| Jeffrey4l | sdake, no idea for this ubuntu gate failure. I debugged that, but found nothing. | 00:18 |
|---|---|---|
| sdake | Jeffrey4l thanksk | 00:18 |
| sdake | we need to fix it but i am in the same boat as you | 00:18 |
| sdake | (not knowing what causes the problem) | 00:18 |
| sdake | not too sure what to do about it | 00:19 |
| Jeffrey4l | sdake, i will try to debug it again, and try to see what's going on. | 00:21 |
| *** banix has quit IRC | 00:22 | |
| *** achanda has joined #openstack-kolla | 00:39 | |
| *** michauds has quit IRC | 00:39 | |
| *** jtriley has joined #openstack-kolla | 00:39 | |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Adds lbaas support to horizon https://review.openstack.org/349314 | 00:44 |
| *** phuongnh has joined #openstack-kolla | 00:46 | |
| *** jtriley has quit IRC | 00:49 | |
| *** yuanying has quit IRC | 00:54 | |
| *** zhiwei has joined #openstack-kolla | 00:54 | |
| *** imcsk8 has joined #openstack-kolla | 00:56 | |
| *** inc0 has quit IRC | 00:59 | |
| *** yuanying has joined #openstack-kolla | 01:02 | |
| *** banix has joined #openstack-kolla | 01:04 | |
| *** rhallisey has quit IRC | 01:05 | |
| *** zhurong has joined #openstack-kolla | 01:08 | |
| *** achanda has quit IRC | 01:10 | |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Add full support for fernet [WIP] https://review.openstack.org/349366 | 01:10 |
| *** sdake has quit IRC | 01:14 | |
| *** banix has quit IRC | 01:19 | |
| *** HyperJohnGraham has joined #openstack-kolla | 01:25 | |
| *** banix has joined #openstack-kolla | 01:29 | |
| *** achanda has joined #openstack-kolla | 01:31 | |
| openstackgerrit | Lu lei proposed openstack/kolla: Using sys.exit(main()) instead of main() https://review.openstack.org/349353 | 01:33 |
| *** banix has quit IRC | 01:39 | |
| *** unicell1 has quit IRC | 01:39 | |
| *** unicell has joined #openstack-kolla | 01:39 | |
| *** signed8bit_Zzz is now known as signed8bit | 01:42 | |
| *** yuanying has quit IRC | 01:45 | |
| *** yuanying has joined #openstack-kolla | 01:46 | |
| *** ssurana has quit IRC | 01:50 | |
| *** yuanying has quit IRC | 01:55 | |
| *** zhurong has quit IRC | 01:57 | |
| *** zhurong has joined #openstack-kolla | 01:58 | |
| *** jtriley has joined #openstack-kolla | 01:59 | |
| *** unicell has quit IRC | 02:02 | |
| *** yuanying has joined #openstack-kolla | 02:04 | |
| *** signed8bit is now known as signed8bit_Zzz | 02:06 | |
| *** jtriley has quit IRC | 02:11 | |
| *** zhurong has quit IRC | 02:24 | |
| *** yuanying has quit IRC | 02:24 | |
| *** yuanying has joined #openstack-kolla | 02:25 | |
| *** zhurong has joined #openstack-kolla | 02:27 | |
| *** achanda has quit IRC | 02:32 | |
| *** yuanying has quit IRC | 02:32 | |
| *** klint has joined #openstack-kolla | 02:33 | |
| *** dwalsh_ has joined #openstack-kolla | 02:36 | |
| *** dwalsh has joined #openstack-kolla | 02:36 | |
| *** achanda has joined #openstack-kolla | 02:44 | |
| *** harlowja_at_home has joined #openstack-kolla | 02:49 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: TEST the ubuntu gate https://review.openstack.org/349761 | 02:50 |
| *** harlowja_at_home has quit IRC | 02:57 | |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Add full support for fernet [WIP] https://review.openstack.org/349366 | 03:03 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: Make the vncserver_listen to bind all interface https://review.openstack.org/349764 | 03:13 |
| *** achanda has quit IRC | 03:14 | |
| *** daneyon_ has joined #openstack-kolla | 03:20 | |
| *** daneyon_ has quit IRC | 03:25 | |
| *** coolsvap has joined #openstack-kolla | 03:28 | |
| openstackgerrit | Md Nadeem proposed openstack/kolla: Added Test case for attribute comp https://review.openstack.org/349771 | 03:29 |
| openstackgerrit | Md Nadeem proposed openstack/kolla: Added Test case for attribute comp https://review.openstack.org/349771 | 03:35 |
| *** godleon has quit IRC | 03:39 | |
| *** yuanying has joined #openstack-kolla | 03:48 | |
| *** zhubingbing has joined #openstack-kolla | 03:51 | |
| zhubingbing | hi all | 03:51 |
| *** dave-mccowan has quit IRC | 03:52 | |
| zhubingbing | http://logs.openstack.org/51/349351/11/check/gate-kolla-tox-bandit/ffbffbc/console.html can i help me look the about tox error | 03:52 |
| *** achanda has joined #openstack-kolla | 03:55 | |
| *** achanda has quit IRC | 03:55 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: TEST the ubuntu gate https://review.openstack.org/349761 | 03:56 |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Adds lbaas support to horizon https://review.openstack.org/349314 | 03:59 |
| openstackgerrit | Ken Wronkiewicz proposed openstack/kolla-kubernetes: Neutron control bootstraps and runs without host network https://review.openstack.org/343116 | 04:00 |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Add full support for fernet [WIP] https://review.openstack.org/349366 | 04:02 |
| coolsvap | zhubingbing: this point forward are all the errors http://logs.openstack.org/51/349351/11/check/gate-kolla-tox-bandit/ffbffbc/console.html#_2016-08-01_16_28_37_410173 | 04:02 |
| zhubingbing | coolsvap thinks | 04:03 |
| *** dwalsh has quit IRC | 04:11 | |
| *** dwalsh_ has quit IRC | 04:11 | |
| *** berendt has quit IRC | 04:15 | |
| Jeffrey4l | coolsvap, hi this issue is not introduced by zhubingbing | 04:20 |
| Jeffrey4l | zhubingbing, no need to worry about this. this job is failed for sure. | 04:21 |
| zhubingbing | ok thinks Jeffrey41 | 04:29 |
| *** signed8bit_Zzz is now known as signed8bit | 04:30 | |
| *** unicell has joined #openstack-kolla | 04:32 | |
| *** berendt has joined #openstack-kolla | 04:34 | |
| openstackgerrit | zhubingbing proposed openstack/kolla: test follow-up https://review.openstack.org/349788 | 04:36 |
| *** dwalsh has joined #openstack-kolla | 04:42 | |
| *** dwalsh_ has joined #openstack-kolla | 04:43 | |
| *** harlowja_at_home has joined #openstack-kolla | 04:49 | |
| *** harlowja_at_home has quit IRC | 04:55 | |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Add full support for fernet [WIP] https://review.openstack.org/349366 | 04:55 |
| *** berendt has quit IRC | 04:57 | |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: TEST the ubuntu gate https://review.openstack.org/349761 | 05:18 |
| *** unicell has quit IRC | 05:22 | |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Adds lbaas support to horizon https://review.openstack.org/349314 | 05:23 |
| *** unicell has joined #openstack-kolla | 05:27 | |
| *** derekjhyang has quit IRC | 05:42 | |
| *** HyperJohnGraham has quit IRC | 05:44 | |
| *** unicell has quit IRC | 05:44 | |
| *** unicell has joined #openstack-kolla | 05:45 | |
| *** achanda has joined #openstack-kolla | 05:46 | |
| *** berendt has joined #openstack-kolla | 05:48 | |
| openstackgerrit | Md Nadeem proposed openstack/kolla: Added Test case for env and volume comp https://review.openstack.org/349808 | 05:55 |
| *** Mr_Broke_ has quit IRC | 05:57 | |
| *** Mr_Broken has joined #openstack-kolla | 05:58 | |
| *** Mr_Broken has quit IRC | 06:02 | |
| *** mewald has joined #openstack-kolla | 06:03 | |
| *** unicell has quit IRC | 06:04 | |
| *** unicell has joined #openstack-kolla | 06:07 | |
| *** unicell has quit IRC | 06:22 | |
| *** unicell1 has joined #openstack-kolla | 06:22 | |
| *** dwalsh_ has quit IRC | 06:29 | |
| *** dwalsh has quit IRC | 06:29 | |
| openstackgerrit | Md Nadeem proposed openstack/kolla: Added Test case for env and volume comp https://review.openstack.org/349808 | 06:36 |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Adds lbaas support to horizon https://review.openstack.org/349314 | 06:42 |
| *** Mr_Broken has joined #openstack-kolla | 06:44 | |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Add full support for fernet [WIP] https://review.openstack.org/349366 | 06:46 |
| *** bootsha has joined #openstack-kolla | 06:48 | |
| *** belmoreira has joined #openstack-kolla | 06:50 | |
| *** duonghq has joined #openstack-kolla | 06:58 | |
| *** Serlex has joined #openstack-kolla | 07:02 | |
| *** b_bezak has joined #openstack-kolla | 07:15 | |
| *** achanda has quit IRC | 07:15 | |
| *** bmace has quit IRC | 07:22 | |
| *** shardy has joined #openstack-kolla | 07:30 | |
| *** duonghq has quit IRC | 07:30 | |
| *** anbehl has joined #openstack-kolla | 07:34 | |
| *** mbound has joined #openstack-kolla | 07:35 | |
| *** bmace has joined #openstack-kolla | 07:37 | |
| *** duonghq has joined #openstack-kolla | 07:38 | |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Add full support for fernet [WIP] https://review.openstack.org/349366 | 07:46 |
| *** achanda has joined #openstack-kolla | 08:16 | |
| *** zhiwei has quit IRC | 08:16 | |
| *** zhiwei has joined #openstack-kolla | 08:17 | |
| *** v1k0d3n has quit IRC | 08:17 | |
| *** v1k0d3n has joined #openstack-kolla | 08:18 | |
| *** strigazi has quit IRC | 08:18 | |
| *** pbourke has quit IRC | 08:18 | |
| *** kong has quit IRC | 08:19 | |
| *** DuncanT has quit IRC | 08:19 | |
| *** ppowell has quit IRC | 08:20 | |
| *** DuncanT has joined #openstack-kolla | 08:20 | |
| *** kong_ has joined #openstack-kolla | 08:21 | |
| *** kong_ has quit IRC | 08:21 | |
| *** kong_ has joined #openstack-kolla | 08:21 | |
| *** ppowell has joined #openstack-kolla | 08:21 | |
| *** kong_ is now known as kong | 08:21 | |
| *** achanda has quit IRC | 08:22 | |
| *** pbourke has joined #openstack-kolla | 08:24 | |
| *** strigazi has joined #openstack-kolla | 08:26 | |
| openstackgerrit | Md Nadeem proposed openstack/kolla: Fix the variable name mismatch https://review.openstack.org/349865 | 08:36 |
| *** nradojevic has joined #openstack-kolla | 08:37 | |
| *** ntpttr has quit IRC | 08:38 | |
| *** ntpttr has joined #openstack-kolla | 08:38 | |
| *** mbound has quit IRC | 08:47 | |
| duonghq | hello, I'm trying deploy kolla follow "Quick start" guide but I got error in Pulling kolla-toolbox image task: it cannot found tag 3.0.0 in repository kolla/ubuntu-binary-kolla-toolbox | 08:48 |
| duonghq | I checked this repository and found that the newest tag is 2.0.2, but in source repo it has 3.0.0 tag | 08:48 |
| duonghq | Anybody got this error like me? | 08:48 |
| Mech422 | duonghq: 2.0.2 is for stable/mitaka - 3.0.0 is for master/newton branch | 08:49 |
| duonghq | Mech422: thank you, but I want to test and develop on master branch, so I guess that I should use source repo? | 08:51 |
| Mech422 | duonghq: yes | 08:51 |
| duonghq | Do we have any policy on update docker hub binary repo and source repo? | 08:52 |
| Mech422 | duonghq: I honestly couldn't tell you - I always build from source and run a local repo | 08:52 |
