09:42 | |
| *** eezhova has joined #openstack-neutron | 09:48 | |
| ajo | vikram, gsagie , you actually agree about preparing a QoS presentation for next summit ;) | 09:48 |
| gsagie | i think moshele/irena will also want to join | 09:50 |
| ajo | of course :) it makes sense | 09:51 |
| *** yfried_ is now known as yfried|afk | 09:52 | |
| *** aix has joined #openstack-neutron | 09:53 | |
| *** openstackgerrit has quit IRC | 09:53 | |
| *** openstackgerrit has joined #openstack-neutron | 09:53 | |
| *** Marga_ has joined #openstack-neutron | 09:57 | |
| *** yfried|afk is now known as yfried_ | 09:58 | |
| *** sudipto has joined #openstack-neutron | 10:00 | |
| *** sdake_ is now known as sdake | 10:01 | |
| *** pc_m has joined #openstack-neutron | 10:03 | |
| *** uschreiber_ has joined #openstack-neutron | 10:03 | |
| *** john-davidge has joined #openstack-neutron | 10:04 | |
| *** uschreiber_ has quit IRC | 10:05 | |
| *** Jabadia has joined #openstack-neutron | 10:08 | |
| *** mmdurrant has quit IRC | 10:09 | |
| *** red_trela has quit IRC | 10:10 | |
| vikram | thanks ajo :) i will.. | 10:15 |
| *** yamahata has quit IRC | 10:16 | |
| *** Murali has joined #openstack-neutron | 10:19 | |
| openstackgerrit | Elena Ezhova proposed openstack/neutron: Switch to oslo.service https://review.openstack.org/192999 | 10:21 |
| *** aranjan has joined #openstack-neutron | 10:24 | |
| *** lukasa has joined #openstack-neutron | 10:26 | |
| *** ronis has joined #openstack-neutron | 10:28 | |
| *** aranjan has quit IRC | 10:29 | |
| *** Piet has quit IRC | 10:30 | |
| *** markvoelker has joined #openstack-neutron | 10:34 | |
| *** rossella_s has quit IRC | 10:37 | |
| *** dims has joined #openstack-neutron | 10:37 | |
| *** rossella_s has joined #openstack-neutron | 10:37 | |
| *** markvoelker has quit IRC | 10:40 | |
| *** Marga_ has quit IRC | 10:40 | |
| *** subscope has joined #openstack-neutron | 10:42 | |
| *** jistr_ has joined #openstack-neutron | 10:49 | |
| *** afazekas__ has joined #openstack-neutron | 10:49 | |
| *** jlibosva1 has joined #openstack-neutron | 10:49 | |
| *** jistr has quit IRC | 10:51 | |
| *** stanzgy has quit IRC | 10:52 | |
| *** subscope has quit IRC | 10:52 | |
| *** apuimedo has joined #openstack-neutron | 10:52 | |
| *** jlibosva has quit IRC | 10:53 | |
| *** afazekas_ has quit IRC | 10:53 | |
| *** fawadkhaliq has quit IRC | 10:53 | |
| *** b3nt_pin has joined #openstack-neutron | 10:54 | |
| *** jlibosva1 has quit IRC | 10:54 | |
| *** jistr_ has quit IRC | 10:54 | |
| *** b3nt_pin is now known as Guest38670 | 10:54 | |
| *** afazekas__ has quit IRC | 10:54 | |
| *** apuimedo has quit IRC | 10:57 | |
| *** apuimedo has joined #openstack-neutron | 10:57 | |
| *** iwamoto has quit IRC | 11:03 | |
| *** jistr_ has joined #openstack-neutron | 11:06 | |
| *** jlibosva1 has joined #openstack-neutron | 11:07 | |
| *** afazekas__ has joined #openstack-neutron | 11:07 | |
| *** yamamoto has quit IRC | 11:08 | |
| *** ildikov has quit IRC | 11:08 | |
| *** coolsvap is now known as coolsvap|away | 11:10 | |
| *** Ramanjaneya has quit IRC | 11:12 | |
| *** Guest38670 is now known as beagles | 11:12 | |
| *** claudiub has joined #openstack-neutron | 11:17 | |
| gongysh_ | gasgie, ping | 11:18 |
| gongysh_ | gsagie: ping | 11:18 |
| gsagie | ? | 11:21 |
| gsagie | gongysh_ : ? | 11:21 |
| *** ildikov has joined #openstack-neutron | 11:22 | |
| *** aranjan has joined #openstack-neutron | 11:25 | |
| *** jlibosva1 is now known as jlibosva | 11:27 | |
| *** baoli has joined #openstack-neutron | 11:28 | |
| *** dguerri is now known as dguerri` | 11:28 | |
| *** aranjan has quit IRC | 11:29 | |
| *** baoli has quit IRC | 11:30 | |
| *** baoli has joined #openstack-neutron | 11:30 | |
| *** markvoelker has joined #openstack-neutron | 11:35 | |
| *** fawadkhaliq has joined #openstack-neutron | 11:36 | |
| *** moshele has quit IRC | 11:36 | |
| *** moshele has joined #openstack-neutron | 11:37 | |
| *** markvoelker has quit IRC | 11:40 | |
| *** baoli has quit IRC | 11:41 | |
| *** sdake has quit IRC | 11:43 | |
| *** yfried_ is now known as yfried|afk | 11:43 | |
| *** yfried|afk is now known as yfried_ | 11:45 | |
| moshele | hi, I think there is a problem with gate-neutron-dsvm-functional in the gate | 11:46 |
| *** fawadkhaliq has quit IRC | 11:46 | |
| *** dguerri` is now known as dguerri | 11:48 | |
| *** dave-mccowan has joined #openstack-neutron | 11:49 | |
| jlibosva | yep | 11:50 |
| *** amuller has joined #openstack-neutron | 11:55 | |
| *** markvoelker has joined #openstack-neutron | 11:55 | |
| *** ildikov_ has joined #openstack-neutron | 11:56 | |
| *** mohankumar has quit IRC | 11:57 | |
| *** ildikov has quit IRC | 11:57 | |
| jlibosva | moshele: I think https://review.openstack.org/#/c/190807/ caused that | 11:58 |
| *** mmdurrant has joined #openstack-neutron | 11:59 | |
| moshele | jlibosva: ok thank I will see if I can fix this | 11:59 |
| *** ohno13 has joined #openstack-neutron | 11:59 | |
| amuller | hrmph | 12:01 |
| *** ihrachyshka has quit IRC | 12:01 | |
| *** yamamoto has joined #openstack-neutron | 12:01 | |
| amuller | http://graphite.openstack.org/render/?from=-10days&height=500&until=now&width=1200&bgcolor=ffffff&fgcolor=000000&yMax=100&yMin=0&target=color(alias(movingAverage(asPercent(stats.zuul.pipeline.check.job.check-neutron-dsvm-functional.FAILURE,sum(stats.zuul.pipeline.check.job.check-neutron-dsvm-functional.{SUCCESS,FAILURE})),%2736hours%27),%20%27check-neutron-dsvm-functional%27),%27blue%27) | 12:02 |
| amuller | oops | 12:02 |
| amuller | http://goo.gl/FJPmAF | 12:02 |
| amuller | graph of functional job failure rate | 12:02 |
| amuller | it spiked on the 29th around 1AM | 12:02 |
| amuller | jlibosva: ^ | 12:03 |
| *** nmagnezi_ has joined #openstack-neutron | 12:03 | |
| *** abregman has quit IRC | 12:03 | |
| *** nmagnezi has quit IRC | 12:03 | |
| *** thorst has joined #openstack-neutron | 12:03 | |
| jlibosva | amuller: chmm, that's weird. And you need to teach me how to get such nice graphs | 12:04 |
| amuller | jlibosva: it's a bit odd, we merged 17 patches after the SafeFixture one | 12:04 |
| amuller | jlibosva: maybe one of them added a functional test, which changed the ordering so that the bug appears | 12:04 |
| jlibosva | amuller: another interesting thing is that if I run it with tox locally, it fails | 12:05 |
| amuller | really? let me try | 12:05 |
| jlibosva | amuller: if I run it with python -m testools.run , it passes | 12:06 |
| *** matrohon has joined #openstack-neutron | 12:06 | |
| *** Sanjeev has joined #openstack-neutron | 12:07 | |
| *** gampel has quit IRC | 12:08 | |
| openstackgerrit | Nikolay Fedotov proposed openstack/neutron: Do not mock arping in L3AgentTestFramework functional tests https://review.openstack.org/196608 | 12:09 |
| *** Sanjeev has quit IRC | 12:09 | |
| *** jpena is now known as jpena|lunch | 12:12 | |
| *** mohankumar has joined #openstack-neutron | 12:15 | |
| *** yfried_ has quit IRC | 12:17 | |
| *** sudipto has quit IRC | 12:18 | |
| *** ecbanks has joined #openstack-neutron | 12:21 | |
| *** yamamoto has quit IRC | 12:22 | |
| *** banix has joined #openstack-neutron | 12:23 | |
| openstackgerrit | vikram.choudhary proposed openstack/python-neutronclient: Support Command line changes for Address Scope https://review.openstack.org/194635 | 12:24 |
| *** Marga_ has joined #openstack-neutron | 12:25 | |
| *** raildo has joined #openstack-neutron | 12:25 | |
| *** yfried_ has joined #openstack-neutron | 12:25 | |
| *** yfried_ is now known as yfried|afk | 12:25 | |
| *** annemccormick has joined #openstack-neutron | 12:25 | |
| *** yfried|afk has quit IRC | 12:25 | |
| *** aranjan has joined #openstack-neutron | 12:26 | |
| *** kawa2014 has quit IRC | 12:26 | |
| *** kawa2014 has joined #openstack-neutron | 12:27 | |
| *** edmondsw has joined #openstack-neutron | 12:27 | |
| *** dguerri is now known as dguerri` | 12:28 | |
| pavel_bondar | hi, do we have a bug/review to address this failure in gate-neutron-dsvm-functional? | 12:29 |
| amuller | pavel_bondar: I don't think so | 12:30 |
| *** yamamoto has joined #openstack-neutron | 12:30 | |
| *** aranjan has quit IRC | 12:31 | |
| pavel_bondar | amuller: ack | 12:31 |
| *** dguerri` is now known as dguerri | 12:32 | |
| *** devvesa has quit IRC | 12:35 | |
| *** yamamoto has quit IRC | 12:35 | |
| *** vikram has quit IRC | 12:36 | |
| *** dave-mccowan has quit IRC | 12:36 | |
| *** mtreinish has quit IRC | 12:40 | |
| *** mtreinish has joined #openstack-neutron | 12:40 | |
| *** kiran-r has quit IRC | 12:44 | |
| *** fawadkhaliq has joined #openstack-neutron | 12:47 | |
| *** baoli has joined #openstack-neutron | 12:48 | |
| *** devvesa has joined #openstack-neutron | 12:49 | |
| *** numan has quit IRC | 12:50 | |
| amuller | jlibosva: There was a new fixtures released today | 12:51 |
| *** ildikov_ has quit IRC | 12:51 | |
| *** fawadkhaliq has quit IRC | 12:51 | |
| *** yamamoto has joined #openstack-neutron | 12:52 | |
| *** banix has quit IRC | 12:52 | |
| amuller | jlibosva: this patch passed the functional job on the 28th, then the same code based on the same patch failed when I reran the test today | 12:52 |
| jlibosva | amuller++ | 12:53 |
| amuller | jlibosva: and there was a fixtures release in between | 12:53 |
| amuller | so that's my conclusion | 12:53 |
| amuller | and it's very possible that the SafeFixture class doesn't play nice with the new fixtures release | 12:53 |
| amuller | lifeless: ^ | 12:54 |
| *** banix has joined #openstack-neutron | 12:54 | |
| openstackgerrit | Wim De Clercq proposed openstack/neutron: Add Nuage plugin configuration parameters. https://review.openstack.org/187922 | 12:55 |
| jlibosva | amuller: I see "Deal with resource leaks during setUp" patch in fixtures. That sounds like a patch to blame | 12:55 |
| amuller | either blame or, we can now revert SafeFixture in Neutron, depends :) | 12:56 |
| jlibosva | amuller: I think better to fix fixtures. The change is about setting _details (the attribute I told you about and couldn't remember the name) is set to None | 12:57 |
| amuller | if the issue is truly fixed in Fixtures... I'll test reverting SafeFixture locally and see if it cleans up resources properly without SafeFixture with the new Fixtures version | 12:57 |
| *** josecastroleon has quit IRC | 12:57 | |
| jlibosva | amuller: and the 'type' prior to that patch is a dict | 12:57 |
| *** marrusl has joined #openstack-neutron | 12:57 | |
| jlibosva | amuller: yep, actually when I was reading fixtures code this morning I was thinking why we do SafeFixture :) | 12:58 |
| amuller | it's a workaround for behavior that we think is a bug in Fixtures | 12:59 |
| amuller | it doesn't call cleanUp if your setUp failed | 12:59 |
| jlibosva | amuller: yeah but when I saw fixtures code it didn't make sense to me | 12:59 |
| amuller | you looked at it after the 1.3 release I guess | 12:59 |
| jlibosva | amuller: because I was looking at the new versino | 12:59 |
| jlibosva | amuller: yep | 12:59 |
| amuller | yeah | 12:59 |
| amuller | that's funny :) | 12:59 |
| jlibosva | bad luck | 12:59 |
| *** rpothier has joined #openstack-neutron | 13:00 | |
| *** krtaylor has quit IRC | 13:04 | |
| *** yfried|afk has joined #openstack-neutron | 13:04 | |
| *** yfried|afk has quit IRC | 13:05 | |
| *** yfried|afk has joined #openstack-neutron | 13:05 | |
| *** fifieldt has joined #openstack-neutron | 13:06 | |
| *** dave-mccowan has joined #openstack-neutron | 13:06 | |
| *** ildikov_ has joined #openstack-neutron | 13:07 | |
| amuller | dougwig: blogan: Yo, check this bug out: https://bugs.launchpad.net/neutron/+bug/1469597 | 13:07 |
| openstack | Launchpad bug 1469597 in neutron "LBaasV2 - Creating a listener fails when specifying session persistence" [High,New] | 13:07 |
| *** dave-mcc_ has joined #openstack-neutron | 13:08 | |
| *** gsagie has quit IRC | 13:11 | |
| *** dave-mccowan has quit IRC | 13:11 | |
| *** dedery_ has quit IRC | 13:12 | |
| *** zhenguo has quit IRC | 13:13 | |
| *** gampel has joined #openstack-neutron | 13:14 | |
| *** regXboi has joined #openstack-neutron | 13:14 | |
| *** yamamoto has quit IRC | 13:15 | |
| *** krtaylor has joined #openstack-neutron | 13:17 | |
| *** jpena|lunch is now known as jpena | 13:18 | |
| *** baoli has quit IRC | 13:20 | |
| *** sridhar_ram has joined #openstack-neutron | 13:22 | |
| *** irdkehn has joined #openstack-neutron | 13:23 | |
| *** julim has joined #openstack-neutron | 13:24 | |
| *** dave-mcc_ has quit IRC | 13:24 | |
| *** ihrachyshka has joined #openstack-neutron | 13:24 | |
| *** yamamoto has joined #openstack-neutron | 13:24 | |
| *** kawa2014 has quit IRC | 13:25 | |
| *** sridhar_ram has quit IRC | 13:27 | |
| *** aranjan has joined #openstack-neutron | 13:27 | |
| *** sudipto has joined #openstack-neutron | 13:27 | |
| *** josecastroleon has joined #openstack-neutron | 13:28 | |
| *** ferest has joined #openstack-neutron | 13:29 | |
| *** aranjan has quit IRC | 13:31 | |
| *** nllrte has joined #openstack-neutron | 13:32 | |
| *** edmondsw has quit IRC | 13:33 | |
| *** zz_jgrimm is now known as jgrimm | 13:33 | |
| *** absubram has quit IRC | 13:36 | |
| *** absubram has joined #openstack-neutron | 13:36 | |
| *** dave-mccowan has joined #openstack-neutron | 13:37 | |
| *** irdkehn has quit IRC | 13:37 | |
| *** ferest has quit IRC | 13:37 | |
| *** prad has joined #openstack-neutron | 13:38 | |
| *** gberginc has quit IRC | 13:38 | |
| *** afazekas__ has quit IRC | 13:38 | |
| *** irdkehn has joined #openstack-neutron | 13:39 | |
| *** kawa2014 has joined #openstack-neutron | 13:39 | |
| *** sudipto has quit IRC | 13:40 | |
| *** lnicolas has quit IRC | 13:41 | |
| *** ZZelle has joined #openstack-neutron | 13:42 | |
| *** prad has quit IRC | 13:43 | |
| *** edmondsw has joined #openstack-neutron | 13:44 | |
| *** shwetaap has joined #openstack-neutron | 13:44 | |
| *** sudipto has joined #openstack-neutron | 13:46 | |
| *** irenab has quit IRC | 13:48 | |
| ihrachyshka | HenryG, hi! I wonder how we proceed with https://review.openstack.org/#/c/192937/3 | 13:50 |
| ihrachyshka | (the zzzeek's online schema migration spec) | 13:50 |
| openstackgerrit | Ihar Hrachyshka proposed openstack/neutron: [neutron-db-manage] support separate migration subtrees https://review.openstack.org/194198 | 13:51 |
| *** Marga_ has quit IRC | 13:51 | |
| sudipto | amotoki, regarding the review of b737babd0afe4e1a i tried the wrapper approach, but may i didn't do it right. | 13:51 |
