| *** raddaoui__ has quit IRC | 00:01 | |
| *** raddaoui_ has quit IRC | 00:01 | |
| *** asettle has joined #openstack-ansible | 00:04 | |
| *** asettle has quit IRC | 00:09 | |
| *** raddaoui_ has joined #openstack-ansible | 00:13 | |
| *** raddaoui has joined #openstack-ansible | 00:13 | |
| *** weezS has quit IRC | 00:14 | |
| *** baker has joined #openstack-ansible | 00:15 | |
| *** jaypipes has quit IRC | 00:20 | |
| *** weezS has joined #openstack-ansible | 00:22 | |
| *** baker has quit IRC | 00:24 | |
| *** baker has joined #openstack-ansible | 00:25 | |
| *** woodard has joined #openstack-ansible | 00:25 | |
| *** galstrom is now known as galstrom_zzz | 00:28 | |
| *** baker has quit IRC | 00:35 | |
| *** baker has joined #openstack-ansible | 00:37 | |
| *** baker has quit IRC | 00:48 | |
| *** sdake_ has joined #openstack-ansible | 00:49 | |
| *** sdake_ has quit IRC | 00:49 | |
| *** sdake_ has joined #openstack-ansible | 00:49 | |
| *** sdake has quit IRC | 00:51 | |
| openstackgerrit | Merged openstack/openstack-ansible-memcached_server: Only update apt cache if necessary https://review.openstack.org/280234 | 00:52 |
|---|---|---|
| openstackgerrit | Hector I Gonzalez Mendoza proposed openstack/openstack-ansible: Backport: Store openstack_host_sysstat_enabled as lower case https://review.openstack.org/283337 | 00:53 |
| *** weezS has quit IRC | 00:54 | |
| *** baker has joined #openstack-ansible | 00:54 | |
| *** sdake_ has quit IRC | 00:56 | |
| *** eil397 has quit IRC | 01:05 | |
| *** baker has quit IRC | 01:14 | |
| *** baker has joined #openstack-ansible | 01:17 | |
| *** baker has quit IRC | 01:21 | |
| *** baker has joined #openstack-ansible | 01:24 | |
| *** baker has quit IRC | 01:26 | |
| *** nn_brian has joined #openstack-ansible | 01:32 | |
| *** nn_brian has quit IRC | 01:33 | |
| *** nn_brian has joined #openstack-ansible | 01:34 | |
| *** baker has joined #openstack-ansible | 01:37 | |
| *** nn_brian has quit IRC | 01:38 | |
| *** kencjohnston has joined #openstack-ansible | 01:41 | |
| *** baker has quit IRC | 01:44 | |
| *** baker has joined #openstack-ansible | 01:46 | |
| *** baker has quit IRC | 01:54 | |
| *** baker has joined #openstack-ansible | 01:55 | |
| *** sdake has joined #openstack-ansible | 01:57 | |
| *** Bjoern_ has joined #openstack-ansible | 01:58 | |
| *** baker has quit IRC | 02:00 | |
| *** sdake has quit IRC | 02:04 | |
| *** kencjohnston has quit IRC | 02:05 | |
| *** sdake has joined #openstack-ansible | 02:07 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible: Clarify user config loading in dynamic inventory https://review.openstack.org/283313 | 02:10 |
| *** rohanp_ has quit IRC | 02:20 | |
| openstackgerrit | Merged openstack/openstack-ansible: DOCS: Correct location of variable overrides https://review.openstack.org/282426 | 02:23 |
| openstackgerrit | Merged openstack/openstack-ansible: Docs: Keystone SSL configuration https://review.openstack.org/281288 | 02:23 |
| *** nn_brian has joined #openstack-ansible | 02:24 | |
| *** galstrom_zzz is now known as galstrom | 02:31 | |
| *** sdake has quit IRC | 02:38 | |
| *** persia has quit IRC | 02:38 | |
| *** sdake has joined #openstack-ansible | 02:39 | |
| *** sdake has quit IRC | 02:41 | |
| *** ravikamachi has quit IRC | 02:41 | |
| *** persia has joined #openstack-ansible | 02:42 | |
| *** gfa is now known as gfa_ | 02:44 | |
| *** phalmos has joined #openstack-ansible | 02:44 | |
| *** gfa_ is now known as gfa | 02:44 | |
| *** phalmos has quit IRC | 02:45 | |
| *** phalmos has joined #openstack-ansible | 02:45 | |
| *** nn_brian has quit IRC | 02:48 | |
| *** phalmos has quit IRC | 02:50 | |
| *** baker has joined #openstack-ansible | 02:52 | |
| *** fawadkhaliq has joined #openstack-ansible | 02:52 | |
| *** galstrom is now known as galstrom_zzz | 02:55 | |
| *** ShannonM has joined #openstack-ansible | 02:57 | |
| *** TheIntern has joined #openstack-ansible | 02:58 | |