| *** bootsha has quit IRC | 08:53 | |
| *** gfidente has joined #openstack-kolla | 08:53 | |
| *** gfidente has joined #openstack-kolla | 08:53 | |
| *** bootsha has joined #openstack-kolla | 08:53 | |
| Mech422 | duonghq: to be honest, its about 30-45 minutes (for me) to build and deploy the docker containers - its worth it to get all the newest fixes/prevent problems | 08:53 |
| *** bootsha has quit IRC | 08:54 | |
| duonghq | Mech422: it take same amount of time to deploy binary one too | 08:55 |
| duonghq | but let me check if default config is binary or source | 08:55 |
| Mech422 | default is binary I believe.. you can change it in globals.yml | 08:55 |
| duonghq | I see, thank you | 08:56 |
| *** mbound has joined #openstack-kolla | 08:56 | |
| *** bootsha has joined #openstack-kolla | 08:58 | |
| *** bootsha has quit IRC | 08:59 | |
| *** bootsha has joined #openstack-kolla | 09:00 | |
| *** bootsha has quit IRC | 09:01 | |
| *** bootsha has joined #openstack-kolla | 09:01 | |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Adds lbaas support to horizon https://review.openstack.org/349314 | 09:03 |
| duonghq | Mech422: I don't know what wrong with me but for heka (centos base), tag 3.0.0 is available on binary repository but not source repo | 09:09 |
| Mech422 | duonghq: hmm - wonder if the version tag increased ? 3.0.1 ? | 09:09 |
| duonghq | It seem that 2 repo type is not in sync for centos base | 09:09 |
| duonghq | no, it is just 2.x.x | 09:10 |
| duonghq | https://hub.docker.com/r/kolla/centos-source-heka/tags/ | 09:10 |
| Mech422 | duonghq: thats an image build...not source... | 09:10 |
| Mech422 | duonghq: you should be building all the containers yourself if your doing source | 09:10 |
| duonghq | thanks, I'm trying | 09:11 |
| Mech422 | I do ubuntu base source type builds | 09:13 |
| duonghq | But, what is below repo purpose? | 09:13 |
| duonghq | the <base>-source-<module> | 09:14 |
| Mech422 | kolla_base_distro: "ubuntu" | 09:14 |
| Mech422 | kolla_install_type: "source" | 09:14 |
| Mech422 | that appears to be all I need in globals.yml ? | 09:14 |
| Mech422 | I have openstack_release commented out | 09:15 |
| duonghq | actually, I should add args to kolla-build to (-t ...) | 09:15 |
| duonghq | the openstack_release is set to 3.0.0 in ansible default | 09:15 |
| *** klint has quit IRC | 09:15 | |
| *** dwalsh_ has joined #openstack-kolla | 09:16 | |
| Mech422 | duonghq: I have it commented out in globals.yml | 09:16 |
| Mech422 | duonghq: I actually drive kolla via an ansible playbook... | 09:16 |
| duonghq | Mech422: it is commented out by default, then it use value from playbook | 09:16 |
| Mech422 | duonghq: let me see if I can figure out what my kolla build line looks like | 09:16 |
| duonghq | the globals.yml settings override playbook one, | 09:17 |
| duonghq | kolla/ansible/group_vars/all.yml contains default value for many settings | 09:17 |
| berendt | is anybody actually using the lbaas integration? | 09:17 |
| Mech422 | duonghq: I don't touch that file as far as I remember | 09:18 |
| *** dwalsh has joined #openstack-kolla | 09:18 | |
| duonghq | Mech422: we should not change this file, indeed | 09:18 |
| Mech422 | duonghq: setting it to source build in globals.yml should be enough AFAIR | 09:18 |
| duonghq | so I leave openstack_release as is (3.0.0 by default) | 09:19 |
| *** bootsha has quit IRC | 09:19 | |
| Mech422 | duonghq: I have it commented out in globals.yml, so whatever is there is what I use | 09:19 |
| duonghq | I can change globals.yml to 2.0.2 but I just want to see if I can run with default value | 09:20 |
| duonghq | Mech422: the error come up when I try kolla-ansible pull | 09:20 |
| duonghq | but not when I use kolla-build | 09:20 |
| duonghq | I'm rebuilding with -t source | 09:20 |
| Mech422 | duonghq: you can't pull the images till you build them ? | 09:20 |
| Mech422 | and you need a repo to pull them from I think... | 09:21 |
| Mech422 | duonghq: did you change globals.yml to specify source build ? | 09:21 |
| Mech422 | duonghq: IIRC, you NEED a local repo to do a source build for MULTINODE, but NOT for all-in-one | 09:22 |
| duonghq | Mech422: If I use kolla_install_type="source", kolla-pull tell that heka source has not tag 3.0.0, but for type="binany" kolla-toolbox has trouble | 09:23 |
| duonghq | Mech422: I use all-in-one | 09:23 |
| Mech422 | duonghq: ok - after you set kolla_install_type to source, did you rebuild all the containers ? | 09:24 |
| Mech422 | it should take a while to rebuild them all - 15 minutes or so at least | 09:24 |
| duonghq | I'm rebuilding, | 09:24 |
| duonghq | I take some time | 09:25 |
| duonghq | but I think I should double-check the source code for what run under-the-hood | 09:27 |
| duonghq | Mech422: I take a look at kolla/ansible/roles/common/defaults/main.yml | 09:31 |
| duonghq | (for variable) | 09:31 |
| duonghq | and the correspoding playbook file | 09:31 |
| duonghq | It seem that it is doesn't matter if I rebuild or not | 09:32 |
| duonghq | I just pull from kolla namespace in docker hub | 09:32 |
| Mech422 | duonghq: hmm - do you even need to pull for all-in-one ? you already have the container images on the machine ? | 09:32 |
| Mech422 | duonghq: what would you pull ? | 09:32 |
| Mech422 | duonghq: maybe after you build - run a docker images and see if the images are already in there | 09:33 |
| duonghq | I'm following guide at http://docs.openstack.org/developer/kolla/quickstart.html | 09:33 |
| *** dwalsh has quit IRC | 09:33 | |
| *** bootsha has joined #openstack-kolla | 09:34 | |
| duonghq | It's write for binary setting | 09:34 |
| duonghq | I'm running kolla-build -t source but it has not finished yet | 09:34 |
| *** dwalsh_ has quit IRC | 09:34 | |
| *** Serlex has quit IRC | 09:35 | |
| *** hieulq has joined #openstack-kolla | 09:35 | |
| Mech422 | duonghq: yeah - it takes a while | 09:36 |
| *** dwalsh has joined #openstack-kolla | 09:57 | |
| *** dwalsh_ has joined #openstack-kolla | 09:57 | |
| *** mnasiadka has joined #openstack-kolla | 09:59 | |
| *** Serlex has joined #openstack-kolla | 10:06 | |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Add full support for fernet [WIP] https://review.openstack.org/349366 | 10:10 |
| *** zhurong has quit IRC | 10:10 | |
| mewald | We have a strange problem: Our nova-compute containers report wrong hostnames when registering with nova. They all report "kvm00" … any ideas? | 10:12 |
| openstackgerrit | Christian Berendt proposed openstack/kolla: Add neutron-vpnaas-agent container https://review.openstack.org/335495 | 10:14 |
| openstackgerrit | Christian Berendt proposed openstack/kolla: Integrate neutron-vpnaas-agent into the neutron Ansible role https://review.openstack.org/335502 | 10:14 |
| berendt | I need some reviews... :) | 10:14 |
| *** phuongnh has quit IRC | 10:15 | |
| *** phuongnh has joined #openstack-kolla | 10:16 | |
| *** zhiwei has quit IRC | 10:16 | |
| *** achanda has joined #openstack-kolla | 10:20 | |
| *** phuongnh has quit IRC | 10:22 | |
| *** achanda has quit IRC | 10:26 | |
| openstackgerrit | Prithiv proposed openstack/kolla: Add Networking-SFC roles in ansible https://review.openstack.org/346842 | 10:27 |
| *** bootsha has quit IRC | 10:28 | |
| *** bootsha has joined #openstack-kolla | 10:31 | |
| *** daneyon_ has joined #openstack-kolla | 10:34 | |
| *** duonghq has quit IRC | 10:38 | |
| *** daneyon_ has quit IRC | 10:38 | |
| *** mewald has quit IRC | 10:42 | |
| *** zhubingbing has quit IRC | 10:49 | |
| *** berendt has quit IRC | 10:51 | |
| *** mbound has quit IRC | 10:58 | |
| *** berendt has joined #openstack-kolla | 11:01 | |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Adds lbaas support to horizon https://review.openstack.org/349314 | 11:01 |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Adds lbaas support to horizon https://review.openstack.org/349314 | 11:04 |
| berendt | Please merge https://review.openstack.org/#/c/349340, LBaaS plugin in Neutron is actually broken | 11:05 |
| patchbot | berendt: patch 349340 - kolla - Change config for lbaas service_plugins | 11:05 |
| berendt | smekel_ Can you please check and extend https://review.openstack.org/#/c/336943/ | 11:06 |
| patchbot | berendt: patch 336943 - kolla - Install neutron-lbaas-dashboard from source in hor... | 11:06 |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Adds lbaas support to horizon https://review.openstack.org/336943 | 11:09 |
| *** dwalsh_ has quit IRC | 11:10 | |
| *** dwalsh has quit IRC | 11:10 | |
| berendt | smekel_ What have you changed with the latest patchset? | 11:10 |
| berendt | smekel_ Please revert patch set 3 and do not overwrite existing reviews! | 11:13 |
| *** shardy is now known as shardy_lunch | 11:14 | |
| openstackgerrit | Christian Berendt proposed openstack/kolla: Install neutron-lbaas-dashboard from source in horizon container https://review.openstack.org/336943 | 11:17 |
| *** mliima has joined #openstack-kolla | 11:23 | |
| *** Mr_Broken has quit IRC | 11:25 | |
| *** Mr_Broken has joined #openstack-kolla | 11:26 | |
| *** sdake has joined #openstack-kolla | 11:27 | |
| *** pbourke has quit IRC | 11:29 | |
| *** sdake_ has joined #openstack-kolla | 11:30 | |
| *** Mr_Broken has quit IRC | 11:30 | |
| *** pbourke has joined #openstack-kolla | 11:30 | |
| *** rhallisey has joined #openstack-kolla | 11:30 | |
| *** sdake has quit IRC | 11:31 | |
| *** dwalsh has joined #openstack-kolla | 11:34 | |
| *** dwalsh_ has joined #openstack-kolla | 11:34 | |
| *** Mr_Broken has joined #openstack-kolla | 11:36 | |
| *** matrohon has joined #openstack-kolla | 11:39 | |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Install neutron-lbaas-dashboard from binary in horizon container https://review.openstack.org/349314 | 11:44 |
| *** sbezverk_ has quit IRC | 11:48 | |
| sdake_ | morning | 11:51 |
| *** sdake_ is now known as sdake | 11:51 | |
| sdake | smekel_ around? | 11:51 |
| pbourke | hey sdake | 11:52 |
| sdake | sup pbourke | 11:52 |
| pbourke | back in the land of the living :) | 11:53 |
| rhallisey | hello | 11:53 |
| rhallisey | pbourke, how was the vacation? | 11:53 |
| pbourke | rhallisey: great thanks | 11:53 |
| pbourke | nice to get away | 11:53 |
| rhallisey | good to hear | 11:53 |
| pbourke | being catching up on things | 11:53 |
| sdake | pbourke didn't know you were on pto | 11:54 |
| sdake | pbourke hope it was enjoyable :) | 11:54 |
| pbourke | thanks thought I told you but you probably had about 100 other things going on | 11:54 |
| pbourke | :) | 11:54 |
| sdake | its ok im not your master you don't have to tell me when you go on pto :) | 11:55 |
| sdake | i was a bit surprised to see you gone for such a time :) | 11:55 |