| ihrachyshka | I think I have the code ready for that ^^, so keeping in mind that it's tight on timing of the merge, I kindly ask to check the spec again and maybe the code above. should finally pass gate and be free from hackarounds | 13:51 |
| sudipto | amotoki, even with the wrapper - it seems like the signature of the find_resourceid_by_id had to be changed? | 13:52 |
| *** shwetaap1 has joined #openstack-neutron | 13:52 | |
| *** kbringard has joined #openstack-neutron | 13:52 | |
| *** fzdarsky has joined #openstack-neutron | 13:52 | |
| sudipto | amotoki, as in we would now need to pass a parameter to query instead of the hardcoded 'id'...do you agree? | 13:52 |
| *** nlahouti has quit IRC | 13:52 | |
| *** jecarey has joined #openstack-neutron | 13:52 | |
| amotoki | sudipto: why do you need to change the signature of find_resourceid_by_name_or_id? | 13:52 |
| sudipto | amotoki, not the name_or_id method | 13:53 |
| sudipto | the find_resourceid_by_id method. | 13:53 |
| amotoki | sudipto: what I think is to implement find_resource_by_id which returns a whole resource object. | 13:54 |
| *** dboik has joined #openstack-neutron | 13:54 | |
| amotoki | sudipto: I think find_resource_by_id does not need to extract a single field. | 13:54 |
| *** igordcard_ has quit IRC | 13:55 | |
| *** shwetaap has quit IRC | 13:55 | |
| *** prad has joined #openstack-neutron | 13:55 | |
| sudipto | amotoki, per your comment: Change find_resource_by_id and find_resource_by_name to return data[collection][0] or info[0] instead of returning xxx['id'] | 13:56 |
| sudipto | amotoki, i am guessing you meant find_resourceid_by_id - that method? | 13:56 |
| sudipto | instead of find_resource_by_id | 13:57 |
| *** superdan is now known as dansmith | 13:57 | |
| amotoki | sudipto: my above comment is based on my comment in patch set 2. | 13:57 |
| amotoki | sudipto: I think we don't need to stick find_resource*id*_by_id | 13:58 |
| sudipto | amotoki, ah got it. | 13:58 |
| *** boris-42 has joined #openstack-neutron | 13:59 | |
| *** amotoki has quit IRC | 13:59 | |
| *** subscope has joined #openstack-neutron | 13:59 | |
| *** fawadkhaliq has joined #openstack-neutron | 14:03 | |
| *** subscope has quit IRC | 14:04 | |
| *** lukasa has quit IRC | 14:05 | |
| *** fawadkhaliq has quit IRC | 14:08 | |
| openstackgerrit | Jakub Libosvar proposed openstack/neutron: Don't use fixture 1.3.0 yet https://review.openstack.org/196677 | 14:09 |
| jlibosva | amuller: I created a bug and proposed a workaround ^^ | 14:09 |
| jlibosva | amuller: I tested that 1.3.0 still doesn't call the cleanup in case setup crashes | 14:10 |
| amuller | jlibosva: that's dissapointing | 14:10 |
| openstackgerrit | Elena Ezhova proposed openstack/neutron: Handle port_other_config=None in port_bound https://review.openstack.org/183443 | 14:10 |
| *** absubram has quit IRC | 14:11 | |
| jlibosva | amuller: I'm looking at fixtures now | 14:11 |
| ZZelle | enikanorov2, hi | 14:12 |
| *** abregman has joined #openstack-neutron | 14:13 | |
| jlibosva | amuller: I think I have a solution | 14:14 |
| *** absubram has joined #openstack-neutron | 14:15 | |
| *** dtynan has quit IRC | 14:15 | |
| openstackgerrit | Sudipta Biswas proposed openstack/python-neutronclient: Determine ip version during subnet create. https://review.openstack.org/194078 | 14:15 |
| *** absubram has quit IRC | 14:15 | |
| *** busterswt has joined #openstack-neutron | 14:16 | |
| *** dtynan has joined #openstack-neutron | 14:16 | |
| openstackgerrit | Pavel Bondar proposed openstack/neutron: Add Pluggable IPAM Backend https://review.openstack.org/183099 | 14:17 |
| openstackgerrit | Pavel Bondar proposed openstack/neutron: Enable pluggable IPAM implementation by default https://review.openstack.org/181023 | 14:18 |
| *** lori|away is now known as lori | 14:18 | |
| *** jecarey has quit IRC | 14:18 | |
| *** r-daneel has joined #openstack-neutron | 14:19 | |
| *** dave-mccowan has quit IRC | 14:20 | |
| *** dims has quit IRC | 14:21 | |
| ajo | garyk, ihrachyshka : https://review.openstack.org/189628 and https://review.openstack.org/193584 are basic stubs we could merge, to create an experimental job loading the service | 14:21 |
| ajo | they don't do nothing, and they just need follow up patches to address a few of the comments (I addressed more) but that will happen during the next 3 days :) | 14:22 |
| *** localloop127 has joined #openstack-neutron | 14:24 | |
| *** baoli has joined #openstack-neutron | 14:24 | |
| *** dims has joined #openstack-neutron | 14:24 | |
| *** subscope has joined #openstack-neutron | 14:24 | |
| *** yamahata has joined #openstack-neutron | 14:25 | |
| *** lindycoder has quit IRC | 14:27 | |
| *** lindycoder has joined #openstack-neutron | 14:27 | |
| *** aranjan has joined #openstack-neutron | 14:28 | |
| *** kir_ has quit IRC | 14:29 | |
| *** kir_ has joined #openstack-neutron | 14:29 | |
| *** aranjan has quit IRC | 14:32 | |
| amuller | jlibosva: so fixtures 1.3 and reverting SafeFixture, we still don't clean up properly in case setUp fails? | 14:34 |
| *** dave-mccowan has joined #openstack-neutron | 14:34 | |
| jlibosva | amuller: According https://github.com/testing-cabal/fixtures/blob/master/README#L82 we should use _setUp() instead of setUp() | 14:34 |
| jlibosva | amuller: then we *should* cleanup properly | 14:35 |
| jlibosva | amuller: I'm testing that | 14:35 |
| amuller | jlibosva: ok awesome | 14:35 |
| jlibosva | amuller: but I'm still getting the None type error :-/ | 14:36 |
| amuller | jlibosva: wow this means changing every single setUp we have to _setUp, not a very friendly change I have to say | 14:36 |
| jlibosva | amuller: only setUp for fixtures | 14:36 |
| jlibosva | "only" | 14:36 |
| amuller | we have a few dozen of those | 14:36 |
| ZZelle | jlibosva, easy | 14:36 |
| *** rossella_s has quit IRC | 14:36 | |
| *** rossella_s has joined #openstack-neutron | 14:37 | |
| ZZelle | jlibosva, amuller, fixtures are located in few modules mustly without testcases | 14:37 |
| *** haleyb has joined #openstack-neutron | 14:37 | |
| *** dguitarbite has quit IRC | 14:37 | |
| *** sdake has joined #openstack-neutron | 14:37 | |
| jlibosva | yeah, I don't think there are that many fixtures | 14:37 |
| *** annegentle has joined #openstack-neutron | 14:37 | |
| ZZelle | jlibosva, ~ 30 | 14:38 |
| *** annegentle has quit IRC | 14:38 | |
| jlibosva | =-O | 14:38 |
| *** annegentle has joined #openstack-neutron | 14:39 | |
| *** lucas__ has joined #openstack-neutron | 14:39 | |
| jlibosva | $ git grep "class .*Fixture" | wc -l | 14:39 |
| jlibosva | 40 | 14:39 |
| *** sdake_ has joined #openstack-neutron | 14:40 | |
| *** yamamoto has quit IRC | 14:41 | |
| amuller | jlibosva: but you tested it with a specific fixture and it didn't solve the problem? | 14:41 |
| amuller | I mean inheriting directly from fixtures.Fixture and implementing _setUp | 14:41 |
| jlibosva | amuller: yep, I still get TypeError | 14:41 |
| jlibosva | amuller: also, look at this | 14:41 |
| amuller | jlibosva: not the functional test | 14:42 |
| *** lucas__ has quit IRC | 14:42 | |
| amuller | jlibosva: just checking manually if clean ups are called | 14:42 |
| *** lucas__ has joined #openstack-neutron | 14:42 | |
| *** yog__ has quit IRC | 14:42 | |
| jlibosva | amuller: https://github.com/testing-cabal/fixtures/blob/master/fixtures/tests/test_fixture.py#L239 | 14:42 |
| ZZelle | amuller, according to the change i reviewed, fixtures cleanup correctly | 14:43 |
| jlibosva | amuller: I rewrote the test we have for SafeFixture | 14:43 |
| jlibosva | ZZelle: do you understand the line I sent to Assaf above ^^ ? | 14:43 |
| *** gsagie has joined #openstack-neutron | 14:43 | |
| *** yamamoto has joined #openstack-neutron | 14:43 | |
| *** sdake has quit IRC | 14:43 | |
| ZZelle | jlibosva, tests to test test tools :) | 14:44 |
| jlibosva | ZZelle: I mean why should cleanUp() raise TypeError? :-/ | 14:44 |
| *** fzdarsky has quit IRC | 14:45 | |
| ZZelle | jlibosva, fixtures behave incorrectly when cleanUp is called without Fxiture.setUp not called (because of details not initialized) | 14:45 |
| jlibosva | ZZelle: what does it mean? Does it mean we can't use cleanup if setup fails? I thought that was the aim | 14:47 |
| jlibosva | the goal | 14:47 |
| *** ihrachyshka has quit IRC | 14:47 | |
| ZZelle | ok, one sec a reference to the code will help | 14:47 |
| amuller | jlibosva: I think it's assumed that _setUp will call super().setUp? | 14:47 |
| amuller | as the first line | 14:47 |
| amuller | ZZelle: ya? | 14:47 |
| jlibosva | amuller: nah | 14:48 |
| jlibosva | amuller: I think setUp() calls _setUp() and wraps it with try block | 14:48 |
| *** shausy has quit IRC | 14:48 | |
| jlibosva | amuller: at least that would make sense to me | 14:48 |
| amuller | uh huh | 14:48 |
| jlibosva | amuller: /me is right | 14:48 |
| jlibosva | amuller: proof: https://github.com/testing-cabal/fixtures/blob/master/fixtures/fixture.py#L195 | 14:49 |
| amuller | jlibosva: so I don't understand the line you pasted | 14:49 |
| amuller | line 239 | 14:49 |
| *** timsim has left #openstack-neutron | 14:49 | |
| *** fawadkhaliq has joined #openstack-neutron | 14:49 | |
| jlibosva | amuller: me neither :-/ IIUC the line says "if setup throws up exception and you call cleanUp() then cleanUp() will puke TypeError" - and I miss the point of doing so | 14:50 |
| ZZelle | jlibosva, it ensures to crash? | 14:51 |
| jlibosva | ZZelle: who? cleanUp? | 14:51 |
| *** apuimedo has quit IRC | 14:52 | |
| *** ildikov_ has quit IRC | 14:52 | |
| jlibosva | lifeless: are you present? | 14:52 |
| *** pckizer_ is now known as pckize | 14:52 | |
| *** pckize is now known as pckizer | 14:52 | |
| *** hichihara has joined #openstack-neutron | 14:52 | |
| ZZelle | jlibosva, "getDetails tries to dereference self._details" from https://github.com/testing-cabal/fixtures/pull/11/files | 14:54 |
| *** Marga_ has joined #openstack-neutron | 14:55 | |
| jlibosva | ZZelle: yes and it results in TypeError and according UT from fixtures I sent you above it seems the TypeError is expected, which I don't understand why | 14:56 |
| ZZelle | jlibosva, now i am lost | 14:56 |
| jlibosva | ZZelle: ok, so let the story begins now :) | 14:56 |
| *** mitchjameson has joined #openstack-neutron | 14:56 | |
| jlibosva | ZZelle: fixtures 1.3.0 was released including the change you sent me minute ago | 14:56 |
| jlibosva | ZZelle: IIUC that change is supposed to fix leakes that you fixed in Neutron two days ago | 14:57 |
| *** fawadk has joined #openstack-neutron | 14:57 | |
| jlibosva | s/leakes/leaks/ | 14:57 |
| jlibosva | ZZelle: we have problem on gate that these two changes seem incompatible, meaning new 1.3.0 fixtures release broke our tests. | 14:57 |
| *** pradipta has joined #openstack-neutron | 14:58 | |
| *** pckizer is now known as pck | 14:58 | |
| jlibosva | ZZelle: according the change in fixtures, we should use _setUp() method instead of setUp() for all classes inheriting from Fixtures() class in order to avoid leaks | 14:58 |
| *** SridharG has left #openstack-neutron | 14:59 | |
| jlibosva | ZZelle: stated above describes quite ideal state :) Now - I'm testing locally whether replacing setUp() to _setUp() will actually call cleanUp() method in case _setUp() raises an exception | 14:59 |
| *** irenab has joined #openstack-neutron | 14:59 | |
| jlibosva | ZZelle: still on the track? | 14:59 |
| ZZelle | jlibosva, yes, i am doing tests in parallel | 15:00 |
| *** patrickeast has joined #openstack-neutron | 15:00 | |
| *** nyechiel has joined #openstack-neutron | 15:00 | |
| *** ijw has joined #openstack-neutron | 15:00 | |
| *** fawadkhaliq has quit IRC | 15:00 | |
| *** kiran-r has joined #openstack-neutron | 15:01 | |
| *** agireud has joined #openstack-neutron | 15:01 | |
| jlibosva | ZZelle: ok, now comes the problem I am observing - that is cleanUp() is called, ineed. But it still raises TypeError because of _details in fixture instance is set to None even though I use _setUp() method | 15:01 |
| * jlibosva has a typo day | 15:01 | |
| jlibosva | s/ineed/indeed/ | 15:01 |
| ZZelle | jlibosva, TypeError is raised after cleanUp is done? | 15:02 |
| *** ihrachyshka has joined #openstack-neutron | 15:03 | |
| *** carl_baldwin has joined #openstack-neutron | 15:03 | |
| *** Murali has quit IRC | 15:04 | |
| *** absubram has joined #openstack-neutron | 15:04 | |
| jlibosva | ZZelle: yes | 15:04 |
| *** localloop127 has quit IRC | 15:04 | |
| jlibosva | ZZelle: or wait, maybe not | 15:05 |
| jlibosva | ZZelle: maybe before | 15:05 |
| *** ildikov_ has joined #openstack-neutron | 15:05 | |
| jlibosva | ZZelle: aha, so TypeError comes from useFixture, so maybe the line from UT is not even relevant and just makes sure that cleanup cannot be called twice | 15:06 |
| * jlibosva becomes more confused | 15:06 | |
| *** zzzeek has joined #openstack-neutron | 15:07 | |
| ZZelle | jlibosva, raising errors in _setUp behaves as expected: cleanUp is called | 15:07 |
| *** moshele has quit IRC | 15:07 | |
| *** annp has joined #openstack-neutron | 15:07 | |
| jlibosva | ZZelle: how do you call _setUp()? | 15:08 |
| ZZelle | jlibosva, using setUp | 15:08 |
| jlibosva | ZZelle: can you try it with useFixture? | 15:08 |
| ZZelle | jlibosva, yes | 15:10 |
| *** claudiub has quit IRC | 15:11 | |
| *** Jabadia has quit IRC | 15:11 | |
| ZZelle | jlibosva, "TypeError: 'NoneType' object is not iterable" | 15:11 |
| *** itzikb has quit IRC | 15:12 | |
| jlibosva | ZZelle: you made me happy, I'm not crazy :) | 15:12 |
| *** jckasper has joined #openstack-neutron | 15:13 | |
| ZZelle | with testtools.useFxiture ... fixtures.useFixture behaves differently! | 15:13 |
| ZZelle | lol | 15:13 |
| *** sudipto has quit IRC | 15:13 | |
| amuller | so this is essentially a fixtures bug still | 15:13 |
| jlibosva | k, I'm reporting a bug | 15:14 |
| ZZelle | jlibosva, amuller, it's not a NEW bug | 15:14 |
| jlibosva | ZZelle: I think it's a regression caused by that patch to support cleanups | 15:14 |
| *** mitchjameson has quit IRC | 15:15 | |
| jlibosva | ZZelle: that's the place where _details is set to None | 15:15 |
| ZZelle | jlibosva, 1sec | 15:15 |
| *** dannywilson has joined #openstack-neutron | 15:16 | |
| scheuran | mestery: Kyle, could you please give your +1 on https://review.openstack.org/189644 some time this day, that the infra team can continue to set up the repository? Thx. | 15:17 |