| sdake | but figured you were just taking a break | 11:55 |
| sdake | which you were ;) | 11:55 |
| pbourke | yeah, just in case you thought Id disappeared completely or something | 11:56 |
| sdake | i would have sent an email after a couple weeks :) | 11:56 |
| sdake | making sure you are ok :) | 11:56 |
| sdake | entirely out of concern not peoples choices to work or not work on kolla | 11:57 |
| sdake | registration = 1700$ | 11:57 |
| * sdake groans | 11:57 | |
| pbourke | what for | 11:58 |
| sdake | replacement 845 tubes = 1600$ | 11:58 |
| sdake | groan | 11:58 |
| sdake | pbourke in the us you have to register your vehicle with the state for the privielge of driving on their roads | 11:59 |
| pbourke | as well as car tax? | 11:59 |
| sdake | there is sales tax when ou buy the car | 11:59 |
| pbourke | we have to pay yearly tax | 11:59 |
| sdake | the registration is like a car tax yes | 11:59 |
| sdake | yes registration is yearly | 11:59 |
| sdake | in arizona is based entirely on the original MSRp of the vehicle | 12:00 |
| pbourke | as well as NCT (test to prove your car is road worthy) | 12:00 |
| sdake | we have to test emmisisoins after 5 years | 12:00 |
| sdake | the US is a throw away society so most cars are replaced beore they even get to emissions testing | 12:00 |
| sdake | the older the car, the cheaper the registration | 12:01 |
| sdake | which is totally counter to how it should work - new cars don't tear up the roads - old cars do | 12:01 |
| pbourke | yeah here the older the car the higher the tax | 12:02 |
| sdake | its opposite in airozna | 12:02 |
| sdake | every state is different | 12:02 |
| *** shardy_lunch is now known as shardy | 12:05 | |
| Daviey | sdake: I think the theory is that old cars have paid their fair share | 12:05 |
| Daviey | in the UK, cars (IIRC) > 25 years, get free road tax | 12:05 |
| sdake | rhallisey mind abandoning this patch https://review.openstack.org/#/c/315068/ | 12:06 |
| patchbot | sdake: patch 315068 - kolla - WIP - Add support for third party plugins for Keys... | 12:06 |
| sdake | Daviey in the us classic cars are free registration | 12:06 |
| sdake | (25 years = classic) | 12:06 |
| Daviey | Ah yes | 12:06 |
| Daviey | sdake: would you mind looking at https://review.openstack.org/#/c/337559/ .. it has been 10 days without a meaningful review | 12:07 |
| patchbot | Daviey: patch 337559 - kolla - Add Watcher default settings | 12:07 |
| Daviey | Last one in my patch series | 12:07 |
| rhallisey | sure | 12:07 |
| Daviey | (for watcher) | 12:07 |
| *** zhurong has joined #openstack-kolla | 12:07 | |
| sdake | what does watcher do | 12:08 |
| sdake | Daviey the review queue is jammed up | 12:09 |
| sdake | Daviey we have like 10 pages of reviews - and every one of em takes 30 mins+ | 12:10 |
| *** zhurong has quit IRC | 12:11 | |
| Daviey | sdake: This is just the config for the container and ansible stuff (which has already landed) | 12:11 |
| Daviey | It looks at the vm's.. and the relationship with the hypervisor from ceilometer data.. and decides if your cloud is well balanaced | 12:12 |
| Daviey | big tent project | 12:12 |
| *** matrohon has quit IRC | 12:12 | |
| Daviey | does live migrations to shuffle stuff around | 12:12 |
| *** Mech422 has quit IRC | 12:13 | |
| *** Mech422 has joined #openstack-kolla | 12:13 | |
| openstackgerrit | Shaun Smekel proposed openstack/kolla: Add full support for fernet [WIP] https://review.openstack.org/349366 | 12:15 |
| sdake | pbourke need another review on https://review.openstack.org/#/c/337559/2 - should take about 3 minutes | 12:16 |
| patchbot | sdake: patch 337559 - kolla - Add Watcher default settings | 12:16 |
| sdake | or rhallisey :) | 12:17 |
| Daviey | (thanks sdake) | 12:17 |
| sdake | Daviey fwiw we dont necessarily require big tent to approve mergess | 12:17 |
| Daviey | sdake: No, i know that... but it is towards the goal, right? :) | 12:18 |
| openstackgerrit | Jeffrey Zhang proposed openstack/kolla: TEST the ubuntu gate https://review.openstack.org/349761 | 12:18 |
| sdake | right | 12:18 |
| sdake | oh connecticut wrappers - best in the world | 12:18 |
| sdake | nothing like rain all night to kill the heat enough to smoke a proper cigar :) | 12:18 |
| *** zhurong has joined #openstack-kolla | 12:19 | |
| *** achanda has joined #openstack-kolla | 12:24 | |
| *** achanda has quit IRC | 12:29 | |
| *** sbezverk has joined #openstack-kolla | 12:29 | |
| *** zhiwei has joined #openstack-kolla | 12:32 | |
| *** zhubingbing has joined #openstack-kolla | 12:32 | |
| zhubingbing | hi | 12:33 |
| *** Mr_Broken has quit IRC | 12:34 | |
| *** Mr_Broken has joined #openstack-kolla | 12:34 | |
| *** godleon has joined #openstack-kolla | 12:35 | |
| sdake | rhallisey ping | 12:35 |
| rhallisey | hey | 12:35 |
| *** Mr_Broke_ has joined #openstack-kolla | 12:35 | |
| *** zhiwei has quit IRC | 12:36 | |
| *** Mr_Broken has quit IRC | 12:39 | |
| openstackgerrit | Merged openstack/kolla: Change config for lbaas service_plugins https://review.openstack.org/349340 | 12:41 |
| *** Mr_Broken has joined #openstack-kolla | 12:42 | |
| sdake | hey folks - I'd like to point people at https://review.openstack.org/#/c/349366/15 | 12:44 |
| patchbot | sdake: patch 349366 - kolla - Add full support for fernet [WIP] | 12:44 |
| sdake | I think this is not really WIP all that much | 12:44 |
| *** Mr_Broke_ has quit IRC | 12:44 | |
| sdake | and we have serious motivation to merge this work prior to newton | 12:44 |
| sdake | so if CRs could weigh in on the next rev would be appreciated - I'll star the review | 12:44 |
| sdake | smekel_ - fantastic work on fernet key rotation - you hit all the endpoints I think we need for this work | 12:46 |
| sdake | smekel_ just one semi-obnoxious request to split out the dockercontainers from the ansible work | 12:46 |
| sdake | i dont know if obnoxious is the right word there | 12:46 |
| sdake | anyway fill in the blank from my end - I know it seems counter to getting real work done, but we want to split out the ansible from containers for backport experimentation | 12:47 |
| sdake | and again - fantastic ! | 12:47 |
| *** jtriley has joined #openstack-kolla | 12:48 | |
| *** signed8bit has joined #openstack-kolla | 12:48 | |
| sdake | speaking of stars - if core reviewers can use stars on reviews that are high priority - it would be appreciated by others | 12:48 |
| sdake | is inc0 kicking around | 12:51 |
| *** b_bezak has quit IRC | 12:54 | |
| *** dmsimard|afk is now known as dmsimard | 12:55 | |
| sdake | hey dmsimard | 13:00 |
| sdake | pbourke our review queue is very long - any help you can provide there would be greatly appreciated (over actual implementation) | 13:00 |
| dmsimard | hey sdake | 13:00 |
| sdake | hey heads up - not sure if everyone nows, but we have access to the OSIC cluster - last I heard inc0 was doing the bare metal deploy so we could get our test cases in work. | 13:01 |
| sdake | Note we need to produce a document / whitepaper on our efforts there, so please take copious notes while doing tests. | 13:02 |
| sdake | nows/knows | 13:02 |
| sdake | pbourke if your looking for work to do, helping with the test cases/analsys we have defined thus far would be a big help since your in EMEA and only you and alicja will likely get access to osic from EMEA | 13:03 |
| pbourke | sdake: sounds good, was hoping to help on that | 13:04 |
| sdake | the issue there is OSIC really only wants core reviewers running the tests because we only get one set of credentails | 13:04 |
| sdake | pbourke thanks dude :) | 13:04 |
| pbourke | sdake: wonder can inc0 send a howto to the ML? | 13:04 |
| openstackgerrit | Prithiv proposed openstack/kolla: Add Networking-SFC roles in ansible https://review.openstack.org/346842 | 13:04 |
| openstackgerrit | Prithiv proposed openstack/kolla: Dockerfiles for Networking-SFC Container https://review.openstack.org/344930 | 13:04 |
| sdake | pbourke i've asked him to do so - I think he is sorting out the bare metal there and will send a note to the cr team off-list (since it will contain credential information) | 13:05 |
| pbourke | sounds good | 13:05 |
| sdake | another option is to post that publoically and just keep the credentials among the core review team | 13:05 |
| sdake | i'd prefer that - i'll speak to him when he wakes up :) | 13:05 |
| sdake | also sean-k-mooney has some reviews in queue for bifrost | 13:06 |
| sdake | imo bifrost is our #1 objective for this cycle to land | 13:06 |
| sdake | if anyone can help on those reviews, it would be greatly appreciated as well | 13:06 |
| sdake | ok - I've sent about 10 requests in the last 10 minutes, i guess I'll stop nwo :) | 13:07 |
| *** mkoderer_cloud has joined #openstack-kolla | 13:11 | |
| *** dwalsh has quit IRC | 13:14 | |
| *** dwalsh_ has quit IRC | 13:15 | |
| *** daneyon_ has joined #openstack-kolla | 13:16 | |
| *** daneyon_ has quit IRC | 13:21 | |
| *** bootsha has quit IRC | 13:21 | |
| *** qwang has joined #openstack-kolla | 13:24 | |
| *** bootsha has joined #openstack-kolla | 13:24 | |
| *** b_bezak has joined #openstack-kolla | 13:27 | |
| *** bootsha has quit IRC | 13:30 | |
| *** mbound has joined #openstack-kolla | 13:35 | |
| *** mewald has joined #openstack-kolla | 13:37 | |
| *** berendt has quit IRC | 13:40 | |
| mewald | Now I am getting this error message about denied access to KVM kernel module: https://gist.github.com/mewald1/e5281805aaa06f8924fc4d1c0828a285 | 13:40 |
| *** mbound has quit IRC | 13:40 | |
| mewald | any suggestions? | 13:40 |
| pbourke | sdake: did anyone ever get to the bottom of that ubuntu gate job problem? | 13:41 |
| *** zhurong has quit IRC | 13:43 | |
| sbezverk | mewald can you manually launch VM using virsh on the node where you see this issue? | 13:46 |
| *** mgoddard_ has joined #openstack-kolla | 13:46 | |
| mewald | sbezverk: no that doesnt work either | 13:49 |
| *** mgoddard has quit IRC | 13:50 | |
| sdake | pbourke no but Jeffrey4l is giving it a second dgo | 13:51 |
| *** inc0 has joined #openstack-kolla | 13:51 | |
| sdake | pbourke when I finish with my backlog (whenever that is) I plan to take a look if the problem isn't solved by then. | 13:51 |
| *** zhurong has joined #openstack-kolla | 13:51 | |
| sbezverk | mewald: what does it say if you run docker exec {nova container name} whoami | 13:52 |
| inc0 | good morning | 13:52 |
| mag009 | morning yall | 13:52 |
| sdake | mewald with which patch? | 13:52 |
| *** thumpba has quit IRC | 13:52 | |
| sdake | mewald if you have the patch that causes the problem I can take a look | 13:53 |
| sdake | have the patch-have a link to the patch | 13:53 |
| *** thumpba has joined #openstack-kolla | 13:53 | |
| sdake | mewald did you get my message about the concept of creating a user user and user role | 13:53 |
| openstackgerrit | Patrick Powell proposed openstack/kolla: Add Nova sanity check https://review.openstack.org/347416 | 13:54 |
| pbourke | hey inc0 | 13:54 |
| sdake | mewald the general consensus on the ops list was that would be confusing and operators would prefer a bare install so they could do their own credentials setup (minus fixing the horizon issue which they said we should do to match whatever is standardized in the default config files) | 13:54 |
| inc0 | hey good morning Paul, back from vacation I see:) | 13:54 |
| sdake | inc0 yo bro | 13:54 |
| pbourke | inc0: yup finally :) | 13:54 |
| sdake | inc0 we have request for you | 13:54 |
| inc0 | finally? you sound almost happy | 13:54 |
| Jeffrey4l | sdake, i debugged the ubuntu gate today. it has the VIR_MIGRATE_AUTO_CONVERGE variable. no idea why it happened. | 13:54 |
| inc0 | ow? | 13:54 |
| Jeffrey4l | see the log http://logs.openstack.org/61/349761/4/check/gate-kolla-dsvm-deploy-ubuntu-source/7114139/console.html#_2016-08-02_12_46_50_444427 | 13:55 |
| sdake | Jeffrey4l oh right - brain coming back online on that problem | 13:55 |
| *** rhallisey has quit IRC | 13:55 | |
| inc0 | Jeffrey4l, so OSA had something like that apperantly | 13:55 |
| pbourke | inc0: great to get away but also nice to be back | 13:55 |
| sdake | Jeffrey4l i remember starting a thread on that with the nova folks on the mailing list | 13:55 |
| inc0 | and it was due to strange version of libvirt in infra mirrors | 13:55 |
| Jeffrey4l | inc0, sure? | 13:55 |
| inc0 | not sure, best bet | 13:56 |
| sdake | Jeffrey4l can you join openstack-ansible | 13:56 |
| sdake | lets ask how htey fixed it | 13:56 |
| Jeffrey4l | the libvirt.py file has the variables. | 13:56 |
| inc0 | however there is one more thing I wanted to test out | 13:56 |
| Jeffrey4l | sdake, how? irc? | 13:56 |
| sdake | ya irc | 13:56 |
| inc0 | it's problem of libvirt-bin not libvirt | 13:56 |
| sdake | #openstack-ansible | 13:56 |
| inc0 | sdake, I know | 13:56 |
| inc0 | how they fixed it | 13:56 |
| sdake | oh do tell :) | 13:56 |
| Jeffrey4l | i am in | 13:56 |
| inc0 | they added pip install --no-binary python-libvirt | 13:56 |
| inc0 | didnt work for us | 13:56 |
| Jeffrey4l | if it is a bug in ifra mirrors. why other deploy tool haven't such a issue. | 13:57 |
| inc0 | we're special cats on that field aren't we? | 13:57 |
| *** thumpba has quit IRC | 13:57 | |
| inc0 | only ones deploying libvirt in container | 13:57 |
| Jeffrey4l | Or, if it is a bug in ifra mirror, how about remove the infra mirror in CI and see what happen | 13:58 |
| inc0 | didt work as well | 13:58 |
| inc0 | but I have another idea | 13:58 |
| inc0 | and that is libvirt-bin is installed in image itself | 13:58 |
| mag009 | inc0: got a sec to talk in pm ? | 13:59 |
| inc0 | so I'm about to purge | 13:59 |
| inc0 | sure mag | 13:59 |
| *** JoseMello has joined #openstack-kolla | 13:59 | |
| sdake | inc0 so related to requeset, can you send a note to ml with link to test cases we re producing for osic cluster | 13:59 |
| sdake | and privately send credentials to core review team? | 13:59 |
| Jeffrey4l | still haven't idea, why the libvirt-bin infect a static variable? libvirt.py is just a plain python file. no livbvirt.so file. | 13:59 |
| sdake | so we can geet the ball rolling on coordination | 13:59 |
| inc0 | sdake, I should get them today | 13:59 |
| inc0 | they had to reschedule | 14:00 |
| inc0 | however we get additional time every time it happends;) | 14:00 |
| Jeffrey4l | is any one reproduce this issue locally? | 14:01 |
| *** diogogmt has quit IRC | 14:02 | |
| inc0 | Jeffrey4l, so libvirt-python fills these flags from libvirt itself | 14:02 |
| Jeffrey4l | no. you can check the libvirt.py file. it just static plain variables. | 14:03 |
| inc0 | so I've seen it before, if libvirt throws an error, it will throw this error too | 14:04 |
| Jeffrey4l | fyi: i found such error message in the auth.log file: Aug 2 12:43:35 ubuntu sudo: PAM audit_log_acct_message() failed: Operation not permitted | 14:04 |
| Jeffrey4l | do not sure it is matter. | 14:04 |
| Jeffrey4l | this is a issue from host os kernel. | 14:04 |
| *** mnasiadka has quit IRC | 14:05 | |
| *** zhurong has quit IRC | 14:06 | |
| inc0 | https://review.openstack.org/#/c/348551/4/tools/setup_Debian.sh let see if this helps | 14:07 |
| patchbot | inc0: patch 348551 - kolla - Testing gate | 14:07 |
| sdake | Jeffrey4l auth log seems important | 14:07 |
| *** zhurong has joined #openstack-kolla | 14:07 | |
| sdake | Jeffrey4l i'd chase that one down - it could be the root cause | 14:07 |
| Jeffrey4l | yep. ci collect the auth.log now in master branch. | 14:07 |
| sdake | inc0 i am in need of a review from you please | 14:08 |
| inc0 | sdake, shoot | 14:08 |
| Jeffrey4l | inc0, hmm. remove the libvirt-bin in the host OS? isn't it meaningless? because we never install the libvirt-bin on host OS> | 14:08 |
| inc0 | customizations? | 14:08 |
| sdake | https://review.openstack.org/#/c/348567/ | 14:08 |
| patchbot | sdake: patch 348567 - kolla - enhancements to install_packages macro | 14:08 |
| inc0 | Jeffrey4l, but it can be preinstalled | 14:08 |
| *** rhallisey has joined #openstack-kolla | 14:09 | |
| sdake | yes one specific patch related to customizations | 14:09 |
| Jeffrey4l | ok. | 14:09 |
| inc0 | so we use tcp for libvirt communication | 14:09 |
| inc0 | and if libvirt is actually running on host | 14:09 |
| inc0 | this might conflict with one we have | 14:09 |
| inc0 | and then version could differ | 14:10 |
| inc0 | sdake, ok, I'll need to take longer time on this one | 14:10 |
| inc0 | will review it later today | 14:11 |
| sdake | inc0 understood | 14:11 |
| sdake | i'll star it | 14:11 |
| sdake | inc0 the issue is it is blocking heka (and anything that uses patterns similar to heka) | 14:11 |
| inc0 | http install you mean? | 14:11 |
| sdake | inc0 id review myself but looks like greek to me | 14:12 |
| inc0 | well technically you still can do this | 14:12 |
| sdake | inc0 i think that is the problem | 14:12 |
| inc0 | you don't *need* to use macro to install stuff | 14:12 |
| sdake | there was some debate on the heka review | 14:12 |
| sdake | inc0 yes but that would break customizations no? | 14:12 |
| inc0 | and in fact, if something is non-standard, I'd even specifically not use it | 14:12 |
| inc0 | nah | 14:12 |
| inc0 | macro is only for convenience | 14:12 |
| inc0 | you still can use | customizable | 14:12 |
| inc0 | and this is what matters for customizations | 14:13 |
| sdake | well if its a repeated operation across our containers, i htink we should try to make it easy to implement and use a macro in some way for it | 14:13 |
| sdake | but ultimately decision up to you | 14:13 |
| inc0 | brb need food | 14:13 |
| *** dave-mccowan has joined #openstack-kolla | 14:14 | |
| Daviey | I do think that functional is better than perfect.. quite a few of the reviews are slowed down by small things that could quickly be fixed in subsequent updates | 14:15 |
| Daviey | i think it is impeding velocity, with only a partial improvement to cleanliness | 14:15 |
| Daviey | (I agree quality is important) | 14:15 |
| *** banix has joined #openstack-kolla | 14:17 | |
| *** Mr_Broken has quit IRC | 14:18 | |
| sdake | Daviey the issue there is tracking the tech debt - we have several hundred bugs which are too much to handle atm | 14:19 |
| sdake | Daviey tell me how to solve the tech debt tracking and i'll ask the core team to relax a little bit on the reviews | 14:21 |
| mewald | sdake: is there a way to find out which commit caused the problem without rebuilding the images all the time? :D | 14:22 |
| inc0 | I'm back | 14:23 |
| inc0 | sdake, agree, but if macro becomes overly complex on it;s own, we'll simply trade one problem for another | 14:23 |
| inc0 | and we don't have that much http based installations | 14:23 |
| inc0 | the macro that Abe did is pretty complex code as opposed to what I had | 14:24 |
| inc0 | which was literally single if | 14:24 |
| zhubingbing | hi all，Can you help me review code | 14:26 |
| zhubingbing | https://review.openstack.org/#/c/349351/ | 14:26 |
| patchbot | zhubingbing: patch 349351 - kolla - Add gnocchi ansible role | 14:26 |
| *** HyperJohnGraham has joined #openstack-kolla | 14:26 | |
| zhubingbing | https://review.openstack.org/#/c/348388/ | 14:26 |
| patchbot | zhubingbing: patch 348388 - kolla - repair gnocchi image dockerfile | 14:26 |
| zhubingbing | I have finished it | 14:27 |
| inc0 | code is finished when it's last user dies;) | 14:27 |
| zhubingbing | What else to improve | 14:28 |
| zhubingbing | thinks | 14:28 |
| zhubingbing | inc0 ok tkinks | 14:28 |
| zhubingbing | thinks | 14:29 |
| *** mgoddard_ has quit IRC | 14:29 | |
| *** mgoddard has joined #openstack-kolla | 14:29 | |
| inc0 | zhubingbing, just kidding, I'll take a look at it later today | 14:30 |
| inc0 | thanks for your contribution! | 14:30 |
| zhubingbing | ok，think you very much | 14:30 |
| sdake | zhubingbing which do you need a look at | 14:33 |
| mewald | I just installed libvirt on the native system and get the same error … wtf ... | 14:33 |
| sdake | speculation towards the selinux equivalent in ubuntu | 14:33 |
| sdake | forget what it is called | 14:34 |
| mewald | apparmor … | 14:34 |
| mewald | checking | 14:34 |
| sdake | inc0 i don't totally disagree, but the macros are an abstraction - abstraction hides complexity :) | 14:35 |
| inc0 | hidden complexity might be worse than apparent one, but we're getting philosophical | 14:35 |
| inc0 | mewald, sooo...I think this is mix of several things | 14:36 |
| *** achanda has joined #openstack-kolla | 14:36 | |
| inc0 | we used to talk to libvirt via socket, but we replaced it with tcp during drop-root stuff | 14:37 |
| sdake | zhubingbing the review https://review.openstack.org/#/c/348329/ is abandoned, any reason why? | 14:37 |
| patchbot | sdake: patch 348329 - kolla - Build gnocchi image using source type base on centos (ABANDONED) | 14:37 |
| sdake | mewald i don't understand your question about which commit caused which problem | 14:38 |
| zhubingbing | this https://review.openstack.org/#/c/348388/ | 14:38 |
| patchbot | zhubingbing: patch 348388 - kolla - repair gnocchi image dockerfile | 14:38 |
| sdake | i'm not sure what problem your talking about | 14:38 |
| sdake | one way is git bisect | 14:38 |
| zhubingbing | I put it together. | 14:38 |
| sdake | mewald would you mind expanding? | 14:38 |
| inc0 | so if libvirt exists on host, it might very well work and our containers might talk to this one | 14:39 |