| *** dannywilson has quit IRC | 15:18 | |
| ZZelle | jlibosva, https://bugs.launchpad.net/python-fixtures/+bug/1323566 | 15:18 |
| openstack | Launchpad bug 1323566 in Python Fixtures "details fail to be collected from fixture " [High,Triaged] | 15:18 |
| *** dannywilson has joined #openstack-neutron | 15:18 | |
| ZZelle | jlibosva, not this one | 15:18 |
| *** ronis has quit IRC | 15:19 | |
| mestery | scheuran: Done, sorry it took so long, I was traveling last week. | 15:20 |
| *** yamamoto has quit IRC | 15:20 | |
| *** sudipto has joined #openstack-neutron | 15:21 | |
| jlibosva | ZZelle: I think it's a new one. cleanUp() calls _remove_state() that sets _details to None. This is behavior is introduced by the patch with _setUp() | 15:21 |
| *** nmagnezi_ has quit IRC | 15:21 | |
| jlibosva | ZZelle: _details is later used in useFixture() creating a dict() out of the _details attribute | 15:22 |
| ZZelle | strange, i was sure there was a bug opened because details fail when setUp fails | 15:22 |
| *** kiran-r has quit IRC | 15:22 | |
| openstackgerrit | Sergey Belous proposed openstack/neutron: Adds base in-tree functional testing of the dhcp agent (OVS) https://review.openstack.org/136834 | 15:23 |
| ZZelle | jlibosva, ok, it's not a new bug, just a bug i found without reporting it O_o | 15:23 |
| *** yamamoto has joined #openstack-neutron | 15:23 | |
| jlibosva | ZZelle: then I took from you the pleasure to report it ;) https://bugs.launchpad.net/python-fixtures/+bug/1469759 | 15:24 |
| openstack | Launchpad bug 1469759 in Python Fixtures "if _setUp() fails in useFixture(fixture) then TypeError is raised" [Undecided,New] | 15:24 |
| amuller | we need a short term solution to unblock the gate | 15:25 |
| ZZelle | jlibosva, so now the neutron trouble | 15:25 |
| amuller | ya | 15:25 |
| jlibosva | amuller: the short term solution is https://review.openstack.org/#/c/196677/ | 15:25 |
| ZZelle | amuller, revert my change! | 15:25 |
| jlibosva | amuller: but it's too punky and breaks politics | 15:26 |
| amuller | jlibosva: that depends if we need a global change or not | 15:26 |
| scheuran | mestery: great, thanks! | 15:26 |
| jlibosva | amuller: I see, it depends how much do we like punk | 15:26 |
| jlibosva | amuller: but maybe revert is also an option. But that will lead to leaks | 15:27 |
| scheuran | mestery: could you please also check the governance patch? https://review.openstack.org/#/c/195907 | 15:27 |
| *** david-ly_ is now known as david-lyle | 15:27 | |
| *** anshul has quit IRC | 15:27 | |
| openstackgerrit | Merged openstack/neutron-specs: "Get me a network" spec https://review.openstack.org/184857 | 15:27 |
| mestery | scheuran: Done | 15:28 |
| jlibosva | amuller: anyway, if it's a bug in 1.3.0 then maybe global requirement for !=1.3.0 makes sene | 15:28 |
| scheuran | mestery: :) | 15:28 |
| ZZelle | amuller, jlibosva, what is the trouble we get currently? | 15:28 |
| jlibosva | ZZelle: functional gives -1 | 15:28 |
| *** aranjan has joined #openstack-neutron | 15:29 | |
| amuller | ZZelle: jlibosva: We could revert the SafeFixture patch, unless there's an easy way for SafeFixture to work with fixtures 1.3.0 AND solve the leaks problem | 15:29 |
| jlibosva | ZZelle: because of _details set to None in fixtures | 15:29 |
| *** armax has joined #openstack-neutron | 15:29 | |
| *** jbell8 has joined #openstack-neutron | 15:30 | |
| jlibosva | amuller: I think we need lifeless to explain whether it's a fixture bug or not. I have a feeling we'll still need to avoid 1.3.0 version | 15:30 |
| ZZelle | jlibosva, amuller, we can revert SafeFixture or just its tests | 15:30 |
| jlibosva | ZZelle: yep | 15:31 |
| amuller | I didn't have time to test this myself, just finished a bunch of meetings, SafeFixture + fixtures 1.3.0, the issue we have, is this only in the functional test for SafeFixture, or is this a problem with SafeFixture just not working? | 15:32 |
| ZZelle | jlibosva, amuller, correct SafeFixture tests is possible also | 15:32 |
| *** thedodd has joined #openstack-neutron | 15:32 | |
| jlibosva | ZZelle: I need to go, I'm already late :( | 15:32 |
| ZZelle | jlibosva, me too | 15:32 |
| amuller | ok | 15:32 |
| jlibosva | amuller: well | 15:32 |
| ZZelle | amuller, i can correct this in 3/4h | 15:33 |
| jlibosva | amuller: it's a problem with our functional test testing SafeFixtures | 15:33 |
| amuller | ok | 15:33 |
| amuller | now I have time I'll look in to this and propose a fix | 15:33 |
| jlibosva | amuller: but anyway if we'll use fixtures correctly how it's designed, we'll still hit it eventually | 15:33 |
| *** ajmiller has joined #openstack-neutron | 15:33 | |
| *** aranjan has quit IRC | 15:34 | |
| ZZelle | jlibosva, when the setUp fails so it's fine as the test is already considered failed | 15:34 |
| jlibosva | ZZelle: right | 15:34 |
| ZZelle | amuller, you need to update https://github.com/openstack/neutron/blob/master/neutron/tests/functional/test_tools.py#L76-L81 | 15:34 |
| jlibosva | ZZelle: the previous one | 15:35 |
| jlibosva | test_error_after_root_setup | 15:35 |
| jlibosva | I think ... | 15:35 |
| *** jlibosva has quit IRC | 15:35 | |
| amuller | I just want to confirm this: When using SafeFixture and fixtures 1.3, does SafeFixture call cleanups properly when setUp fails? | 15:35 |
| amuller | If so I'll just remove the functional tests for SafeFixture and call it a day | 15:35 |
| amuller | ZZelle: ^ | 15:36 |
| ZZelle | amuller, yes | 15:36 |
| amuller | ZZelle: alright | 15:36 |
| amuller | I'll remove the tests for it then | 15:36 |
| amuller | ZZelle: And if you feel like it you can add a fixed version back in your free time | 15:36 |
| ZZelle | amuller, it's ensured by https://github.com/openstack/neutron/blob/master/neutron/tests/tools.py#L40-L61 ... the 1st method ensures to call safe_cleanup, the 2nd to clean up when needed | 15:37 |
| ZZelle | amuller, will do | 15:37 |
| *** achanda has joined #openstack-neutron | 15:37 | |
| *** kiran-r has joined #openstack-neutron | 15:38 | |
| ZZelle | amuller, we need to support fixtures 1.3 and lower as the requirement is fixtures >.3.14 :s | 15:38 |
| *** flashgordon has quit IRC | 15:39 | |
| *** moha_hunt has joined #openstack-neutron | 15:40 | |
| ZZelle | amuller, when fixtures>1.3 will be updated in requirements .txt, i will remove SafeFixture and moves to _setUp | 15:40 |
| *** Aless has joined #openstack-neutron | 15:40 | |
| amuller | ZZelle: yes assuming fixtures 1.3.1 or whatever fixes the issue completely | 15:40 |
| amuller | then we can move to it | 15:41 |
| *** gampel has quit IRC | 15:42 | |
| *** mitchjameson has joined #openstack-neutron | 15:42 | |
| rossella_s | amuller hi! you asked me regarding the func tests of the OVS agent. I tried to reproduce the failure locally and I failed...I tried like 20 times with a loaded host | 15:44 |
| *** Marga_ has quit IRC | 15:44 | |
| openstackgerrit | Assaf Muller proposed openstack/neutron: Remove failing SafeFixture tests https://review.openstack.org/196711 | 15:44 |
| *** yamamoto has quit IRC | 15:46 | |
| *** Sukhdev has joined #openstack-neutron | 15:46 | |
| amuller | mestery: can you ninja merge that to unbreak the gate? | 15:46 |
| amuller | A self ninja merge is too cheeky even for me | 15:46 |
| *** scheuran has quit IRC | 15:46 | |
| *** kiran-r has quit IRC | 15:47 | |
| *** annegentle has quit IRC | 15:48 | |
| *** nyechiel has quit IRC | 15:48 | |
| *** annegentle has joined #openstack-neutron | 15:49 | |
| *** matrohon has quit IRC | 15:49 | |
| openstackgerrit | Pavel Bondar proposed openstack/neutron: Add Pluggable IPAM Backend https://review.openstack.org/183099 | 15:50 |
| *** vikram has joined #openstack-neutron | 15:50 | |
| amuller | rossella_s: I think I know what the problem is | 15:51 |
| *** tidwellr has joined #openstack-neutron | 15:51 | |
| amuller | rossella_s: This patch: https://review.openstack.org/#/c/151501/ was merged on June 19th 11:23AM, your patch passed the check queue before 11:23AM but failed the gate queue after 11:23AM | 15:52 |
| amuller | rossella_s: And that patch I linked screws up ancillary bridges | 15:52 |
| amuller | rossella_s: and your patch was failing in the setup ancillary bridges method | 15:52 |
| amuller | rossella_s: Rebase your patch on the revert I pushed here: https://review.openstack.org/#/c/196392/ | 15:53 |
| *** mitchjameson has quit IRC | 15:53 | |
| amuller | then I bet that if we recheck it 20 times it'll pass just fine | 15:53 |
| *** hichihara has quit IRC | 15:54 | |
| *** flashgordon has joined #openstack-neutron | 15:55 | |
| rossella_s | amuller good catch! let me do that | 15:57 |
| openstackgerrit | Rossella Sblendido proposed openstack/neutron: Adds base in-tree functional testing of the ovs_neutron_agent https://review.openstack.org/140042 | 15:57 |
| amuller | rossella_s: oh but the revert I proposed was not merged yet | 15:58 |
| amuller | rossella_s: so rebasing on HEAD is not good enough | 15:58 |
| amuller | you can just wait a few hours and the revert will probably me merged | 15:58 |
| amuller | yeah markmcclain just approved it | 15:59 |
| Aless | amuller: i just came from reading your DVR Articles, they are excellent | 15:59 |
| *** jistr_ has quit IRC | 15:59 | |
| amuller | rossella_s: I'd just wait a couple of hours until that patch is merged then rebase on updated HEAD | 15:59 |
| *** phil_h has joined #openstack-neutron | 16:00 | |
| amuller | rossella_s: otherwise I *think* if you rebase on an unmerged patch, then when that patch gets merged Gerrit becomes stupid and says that you have to rebase your patch | 16:00 |
| Aless | but there i still something i dont get - how do i handle two Routers in a Clients Network in a DVR Environment? | 16:00 |
| amuller | Aless: Thanks I'm glad it helped you | 16:00 |
| amuller | Aless: I'm not following | 16:01 |
| Aless | I have 192.168.1.1 as a Distributed Router and 192.168.1.4 as SNAT Router.... do i have to have a router in front of the routers? :-) | 16:01 |
| *** dave-mccowan has quit IRC | 16:01 | |
| *** fawadk has quit IRC | 16:01 | |
| rossella_s | amuller: by mistake I rebased few minutes ago...anyway to get your patch I will rebase in a couple of hours thenthen... | 16:01 |
| *** yamamoto has joined #openstack-neutron | 16:01 | |
| *** mitchjameson has joined #openstack-neutron | 16:02 | |
| mestery | amuller: need the ninja merge still? | 16:02 |
| amuller | mestery: yes, here: https://review.openstack.org/#/c/196711/ | 16:03 |
| Aless | amuller: in you're article your mentioning a router with two different weighed routes, as from what i understand one ist the distributed one (L3 of the compute) and the other one is the central SNAT am I assuming correctly? | 16:03 |
| *** sthillma has joined #openstack-neutron | 16:03 | |
| mestery | amuller: done :) | 16:03 |
| *** localloop127 has joined #openstack-neutron | 16:03 | |
| amuller | thanks, we'll need to recheck all of the recently approved patches after that one's merged | 16:04 |
| *** jlanoux has quit IRC | 16:07 | |
| *** kiran-r has joined #openstack-neutron | 16:07 | |
| *** sridhar_ram has joined #openstack-neutron | 16:07 | |
| *** Marga_ has joined #openstack-neutron | 16:07 | |
| *** annegentle has quit IRC | 16:07 | |
| openstackgerrit | Ihar Hrachyshka proposed openstack/neutron: [neutron-db-manage] support separate migration subtrees https://review.openstack.org/194198 | 16:08 |
| *** rwsu has joined #openstack-neutron | 16:08 | |
| *** busterswt has quit IRC | 16:08 | |
| *** busterswt has joined #openstack-neutron | 16:11 | |
| *** anilvenkata has quit IRC | 16:11 | |
| *** kiran-r has quit IRC | 16:11 | |
| mestery | amuller: Got it | 16:11 |
| *** lazy_prince has joined #openstack-neutron | 16:13 | |
| *** lucas__ has quit IRC | 16:16 | |
| *** madhu_ak has joined #openstack-neutron | 16:17 | |
| *** claudiub has joined #openstack-neutron | 16:19 | |
| *** s3wong has joined #openstack-neutron | 16:20 | |
| *** dave-mccowan has joined #openstack-neutron | 16:21 | |
| *** mlavalle has joined #openstack-neutron | 16:21 | |
| *** gsagie has quit IRC | 16:22 | |
| *** busterswt has quit IRC | 16:24 | |
| *** ygbo has quit IRC | 16:25 | |
| *** kiran-r has joined #openstack-neutron | 16:25 | |
| *** packet has joined #openstack-neutron | 16:25 | |
| *** fawadkhaliq has joined #openstack-neutron | 16:27 | |
| *** spligak has joined #openstack-neutron | 16:27 | |
| *** gsagie has joined #openstack-neutron | 16:28 | |
| *** doude has quit IRC | 16:28 | |
| *** baoli has quit IRC | 16:29 | |
| *** lucas__ has joined #openstack-neutron | 16:29 | |
| *** aranjan has joined #openstack-neutron | 16:29 | |
| *** kiran-r has quit IRC | 16:30 | |
| *** kiran-r has joined #openstack-neutron | 16:30 | |
| *** vikram has quit IRC | 16:30 | |
| *** baoli has joined #openstack-neutron | 16:31 | |
| *** jpena has quit IRC | 16:31 | |
| Aless | amuller: Or everyone really, what could be wrong when http://www.directupload.net/file/d/4033/8735wkan_png.htm my router looks like that? | 16:31 |
| Aless | why do I have two routing interfaces within my VM Network and how am I supposed to deal with that? | 16:31 |
| amuller | Aless: That is how it's supposed to look like | 16:32 |
| *** _cjones_ has joined #openstack-neutron | 16:32 | |
| *** yamahata has quit IRC | 16:32 | |
| amuller | Aless: your 192.168.1.1 is the subnet's default gateway | 16:32 |
| amuller | that is the neutron port on the compute nodes, the distributed part | 16:32 |
| regXboi | carl_baldwin, mestery: ping? wanted to catch up on routed networks... where did we come out from the sprint? | 16:33 |
| amuller | 192.168.1.4 is the SNAT port, present only where the SNAT portion of the router is scheduled to | 16:33 |
| Aless | amuller: how do I tell the machines that are not using the distributed router that 192.168.1.4 is the main gateway? | 16:33 |
| amuller | Aless: you don't | 16:33 |
| amuller | Aless: all VMs connected to that subnet will use that distributed router | 16:33 |
| Aless | amuller: I see so I just base all on the distributed one. | 16:33 |
| Aless | i see | 16:33 |
| amuller | yeah | 16:34 |
| Aless | may I ask why I even see it then? :-) | 16:34 |
| Aless | because of the port assignment? | 16:34 |
| amuller | when you create another VM on another hypervisor, the distributed portion of the router will be auto scheduled to that hypervisor | 16:34 |
| *** aranjan has quit IRC | 16:34 | |
| amuller | and routing will occur locally with 192.168.1.1 as the gateway | 16:34 |