| sdake | we have an experimenatal policy of separating dockerfiles and ansible code atm | 14:39 |
| sdake | inc0 ++ that sounds plausable | 14:39 |
| sdake | zhubingbing ^^ the experimental policy msg was for you above | 14:39 |
| sdake | its related to our backport ideas and following stable:follows-policy | 14:40 |
| sdake | zhubingbing can you unabandon that patch if the reason wa sto merge the work into one patch? | 14:44 |
| zhubingbing | ok | 14:44 |
| *** dwalsh has joined #openstack-kolla | 14:45 | |
| sdake | zhubingbing as a positive side effect, we end up getting faster reviews when a patch doesn't have 50 files affected :) | 14:45 |
| sdake | (50 random number picked out of thin air) | 14:45 |
| zhubingbing | iHow should I do it | 14:45 |
| sdake | ansible in one patch dockerfiles in another | 14:45 |
| sdake | there is an unabandon button in the gerrit ui | 14:46 |
| *** achanda has quit IRC | 14:48 | |
| openstackgerrit | Patrick Powell proposed openstack/kolla: Add Heat sanity check https://review.openstack.org/350071 | 14:48 |
| zhubingbing | sdake， thinks | 14:49 |
| sdake | zhubingbing thinks/thanks? | 14:49 |
| sdake | or do you mean your thinking | 14:49 |
| zhubingbing | thanks | 14:49 |
| sdake | roger | 14:50 |
| zhubingbing | - - | 14:50 |
| zhubingbing | My first contribution to the code, I hope to be able to contribute more in the future, I hope you do not mind my new | 14:50 |
| sdake | contributions entirely welcome | 14:51 |
| sdake | we want gnocchi | 14:51 |
| zhubingbing | ok | 14:51 |
| sdake | we want all big tent projects deployable, and maybe even some that aren't big tent if they are server projects | 14:51 |
| sdake | one I want is vmtp - to help validate the cloud - similar to the sanity blueprint | 14:52 |
| *** michauds has joined #openstack-kolla | 14:53 | |
| *** dwalsh has quit IRC | 14:53 | |
| inc0 | damn libvirt has same issue | 14:53 |
| inc0 | mewald, you managed to reproduce it on localhost? | 14:53 |
| inc0 | the VIR_MIGRATE thing | 14:54 |
| zhubingbing | sdake thinks | 14:54 |
| inc0 | zhubingbing, by the way, I think you meant thanks:) thinks is a different word | 14:54 |
| sdake | inc0 on ubuntu? | 14:54 |
| zhubingbing | thanks | 14:54 |
| inc0 | sdake, yeah | 14:54 |
| zhubingbing | - | 14:54 |
| sdake | i don't deploy ubuntu particluarly often | 14:54 |
| inc0 | :) | 14:54 |
| sdake | so ubuntu is completely broken???? | 14:54 |
| inc0 | sdake, I was asking mewald | 14:55 |
| inc0 | just gates, don't sweat it | 14:55 |
| sdake | ok thanks :) | 14:55 |
| inc0 | however we can't get deploy gate to pass | 14:55 |
| inc0 | and locally it's all rainbows and raspberries | 14:55 |
| sdake | gates failing is an entirely different problem then ubuntu doesn't deploy :) | 14:55 |
| inc0 | yeah, it's been failing for few weeks | 14:56 |
| inc0 | and no luck in making them work | 14:56 |
| sdake | one challenge we face is we have ALOT of upstreams | 14:56 |
| sdake | any one of those upstreams change something, and we are prone to breakage | 14:56 |
| sdake | we are basically ci for hundreds of projects | 14:56 |
| sdake | :( | 14:56 |
| inc0 | so it was either changing to infra mirrors or change of infra image | 14:57 |
| *** dwalsh has joined #openstack-kolla | 14:57 | |
| inc0 | that's my bets | 14:57 |
| sdake | the infra mirrors were not changed in our code base - they were already in place iirc | 14:57 |
| inc0 | well we started to use them | 14:57 |
| sdake | i did add ubuntu-cloud-archive, but iirc deploy was busted prior to that | 14:57 |
| *** diogogmt has joined #openstack-kolla | 14:58 | |
| sdake | let me remove ubuntu cloud archive mirror and see if that fixes things - moment | 14:59 |
| zhubingbing | excuse me, is this the next aodh to do it, no one to do me to submit a blueprint | 15:00 |
| Jeffrey4l | inc0, how could u sure is it is a issue from infra mirror? | 15:00 |
| inc0 | Jeffrey4l, I'm not sure, I speculate | 15:01 |
| Jeffrey4l | ok. | 15:01 |
| openstackgerrit | Steven Dake proposed openstack/kolla: DO NOT MERGE - test of Revert "Use local mirror of ceph and uca" https://review.openstack.org/350082 | 15:02 |
| inc0 | issue is, I run ubuntu and I don't have this issue | 15:02 |
| inc0 | btw we need to use newer ceph | 15:02 |
| inc0 | we're 2 releases behind | 15:02 |
| Jeffrey4l | inc0, what's you kernel number? | 15:02 |
| Jeffrey4l | kernel version? | 15:02 |
| sdake | inc0 agree - jewel is already mirrored so that part is easy | 15:03 |
| inc0 | 4,4,0-28 for me | 15:03 |
| sdake | the hard part is porting to jewel | 15:03 |
| *** mliima has quit IRC | 15:03 | |
| inc0 | I really doubt it's kernel issue, but never know | 15:03 |
| inc0 | I blame libvirt now | 15:03 |
| Jeffrey4l | could u try 3.13-92, the same like the ci ubuntu kernel image. | 15:03 |
| inc0 | this will be ubuntu 14.04 with upgraded kernel? | 15:04 |
| Jeffrey4l | "Linux ubuntu-trusty-ovh-bhs1-3097158 3.13.0-92-generi" | 15:04 |
| inc0 | I have vanilla xenial | 15:04 |
| inc0 | ah, so trusty | 15:04 |
| sdake | Jeffrey4l i think that kernel wont work on inc0's system | 15:04 |
| inc0 | you know what? even better than that, I'll try to get actual infra image | 15:04 |
| Jeffrey4l | hm. need trusty. | 15:04 |
| Jeffrey4l | cool | 15:04 |
| *** daneyon_ has joined #openstack-kolla | 15:04 | |
| inc0 | sdake, I run vms and have large quota | 15:04 |
| inc0 | will find a way | 15:05 |
| sdake | inc0 cool | 15:05 |
| sdake | thanks :) | 15:05 |
| inc0 | but later, meeting charlie foxtrot approaching | 15:05 |
| sdake | would be helpful to get that gate pasing, alot of people have concerns about it when they submit patches | 15:05 |
| Jeffrey4l | i doubt the kernel version. | 15:05 |
| sdake | inc0 whisky tango foxtrot ftw :) | 15:05 |
| inc0 | I need to learn better military alphabet | 15:06 |
| inc0 | as it is actually useful thing | 15:06 |
| *** ppowell has quit IRC | 15:07 | |
| sdake | it all starts with alpha :) | 15:07 |
| *** b_bezak has quit IRC | 15:07 | |
| sdake | i have to nato-ize my last name on the phone - otherwise poeple never get it | 15:07 |
| inc0 | yeah, trust me I know | 15:08 |
| sdake | which is hwy i learned it | 15:08 |
| sdake | ya - you have it rough inc0 :) | 15:08 |
| inc0 | only for me it's Juliet Alpha Sierra Tango Romeo Zulu Echo Bravo Sierra Kilo India | 15:09 |
| *** daneyon_ has quit IRC | 15:09 | |
| *** mbound has joined #openstack-kolla | 15:10 | |
| sdake | i hope zulu is an option for openstack name for Z :) | 15:12 |
| inc0 | I think there are better names;) | 15:12 |
| zhubingbing | https://blueprints.launchpad.net/kolla/+spec/aodh-role | 15:17 |
| sdake | zhubingbing need something with that blueprint? | 15:17 |
| sdake | aodh containers are built - not sure if they are functional as no ansible playbook exists for them | 15:18 |
| *** mliima has joined #openstack-kolla | 15:18 | |
| *** mkoderer_cloud has quit IRC | 15:18 | |
| *** pbourke has quit IRC | 15:18 | |
| sdake | zhubingbing you just need it approved? | 15:19 |
| zhubingbing | yes | 15:19 |
| sdake | low priority - but will approve | 15:19 |
| sdake | blueprints need to be in good progress by 15th or they will be pushed to occata | 15:19 |
| *** ppowell has joined #openstack-kolla | 15:19 | |
| zhubingbing | ok，I expect it to be completed in 3 days. | 15:20 |
| *** belmoreira has quit IRC | 15:21 | |
| sdake | ya shouldn't be too hard if you are familiar with aodh | 15:21 |
| zhubingbing | I'm going to study now. | 15:24 |
| *** zhiwei has joined #openstack-kolla | 15:26 | |
| *** zhurong has quit IRC | 15:30 | |
| *** zhiwei has quit IRC | 15:30 | |
| *** thumpba has joined #openstack-kolla | 15:32 | |
| *** thumpba has quit IRC | 15:32 | |
| *** thumpba has joined #openstack-kolla | 15:33 | |
| *** mark-casey has joined #openstack-kolla | 15:33 | |
| *** nradojevic has quit IRC | 15:38 | |
| *** dwalsh has quit IRC | 15:40 | |
| *** dwalsh has joined #openstack-kolla | 15:40 | |
| *** dave-mcc_ has joined #openstack-kolla | 15:43 | |
| *** ppowell has quit IRC | 15:43 | |
| *** dave-mccowan has quit IRC | 15:44 | |
| *** banix has quit IRC | 15:53 | |
| *** pbourke has joined #openstack-kolla | 15:54 | |
| *** ppowell has joined #openstack-kolla | 15:58 | |
| *** bootsha has joined #openstack-kolla | 15:59 | |
| *** godleon has quit IRC | 15:59 | |
| *** HyperJohnGraham has quit IRC | 16:00 | |
| jogam | currently running OS-KOLLA from tag 2.0.1 any thoughts/comments on what to deploy TODAY (newton-1 || 2.0.2)? | 16:07 |
| jogam | sdake: any comments for which version of kolla to use nowadays? | 16:16 |
| sdake | 2.0.1 or 2.0.2 | 16:17 |
| sdake | 2.0.2 is very close to 2.0.1 | 16:17 |
| sdake | the changes are here: | 16:17 |
| sdake | https://launchpad.net/kolla/+milestone/2.0.2 | 16:18 |
| *** nradojevic has joined #openstack-kolla | 16:18 | |
| *** ssurana has joined #openstack-kolla | 16:18 | |
| *** bootsha has quit IRC | 16:19 | |
| zhubingbing | hi，Aodh no expirer this service，this is package，so i think，/docker/aodh/aodh-expirer dockerfile should not be deleted？ | 16:19 |
| zhubingbing | this link ：https://github.com/openstack/aodh/blob/master/install-guide/source/install-rdo.rst | 16:19 |
| zhubingbing | ithink a service is a container | 16:20 |
| zhubingbing | so，/docker/aodh/aodh-expirer dockerfile should not be deleted？ | 16:20 |
| *** mbound has quit IRC | 16:21 | |
| *** unicell has joined #openstack-kolla | 16:21 | |
| *** dwalsh has quit IRC | 16:22 | |
| *** bootsha has joined #openstack-kolla | 16:22 | |
| zhubingbing | sdake | 16:22 |
| *** nradojevic has quit IRC | 16:23 | |
| sdake | zhubingbing i don't know anything about aodh | 16:23 |
| *** unicell1 has quit IRC | 16:24 | |
| zhubingbing | ok | 16:24 |
| *** dwalsh has joined #openstack-kolla | 16:24 | |
| mark-casey | inc0: wirehead_: kfox1111: that Mysql and Ceph webinar redhat and percona are putting on starts in 45 min. If you register and have to skip they usually send a recording link after ...was trying to remember who all was interested | 16:27 |
| mark-casey | https://www.percona.com/resources/webinars/mysql-and-ceph | 16:27 |
| jogam | sdake: thanks, e.g. newton is a way off yet? | 16:29 |
| sdake | few months | 16:29 |
| sdake | need exct schedule? | 16:29 |
| jogam | sdake: looking at launchpad timeline, that exact? | 16:30 |
| sdake | historically a big chunk of kolla conribution comes in milestone 3 and the rcs.. | 16:30 |
| *** JoseMello has quit IRC | 16:30 | |
| sdake | what date does it say | 16:30 |
| jogam | sdake: 2016-10-21 | 16:30 |
| jogam | fpor 3.0.0 | 16:30 |
| sdake | yup that is our targt | 16:30 |
| sdake | i think | 16:31 |