| regXboi | Aless: I could argue that as a tenant, maybe I don't want to see it, but as an operator, I really *do* want to see it | 16:34 |
| amuller | Aless: Good question, you know I thought we did hide that | 16:34 |
| carl_baldwin | regXboi: We change the proposed model a bit. I plan to write it up today as a devref: smaller than the current spec proposed. I will split out the spec a bit. | 16:34 |
| regXboi | carl_baldwin: thx | 16:35 |
| amuller | regXboi: Aless: We actually hide similar resources with a blank tenant_id | 16:35 |
| *** krtaylor has quit IRC | 16:35 | |
| regXboi | amuller: yeah, I'm not convinced that is a good idea | 16:35 |
| *** Kiall has joined #openstack-neutron | 16:35 | |
| *** _kiran_ has joined #openstack-neutron | 16:35 | |
| amuller | I thought we did the same for DVR SNAT ports | 16:35 |
| regXboi | amuller: I'd prefer a magic tenant for that - because having empty tenant-ids can do nasty things downstream | 16:35 |
| *** kiran-r has quit IRC | 16:36 | |
| amuller | regXboi: can you expand on that? What issues are you seeing? | 16:36 |
| Aless | amuller: i stumbled upon it because my VMs who have a floating IP are available through that IP but cannot get "outside"... | 16:36 |
| Aless | is there an easy way to "track" the point where they deal with eachother? | 16:37 |
| *** timsim has joined #openstack-neutron | 16:37 | |
| *** timsim has left #openstack-neutron | 16:37 | |
| regXboi | amuller: so, I have a downstream mechanism different from the reference implementation, which wants to pick up the SNAT address and use it | 16:38 |
| *** kawa2014 has quit IRC | 16:38 | |
| amuller | I see | 16:38 |
| regXboi | amuller: prior to DVR, that router with SNAT was in the tenant context | 16:38 |
| *** yamamoto has quit IRC | 16:38 | |
| *** jckasper has quit IRC | 16:38 | |
| regXboi | moving the SNAT out of the tenant context *without telling people* isn't really polite | 16:38 |
| * regXboi admits that he's whining about documentation lag | 16:39 | |
| *** dguerri is now known as dguerri` | 16:40 | |
| amuller | regXboi: It's not an easy sell to move the SNAT ports out of the tenant context after 2 releases | 16:40 |
| *** devvesa has quit IRC | 16:40 | |
| Sam-I-Am | regXboi: what documentation? | 16:41 |
| amuller | also we'll have a compatability issue in the code since some ports will be with a tenant_id and some without | 16:41 |
| amuller | so I don't see this happening | 16:41 |
| *** yamamoto has joined #openstack-neutron | 16:41 | |
| amuller | with L3 HA we did that from the get go | 16:41 |
| regXboi | amuller: I don't want to move them out - I want the ones that don't have a tenant_id to get one | 16:41 |
| *** neeti has quit IRC | 16:41 | |
| amuller | regXboi: Yeah I understand :) I'm just explaining why DVR SNAT ports won't move out, it's not feasible | 16:41 |
| Sam-I-Am | what about that flag to mark certain networks as special? | 16:42 |
| regXboi | amuller: I can live with that | 16:42 |
| *** agireud_ has joined #openstack-neutron | 16:42 | |
| amuller | so we're talking about L3 HA router ports, and the per-tenant HA network. Also the port backing a floating IP. | 16:43 |
| *** Swami has joined #openstack-neutron | 16:43 | |
| *** jgrimm has quit IRC | 16:43 | |
| amuller | I personally think that hiding them is a good idea, but giving the admin a nice way to track which resources belong where would be nice also | 16:43 |
| *** _cjones_ has quit IRC | 16:43 | |
| *** annp has quit IRC | 16:43 | |
| amuller | Sam-I-Am: a 'hidden' flag would solve that | 16:43 |
| *** _cjones_ has joined #openstack-neutron | 16:43 | |
| Sam-I-Am | yezzz | 16:43 |
| Aless | btw what is blocking ha l3 from working with DVR? the concerns sound orthogonal to the outsider | 16:44 |
| Sam-I-Am | seems like it might solve a lot of things | 16:44 |
| amuller | so the resource would belong to a tenant but if it were marked with 'hidden' it would not be returned by API calls | 16:44 |
| regXboi | amuller: yeah, I wasn't really happy with the null tenant-id because you've said what I wanted - hide them yes, but let the operator know how is with what | 16:44 |
| amuller | so only admins could see them | 16:44 |
| *** agireud has quit IRC | 16:44 | |
| *** agireud_ is now known as agireud | 16:44 | |
| regXboi | amuller, Sam-I-Am: isn't there some bp around RBAC and wouldn't that solve this problem? | 16:44 |
| Sam-I-Am | there was also some discussion at the sprint about marking 'preferred' networks | 16:44 |
| amuller | Aless: we have patches up for review that solve the L3 HA and DVR integration issue, had for a while, we hope to land it in L | 16:44 |
| Sam-I-Am | seems like L3HA would only benefit the fixed-ip (central SNAT) problem with DVR | 16:45 |
| Sam-I-Am | and even then its only failover | 16:45 |
| amuller | that's the goal yes | 16:45 |
| amuller | there is no HA to provide to the distributed portion since it's distributed... | 16:45 |
| Sam-I-Am | not sure why dvr just didnt use an ip on each compute node for snat. it eats them up anyway. | 16:45 |
| amuller | but we'll provide fast failover for the SNAT portion | 16:45 |
| regXboi | Sam-I-Am: that doesn't work | 16:46 |
| Sam-I-Am | it does for nova-net | 16:46 |
| *** nlahouti has joined #openstack-neutron | 16:46 | |
| amuller | Sam-I-Am: because you can't put a unique ipv4 per host | 16:46 |
| amuller | we could with ipv6 | 16:46 |
| Swami | Sam-I-Am: The reason we did not distribute SNAT is because, there were VPN services centralized and part of the router-namespace | 16:46 |
| Sam-I-Am | Swami: makes sense | 16:46 |
| Swami | Sam-I-Am: We don't want to distribute the centralized services. | 16:46 |
| Sam-I-Am | hopefully we're not natting with v6 | 16:46 |
| Sam-I-Am | that would be horribad | 16:46 |
| regXboi | Swami beat me to the punch | 16:46 |
| *** _kiran_ has quit IRC | 16:47 | |
| *** dyydj has joined #openstack-neutron | 16:48 | |
| amuller | Sam-I-Am: ipv6 with Neutron supports routing without NAT | 16:48 |
| amuller | Sam-I-Am: As you'd expect | 16:48 |
| Sam-I-Am | isnt l3 with v6 new in kilo? | 16:48 |
| amuller | kilo is months old :) | 16:49 |
| Sam-I-Am | gee thanks :) | 16:49 |
| amuller | hehe | 16:49 |
| amuller | it's fun being obnoxious | 16:49 |
| Sam-I-Am | i'm used to it | 16:49 |
| *** zz_jgrimm has joined #openstack-neutron | 16:49 | |
| Sam-I-Am | currently updating the networking guide scenarios for kilo, then need to consider additional content | 16:49 |
| Sam-I-Am | perhaps v6 | 16:49 |
| *** ildikov_ has quit IRC | 16:50 | |
| amuller | you're a saint, as they say | 16:50 |
| Aless | would distributing SNAT work at all? In case where you move instance from host to host this could be cumbersome anyways | 16:50 |
| *** yamamoto has quit IRC | 16:51 | |
| *** patrickeast has quit IRC | 16:51 | |
| *** ohno13 has quit IRC | 16:51 | |
| Sam-I-Am | Aless: it somehow works in nova-net | 16:51 |
| Aless | I like the somehow ;-) | 16:52 |
| Sam-I-Am | magic. it doesnt track connections among hosts. | 16:52 |
| Sam-I-Am | its hacky but people are used to it | 16:52 |
| Aless | So moving from that to todays neutron must have been a step back hasnt it? | 16:53 |
| Aless | /wasnt | 16:53 |
| *** changbl has quit IRC | 16:53 | |
| Sam-I-Am | its a sticking point for a lot of nova-net users | 16:53 |
| Sam-I-Am | dvr solves the 'multi-host' issue with floating IPs, but not fixed IPs | 16:53 |
| *** yamamoto has joined #openstack-neutron | 16:54 | |
| Sam-I-Am | dvr also scares the shit out of people, so thats probably why we're seeing an uptick in linuxbridge and provider networks | 16:54 |
| amuller | Neutron does a million things that Nova net didn't, so there's that... | 16:54 |
| Sam-I-Am | the conversation around dvr usually goes like... 'this was really simple in nova-net, why is it so complex in neutron?' | 16:55 |
| *** nlahouti1 has joined #openstack-neutron | 16:55 | |
| amuller | Sam-I-Am: Do you think that mixes in people that are not familiar with OVS? | 16:55 |
| Swami | Sam-I-Am: DVR just utilized whatever was in neutron to achieve the distributed way. | 16:55 |
| *** mutoulbj has joined #openstack-neutron | 16:56 | |
| Sam-I-Am | amuller: ovs is also scary for nova-net people | 16:57 |
| amuller | yeah that's what I'm saying | 16:57 |
| *** Murali has joined #openstack-neutron | 16:57 | |
| Sam-I-Am | its an extra layer of stuff | 16:57 |
| *** Aless has quit IRC | 16:57 | |
| Sam-I-Am | which you can see side-by-side now by comparing network guide scenario 1a and 1b | 16:57 |
| *** spandhe has joined #openstack-neutron | 16:57 | |
| amuller | network admins will have to get used to OVS anyway, it's being used so heavily in so many SDN and SDN like solutions | 16:58 |
| amuller | it's crazy how popular it got in the last few years | 16:58 |
| *** mickeys has joined #openstack-neutron | 16:58 | |
| *** nlahouti has quit IRC | 16:58 | |
| amuller | but network admins are very traditional people :) | 17:00 |
| Sam-I-Am | hopefully the network guide is helping a bit | 17:00 |
| Sam-I-Am | i think most people just needed a good visual of whats going on | 17:00 |
| Sam-I-Am | rather than the black box of dooooom | 17:00 |
| amuller | Sam-I-Am++ | 17:01 |
| amuller | I really have to start contributing there | 17:01 |
| amuller | Sam-I-Am: gone are the days of that xml stuff I heard somewhere | 17:01 |
| amuller | now it's just .rsts | 17:01 |
| Sam-I-Am | thankfully | 17:01 |
| Sam-I-Am | we did the networking guide rst from the start | 17:01 |
| amuller | yeah that's good | 17:01 |
| Sam-I-Am | because even as a docs writer i want none of that xml | 17:01 |
| amuller | barrier of entry is much lower | 17:01 |
| *** nllrte has quit IRC | 17:02 | |
| *** Murali has quit IRC | 17:02 | |
| *** armax has quit IRC | 17:03 | |
| *** mitchjameson has quit IRC | 17:03 | |
| pc_m | dougwig: ping | 17:03 |
| *** nllrte has joined #openstack-neutron | 17:03 | |
| *** krtaylor has joined #openstack-neutron | 17:04 | |
| *** armax has joined #openstack-neutron | 17:05 | |
| *** armax has quit IRC | 17:05 | |
| *** spandhe has quit IRC | 17:05 | |
| Sam-I-Am | amuller: still waiting for more contributors :) | 17:05 |
| *** yamamoto has quit IRC | 17:05 | |
| *** mwagner_afk has quit IRC | 17:06 | |
| *** ronis has joined #openstack-neutron | 17:06 | |
| *** yamamoto has joined #openstack-neutron | 17:06 | |
| *** ivar-lazzaro has joined #openstack-neutron | 17:07 | |
| dougwig | pc_m: ack | 17:07 |
| amuller | Sam-I-Am: I see some stuff I can clean up with the L3 HA docs | 17:07 |
| pc_m | dougwig: Having issues with getting tox in VPN repo to use local Neutron repo. | 17:07 |
| amuller | I'll try to send some patches | 17:07 |
| dougwig | pc_m: ok, what kind of issues? | 17:08 |
| Sam-I-Am | amuller: just file a bug | 17:08 |
| pc_m | dougwig: Looks like, with the tox_install.sh, when I run the functional tests for VPN, it pip installs neutron-vpnaas (which is in the {packages}) | 17:08 |
| amuller | Sam-I-Am: you wouldn't prefer me to send the actual patch? | 17:08 |
| amuller | less work for you :) | 17:08 |
| Sam-I-Am | amuller: oh, i want that too | 17:08 |
| Sam-I-Am | amuller: but docs is pretty big about tracking and bug triage | 17:09 |
| *** Djinnn has joined #openstack-neutron | 17:09 | |
| amuller | Sam-I-Am: ack | 17:09 |
| pc_m | dougwig: I tried adding "-e ../neutron" and it does the pip install, but the neutron code is not placed under .tox/dsvm-functional/src | 17:09 |
| pc_m | dougwig: Been trying a few things w/o joy | 17:09 |
| *** yamamoto has quit IRC | 17:09 | |
| dougwig | is neutron-vpnaas in the requirements file or something? how is that ending up in the deps list? | 17:10 |
| dougwig | and why is it looking for neutron by path instead of by import? | 17:10 |
| *** baoli has quit IRC | 17:11 | |
| pc_m | dougwig: no neutron-vpnaas in *requirements.txt | 17:11 |
| pc_m | dougwig: It's not looking for neutron at all. The install_command line has {packages}, which has /opt/stack/neutron-vpnaas | 17:12 |
| *** baoli has joined #openstack-neutron | 17:12 | |
| *** annegentle has joined #openstack-neutron | 17:12 | |
| pc_m | dougwig: When install runs, it hits the "pip install -U $*" which tries to install that package. | 17:12 |
| *** shivrao has joined #openstack-neutron | 17:12 | |
| *** patrickeast has joined #openstack-neutron | 17:13 | |
| pc_m | dougwig: For giggles, I added "-e ../neutron" and it said it was installing the package and running setup.py, but there is no src area in the venv | 17:13 |
| dougwig | pc_m: did you copy the guts of the apiv2 env from lbaas? | 17:14 |
| pc_m | dougwig: apiv2 env? | 17:15 |
| dougwig | yep, go look in lbaas's tox.ini | 17:15 |
| pc_m | Ah, don't have any of that. | 17:15 |
| *** ihrachyshka has quit IRC | 17:16 | |
| *** sdake has joined #openstack-neutron | 17:16 | |
| *** achanda has quit IRC | 17:17 | |
| pc_m | dougwig: where/how does that come into play? | 17:17 |
| pc_m | dougwig: It seems like, for VPN, those settings are applied per target, for the two functional tests. | 17:19 |
| dougwig | pc_m: there's some magic in there about how it finds packages. but as long as it can import, where it is is less important. if you need it in the venv instead of being pathed from the CI location (/opt/stack), you'll have to tweak it a bit. | 17:19 |
| *** sdake_ has quit IRC | 17:20 | |
| *** marcelo_amaral has joined #openstack-neutron | 17:20 | |
| pc_m | dougwig: not sure how to tweak it. | 17:21 |
| *** garyk has quit IRC | 17:21 | |
| marcelo_amaral | Swami: hi | 17:21 |
| Swami | marcelo_amaral:hi | 17:22 |
| Swami | marcelo_amaral: did you see my comments on your patch. | 17:22 |
| pc_m | dougwig: what should {packages} contain? the package being tested, or support packages needed by the package? | 17:23 |
| Sukhdev | mestery: Ping | 17:23 |
| marcelo_amaral | Swami: Yes, I did, I'm trying to run the pep8 test, but I am getting a error in the env | 17:23 |
| mestery | Sukhdev pong | 17:23 |
| Sukhdev | Need a bit of help from you get couple my patches back-ported (forgot to push them last week when you were there :-)) | 17:24 |
| Swami | marcelo_amaral: what errors are you seeing in the env | 17:24 |
| marcelo_amaral | Swami: distutils.errors.DistutilsError: Could not find suitable distribution for Requirement.parse('pbr'). I tried to change the requirements to a older pbr version but dind't work. Have you seen this error | 17:24 |
| Sukhdev | mestery: https://review.openstack.org/#/c/196325/ | 17:24 |
| Sukhdev | mestery: and - https://review.openstack.org/#/c/196326/ | 17:24 |
| Sukhdev | markmcclain: if you are here and you could bless these two patches as well, please | 17:25 |
| mestery | Sukhdev: Looking | 17:25 |
| Sukhdev | mestery: thanks | 17:26 |
| dougwig | pc_m: it should be the dependencies, from *requirements.txt. | 17:26 |
| pc_m | dougwig: it's not. | 17:26 |
| pc_m | dougwig: it seems to be the project path | 17:27 |
| *** abregman has quit IRC | 17:27 | |
| Sukhdev | mestery: Also, while on it, could you please tag this bug as well - https://bugs.launchpad.net/neutron/+bug/1468828 | 17:27 |
| openstack | Launchpad bug 1468828 in neutron "HA router-create breaks ML2 drivers that implement create_network such as Arista" [Medium,Fix committed] - Assigned to Sukhdev Kapur (sukhdev-8) | 17:27 |
| dougwig | pc_m: try adding "set -x; echo > /tmp/tox_install.out 2>&1" to the script and run it with pep8 or the units and see if the args differ. | 17:28 |
| openstackgerrit | Ivar Lazzaro proposed openstack/neutron: Decompose Apic ML2 mechanism driver https://review.openstack.org/193365 | 17:28 |
| mestery | Sukhdev: done | 17:28 |
| Sukhdev | mestery: thank you so much | 17:29 |
| dougwig | err, exec>, not echo> | 17:29 |
| *** changbl has joined #openstack-neutron | 17:30 | |
| Sukhdev | dougwig: I see lbaas api making jenkins unhappy - can you shed some wisdom on this please - https://review.openstack.org/#/c/196325/ | 17:30 |
| dougwig | Sukhdev: standby, looking | 17:30 |
| markmcclain | Sukhdev: once jenkins is happy I'll add the +2s | 17:30 |
| openstackgerrit | Ivar Lazzaro proposed openstack/neutron: Decompose Apic ML2 mechanism driver https://review.openstack.org/193365 | 17:30 |
| *** aranjan has joined #openstack-neutron | 17:31 | |
| Sukhdev | markmcclain: cool - thanks - will ping you | 17:31 |
| marcelo_amaral | Swami: I have the log here http://pastebin.com/pXpUEE0s | 17:31 |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/neutron: Fix DVR interface delete by port when gateway is set https://review.openstack.org/173001 | 17:31 |
| Sukhdev | markmcclain: hopefully dougwig will give me some ideas | 17:31 |
| Swami | marcelo_amaral: can you do a private chat. | 17:31 |
| dougwig | Sukhdev: those jobs are broken on stable. recheck and they should disappear magically. | 17:32 |
| *** dontalton has joined #openstack-neutron | 17:32 | |
| dougwig | Sukhdev: fallout from tempest testing against master, and those tests no longer being in tempest. | 17:32 |
| amuller | Sam-I-Am: I noticed that the L3 HA LB and OVS scenarios have a ton of shared content, but the content is basically copy/pasted in two different RSTs. There's some differences in the content that is supposed to be shared like the intro paragraph mentions the l2pop bug in the LB scenario but not the OVS scenario. Did you guys consider importing RST files in to other RST files like we do in the Neutron dev docs? | 17:32 |
| *** Marga_ has quit IRC | 17:32 | |
| amuller | Sam-I-Am: that way the shared content is maintained in only one place | 17:32 |
| *** Marga_ has joined #openstack-neutron | 17:32 | |
| Sukhdev | dougwig: cool - just a simple recheck or recheck with some bug-id? | 17:32 |
| annegentle | amuller: do you know how to do that across repos? | 17:33 |
| *** yamahata has joined #openstack-neutron | 17:33 | |
| amuller | annegentle: no | 17:33 |
| annegentle | amuller: so far we have automated scripted copies of say glossary.rst | 17:33 |
| annegentle | amuller: but seems like we could do some in-repo reuse | 17:34 |
| *** sridhar_ram has quit IRC | 17:34 | |
| *** ajmiller has quit IRC | 17:34 | |
| amuller | annegentle: indeed | 17:34 |
| *** aranjan has quit IRC | 17:35 | |
| *** ajmiller has joined #openstack-neutron | 17:35 | |
| *** padkrish has joined #openstack-neutron | 17:38 | |
| *** Djinnn has quit IRC | 17:40 | |
| openstackgerrit | Ivar Lazzaro proposed openstack/neutron: Decompose Apic ML2 mechanism driver https://review.openstack.org/193365 | 17:40 |
| *** dsanders has joined #openstack-neutron | 17:41 | |
| amuller | Sam-I-Am: Is the networking guide versioned? There's a new Kilo L3 HA feature I can add | 17:43 |
| Sam-I-Am | amuller: its not versioned | 17:45 |
| Sam-I-Am | we're trying to do CI with it | 17:46 |
| *** Sukhdev has quit IRC | 17:46 | |
| *** achanda has joined #openstack-neutron | 17:46 | |
| *** armax has joined #openstack-neutron | 17:46 | |
| Sam-I-Am | adding occasional notes of 'this only works in X' providing it isnt a major overhaul of something | 17:46 |
| amuller | Sam-I-Am: so if L3 HA was introduced in Juno and I added a feature in Kilo how should I document that? Add a note that it's from Kilo? | 17:46 |
| amuller | Sam-I-Am: ok | 17:46 |
| *** marun has joined #openstack-neutron | 17:46 | |
| *** lazy_prince has quit IRC | 17:47 | |
| Sam-I-Am | for the description, 'as of kilo, thinger X does doohickey Y' | 17:47 |
| Sam-I-Am | if theres config changes or additions, you can add them as notes to the existing snippets | 17:48 |
| *** safchain has quit IRC | 17:48 | |
| *** annegentle has quit IRC | 17:49 | |
| Sam-I-Am | amuller: if we get to the point where architectures change, i suspect i'll just support the newest version | 17:49 |
| pc_m | dougwig: for pep8, the args to the script are "-r/opt/stack/neutron-vpnaas/requirements.txt -r...test-requirements.txt pylint" | 17:51 |
| *** annegentle has joined #openstack-neutron | 17:51 | |
| pc_m | dougwig: for py27, it is just the two -r args with requirements files. | 17:51 |
| pc_m | dougwig: for dsvm-functional the args are "-e /opt/stack/neutron-vpnaas" | 17:55 |
| *** enikanorov2 has quit IRC | 17:55 | |
| *** agireud has quit IRC | 17:55 | |
| pc_m | dougwig: must be some issue with tox.ini | 17:56 |
| *** Aless has joined #openstack-neutron | 17:58 | |
| *** gsagie has quit IRC | 17:58 | |
| Sam-I-Am | amuller: i see your patch | 17:59 |
| Sam-I-Am | amuller: is that specific to kilo? | 17:59 |
| amuller | always been like that | 17:59 |
| amuller | no my question earlier was about something else | 17:59 |
| *** rotbeard has quit IRC | 17:59 | |
| *** agireud has joined #openstack-neutron | 18:00 | |
| Sam-I-Am | amuller: good to know the other nodes need less config. makes my life easier. | 18:00 |
| Sam-I-Am | i think we originally had it there so people could mostly just copy the files over | 18:01 |
| amuller | Sam-I-Am: yeah this applies to the other scenarios that I looked at too | 18:01 |
| Sam-I-Am | but less type typey = mo betta | 18:01 |
| Sam-I-Am | it also applies to the install guide | 18:01 |
| amuller | ahh | 18:01 |
| Sam-I-Am | guess we should update the bug to be more generic | 18:01 |
| Sam-I-Am | and if you could partial-bug with your patch that'd be cool | 18:01 |
| Sam-I-Am | i'm in the legacy/ovs scenario now and can hit some of that stuff | 18:02 |
| *** mwagner_afk has joined #openstack-neutron | 18:02 | |
| amuller | Sam-I-Am: will do | 18:02 |
| openstackgerrit | Assaf Muller proposed openstack/neutron: Remove unused linux bridge agent configuration options https://review.openstack.org/196765 | 18:04 |
| amuller | Sam-I-Am: also removed some unused config opts to the LB agent from the Neutron tree ^ | 18:05 |
| amuller | just noticed those | 18:05 |
| Sam-I-Am | i see. it should read those from the ml2 bits, no? | 18:05 |
| amuller | Sam-I-Am: yeah the ml2 server reads those from its own configuration | 18:06 |
| amuller | Sam-I-Am: and the LB agent doesn't use those values | 18:06 |
| amuller | so, to the trash bin | 18:06 |
| Sam-I-Am | shouldnt lb also read its vxlan bits from ml2? | 18:06 |
| Sam-I-Am | rather than [vxlan] | 18:06 |
| *** annemccormick has quit IRC | 18:07 | |
| amuller | the [ml2] sections are just for the neutron server | 18:07 |
| amuller | Sam-I-Am: in RDO for example we have two conf files, one for the ML2 server and one for OVS agent | 18:07 |
| amuller | (and one for LB agent) | 18:07 |
| amuller | so agent options go in the agent .ini | 18:07 |
| amuller | and server options go in the ml2 plugin .ini | 18:07 |
| amuller | I think this avoids a lot of confusion | 18:07 |
| *** annemccormick has joined #openstack-neutron | 18:07 | |
| *** enikanorov2 has joined #openstack-neutron | 18:08 | |
| *** pradipta has quit IRC | 18:08 | |
| Sam-I-Am | yeah, and the fact that the agent config file was named "plugin" confused a lot of people | 18:08 |
| amuller | yes :) | 18:08 |
| amuller | I know | 18:08 |
| Sam-I-Am | its nice to see that finally resolved | 18:08 |
| amuller | it's so stupid | 18:08 |
| Sam-I-Am | which also means a lot of changes to docs | 18:09 |
| *** ivar-laz_ has joined #openstack-neutron | 18:09 | |
| amuller | upstream and in devtack we still have just the ml2_conf.ini passed to the OVS agent | 18:09 |
| amuller | so one file for both the server and OVS agent opts | 18:09 |
| Sam-I-Am | so, for example... lb has [vxlan] enable_vxlan while ovs has [agent] tunnel_types = vxlan | 18:09 |
| *** ivar-laz_ has quit IRC | 18:10 | |
| Sam-I-Am | l2_population is also in different places | 18:10 |
| *** yamamoto has joined #openstack-neutron | 18:10 | |
| amuller | yeah I'm not sure if there's a backwards compatible way to solve that | 18:10 |
| amuller | I'd have to check if oslo config has some back compat fancy features | 18:10 |
| *** ivar-laz_ has joined #openstack-neutron | 18:10 | |
| amuller | for moving opts from one group to another | 18:10 |
| Sam-I-Am | create new, deprecate other? | 18:10 |
| amuller | probably | 18:10 |
| Sam-I-Am | i've seen it with other services | 18:10 |
| Sam-I-Am | so there's definitely a way to do it | 18:11 |
| Sam-I-Am | in particular 'connection' has moved several times | 18:11 |
| *** ivar-laz_ has quit IRC | 18:11 | |
| amuller | true | 18:11 |
| *** tflynn has joined #openstack-neutron | 18:11 | |
| amuller | Sam-I-Am: off to lunch, ttyl | 18:11 |
| *** ivar-laz_ has joined #openstack-neutron | 18:12 | |
| *** ivar-lazzaro has quit IRC | 18:12 | |
| Sam-I-Am | amuller: this lunch thing, sounds like a good idea | 18:12 |
| Sam-I-Am | going to locate a sandwich, then relocate to a coffee shop | 18:12 |
| dougwig | pc_m: sounds like the deps for your functional env are broken. | 18:13 |
| dougwig | pc_m: the others you listed are correct. | 18:13 |
| pc_m | dougwig: It uses the testenv, so I would think it is inheriting those settings. BTW I'm starting to discuss in infra IRC | 18:14 |
| *** SourabhP has joined #openstack-neutron | 18:14 | |
| pc_m | dougwig: Uh.. qa IRC | 18:14 |
| *** yamamoto has quit IRC | 18:15 | |
| *** moshele has joined #openstack-neutron | 18:15 | |
| *** crc32 has joined #openstack-neutron | 18:15 | |
| *** dsanders1 has joined #openstack-neutron | 18:16 | |
| *** dsanders has quit IRC | 18:16 | |
| *** Marga_ has quit IRC | 18:17 | |
| lifeless | ZZelle: hi | 18:18 |
| lifeless | ZZelle: sorry about the test failures, I didn't realise the safe fixture tests tried to cleanup cleaned up fixtures; ditto tempest | 18:18 |
| *** gridinv_ has joined #openstack-neutron | 18:21 | |
| *** SumitNaiksatam has quit IRC | 18:21 | |
| *** Marga_ has joined #openstack-neutron | 18:23 | |
| *** mutoulbj has quit IRC | 18:24 | |
| *** Sukhdev has joined #openstack-neutron | 18:24 | |
| *** Djinnn has joined #openstack-neutron | 18:25 | |
| *** ivar-laz_ has quit IRC | 18:26 | |
| *** ivar-lazzaro has joined #openstack-neutron | 18:27 | |
| *** openstackgerrit has quit IRC | 18:30 | |
| *** Sukhdev has quit IRC | 18:30 | |
| *** ijw has quit IRC | 18:30 | |
| *** openstackgerrit has joined #openstack-neutron | 18:30 | |
| *** dtynan has quit IRC | 18:31 | |
| *** ivar-lazzaro has quit IRC | 18:31 | |
| *** ivar-lazzaro has joined #openstack-neutron | 18:32 | |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/neutron: Delete FIP agent gateway port with external gw port https://review.openstack.org/194441 | 18:32 |
| *** dtynan has joined #openstack-neutron | 18:32 | |
| *** abregman has joined #openstack-neutron | 18:32 | |
| Aless | btw is there any advantage for GRE currently? VXLAN seems like the better choice in all regards | 18:33 |
| *** dlundquist has joined #openstack-neutron | 18:33 | |
| Sam-I-Am | Aless: not a lot of nics do vxlan offload | 18:34 |
| *** nllrte has quit IRC | 18:34 | |
| Sam-I-Am | so ymmv depending on nic/driver | 18:34 |
| Aless | true, didnt even think about that | 18:34 |
| *** nllrte has joined #openstack-neutron | 18:35 | |
| *** achanda has quit IRC | 18:35 | |
| *** achanda has joined #openstack-neutron | 18:36 | |
| *** padkrish has quit IRC | 18:36 | |
| *** ildikov_ has joined #openstack-neutron | 18:38 | |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/neutron: Delete FIP agent gateway port with external gw port https://review.openstack.org/194441 | 18:38 |
| *** ivar-lazzaro has quit IRC | 18:38 | |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/neutron: Add RPC command and delete if last FIP on Agent https://review.openstack.org/194446 | 18:38 |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/neutron: Cleanup the fip agent gateway port delete routines https://review.openstack.org/194455 | 18:39 |
| *** Marga_ has quit IRC | 18:40 | |
| *** ivar-lazzaro has joined #openstack-neutron | 18:40 | |
| *** Marga_ has joined #openstack-neutron | 18:40 | |
| *** padkrish has joined #openstack-neutron | 18:40 | |
| *** achanda has quit IRC | 18:41 | |
| SourabhP | HenryG: ping | 18:41 |
| *** afazekas_ has joined #openstack-neutron | 18:41 | |
| *** mutoulbj has joined #openstack-neutron | 18:43 | |
| *** krtaylor has quit IRC | 18:44 | |
| *** fawadkhaliq has quit IRC | 18:44 | |
| *** padkrish has quit IRC | 18:44 | |
| *** padkrish has joined #openstack-neutron | 18:45 | |
| amuller | lifeless: heya, I think that Cedric is away | 18:45 |
| amuller | lifeless: I just removed the failing functional tests for the SafeFixture class to unblock the gate: https://review.openstack.org/#/c/196711/ | 18:46 |
| amuller | lifeless: but it looks like we'll have to keep using SafeFixture even with fixtures 1.3 | 18:46 |
| *** mickeys has quit IRC | 18:46 | |
| amuller | jlibosva reported a bug on fixtures today | 18:46 |
| *** john-davidge has quit IRC | 18:47 | |
| *** mutoulbj has quit IRC | 18:48 | |
| amuller | lifeless: trying to find that bug | 18:49 |
| *** padkrish has quit IRC | 18:49 | |
| amuller | lifeless: https://bugs.launchpad.net/python-fixtures/+bug/1469759 | 18:50 |