| sdake | let me double check exact schedule moment | 16:31 |
| jogam | sdake: what do you think about CoreOS Tectonic - Stackanetes? - particularily self-healing capabilities? | 16:31 |
| sdake | http://releases.openstack.org/newton/schedule.html | 16:31 |
| sdake | i have no opinion of tectonic - have never tried it | 16:31 |
| sdake | stackanetes is based upon kolla - but again haven't tried it | 16:32 |
| sdake | my general take on what stackenetes is trying to achieve is similar to what kolla-kubernetes is in progress achieving | 16:32 |
| jogam | sdake: will kolla have self-healing capabilities similar to the advertised Tectonic example: https://www.youtube.com/watch?v=DPYJxYulxO4 | 16:32 |
| sdake | and that is likely 9 months out | 16:32 |
| jogam | sdake: thanks | 16:35 |
| jogam | sdake: so is the kolla team planning on ´merging´ with the kolla-kubernetes team/efforts or are those two projects to co-exist and potentially compete docker vs k8s? | 16:40 |
| sdake | huh? | 16:40 |
| sdake | kolla-kubernetes is a delivery of the kolla project | 16:41 |
| sdake | so its not a merge - we are already one community | 16:41 |
| mark-casey | jogam: rescheduling down/failed pods is a feature of kubernetes itself more than it is specific to any particular service implemented in it. Once you're working within k8s correctly you sort of get that for free; that's why the platform is appealing (one reason) | 16:42 |
| jogam | e.g. kolla is the openstack deployment tool vs. docker/k8s are the underlying container environments? | 16:43 |
| *** berendt has joined #openstack-kolla | 16:43 | |
| jogam | and kolla is trying to use tools (ansible) that allow to deploy to any container env? | 16:43 |
| *** thumpba has quit IRC | 16:44 | |
| *** thumpba has joined #openstack-kolla | 16:45 | |
| sdake | jogam kolla is a collection of components and an ABI that provide operational capabilities for an openstack cloud - currently our implmenetation is based upon ansible + docker + some extra "glue" | 16:45 |
| sdake | jogam we are expanding to adding kubernetes + docker + some extra "glue" as a deployment technique | 16:45 |
| sdake | choice will be on operators as to which they want to use | 16:45 |
| sdake | but because we have a common abi, either way will work well | 16:46 |
| *** banix has joined #openstack-kolla | 16:46 | |
| sdake | jogam currently we lead with ansible and i don't see a strong motive to change that at this time - in the future kuberenetes will be an alternate choice in the kolla project | 16:47 |
| sdake | as for self healing, as mark-casey said, we get that for free with kubernetes | 16:48 |
| sdake | if your question is "is self healing coming to ansible" - no idea | 16:48 |
| sdake | jogam the self-healing is the main motive for making use of k8s as a deployment system | 16:49 |
| *** egonzalez has joined #openstack-kolla | 16:49 | |
| *** thumpba has quit IRC | 16:49 | |
| sdake | if k8s didn't have self healing, it wouldn't offer much value and we likely wouldn't have a kolla-kubernetes project in development | 16:49 |
| jogam | sdake/mark-casey: e.g. kolla on docker via ansible, no k8s will not have self-healing capabilities | 16:49 |
| *** rustyl has joined #openstack-kolla | 16:50 | |
| openstackgerrit | zhubingbing proposed openstack/kolla: Add gnocchi ansible role https://review.openstack.org/349351 | 16:50 |
| jogam | sdake: what is the reason for not developing kolla for k8s from the start, are there technical or performance reasons? | 16:51 |
| sdake | jogam good question | 16:52 |
| openstackgerrit | Prithiv proposed openstack/kolla: Dockerfiles for Networking-SFC Container https://review.openstack.org/344930 | 16:52 |
| sdake | kolla started as a project to deploy openstack on kubernetes on top of fedora atomic | 16:52 |
| sdake | kubernetes had technical limitations that nearly killed the project and made the objective impossible | 16:53 |
| sdake | we pivoted to ansible because it had no such limitations | 16:53 |
| sdake | 3 years have passed - we are giving it another go | 16:53 |
| sdake | we actually pivoted a few times before we go to ansible fwiw ;) | 16:54 |
| sdake | got to that is | 16:54 |
| inc0 | jogam, there is more to this | 16:55 |
| inc0 | k8s is a different beast which many ops doesn't want to deal with | 16:55 |
| inc0 | not everyone is google | 16:55 |
| inc0 | ansible is deterministic, closer to "classical" approach | 16:55 |
| inc0 | many people prefer this over magical-schedule-me-wherever k8s | 16:56 |
| *** unicell has quit IRC | 16:56 | |
| sdake | inc0 makes an excellent point - some folks want ansible some folks want kubernetes - we give both (in the future) and operators can choose their paltform | 16:57 |
| jogam | inc0: but how would you use ansible to enable self-healing? | 16:57 |
| sdake | jogam i answered that question already | 16:58 |
| inc0 | jogam, you would not | 16:58 |
| sdake | "no idea" | 16:58 |
| inc0 | not everybody want's or cares about self healing | 16:58 |
| inc0 | ops runs without this capability for decades now | 16:58 |
| jogam | sdake/inc0: I understand the two different approaches, but my main question with the ansible kolla line is: how do you perform host OS upgrades etc without braking the OpenStack environment as well as how does kolla-ansible react to host failure? | 16:59 |
| sdake | upgrades = done | 17:00 |
| sdake | kolla has nway active ha to protect from host failure | 17:00 |
| sdake | try kolla-ansible upgrade out | 17:00 |
| sdake | try killing a control node with 3 control nodes deployed | 17:00 |
| sdake | both of these work properly | 17:00 |
| jogam | sdake: ok, strength by numbers, that always works, but now I have an instance running on one nova-compute node on host A, if A dies the instance is gone, right? | 17:01 |
| sdake | we do not protect from instance failure | 17:01 |
| sdake | no product has this capability | 17:02 |
| sdake | even with kubernetes you dont get this capabilty | 17:02 |
| sbezverk | mark-casey Have you managed to connect to the mysql-ceph webinar? | 17:02 |
| sdake | the instance is not gone forever | 17:02 |
| sdake | it can be rebooted | 17:02 |
| sdake | but its running state is lost | 17:03 |
| sdake | sbezverk whats that about? | 17:05 |
| sdake | how to run mysql backed by ceph? | 17:05 |
| sbezverk | sdake do not know, trying to join to find out | 17:06 |
| jogam | sdake: I guess instance HA would then be via PaceMaker/CoroSync - https://access.redhat.com/articles/1544823 | 17:12 |
| sdake | jogam that is a possibility - btw its Corosync ) | 17:12 |
| *** thumpba has joined #openstack-kolla | 17:13 | |
| *** dwalsh has quit IRC | 17:13 | |
| jogam | sdake: change of topic -- on kolla-kubernetes one of the first documented services is DNS, for kolla I see that the designate images are build but there are no config/deploy files in the sources? Dead? | 17:14 |
| sdake | huh? | 17:15 |
| sdake | just because nobody has implemented the code doesn't make a project dead | 17:15 |
| sdake | we have 10 pages of outsttanding reviews - review team doing best we can to keep up with reviews | 17:15 |
| sdake | but we are overloaded | 17:15 |
| mark-casey | sbezverk: I had to take my own advice about getting sent the recording :/ got tied up | 17:16 |
| *** mbound has joined #openstack-kolla | 17:16 | |
| sbezverk | mark-casey: if you get the recording, please share it. I cannot access that damn PoC meeting server they are using.. | 17:17 |
| mark-casey | I should get emailed a vid link. I'll ping you with it. They say 24-48 hours but it's often same-day | 17:17 |
| sbezverk | mark-casey: thank you | 17:18 |
| jogam | sdake: ok, wont steal more of your time then :) just noticed that there is very little code for designate and was wondering if it is ´dead´... | 17:20 |
| *** gfidente has quit IRC | 17:21 | |
| sbezverk | jogam for kolla-kubernetes we are using skydns services, but this dns is very specific to internal openstack services running on the cluster needs. | 17:22 |
| sdake | jogam we have a whole slew of containers for which there is no ansible deployment | 17:22 |
| sdake | designate is one of many | 17:22 |
| sdake | afaik designate is alive and doing well | 17:22 |
| *** signed8bit is now known as signed8bit_Zzz | 17:22 | |
| *** thumpba_ has joined #openstack-kolla | 17:22 | |
| *** unicell has joined #openstack-kolla | 17:24 | |
| jogam | sdake | 17:24 |
| jogam | sdake: designate itself is definitely well :) | 17:24 |
| *** thumpba has quit IRC | 17:25 | |
| openstackgerrit | Eduardo Gonzalez proposed openstack/kolla: Customizations for Designate https://review.openstack.org/350172 | 17:25 |
| jogam | surprise, surprise :) | 17:26 |
| sdake | then why did you ask the question? | 17:26 |
| jogam | sdake: because I tried to get DNS working and was searching through the GIT repo and could not find anything that would indicate that there is a flag to enable DNS/designate | 17:27 |
| openstackgerrit | Eduardo Gonzalez proposed openstack/kolla: Customizations for Designate https://review.openstack.org/350172 | 17:31 |
| *** signed8bit_Zzz is now known as signed8bit | 17:32 | |
| *** bdaca has joined #openstack-kolla | 17:42 | |
| *** unicell has quit IRC | 17:44 | |
| *** unicell1 has joined #openstack-kolla | 17:44 | |
| *** inc0 has quit IRC | 17:52 | |
| *** zhiwei has joined #openstack-kolla | 17:53 | |
| *** inc0 has joined #openstack-kolla | 17:54 | |
| sdake | jogam right because its nto implemented | 17:55 |
| jogam | sdake: understood and figured out last week THU or so... | 17:58 |
| *** zhiwei has quit IRC | 17:58 | |
| jogam | sdake: thats why my question if its an open or dead item :) | 17:58 |
| sdake | cool - well 2.0.2 is what i'd recommend | 17:58 |
| sdake | oh definately open item | 17:58 |
| sdake | big tent is one of our many objectives | 17:58 |
| jogam | have the 2.0.1 images built already and in my local registry e.g. I will just use those for now :) | 18:00 |
| *** mbound has quit IRC | 18:02 | |
| sbezverk | rhallisey my computer does not have a mic | 18:03 |
| sbezverk | sorry I will in listen mode | 18:03 |
| rhallisey | https://etherpad.openstack.org/p/kolla-kube-features | 18:05 |
| *** bootsha has quit IRC | 18:09 | |
| sbezverk | we already have it on bare metal | 18:09 |
| sbezverk | :-) | 18:09 |
| *** shardy has quit IRC | 18:10 | |
| *** bootsha has joined #openstack-kolla | 18:12 | |
| wirehead_ | inc0: https://review.openstack.org/#/c/343117/ https://review.openstack.org/#/c/343116/ | 18:25 |
| patchbot | wirehead_: patch 343117 - kolla - Fix interface address for Neutron-server | 18:25 |
| patchbot | wirehead_: patch 343116 - kolla-kubernetes - Neutron control bootstraps and runs without host n... | 18:25 |