| openstack | Launchpad bug 1469759 in Python Fixtures "if _setUp() fails in useFixture(fixture) then TypeError is raised" [Undecided,New] | 18:50 |
| amuller | lifeless: he found that using fixtures.useFixture and testtoolt.useFixture results in different behavior when _setUp fails | 18:50 |
| *** norman has quit IRC | 18:51 | |
| *** markvoelker_ has joined #openstack-neutron | 18:51 | |
| *** john-davidge has joined #openstack-neutron | 18:51 | |
| *** Sukhdev has joined #openstack-neutron | 18:51 | |
| *** markvoelker has quit IRC | 18:52 | |
| *** Marga_ has quit IRC | 18:53 | |
| *** banix has quit IRC | 18:54 | |
| *** banix has joined #openstack-neutron | 18:55 | |
| *** dave-mccowan has quit IRC | 18:56 | |
| *** sudipto has quit IRC | 18:57 | |
| *** markvoelker_ has quit IRC | 18:57 | |
| *** banix has quit IRC | 18:57 | |
| *** markvoelker has joined #openstack-neutron | 18:57 | |
| *** padkrish has joined #openstack-neutron | 19:00 | |
| *** padkrish has quit IRC | 19:00 | |
| *** padkrish has joined #openstack-neutron | 19:01 | |
| *** claudiub has quit IRC | 19:02 | |
| *** busterswt has joined #openstack-neutron | 19:03 | |
| *** jckasper has joined #openstack-neutron | 19:03 | |
| *** ijw has joined #openstack-neutron | 19:05 | |
| *** banix has joined #openstack-neutron | 19:06 | |
| *** Marga_ has joined #openstack-neutron | 19:10 | |
| *** boris-42 has quit IRC | 19:12 | |
| *** afazekas_ has quit IRC | 19:15 | |
| *** fawadkhaliq has joined #openstack-neutron | 19:17 | |
| *** dguerri` is now known as dguerri | 19:18 | |
| *** moshele has quit IRC | 19:19 | |
| *** dguerri is now known as dguerri` | 19:20 | |
| *** dave-mccowan has joined #openstack-neutron | 19:21 | |
| *** achanda has joined #openstack-neutron | 19:22 | |
| *** nlahouti1 has quit IRC | 19:23 | |
| *** sridhar_ram has joined #openstack-neutron | 19:24 | |
| *** ZZelle_ has joined #openstack-neutron | 19:24 | |
| *** ronis has quit IRC | 19:25 | |
| openstackgerrit | Kyle Mestery proposed openstack/neutron-specs: get-me-a-network: Really get a network https://review.openstack.org/196803 | 19:26 |
| *** fawadk has joined #openstack-neutron | 19:26 | |
| *** fawadkhaliq has quit IRC | 19:27 | |
| openstackgerrit | Carl Baldwin proposed openstack/neutron: WIP Add devref for routed network segments https://review.openstack.org/196812 | 19:28 |
| *** fzdarsky has joined #openstack-neutron | 19:29 | |
| *** Djinnn has quit IRC | 19:29 | |
| *** armax has quit IRC | 19:32 | |
| *** dedery has joined #openstack-neutron | 19:34 | |
| *** dtynan has quit IRC | 19:35 | |
| *** busterswt has quit IRC | 19:35 | |
| *** bustersw_ has joined #openstack-neutron | 19:35 | |
| *** padkrish has quit IRC | 19:36 | |
| *** dtynan has joined #openstack-neutron | 19:36 | |
| *** padkrish has joined #openstack-neutron | 19:37 | |
| *** mwagner_afk has quit IRC | 19:39 | |
| *** padkrish_ has joined #openstack-neutron | 19:40 | |
| *** dedery has quit IRC | 19:40 | |
| *** padkrish has quit IRC | 19:41 | |
| *** banix has quit IRC | 19:41 | |
| *** lucas__ has quit IRC | 19:44 | |
| *** annegentle has quit IRC | 19:45 | |
| *** fawadk has quit IRC | 19:46 | |
| openstackgerrit | Kyle Mestery proposed openstack/neutron-specs: get-me-a-network: Really get a network https://review.openstack.org/196803 | 19:46 |
| *** dguerri` is now known as dguerri | 19:47 | |
| *** achanda has quit IRC | 19:48 | |
| *** lucas__ has joined #openstack-neutron | 19:49 | |
| *** gridinv_ has quit IRC | 19:49 | |
| *** annegentle has joined #openstack-neutron | 19:50 | |
| Sukhdev | markmcclain: Hi Mark, this one is ready to go - https://review.openstack.org/#/c/196326/ | 19:51 |
| *** padkrish_ has quit IRC | 19:52 | |
| *** padkrish has joined #openstack-neutron | 19:53 | |
| *** john-davidge has quit IRC | 19:53 | |
| *** matrohon has joined #openstack-neutron | 19:53 | |
| *** lucas__ has quit IRC | 19:53 | |
| *** moshele has joined #openstack-neutron | 19:53 | |
| *** gridinv_ has joined #openstack-neutron | 19:54 | |
| *** ecbanks has quit IRC | 19:55 | |
| *** ecbanks has joined #openstack-neutron | 19:56 | |
| *** dedery has joined #openstack-neutron | 19:57 | |
| *** annegent_ has joined #openstack-neutron | 19:58 | |
| *** bustersw_ has quit IRC | 19:58 | |
| *** nllrte has quit IRC | 19:59 | |
| *** lucas__ has joined #openstack-neutron | 20:00 | |
| *** achanda has joined #openstack-neutron | 20:00 | |
| *** annegentle has quit IRC | 20:01 | |
| *** dedery has quit IRC | 20:04 | |
| *** dedery has joined #openstack-neutron | 20:05 | |
| openstackgerrit | Merged openstack/neutron: Remove failing SafeFixture tests https://review.openstack.org/196711 | 20:05 |
| *** ijw has quit IRC | 20:05 | |
| *** banix has joined #openstack-neutron | 20:09 | |
| openstackgerrit | Moshe Levi proposed openstack/neutron: Fix Consolidate sriov agent and driver code https://review.openstack.org/196433 | 20:12 |
| *** baoli has quit IRC | 20:16 | |
| *** moshele has quit IRC | 20:16 | |
| *** sridhar_ram has quit IRC | 20:16 | |
| *** sridhar_ram has joined #openstack-neutron | 20:18 | |
| *** dedery has quit IRC | 20:19 | |
| Sam-I-Am | amuller: your patch mentions that the ml2 plug-in is not present on the network and compute nodes | 20:20 |
| amuller | Sam-I-Am: yeah | 20:20 |
| *** moshele has joined #openstack-neutron | 20:20 | |
| amuller | Sam-I-Am: I mean, the ml2 plugin is just code in the neutron-server binary | 20:20 |
| Sam-I-Am | on most distros, the agent is tied to the ml2 package | 20:20 |
| amuller | Sam-I-Am: hmm, let me check that using RDO as an example | 20:20 |
| Sam-I-Am | i usually just assume the plug-in is there to make it conceptually easier to understand | 20:21 |
| Sam-I-Am | although in this case, the plugin may not need specific config | 20:21 |
| *** nlahouti has joined #openstack-neutron | 20:21 | |
| *** Longgeek has quit IRC | 20:21 | |
| *** gridinv_ has quit IRC | 20:22 | |
| *** achanda has quit IRC | 20:22 | |
| *** gridinv_ has joined #openstack-neutron | 20:22 | |
| *** dedery has joined #openstack-neutron | 20:22 | |
| *** Swami has quit IRC | 20:23 | |
| amuller | Sam-I-Am: RDO has a separate RPM for neutron-server, ml2 plugin, OVS agent, and LB agent | 20:23 |
| amuller | Sam-I-Am: the agent packages include configuration and binaries I see | 20:23 |
| amuller | Sam-I-Am: I suppose it's mainly a terminology issue: When talking about ML2, OVS, LB etc, the term 'plugin' refers to the server side API implementation | 20:24 |
| amuller | so saying that the ML2 plugin is present on the compute/network nodes is confusing to me | 20:24 |
| jaypipes | mestery: so it's not that mordred and I want auto *creation* of resources. What we want is auto *assignment* of "free" or "available" pre-created resources :) | 20:25 |
| lifeless | amuller: why? | 20:25 |
| *** hitalia has joined #openstack-neutron | 20:25 | |
| lifeless | amuller: they are different, but thats not intrinsically a problem. Reading | 20:25 |
| mestery | jaypipes: Ack, can you comment on that on the spec? I think auto-creation is nice as well, and floats nicely the boat you and mordred are in :) | 20:25 |
| lifeless | amuller: (my why was in reference to still needing SafeFixture) | 20:25 |
| amuller | lifeless: he got a TRACE when using testtools.useFixture, but fixtures.useFixture worked fine | 20:26 |
| Sam-I-Am | amuller: i find consistency providing it isnt "wrong" helps people understand neutron better | 20:26 |
| amuller | this is when the fixture raised an exception during _setUp | 20:26 |
| amuller | Sam-I-Am: I agree consistency in the guide is important | 20:26 |
| *** annemccormick has quit IRC | 20:26 | |
| *** dontalton has quit IRC | 20:26 | |
| amuller | Sam-I-Am: I'll defer to you here | 20:27 |
| Sam-I-Am | if we say neutron uses a plug-in to manage agents, then say there's no plugin on the other nodes... it gets interesting. | 20:27 |
| Sam-I-Am | plus the distro dependency issue | 20:27 |
| *** sridhar_ram has quit IRC | 20:27 | |
| lifeless | amuller: ok, thats not a fixtures bug | 20:27 |
| lifeless | it was inappropriate use from testtools | 20:28 |
| lifeless | and it only fires when a failure has alread occured | 20:28 |
| Sam-I-Am | amuller: so allow_overlapping_ips is required on every node? | 20:28 |
| *** annegent_ has quit IRC | 20:28 | |
| *** julim has quit IRC | 20:29 | |
| *** annegentle has joined #openstack-neutron | 20:29 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/neutron: Updated from global requirements https://review.openstack.org/182746 | 20:29 |
| lifeless | amuller: details added to the bug | 20:29 |
| amuller | Sam-I-Am: only on the controller | 20:29 |
| amuller | Sam-I-Am: did I leave it for the agents? | 20:29 |
| Sam-I-Am | yeah | 20:30 |
| *** Piet has joined #openstack-neutron | 20:30 | |
| Sam-I-Am | so neutron.conf just needs verbose=True (if you want it) plus the bits that make it authenticate, do stuff with nova, etc. | 20:30 |
| Sam-I-Am | we leave the latter out of the guide | 20:31 |
| *** ihrachyshka has joined #openstack-neutron | 20:31 | |
| Sam-I-Am | other stuff says how to do the keystone bits | 20:31 |
| *** boris-42 has joined #openstack-neutron | 20:31 | |
| amuller | Sam-I-Am: you mean neutron.conf on network/compute nodes? | 20:32 |
| amuller | Sam-I-Am: yeah it needs very few things there | 20:32 |
| amuller | Sam-I-Am: by the way I'm working on de-duping the L3 HA content by using the ".. include" directive, just having a snag with an included file and that file displaying an image | 20:33 |
| amuller | relative paths, I dunno something's messed up =D | 20:33 |
| openstackgerrit | Ihar Hrachyshka proposed openstack/neutron: [neutron-db-manage] support separate migration subtrees https://review.openstack.org/194198 | 20:33 |
| amuller | lifeless: But you understand why we still need SafeFixture then yes? | 20:33 |
| lifeless | amuller: no, I don't | 20:34 |
| *** claudiub has joined #openstack-neutron | 20:34 | |
| amuller | lifeless: I'll ask Jakub to paste a minimal explanation on the bug then tomorrow when he gets on | 20:34 |
| lifeless | amuller: the trace shows a cascade failure | 20:34 |
| *** moshele has quit IRC | 20:34 | |
| lifeless | amuller: e.g. neutron's fixture failed to come up. Then testtools tries to interrogate it and barfs. | 20:34 |
| lifeless | amuller: the code path it is in is incapable of causing a test failure | 20:35 |
| lifeless | amuller: (but it is capable of causing a test failure to become inscrutable) | 20:35 |
| Sam-I-Am | amuller: what do you think of this? | 20:36 |
| Sam-I-Am | http://apps1-dfw.osdt.ionosphere80.com/network-guide/scenario_legacy_ovs.html | 20:36 |
| *** dedery has quit IRC | 20:36 | |
| Sam-I-Am | (particularly the changes to the config) | 20:36 |
| *** dedery has joined #openstack-neutron | 20:36 | |
| Sam-I-Am | also doing some other fix-ups, like s/tenant/project and adding the "same network" traffic flow case | 20:36 |
| Sam-I-Am | (yeah, same network seems easy enough, but i get questions about it all the time) | 20:37 |
| *** zz_jgrimm has quit IRC | 20:37 | |
| *** gridinv_ has quit IRC | 20:37 | |
| *** krtaylor has joined #openstack-neutron | 20:38 | |
| *** thedodd has quit IRC | 20:38 | |
| amuller | Sam-I-Am: I think that it's very hard to maintain this stuff when it appears so many times =D | 20:38 |
| amuller | in different .rst files | 20:38 |
| *** krtaylor has quit IRC | 20:39 | |
| *** Djinnn has joined #openstack-neutron | 20:39 | |
| Sam-I-Am | amuller: the config stuff? | 20:39 |
| openstackgerrit | Nikolay Fedotov proposed openstack/neutron: Do not mock arping in L3AgentTestFramework functional tests https://review.openstack.org/196608 | 20:39 |
| amuller | Sam-I-Am: yeah it looks like the config stuff and the validation sections at the end appear (Nearly identical) in a lot of pages | 20:39 |
| amuller | I'd make liberal use of ".. include::" if it were me | 20:40 |
| Sam-I-Am | there are some differences per each scenario | 20:40 |
| *** krtaylor has joined #openstack-neutron | 20:40 | |
| *** Swami has joined #openstack-neutron | 20:40 | |
| *** dedery has quit IRC | 20:40 | |
| Sam-I-Am | i may go back and try includes at some point for some things, but i like self-contained config in this case because changing one thing doesnt inadvertently break other stuff | 20:41 |
| *** edmondsw has quit IRC | 20:43 | |
| openstackgerrit | Dustin Lundquist proposed openstack/neutron: Added config variable to enable IPv6 https://review.openstack.org/196199 | 20:44 |
| *** jgrimm has joined #openstack-neutron | 20:45 | |
| *** gampel has joined #openstack-neutron | 20:46 | |
| *** annegentle has quit IRC | 20:46 | |
| *** armax has joined #openstack-neutron | 20:48 | |
| kevinbenton | Sam-I-Am: i noticed a lot of these diagrams have br-ex as a separate bridge. do you have any that show external networks handled just like normal networks? | 20:49 |
| *** claudiub has quit IRC | 20:49 | |
| Sam-I-Am | kevinbenton: what do you mean? | 20:49 |
| Sam-I-Am | in the provider networks case we use br-ex for everything | 20:49 |
| Sam-I-Am | it could be br-anything | 20:50 |
| kevinbenton | Sam-I-Am: having a separate bridge for external networks vs internal ones isn't necessary | 20:50 |
| Sam-I-Am | oh, you mean use br-ex for both internal and external non-tunnel networks? | 20:50 |
| kevinbenton | Sam-I-Am: if you're external network is VLAN based as well | 20:50 |
| kevinbenton | Sam-I-Am: yeah | 20:50 |
| kevinbenton | Sam-I-Am: except not br-ex | 20:50 |
| kevinbenton | Sam-I-Am: just br-vlan | 20:50 |
| kevinbenton | Sam-I-Am: or whatever your normal VLAN bridge is | 20:51 |
| Sam-I-Am | oh, yeah. i added a bit more 'things' to these scenarios to make it conceptually a bit easier to grasp | 20:51 |
| *** Sukhdev has quit IRC | 20:51 | |
| kevinbenton | Sam-I-Am: Which gets rid of the limitation of one external network per l3 agent | 20:51 |
| amuller | kevinbenton: they do show external_bridge = blank | 20:52 |
| Sam-I-Am | kevinbenton: isnt that a config thing? as in, external_network_bridge = "" | 20:52 |
| amuller | kevinbenton: and using provider bridges | 20:52 |
| amuller | they happen to call that provider bridge br-ex | 20:52 |
| *** jgrimm has quit IRC | 20:52 | |
| Sam-I-Am | i wish the default was blank, not br-ex | 20:52 |
| kevinbenton | amuller: oh, and then reference a differe physnet? | 20:52 |
| kevinbenton | Sam-I-Am: right | 20:52 |
| Sam-I-Am | i find some people comment it out and then it goes back to br-ex | 20:53 |
| kevinbenton | Sam-I-Am: yeah, that was from before provider networking extension | 20:53 |
| pc_m | dougwig: Well dsvm-functional, pep8,a nd py27 tests work now, but the dsvm-functional-sswan tests are failing in rootwrap code (looks like oslo_rootwrap is trying to do a socket_connect and is failing with EINVAL: 22) | 20:53 |
| kevinbenton | Sam-I-Am: it's one of those that would probably be too dangerous to change the default | 20:54 |
| Sam-I-Am | kevinbenton: ah | 20:54 |
| Sam-I-Am | kevinbenton: in general, the scenarios are a bit more 'expanded' because it makes the traffic flow and diagrams easier to follow | 20:54 |
| *** flaviof has quit IRC | 20:54 | |
| kevinbenton | Sam-I-Am: yeah, i see now that the external networks are still normal | 20:55 |
| kevinbenton | that amuller pointed out | 20:55 |
| *** flaviof has joined #openstack-neutron | 20:55 | |
| kevinbenton | as in, they map to a physnet | 20:55 |
| *** jgrimm has joined #openstack-neutron | 20:55 | |
| kevinbenton | so i'm good with that | 20:55 |
| Sam-I-Am | yeah | 20:55 |
| amuller | kevinbenton: yeah this is one of those things we might have to support in code forever | 20:55 |
| kevinbenton | i was just worried we were pushing people to the external_network_bridge = 'br-ex' style setup | 20:55 |
| Sam-I-Am | at some point in time, you could avoid a mapping for the ext net, neutron would create the network as gre, and it would magically Just Work :) | 20:56 |
| Sam-I-Am | ohhhh, no | 20:56 |
| jaypipes | mestery: yup, I will make a comment on the review in question | 20:56 |
| amuller | I don't see how we could deprecate the external_network_bridge = 'br-ex' option | 20:56 |
| Sam-I-Am | we're using br-ex because people are most familiar with it | 20:56 |
| kevinbenton | amuller: yeah, we would need some kind of db migration script | 20:56 |
| Sam-I-Am | btw - https://review.openstack.org/196856 | 20:58 |
| *** marun has quit IRC | 20:58 | |
| Sam-I-Am | going through the fun process of validating the scenarios (again) and trying to improve stuff | 20:58 |
| *** Marga_ has quit IRC | 20:59 | |
| ajo | ihrachyshka: http://logs.openstack.org/50/195050/4/check/check-tempest-dsvm-neutron-full/5c28135/logs/devstacklog.txt.gz#_2015-06-29_09_22_51_093 | 21:00 |
| ajo | error with db migration from qos, do you know what it means? | 21:01 |
| ajo | may be wrong reference? | 21:01 |
| Sam-I-Am | i've also been debating trying to munge external vlan networks into the scenarios, but its a delicate balance with making things too overwhelming | 21:01 |
| ihrachyshka | ajo, yes, a new migration is probably referring to a non existent parent | 21:02 |
| Sam-I-Am | given the config, simply creating a vlan ext net should work | 21:02 |
| ihrachyshka | ajo, I'll check it | 21:02 |
| *** marun has joined #openstack-neutron | 21:02 | |
| *** Steap has quit IRC | 21:02 | |
| ajo | ihrachyshka: https://review.openstack.org/#/c/195050 | 21:02 |
| ajo | bur probably for tomorrow... | 21:03 |
| ajo | zZzZ ;) | 21:03 |
| Sam-I-Am | sleep is optional | 21:03 |
| *** regXboi has quit IRC | 21:03 | |
| pc_m | dougwig: Not sure what is set up wrong for the StrongSwan functional test. | 21:04 |
| Sam-I-Am | Swami: is this updated? https://wiki.openstack.org/wiki/Neutron/DVR_L2_Agent | 21:04 |
| *** ijw has joined #openstack-neutron | 21:04 | |
| *** annegentle has joined #openstack-neutron | 21:05 | |
| *** lori is now known as lori|away | 21:05 | |
| *** dedery has joined #openstack-neutron | 21:06 | |
| *** marrusl has quit IRC | 21:06 | |
| Swami | Sam-I-Am: this was done a year ago, it is not updated from then | 21:06 |
| openstackgerrit | Ihar Hrachyshka proposed openstack/neutron: [WIP] Support Basic QoS CRUD operation https://review.openstack.org/195050 | 21:07 |
| *** rkukura has quit IRC | 21:08 | |
| ihrachyshka | ajo, oh, it's not on feature/qos | 21:08 |
| Sam-I-Am | Swami: ok. should it... go away to avoid confusion? | 21:08 |
| ihrachyshka | I'll move it now | 21:08 |
| *** banix has quit IRC | 21:08 | |
| Swami | Sam-I-Am: I am working on a developer doc, so we can retain it until the developer doc is fully done. | 21:08 |
| ajo | ack ihrachyshka, point them to the new one afterwards, thanks :) | 21:08 |
| Sam-I-Am | Swami: people seem to stumble on defunct info all the time | 21:08 |
| Sam-I-Am | Swami: is it any different than the networking guide? | 21:09 |
| ajo | ihrachyshka: I'm reading oslo.messaging code :) | 21:09 |
| *** dlundquist has quit IRC | 21:09 | |
| ihrachyshka | ajo, what do you mean? it's just another rebase, Change-Id and all is the same | 21:09 |
| *** lucas__ has quit IRC | 21:09 | |
| ajo | ihrachyshka, ahh, ok I thought different branches, same change ID makes different review | 21:09 |
| ajo | ihrachyshka, as when you cherry-pick to stable | 21:09 |
| Swami | Sam-I-Am: Networking guide just covers the Use case scenarios and deals mostly with the floatingIP and default SNAT. It does not discuss in detail about the OVS rules. I hope the design guide wiki covers most of the information. | 21:10 |
| *** Aless has quit IRC | 21:10 | |
| *** dboik has quit IRC | 21:10 | |
| Sam-I-Am | Swami: do you want to put the ovs content into the networking guide? | 21:10 |
| *** nmj3e has quit IRC | 21:11 | |
| Swami | I tried that with the documentation folks and they said it was too much to go into the networking guide. | 21:11 |
| sc68cal | Swami: link? | 21:11 |
| Swami | I think the developer doc will be the right place to add those details. | 21:11 |
| Swami | sc68cal: what link are you asking for? | 21:12 |
| sc68cal | the ovs content | 21:12 |
| Sam-I-Am | Swami: hmm, when was this? i think we wanted to do enough to get the scenarios published in time for kilo, and the ovs stuff was just too much at the time. | 21:12 |
| Sam-I-Am | i dont see a problem adding an 'advanced discussion' section with the ovs crud | 21:13 |
| Swami | I was working with Matt for documentation and that when we decided to just add the Use cases first for the Networking guide. | 21:13 |
| *** jgrimm has quit IRC | 21:13 | |
| *** krtaylor has quit IRC | 21:13 | |
| Sam-I-Am | Swami: <- matt | 21:13 |
| Swami | If you want to add the OVS details and flow rule details in the documentation feel free to let me know I will work. | 21:13 |
| *** apuimedo has joined #openstack-neutron | 21:13 | |
| Sam-I-Am | Swami: can we put it after the other content to keep people from freaking out? | 21:14 |
| kevinbenton | Sam-I-Am: i thought Swami was Swami, not Matt | 21:14 |
| Sam-I-Am | kevinbenton: haha | 21:14 |
| Sam-I-Am | <- matt | 21:14 |
| Sam-I-Am | stupid tab complete fail | 21:14 |
| kevinbenton | :) | 21:14 |
| *** padkrish has quit IRC | 21:15 | |
| *** padkrish has joined #openstack-neutron | 21:15 | |
| kevinbenton | Sam-I-Am: back to your earlier comment, i don't think adding scenarios for external vlan networks is a big priority | 21:15 |
| Sam-I-Am | kevinbenton: cool, makes my life easier | 21:15 |
| kevinbenton | Sam-I-Am: since they are setup in the 'correct' way already, it would be as simple as a command | 21:15 |
| Sam-I-Am | the idea was having the config there, and hopefully people can connect the dots | 21:15 |
| *** abregman has quit IRC | 21:15 | |
| Sam-I-Am | there's a fine line between being useful and overkill | 21:16 |
| *** marrusl has joined #openstack-neutron | 21:16 | |
| kevinbenton | Sam-I-Am: right, as long as they know how to do a VLAN network another way, the external network is no different | 21:16 |
| Sam-I-Am | docs is an interesting role sometimgs | 21:16 |
| Sam-I-Am | er, sometimes | 21:16 |
| *** dlundquist has joined #openstack-neutron | 21:17 | |
| *** matrohon has quit IRC | 21:18 | |
| *** fzdarsky has quit IRC | 21:18 | |
| *** banix has joined #openstack-neutron | 21:18 | |
| Sam-I-Am | aight, back in a few. this coffee shop's network suxxxxx. | 21:18 |
| *** ijw has quit IRC | 21:19 | |
| *** Longgeek has joined #openstack-neutron | 21:22 | |
| *** r-daneel has quit IRC | 21:23 | |
| *** nmj3e has joined #openstack-neutron | 21:24 | |
| *** mwagner_lap has joined #openstack-neutron | 21:26 | |
| *** packet has quit IRC | 21:26 | |
| *** apuimedo has quit IRC | 21:27 | |
| *** lucas__ has joined #openstack-neutron | 21:27 | |
| *** thorst has quit IRC | 21:27 | |
| *** mickeys has joined #openstack-neutron | 21:27 | |
| *** padkrish has quit IRC | 21:27 | |
| *** gampel has quit IRC | 21:27 | |
| *** rpothier has quit IRC | 21:27 | |
| *** Sukhdev has joined #openstack-neutron | 21:27 | |
| *** lucas___ has joined #openstack-neutron | 21:28 | |
| *** lucas__ has quit IRC | 21:29 | |
| *** lucas___ has quit IRC | 21:29 | |
| *** lucas____ has joined #openstack-neutron | 21:29 | |
| *** apuimedo has joined #openstack-neutron | 21:30 | |
| *** kevinbenton has quit IRC | 21:31 | |
| *** jbell8 has quit IRC | 21:31 | |
| *** padkrish has joined #openstack-neutron | 21:32 | |
| *** ivar-laz_ has joined #openstack-neutron | 21:32 | |
| *** jbell8 has joined #openstack-neutron | 21:32 | |
| Sukhdev | Hey Folks, does anybody know about grenade issue in stable/kilo - Jenkins is unhappy - please see this one - https://review.openstack.org/#/c/196325/ | 21:33 |
| *** lucas____ has quit IRC | 21:34 | |
| Sukhdev | dougwig: your idea of recheck worked on stable/juno - but, stable/kilo is still failing | 21:34 |
| *** ivar-lazzaro has quit IRC | 21:35 | |
| *** jbell8 has quit IRC | 21:37 | |
| openstackgerrit | Assaf Muller proposed openstack/neutron: Adds base in-tree functional testing of the ovs_neutron_agent https://review.openstack.org/140042 | 21:37 |
| Sukhdev | amuller: Ping | 21:38 |
| openstackgerrit | Assaf Muller proposed openstack/neutron: Adds base in-tree functional testing of the ovs_neutron_agent https://review.openstack.org/140042 | 21:39 |
| amuller | Sukhdev: yo | 21:39 |
| Sukhdev | amuller: quick question - the HA router feature went into juno or Icehouse? | 21:39 |
| amuller | Sukhdev: juno | 21:39 |
| Sukhdev | amuller: cool - so, I do not need to back port my fix to Icehouse :-) | 21:40 |
| Sukhdev | amuller: thanks | 21:40 |
| amuller | Sukhdev: I don't think that'll be needed :) | 21:40 |
| Sukhdev | amuller: makes sense - | 21:40 |
| ihrachyshka | Sukhdev, upstream wise, there is or should not be icehouse anymore | 21:41 |
| ihrachyshka | it's just that we need to complete cleanup for the branches | 21:41 |
| *** Networkn3rd has quit IRC | 21:41 | |
| Sukhdev | ihrachyshka: oh I see - yes, I should have thought about it - two release cycles rule | 21:41 |
| Sukhdev | ihrachyshka: one less thing to worry about :-) | 21:42 |
| ihrachyshka | Sukhdev, well, there is no clear rule for that, but in any case, icehouse is EOL and there won't be more releases | 21:42 |
| *** crc32 has quit IRC | 21:42 | |
| amuller | ZZelle: yo, check out lifeless's comments here: https://bugs.launchpad.net/testtools/+bug/1469759 | 21:43 |
| openstack | Launchpad bug 1469759 in testtools "if _setUp() fails in useFixture(fixture) then TypeError is raised" [High,Triaged] - Assigned to Robert Collins (lifeless) | 21:43 |
| Sukhdev | ihrachyshka: since I got you here, perhaps you can send me in the right direction - my back port patch keeps failing grenade gate for stable/kilo (see here - https://review.openstack.org/#/c/196325/) who can help? | 21:43 |
| openstackgerrit | lifeless proposed openstack/neutron: Improve fixture usage. https://review.openstack.org/196879 | 21:43 |
| ZZelle_ | amuller, i understand 1st lifeless comment not the 2nd one | 21:45 |
| lifeless | I'm right here :) | 21:45 |
| openstackgerrit | lifeless proposed openstack/neutron: Improve fixture usage. https://review.openstack.org/196879 | 21:46 |
| openstackgerrit | Paul Michali proposed openstack/neutron-vpnaas: VPNaaS: Don't clone neutron automatically for tests https://review.openstack.org/196237 | 21:47 |
| ZZelle_ | lifeless, what do you mean by "hermetic" in https://bugs.launchpad.net/testtools/+bug/1469759/comments/2? | 21:47 |
| openstack | Launchpad bug 1469759 in testtools "if _setUp() fails in useFixture(fixture) then TypeError is raised" [High,Triaged] - Assigned to Robert Collins (lifeless) | 21:47 |
| lifeless | self contained | 21:47 |
| lifeless | not leakin | 21:47 |
| lifeless | leaking | 21:47 |
| *** shwetaap has joined #openstack-neutron | 21:49 | |
| *** shwetaap1 has quit IRC | 21:51 | |
| *** krtaylor has joined #openstack-neutron | 21:52 | |
| *** pc_m has quit IRC | 21:52 | |
| *** zz_jgrimm has joined #openstack-neutron | 21:52 | |
| *** dedery has quit IRC | 21:55 | |
| *** localloop127 has quit IRC | 21:55 | |
| *** dedery has joined #openstack-neutron | 21:55 | |
| *** Longgeek has quit IRC | 21:56 | |
| ihrachyshka | Sukhdev, looking; it seems weird, looking at the code in source trees, keystoneclient should have been bumped to 1.6.0, while logs show that it was not | 21:58 |
| Sukhdev | ihrachyshka: Does this mean it could be some infra related issue? | 21:59 |
| *** oda-g has joined #openstack-neutron | 21:59 | |
| ihrachyshka | Sukhdev, not sure, still looking | 21:59 |
| *** mdurrant has quit IRC | 21:59 | |
| *** dedery has quit IRC | 21:59 | |
| openstackgerrit | Kevin Benton proposed openstack/neutron: Read vif port information in bulk https://review.openstack.org/186734 | 21:59 |
| *** mdurrant has joined #openstack-neutron | 21:59 | |
| Sukhdev | ihrachyshka: cool - appreciate your help in this regard. BTW, I just kicked off another run - will keep my fingers crossed :-) | 22:00 |
| *** claudiub has joined #openstack-neutron | 22:00 | |
| Sukhdev | ihrachyshka: BTW, if it passes, I can use a +2 from you on this one | 22:00 |
| *** sdake_ has joined #openstack-neutron | 22:01 | |
| openstackgerrit | Terry Wilson proposed openstack/neutron: Add support for PluginWorker and Process creation notification https://review.openstack.org/189391 | 22:01 |
| *** aranjan has joined #openstack-neutron | 22:01 | |
| *** mikeymeitbual has joined #openstack-neutron | 22:02 | |
| *** ijw has joined #openstack-neutron | 22:02 | |
| *** ijw has quit IRC | 22:03 | |
| *** rkukura has joined #openstack-neutron | 22:03 | |
| *** ijw has joined #openstack-neutron | 22:03 | |
| amuller | lifeless: I know that jlibosva tried to convert a fixture to inherit from fixtures.Fixture and implement _setUp instead, and he was seeing some issue | 22:03 |
| amuller | lifeless: I'll let him comment tomorrow | 22:03 |
| *** agireud_ has joined #openstack-neutron | 22:04 | |