| *** jrist has quit IRC | 18:27 | |
| *** jrist has joined #openstack-kolla | 18:28 | |
| *** jrist has quit IRC | 18:30 | |
| *** YefimG has joined #openstack-kolla | 18:30 | |
| YefimG | Hello, guys! I am trying to evaluate Kolla all-in-one using Docker Hub images, but both 2.0.0 and 3.0.0 tags are not found: "Unknown error message: Tag 2.0.0 not found in repository docker.io/kolla/centos-binary-kolla-toolbox". Am I missing something? | 18:31 |
| openstackgerrit | Robin Naundorf proposed openstack/kolla: Fix inconsistencies in git url https://review.openstack.org/350222 | 18:39 |
| *** signed8bit is now known as signed8bit_Zzz | 18:43 | |
| *** signed8bit_Zzz is now known as signed8bit | 18:43 | |
| *** unicell1 has quit IRC | 18:49 | |
| rhallisey | inc0, sdake, kolla-kube patches | 18:52 |
| rhallisey | https://etherpad.openstack.org/p/R3rVu81fGU | 18:52 |
| *** unicell has joined #openstack-kolla | 18:53 | |
| *** unicell has joined #openstack-kolla | 18:53 | |
| *** unicell1 has joined #openstack-kolla | 18:55 | |
| *** unicell has quit IRC | 18:55 | |
| *** jtriley has quit IRC | 18:58 | |
| sdake | YefimG odd - let me have a look | 19:00 |
| sdake | YefimG https://hub.docker.com/r/kolla/centos-binary-kolla-toolbox/tags/ | 19:01 |
| sdake | YefimG as you can see, 2.0.0 is not there | 19:02 |
| sdake | YefimG we went through a cleaning of our dockerhub registry a few weeks ago - perhaps we cleaned too much | 19:02 |
| sdake | I'd recommend setting "openstack_release: 2.0.1" in globals.yml | 19:02 |
| sdake | (or if using mitaka branch, 3.0.1) | 19:02 |
| sdake | YefimG in the meantime, we will get the 2.0.0 and 3.0.0 images published | 19:03 |
| *** bootsha has quit IRC | 19:03 | |
| sdake | sorry not 3.0.0 | 19:03 |
| sdake | 3.0.0 is not yet eleased | 19:03 |
| *** mbound has joined #openstack-kolla | 19:03 | |
| sdake | 3.0.0 is master = we dont publish unreleased software as tags | 19:03 |
| sdake | (fwiw the cleaning was removing the 3.0.0 images from dockerhub because they are not part of an official tag of kolla | 19:04 |
| YefimG | Thank you, sdake! :) I was confused because of the quick start guide where it says to use 3.0.0 for Mitaka release. | 19:04 |
| YefimG | 2.0.0, sorry | 19:05 |
| *** mbound has quit IRC | 19:08 | |
| sdake | YefimG ya - docs = imperfect | 19:09 |
| sdake | inc0 re the blueprint third party plugin | 19:09 |
| sdake | inc0 This is a very cool feature. However, there seems no doc or example | 19:10 |
| sdake | about how to customize the Docerfile. For example, if I want to run | 19:10 |
| sdake | ./tools/build.py --template-only --template-override ./heat-extend.j2 | 19:10 |
| sdake | heat, what is the correct format for heat-extend.j2? Would you mind | 19:10 |
| sdake | adding a guide about this? Thanks. - Hui | 19:10 |
| sdake | inc0 when you get some free time before 9/2/2016, can you write some docs or examples and also respond to hui when thoe docs or examples will be in queue? | 19:10 |
| sdake | coolsvap ping re 2.0.0 images | 19:12 |
| *** HyperJohnGraham has joined #openstack-kolla | 19:16 | |
| Daviey | sdake: talking of versioning... did you look at my standalone example for determining image version at runtime? | 19:17 |
| sdake | i have not | 19:17 |
| Daviey | rather than all builds being a fixed version, regardless of patch release? | 19:17 |
| Daviey | sdake: https://github.com/Daviey/tmp-kolla-version | 19:18 |
| Daviey | the README makes it make sense | 19:19 |
| openstackgerrit | Robin Naundorf proposed openstack/kolla: Fix inconsistencies in git url https://review.openstack.org/350222 | 19:19 |
| *** inc0 has quit IRC | 19:21 | |
| *** HyperJohnGraham has quit IRC | 19:22 | |
| rhallisey | wirehead_, https://blueprints.launchpad.net/kolla-kubernetes/+spec/neutron-kubernetes | 19:25 |
| rhallisey | if you can fill out work items that would be awesome | 19:25 |
| *** zhubingbing has quit IRC | 19:25 | |
| *** HyperJohnGraham has joined #openstack-kolla | 19:33 | |
| *** daneyon_ has joined #openstack-kolla | 19:35 | |
| *** HyperJohnGraham has quit IRC | 19:39 | |
| *** daneyon_ has quit IRC | 19:40 | |
| *** HyperJohnGraham has joined #openstack-kolla | 19:40 | |
| *** godleon has joined #openstack-kolla | 19:40 | |
| *** coolsvap has quit IRC | 19:41 | |
| sdake | Daviey looking now | 19:42 |
| *** bdaca has quit IRC | 19:43 | |
| sdake | Daviey cool had a look - can you explain whatproblem that solves | 19:43 |
| sdake | Daviey too many problems floating around in my head to keep context - I know we have already had this discussion but mind refreshing me ? | 19:43 |
| *** zhiwei has joined #openstack-kolla | 19:45 | |
| Daviey | sdake: build images from trunk or stable/* right now... Are you on the same version as me? | 19:45 |
| Daviey | 3.0.0 != 3.0.0 | 19:45 |
| sdake | right! | 19:45 |
| sdake | i like it | 19:45 |
| Daviey | if i deploy 3.0.0 now... and add some more nodes tomorrow.. are they are running the same 3.0.0? | 19:46 |
| sdake | what happens when 3.0.0 comes ut | 19:46 |
| Daviey | upgrade wont detect mismatching 3.0.0's | 19:46 |
| Daviey | sdake: 3.0.0 == exactly 3.0.0 | 19:46 |
| Daviey | ie, the tagged release | 19:46 |
| sdake | our current recomendation for stable branches is to add an alpha version number | 19:46 |
| Daviey | sdake: per commit? | 19:46 |
| sdake | no - the operator adds an alpha version nubmer if they change something | 19:47 |
| Daviey | that adds an incrementing devX (where X = patches since last tag) | 19:47 |
| sdake | they become responsible for their own versioning at that point | 19:47 |
| sdake | that is what the docs suggest yes | 19:47 |
| Daviey | right.. well, with this they still can be | 19:47 |
| sdake | for master this work makes alot of sense to me | 19:47 |
| Daviey | if they declare an openstack_release = "3.0.0aDAVIEYMAGIC1" | 19:48 |
| Daviey | sdake: I actually think it makes more sense post release | 19:48 |
| Daviey | but also useful pre-release for us | 19:48 |
| Daviey | i've been hit multiple times of thinking all images on my deployment were identical | 19:48 |
| sdake | can you explain the post release part | 19:48 |
| Daviey | sdake: If i leave it as openstack_release = "currrent" | 19:48 |
| sdake | i get pre-release - makes e sen to me | 19:49 |
| Daviey | then.. when i update my git checkout (or distro upgrade if packaged kolla) | 19:49 |
| *** zhiwei has quit IRC | 19:49 | |
| Daviey | try and do a deploy, i'll get an error that i don't have the correct image build | 19:49 |
| Daviey | HITN.. do a rebuild, something changed | 19:49 |
| Daviey | HINT* | 19:49 |
| sdake | its not that i dont like the idea and am not listening, i just dont get it | 19:50 |
| Daviey | but.. I am uber user.. so i know i want version FOOBAR.. so i can override that behavior and declare openstack_release = "3.0.0davieymagic" | 19:50 |
| Daviey | sdake: How are distros supposed to patch version number on config files? | 19:50 |
| sdake | ok i get 3.0.0 = current since its prerelease | 19:50 |
| Daviey | we don't support .d directories | 19:50 |
| Daviey | and at the moment WE are patching the persion number by hand, prior to release | 19:51 |
| Daviey | But, as you know, distros (and even stable-maint) shouldn't change config files | 19:51 |
| Daviey | so we have a mismatch | 19:51 |
| sdake | if we do what i think i heard, 1.1.1 would be current 2.0.2 would be current, 3.0.0 would be current | 19:51 |
| sdake | is that true or false? | 19:51 |
| Daviey | no... | 19:51 |
| sdake | running slow today apparently :( | 19:52 |
| Daviey | :) | 19:52 |
| Daviey | So... | 19:52 |
| Daviey | If i leave openstack_release = "current" | 19:52 |
| Daviey | then kolla works that out to be the current installed version of kolla | 19:52 |
| Daviey | when i ran my example.. | 19:53 |
| Daviey | "kolla_release": "3.0.0.0b2.dev182" | 19:53 |
| Daviey | SO... it is 3.0.0 beta 2.. + 182 commits | 19:53 |
| Daviey | when b3 gets cut, the version will be 3.0.0b3 (clean, no dev) | 19:53 |
| Daviey | when release gets cut, it will be 3.0.0 (clean) | 19:53 |
| Daviey | To take it further, post release | 19:54 |
| Daviey | (where config files matter!) | 19:54 |
| Daviey | add a commit, 3.0.0.dev1 | 19:54 |
| Daviey | etc.. | 19:54 |
| Daviey | When we are ready to cut 3.0.1, the "current" will be that version, clean | 19:54 |
| Daviey | No config files changed | 19:54 |
| Daviey | (See why i think it is more interesting post release?) | 19:55 |
| sdake | i would say that upgrades don't work with that | 19:55 |
| sdake | but we fixed that (jeffrey4l did iirc) | 19:55 |
| Daviey | Stops us having to do nasty things like bug 1605061 .. EACH point release | 19:56 |
| openstack | bug 1605061 in kolla "openstack release tag needs to be updated for latest stable mitaka release" [Critical,In progress] https://launchpad.net/bugs/1605061 - Assigned to Rahul Sharma (rahulsharmaait) | 19:56 |
| Daviey | sdake: why wont upgrades work? | 19:56 |
| sdake | they prevously looked at version #s only | 19:56 |
| Daviey | kolla works out the image version locally, automagically | 19:56 |
| sdake | if its always current - that wouldn't work | 19:56 |
| Daviey | If the image isn't built for that version, kolla (ansible) will complain it can't find the image in the registry | 19:56 |
| *** bootsha has joined #openstack-kolla | 19:57 | |
| Daviey | which forces the user to rebuild (or backdate there openstack_release = "current" to openstack_release = "3.0.1" | 19:57 |
| Daviey | But we want to encourage image rebuilds if kolla gets upgraded, surely? | 19:57 |
| *** jrist has joined #openstack-kolla | 19:58 | |
| sdake | Daviey are you in the us? | 19:58 |
| Daviey | sdake: UK | 19:58 |
| sdake | sweet | 19:58 |
| Daviey | But, you are part of us.. right? | 19:58 |
| sdake | can you run your idea by jeffrey4l | 19:58 |
| sdake | yes of course | 19:58 |
| Daviey | I'll send him a mail.. but probably not until the morning now | 19:59 |
| sdake | i am struggling to understand all the implications with upgrade and reconfigure | 19:59 |
| Daviey | i *think* this makes life easier for upstream, distros and users | 19:59 |
| sdake | if Jeffrey4l says it will work - that wfm - he wrote alot of that code | 19:59 |
| Daviey | but i'm happy to be told i'm wrong | 19:59 |
| Daviey | (well, somewhat) | 19:59 |
| sdake | i'd ask inc0 on the upgrade part of things - but he is setting up the osic cluster atm | 20:00 |
| sdake | the first ting i can agree with - that using 3.0.0 when we are not 3.0.0 is completely wrong | 20:00 |
| sdake | that should be current | 20:00 |
| Daviey | The way it looks to me... on upgrade | 20:01 |
| sdake | Daviey the reason i asked about your location is jeffrey4l is on your hours | 20:01 |
| Daviey | "hey node, are you running version X? No? Oh, please upgrade" | 20:01 |
| Daviey | so it should make life easier | 20:01 |
| sdake | our upgrades actually look at the hash and version number in combo iirc | 20:02 |
| Daviey | sdake: smarter people than you or I have tried to work out my hours, and died trying | 20:02 |
| sdake | Daviey yup - i hear ya :) | 20:02 |
| Daviey | sdake: Oh.. i didn't know about the hash check | 20:02 |
| sdake | Jeffrey4l knows for sure | 20:02 |
| sdake | does it make a difference? | 20:03 |
| sdake | (the hash) | 20:03 |
| *** jtriley has joined #openstack-kolla | 20:05 | |
| *** mark-casey has quit IRC | 20:08 | |
| *** inc0 has joined #openstack-kolla | 20:08 | |
| wirehead_ | Well, it’s a matter of tribes. Eastern Standard Tribe. | 20:08 |
| inc0 | sdake, back | 20:08 |
| inc0 | eee? | 20:08 |
| wirehead_ | It’s just that I suspect that sdake is frequently living in the Eastern Standard Tribe regardless of actual location. | 20:08 |
| inc0 | sdake, I'll publish docs for it | 20:09 |
| inc0 | let me just sort out this stackanetes crap | 20:09 |
| sdake | inc0 roger no rush - 9/2 is our deadline | 20:09 |
| sdake | and docs can land during rcs | 20:09 |
| inc0 | as I have hard time focused on anything else | 20:09 |
| inc0 | 9/2 is safe bet | 20:09 |
| *** HyperJohnGraham_ has joined #openstack-kolla | 20:09 | |
| sdake | no rush | 20:10 |
| sdake | i just saw it in a a review - thought it was important | 20:10 |
| sdake | inc0 mind lending some brainpower for 10 mins to Daviey's proposal | 20:10 |
| sdake | or too busy? | 20:10 |
| inc0 | nah, shoot | 20:10 |
| inc0 | which proposal? | 20:10 |
| sdake | Daviey mind starting again with inc0 | 20:11 |
| sdake | inc0 wrote our upgrade design | 20:11 |
| sdake | inc0 here is reo https://github.com/Daviey/tmp-kolla-version | 20:11 |
| *** HyperJohnGraham has quit IRC | 20:11 | |
| sdake | 3 small files | 20:11 |
| sdake | easy to loook at | 20:11 |
| *** HyperJohnGraham_ is now known as HyperJohnGraham | 20:11 | |
| sdake | i don't understand how current is used in stable branches | 20:12 |
| sdake | might read scrollbackt o save him some time typing again | 20:12 |
| inc0 | sdake, soo this "current" will mean latest stable | 20:13 |
| sdake | ya but how do we tag that | 20:13 |
| sdake | when we have multiple stables | 20:14 |
| inc0 | so Daviey you're trying to autodiscover version of kolla and set_fact openstack_release based on this? | 20:16 |
| sdake | Daviey i'm afraid I just dont totallly understand the solution with all the variables - we have our team meeting tomorrow if you can make it - we can discuss then ify ou like | 20:16 |
| inc0 | I have another idea | 20:16 |
| Daviey | sorry chaps.. i have a hard stop | 20:16 |
| Daviey | will read scrollback in 2 hours | 20:16 |
| sdake | roger Daviey | 20:16 |
| inc0 | how about writing super simple ansible module | 20:16 |
| inc0 | that will do this | 20:16 |
| sdake | sorry i can't give you a concrete answer right at this moment :) | 20:16 |
| inc0 | sure, we can talk later | 20:16 |
| sdake | Daviey i think there is potential here to fix atleast the master problem, lets figure out if it works for stable branches too | 20:17 |
| sdake | for that i need inc0 and jeffrey4l to weigh in on reconfigure + upgrade | 20:17 |
| openstackgerrit | sean mooney proposed openstack/kolla: generate bifrost yaml configs https://review.openstack.org/349684 | 20:20 |
| openstackgerrit | sean mooney proposed openstack/kolla: add bifrost deploy ansible support https://review.openstack.org/349685 | 20:20 |
| openstackgerrit | sean mooney proposed openstack/kolla: add deploy-servers command https://review.openstack.org/349729 | 20:20 |
| *** egonzalez has quit IRC | 20:26 | |
| *** jtriley has quit IRC | 20:28 | |
| sdake | rhallisey this patch needs attention: https://review.openstack.org/#/c/336751/6 | 20:31 |
| patchbot | sdake: patch 336751 - kolla - Fix interface address for RabbitMQ and disable clu... | 20:31 |
| sdake | rest ready to merge | 20:31 |
| rhallisey | kk | 20:32 |
| *** openstackgerrit_ has joined #openstack-kolla | 20:35 | |
| *** Jeffrey4l has quit IRC | 20:35 | |
| *** Jeffrey4l has joined #openstack-kolla | 20:35 | |
| *** mliima has quit IRC | 20:35 | |
| *** openstackgerrit_ has quit IRC | 20:36 | |
| openstackgerrit | Ryan Hallisey proposed openstack/kolla: Horizon interface address and memcached override https://review.openstack.org/330828 | 20:41 |
| openstackgerrit | Ryan Hallisey proposed openstack/kolla: Fix interface address for memcached atop kubernetes https://review.openstack.org/327987 | 20:41 |
| openstackgerrit | Ryan Hallisey proposed openstack/kolla: Fix interface address for RabbitMQ and disable cluster for Kubernetes https://review.openstack.org/336751 | 20:41 |
| openstackgerrit | Ryan Hallisey proposed openstack/kolla: Glance interface address, memcached override, and glance override. https://review.openstack.org/332481 | 20:41 |
| openstackgerrit | Ken Wronkiewicz proposed openstack/kolla-kubernetes: Neutron-network services split out into discrete pods https://review.openstack.org/350260 | 20:43 |
| *** YefimG has quit IRC | 20:44 | |
| *** HyperJohnGraham has quit IRC | 20:48 | |
| *** bootsha has quit IRC | 20:48 | |
| *** fragatina has joined #openstack-kolla | 20:50 | |
| *** fragatina has quit IRC | 20:50 | |
| *** fragatina has joined #openstack-kolla | 20:51 | |
| *** kbyrne has quit IRC | 20:58 | |
| *** thumpba_ has quit IRC | 21:01 | |
| *** ppowell has quit IRC | 21:01 | |
| *** HyperJohnGraham has joined #openstack-kolla | 21:03 | |
| *** banix has quit IRC | 21:06 | |
| *** mbound has joined #openstack-kolla | 21:11 | |
| *** jtriley has joined #openstack-kolla | 21:14 | |
| *** kbyrne has joined #openstack-kolla | 21:19 | |
| *** HyperJohnGraham has quit IRC | 21:20 | |
| *** HyperJohnGraham has joined #openstack-kolla | 21:20 | |
| *** rhallisey has quit IRC | 21:22 | |
| *** zhiwei has joined #openstack-kolla | 21:23 | |
| *** signed8bit is now known as signed8bit_Zzz | 21:27 | |
| *** zhiwei has quit IRC | 21:28 | |
| *** signed8bit_Zzz is now known as signed8bit | 21:28 | |
| openstackgerrit | David Wang proposed openstack/kolla-kubernetes: Fix missing dependency python-openstackclient. https://review.openstack.org/350278 | 21:28 |
| *** jtriley has quit IRC | 21:37 | |
| *** fragatina has quit IRC | 21:38 | |
| *** jtriley has joined #openstack-kolla | 21:40 | |
| *** HyperJohnGraham_ has joined #openstack-kolla | 21:41 | |
| *** HyperJohnGraham has quit IRC | 21:42 | |
| *** HyperJohnGraham_ is now known as HyperJohnGraham | 21:42 | |
| *** Serlex has quit IRC | 21:45 | |
| *** nradojevic has joined #openstack-kolla | 21:49 | |
| openstackgerrit | Serguei Bezverkhi proposed openstack/kolla-kubernetes: Glance bootstrap to create service/project/role https://review.openstack.org/346215 | 21:51 |
| openstackgerrit | David Wang proposed openstack/kolla-kubernetes: TrivialFix missing dependency python-openstackclient https://review.openstack.org/350278 | 21:52 |
| *** mark-casey has joined #openstack-kolla | 21:54 | |
| *** nradojevic has quit IRC | 21:55 | |
| *** jtriley has quit IRC | 21:58 | |
| sdake | heads up - ci finished up on my revert on the uca change, and that didn't fix the ubuntu gate | 21:58 |
| sdake | so i abandoned the change, but atleast that proved that speculation is incorrect | 21:58 |
| *** HyperJohnGraham has quit IRC | 22:05 | |
| *** jtriley has joined #openstack-kolla | 22:05 | |
| openstackgerrit | Ken Wronkiewicz proposed openstack/kolla-kubernetes: 3/4 of the Neutron Compute services start correctly https://review.openstack.org/350285 | 22:05 |
| *** inc0 has quit IRC | 22:06 | |
| openstackgerrit | sean mooney proposed openstack/kolla: add deploy-servers command https://review.openstack.org/349729 | 22:09 |
| *** HyperJohnGraham has joined #openstack-kolla | 22:14 | |
| *** daneyon_ has joined #openstack-kolla | 22:18 | |
| *** HyperJohnGraham has quit IRC | 22:19 | |
| *** jtriley has quit IRC | 22:20 | |
| *** daneyon_ has quit IRC | 22:22 | |
| *** mark-casey has quit IRC | 22:26 | |
| *** jtriley has joined #openstack-kolla | 22:26 | |
| *** HyperJohnGraham has joined #openstack-kolla | 22:26 | |
| wirehead_ | sdake: https://blueprints.launchpad.net/kolla-kubernetes/+spec/neutron-kubernetes | 22:30 |
| *** HyperJohnGraham has quit IRC | 22:38 | |
| *** sdake has quit IRC | 22:40 | |
| *** mbound has quit IRC | 22:41 | |
| *** jtriley has quit IRC | 22:44 | |
| *** mbound has joined #openstack-kolla | 22:49 | |
| *** david-lyle has quit IRC | 22:50 | |
| *** david-lyle has joined #openstack-kolla | 22:51 | |
| *** michauds has quit IRC | 22:52 | |
| *** mbound has quit IRC | 22:54 | |
| *** banix has joined #openstack-kolla | 22:59 | |
| godleon | does kolla-kubernetes use the docker images built by kolla ? | 23:01 |
| *** gfidente has joined #openstack-kolla | 23:01 | |
| *** dave-mcc_ has quit IRC | 23:02 | |
| wirehead_ | godleon: yep. | 23:06 |
| *** sdake has joined #openstack-kolla | 23:15 | |
| *** sdake_ has joined #openstack-kolla | 23:17 | |
| *** sdake has quit IRC | 23:20 | |
| sdake_ | hey bmace how you been | 23:20 |
| *** sdake_ is now known as sdake | 23:21 | |
| openstackgerrit | David Wang proposed openstack/kolla-kubernetes: Add Ceph support for Mariadb https://review.openstack.org/350312 | 23:29 |
| *** signed8bit is now known as signed8bit_Zzz | 23:31 | |
| bmace | hey sdake_. very good thanks. you? | 23:35 |
| sdake | rockin | 23:35 |
| sdake | only problem is our review queue is 10+ pages long | 23:36 |
| sdake | but there are worse problems to be had :) | 23:36 |
| bmace | certainly sounds like a growing problem. that mostly due to test time? rejected reviews? insufficient reviewers? a mix perhaps? i find the happiest developer is the one that can commit / iterate at their own speed. | 23:37 |
| sdake | mix | 23:38 |
| sdake | plus input has doubled or more | 23:38 |
| sdake | mostly in the last month | 23:40 |
| bmace | sadly more input has less of an impact if it takes forever to get committed :). less input with more time reviewing may be necessary. | 23:40 |
| sdake | its not the core team doing most of hte input | 23:41 |
| bmace | good to hear you have lots of new interest. | 23:41 |
| sdake | i could ask the core team to slow down utni lwe were caught up if that wer ethe case | 23:41 |
| *** inc0 has joined #openstack-kolla | 23:41 | |
| sdake | yo inc0 | 23:42 |
| inc0 | yo | 23:42 |
| *** berendt has quit IRC | 23:49 | |
| *** jtriley has joined #openstack-kolla | 23:53 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!