| *** mdurrant has quit IRC | 22:04 | |
| *** mikeymeitbual is now known as mdurrant | 22:04 | |
| *** sdake has quit IRC | 22:04 | |
| lifeless | amuller: ok | 22:06 |
| lifeless | amuller: I pushed up a possible mass migration, waiting to see what tests think of it | 22:06 |
| *** agireud has quit IRC | 22:06 | |
| *** agireud_ has quit IRC | 22:09 | |
| ihrachyshka | Sukhdev, I *think* it's https://review.openstack.org/196887 that should fix it (eventually, after version bump is propagated into keystone repo) | 22:09 |
| *** apuimedo has quit IRC | 22:10 | |
| *** padkrish has quit IRC | 22:11 | |
| Sukhdev | ihrachyshka: we need mestery and markmcclain to approve this patch to fix this issue | 22:12 |
| ihrachyshka | Sukhdev, yeah, though I also want to hear from -qa guys on how we got into that incompatibility situation in the first place | 22:13 |
| ihrachyshka | I thought we have guards against it | 22:13 |
| *** Djinnn has quit IRC | 22:14 | |
| markmcclain | ihrachyshka: we probably should propose the change in the upstream requirements repo | 22:14 |
| ihrachyshka | markmcclain, that's what the patch does, no? | 22:15 |
| markmcclain | ihrachyshka: oops.. didn't check the project | 22:15 |
| *** daneyon has joined #openstack-neutron | 22:15 | |
| ihrachyshka | np | 22:16 |
| *** daneyon_ has joined #openstack-neutron | 22:16 | |
| *** hitalia has quit IRC | 22:17 | |
| ihrachyshka | zzzeek, btw the patch for neutron 3 phase migration is there. I wonder whether we need expand_offline now though, maybe it's YAGNI until we get into live data migration business (?) | 22:17 |
| openstackgerrit | lifeless proposed openstack/neutron: Improve fixture usage. https://review.openstack.org/196879 | 22:17 |
| *** aranjan has quit IRC | 22:17 | |
| zzzeek | ihrachyshka: i think expand offline is kind of an optional thing atm unless we haev a need for it | 22:18 |
| ihrachyshka | zzzeek, the patch passes gate, should be ready for review. Not sure how HenryG and company wants to proceed on this, if at all though ;) | 22:18 |
| zzzeek | ihrachyshka: doesnt hurt to have it in the flow | 22:18 |
| zzzeek | ihrachyshka: OK. ive been spending all my time on the alembic extensions | 22:18 |
| zzzeek | maybe later in the week i can have soemtign presentable | 22:18 |
| zzzeek | basic idea works | 22:18 |
| *** absubram has quit IRC | 22:19 | |
| ihrachyshka | zzzeek, yeah, but it complicates a bit since you have +1 head to see in all output from alembic etc. Not a huge deal, but while we have no immediate plans for data migrations at all... | 22:19 |
| *** shwetaap1 has joined #openstack-neutron | 22:19 | |
| zzzeek | ihrachyshka: OK if the actual stream of files isn’t there then you wont see it that is fine for now if you dont need it | 22:19 |
| *** daneyon has quit IRC | 22:20 | |
| *** aranjan has joined #openstack-neutron | 22:20 | |
| *** padkrish has joined #openstack-neutron | 22:21 | |
| openstackgerrit | Dustin Lundquist proposed openstack/neutron: Added config variable to enable IPv6 https://review.openstack.org/196199 | 22:22 |
| *** annegentle has quit IRC | 22:22 | |
| openstackgerrit | lifeless proposed openstack/neutron: Updated from global requirements https://review.openstack.org/182746 | 22:23 |
| *** shwetaap has quit IRC | 22:23 | |
| *** annegentle has joined #openstack-neutron | 22:23 | |
| lifeless | ^ that probably needs to land sooner rather than later, assuming everything is happy with it, as the global requirements proposal will overwrite it otherwise | 22:23 |
| *** kbringard has quit IRC | 22:23 | |
| openstackgerrit | Kevin Benton proposed openstack/neutron: Update DVR agent to use get_vifs_by_id https://review.openstack.org/196568 | 22:24 |
| *** dlundquist has left #openstack-neutron | 22:24 | |
| *** madhu_ak has quit IRC | 22:24 | |
| *** aranjan has quit IRC | 22:25 | |
| *** madhu_ak has joined #openstack-neutron | 22:25 | |
| *** zzzeek has quit IRC | 22:26 | |
| openstackgerrit | Adolfo Duarte proposed openstack/neutron: L3 Agent support for routers with HA and DVR https://review.openstack.org/196893 | 22:26 |
| *** prad has quit IRC | 22:31 | |
| *** sthillma has quit IRC | 22:31 | |
| *** mwagner_lap has quit IRC | 22:32 | |
| openstackgerrit | lifeless proposed openstack/neutron: Updated from global requirements https://review.openstack.org/182746 | 22:32 |
| amuller | lifeless: I cherry picked the patch and it looks like we're back to leaking fixtures | 22:32 |
| lifeless | amuller: oh ? | 22:33 |
| *** SourabhP has quit IRC | 22:33 | |
| *** sridhar_ram has joined #openstack-neutron | 22:33 | |
| lifeless | amuller: that makes me very sad. I want to fix this! | 22:33 |
| amuller | lifeless: check out neutron/tests/fullstack/test_l3_agent.py, there's one test there | 22:33 |
| *** SourabhP has joined #openstack-neutron | 22:35 | |
| *** sridhar_ram1 has joined #openstack-neutron | 22:35 | |
| *** ivar-laz_ has quit IRC | 22:35 | |
| amuller | lifeless: it passes SingleNodeEnvironment to its parent | 22:35 |
| *** sthillma has joined #openstack-neutron | 22:35 | |
| *** annegentle has quit IRC | 22:35 | |
| *** ivar-lazzaro has joined #openstack-neutron | 22:36 | |
| ZZelle_ | amuller, we should rename setUp in _setUp | 22:36 |
| *** Djinnn has joined #openstack-neutron | 22:36 | |
| *** rossella_s has quit IRC | 22:36 | |
| *** rossella_s has joined #openstack-neutron | 22:37 | |
| amuller | lifeless: and the test's parent (neutron/tests/fullstack/base.py.BaseFullStackTestCase) implements setUp and in it calls self.useFixture(self.environment) | 22:37 |
| amuller | which is an instance of SingleNodeEnvironment | 22:37 |
| *** sridhar_ram has quit IRC | 22:37 | |
| amuller | SingleNodeEnvironment._setUp instnatiates an OVS agent fixture, it inside creates an OVS bridge | 22:37 |
| amuller | if environment.setUp fails, that bridge is not cleaned | 22:38 |
| amuller | this is the canonical example I'm testing locally | 22:38 |
| lifeless | ok | 22:38 |
| lifeless | I'm just bringing up a test env | 22:38 |
| amuller | ZZelle_: ZZelle_: I changed SingleNodeEnvironment setUp to _setUp locally | 22:38 |
| lifeless | I presume a container is sufficiently namespaced to do this ? | 22:38 |
| amuller | otherwise the tests don't pass | 22:38 |
| ijw | Go sort out https://review.openstack.org/#/c/190917/, peeps | 22:39 |
| ijw | It's not on Neutron so you're not even making work for yourselves if you +1 it | 22:39 |
| ijw | (Nova plugging interface stuff) | 22:39 |
| amuller | lifeless: never tried running the functional or fullstack tests in a container | 22:39 |
| amuller | lifeless: I use a VM | 22:39 |
| *** tidwellr has left #openstack-neutron | 22:40 | |
| ijw | amuller: that means you | 22:40 |
| ZZelle_ | amuller, lifeless, a container is fine (i use one) | 22:40 |
| amuller | lifeless: you'll have to git clone neutron, setup devstack, stack.sh, unstack, keep mariadb and rabbitmq-server running | 22:40 |
| amuller | lifeless: then you can run tox -e dsvm-fullstack | 22:40 |
| lifeless | amuller: wait-what. | 22:41 |
| lifeless | amuller: that test is overriding __init__!?!?!? | 22:41 |
| ZZelle_ | the easy way: OS_SUDO_TESTING=1 tox -efunctional ... devstack in lxc requires some customization | 22:41 |
| * lifeless closes eyes and moves past the madness | 22:41 | |
| lifeless | ZZelle_: thanks | 22:41 |
| amuller | lifeless: it's calling super =D I have a bunch of tests that heavily modify fullstack tests, one of the changes gets rid of us overriding __init__ | 22:42 |
| amuller | lifeless: if that eases you | 22:42 |
| amuller | ijw: I don't see myself getting in to the nova<-->neutron vif specs any time soon sadly | 22:42 |
| amuller | no time | 22:42 |
| lifeless | amuller: \o/ | 22:42 |
| amuller | lifeless: I meant a bunch of patches | 22:42 |
| ijw | amuller: that's fine, don't want you to, I just want a +1 ;) | 22:42 |
| amuller | ijw: It's hard for me to +1 something I never read about :) | 22:43 |
| ijw | You know how you could fix that? | 22:43 |
| ijw | Read the spec | 22:43 |
| lifeless | amuller: so the bridge | 22:43 |
| ijw | It's like self explaining poetry | 22:43 |
| amuller | ijw: I need time to do that | 22:43 |
| lifeless | amuller: its coming from OVSAgentFixture ? | 22:43 |
| amuller | lifeless: yes | 22:43 |
| amuller | lifeless: OVSAgentFixture._setUp is calling self.useFixture(net_helpers.OVSBridgeFixture(self._get_br_int_name())) | 22:44 |
| *** lucas__ has joined #openstack-neutron | 22:44 | |
| Sam-I-Am | moo. | 22:45 |
| openstackgerrit | lifeless proposed openstack/neutron: Improve fixture usage. https://review.openstack.org/196879 | 22:46 |
| amuller | lifeless: What I do to simulate a _setUp failure that developers would often encounter while they're fiddling around with fullstack tests locally is change self.wait_until_env_is_up(agents_count=3) (from 2) in SingleNodeEnvironment._setUp | 22:46 |
| amuller | lifeless: and you can change the timeout parameter to wait_until_true to something shorter so your life would be happier | 22:46 |
| amuller | lifeless: then you run the test, it'll fail, and you can see if resources were cleaned up or not | 22:47 |
| lifeless | (thats just the additional fixture changed) | 22:47 |
| amuller | lifeless: 'rabbitmqctl list_vhosts' is another resource that's leaking, we're creating a vhost per test | 22:47 |
| lifeless | btw | 22:48 |
| lifeless | -r{toxinidir}/neutron/tests/functional/requirements.txt | 22:48 |
| lifeless | ^- that is not synced by the requirements job | 22:48 |
| lifeless | its a bad anti pattern | 22:48 |
| lifeless | The easiest thing to do is to fold it in with test-requirements.txt | 22:48 |
| amuller | lifeless: yeah I'm not sure why that file exists | 22:48 |
| *** lucas__ has quit IRC | 22:49 | |
| *** imcsk8 has joined #openstack-neutron | 22:49 | |
| *** annegentle has joined #openstack-neutron | 22:50 | |
| amuller | let's git blame our way to win | 22:51 |
| *** irdkehn is now known as dkehn | 22:51 | |
| *** hitalia has joined #openstack-neutron | 22:51 | |
| imcsk8 | hello i have a problem while tying to create a rule: http://fpaste.org/237849/56182321/ can somebody give me a hint? | 22:52 |
| amuller | lifeless: also leaking the processes started by the fixtures: neutron-server, neutron-l3-agent, neutron-openvswitch-agent. Basically anything that happens before wait_until_env_is_up as that's the final line of code before the test starts pretty much | 22:53 |
| ijw | imcsk8: have you enabled FWaaS in your configuration? Looks like it's turned off by the way the backend is saying 'I dont know what you're saying' | 22:54 |
| imcsk8 | ijw: let me check that might be my problem | 22:55 |
| *** dims_ has joined #openstack-neutron | 22:55 | |
| amuller | lifeless: ZZelle_: I have to wrap up for today, I'll ping you tomorrow morning | 22:56 |
| amuller | lifeless: I also would like to get rid of SafeFixture, as nice as it is :) | 22:56 |
| *** ihrachyshka has quit IRC | 22:56 | |
| lifeless | amuller: right, I now have a fast test environment to poke at this | 22:56 |
| lifeless | amuller: its pdb time :) | 22:57 |
| ZZelle_ | amuller, ok | 22:57 |
| *** dims has quit IRC | 22:59 | |
| *** lnicolas has joined #openstack-neutron | 22:59 | |
| *** daneyon_ has quit IRC | 23:00 | |
| *** dims_ has quit IRC | 23:00 | |
| *** sdake has joined #openstack-neutron | 23:02 | |
| *** amuller has quit IRC | 23:02 | |
| openstackgerrit | Merged openstack/neutron: Revert "Removed test_lib module" https://review.openstack.org/196580 | 23:02 |
| *** zzzeek has joined #openstack-neutron | 23:03 | |
| *** amuller has joined #openstack-neutron | 23:03 | |
| amuller | lifeless: was just on my way out when I thought to tell you: Check out the base class for the fullstack test | 23:03 |
| amuller | lifeless: it comes from oslo DB | 23:04 |
| lifeless | amuller: k | 23:04 |
| amuller | now I'm really out, tata | 23:04 |
| *** amuller has quit IRC | 23:04 | |
| *** ijw has quit IRC | 23:05 | |
| *** mlavalle has quit IRC | 23:05 | |
| *** sdake_ has quit IRC | 23:05 | |
| *** ijw has joined #openstack-neutron | 23:05 | |
| *** ijw has quit IRC | 23:07 | |
| *** ijw has joined #openstack-neutron | 23:07 | |
| Sam-I-Am | whoa, people actually sign off irc? | 23:08 |
| *** ijw has quit IRC | 23:10 | |
| *** sbalukoff has quit IRC | 23:10 | |
| *** ZZelle_ has quit IRC | 23:13 | |
| *** markvoelker has quit IRC | 23:13 | |
| *** annegentle has quit IRC | 23:13 | |
| *** dguerri is now known as dguerri` | 23:17 | |
| *** ijw has joined #openstack-neutron | 23:18 | |
| *** hichihara has joined #openstack-neutron | 23:23 | |
| openstackgerrit | Merged openstack/neutron: Fix Consolidate sriov agent and driver code https://review.openstack.org/196433 | 23:24 |
| *** yamamoto has joined #openstack-neutron | 23:24 | |
| *** edmondsw has joined #openstack-neutron | 23:24 | |
| *** lucas__ has joined #openstack-neutron | 23:25 | |
| *** Sukhdev has quit IRC | 23:27 | |
| *** lucas__ has quit IRC | 23:30 | |
| *** vivek-ebay has joined #openstack-neutron | 23:30 | |
| *** yuanying has joined #openstack-neutron | 23:30 | |
| *** aranjan has joined #openstack-neutron | 23:31 | |
| *** jasondotstar has joined #openstack-neutron | 23:31 | |
| *** boris-42 has quit IRC | 23:32 | |
| *** lucas__ has joined #openstack-neutron | 23:33 | |
| *** lucas__ has quit IRC | 23:38 | |
| *** padkrish has quit IRC | 23:38 | |
| *** padkrish has joined #openstack-neutron | 23:39 | |
| *** claudiub has quit IRC | 23:42 | |
| *** padkrish has quit IRC | 23:43 | |
| openstackgerrit | Swaminathan Vasudevan proposed openstack/neutron: Fix update_port_postcommit errors with DVR https://review.openstack.org/196908 | 23:49 |
| *** markvoelker has joined #openstack-neutron | 23:51 | |
| *** edmondsw has quit IRC | 23:51 | |
| *** mickeys has quit IRC | 23:53 | |
| *** Longgeek has joined #openstack-neutron | 23:54 | |
| *** marun has quit IRC | 23:54 | |
| imcsk8 | ijw: that was the problem, once i enabled fwaas it worked fine, thanks! | 23:55 |
| ijw | welcome | 23:55 |
| ijw | Also, just as well it was, cos my knowledge of FWaaS is zero | 23:55 |
| *** zzzeek has quit IRC | 23:56 | |
| *** yamamoto has quit IRC | 23:56 | |
| *** hitalia has quit IRC | 23:57 | |
| *** dannywil_ has joined #openstack-neutron | 23:57 | |
| *** dsneddon has quit IRC | 23:58 | |
| *** Swami has quit IRC | 23:58 | |
| *** Sukhdev has joined #openstack-neutron | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!