| *** Bjoern_ has quit IRC | 03:05 | |
| *** phalmos has joined #openstack-ansible | 03:06 | |
| *** ShannonM has quit IRC | 03:24 | |
| *** fawadkhaliq has quit IRC | 03:40 | |
| *** Bjoern_ has joined #openstack-ansible | 03:42 | |
| *** TheIntern has quit IRC | 03:43 | |
| *** fawadkhaliq has joined #openstack-ansible | 03:44 | |
| *** fawadkhaliq has quit IRC | 03:44 | |
| *** fawadkhaliq has joined #openstack-ansible | 03:45 | |
| *** fawadkhaliq has quit IRC | 03:45 | |
| *** nn_brian has joined #openstack-ansible | 03:46 | |
| *** nn_brian has quit IRC | 03:48 | |
| *** nn_brian has joined #openstack-ansible | 03:48 | |
| *** sdake has joined #openstack-ansible | 04:00 | |
| prometheanfire | uhh, can we run nova in a container now? | 04:03 |
| prometheanfire | tgt (1:1.0.63-1) unstable; urgency=medium | 04:03 |
| prometheanfire | * New upstream release: | 04:03 |
| prometheanfire | + Do not fail on EACCESS on oom_adjust; tgtd can now run in a different | 04:03 |
| prometheanfire | user namespace (Closes: #813368). | 04:04 |
| prometheanfire | that's user namespace at least | 04:04 |
| prometheanfire | probably not netns | 04:04 |
| prometheanfire | got excited there for a sec :P | 04:05 |
| *** shausy has joined #openstack-ansible | 04:15 | |
| *** baker_ has joined #openstack-ansible | 04:17 | |
| *** baker has quit IRC | 04:20 | |
| *** baker has joined #openstack-ansible | 04:26 | |
| *** asettle has joined #openstack-ansible | 04:27 | |
| *** sdake has quit IRC | 04:28 | |
| *** baker_ has quit IRC | 04:29 | |
| *** asettle has quit IRC | 04:32 | |
| *** nn_brian has quit IRC | 04:36 | |
| *** baker has quit IRC | 04:42 | |
| *** shausy has quit IRC | 05:07 | |
| *** shausy has joined #openstack-ansible | 05:07 | |
| openstackgerrit | Michael Davies proposed openstack/openstack-ansible-ironic: Fix tests for os_ironic https://review.openstack.org/283384 | 05:08 |
| *** javeriak has joined #openstack-ansible | 05:12 | |
| *** javeriak has quit IRC | 05:13 | |
| *** javeriak_ has joined #openstack-ansible | 05:13 | |
| *** woodard has quit IRC | 05:18 | |
| *** javeriak_ has quit IRC | 05:33 | |
| *** sdake has joined #openstack-ansible | 05:41 | |
| *** swati has joined #openstack-ansible | 05:46 | |
| swati | HI , I am working on ansible liberty | 05:46 |
| swati | I have some query regarding getting the services of DNS component up on Horizon...anybody available? | 05:47 |
| cloudnull | prometheanfire: whats going ? | 05:49 |
| cloudnull | swati: what are you looking for ? | 05:50 |
| swati | Hi <cloudnull>....like if I have to enable designate functionality on "Devstack" Horizon, I enable those services in local.conf file... http://docs.openstack.org/developer/designate/devstack.html | 05:51 |
| swati | This is new in Liberty | 05:51 |
| swati | In ansible also, I have setup designate and its services are also up.......Now if I have to get its Horizon part working...in which file should I enable these services? | 05:52 |
| cloudnull | you'd have to edit the horizon local_config.py to enable the designate pane | 05:52 |
| cloudnull | ah looks like the desegnate panel is not in horizon | 05:54 |
| cloudnull | you need | 05:54 |
| cloudnull | https://github.com/openstack/designate-dashboard | 05:54 |
| cloudnull | specifically. https://github.com/openstack/designate-dashboard/tree/stable/liberty | 05:54 |
| cloudnull | so to do the needful, you need to install that package in the venv and add the required config options | 05:55 |
| prometheanfire | cloudnull: I maintain tgt and was looking at the changelog, looks like it got some namespace support | 05:56 |
| prometheanfire | not netns though | 05:56 |
| prometheanfire | ust a userns bugfix | 05:57 |
| swati | Thanks <cloudnull>... il check this now | 05:57 |
| *** sdake has quit IRC | 05:57 | |
| *** javeriak has joined #openstack-ansible | 05:57 | |
| swati | Kevin_ should I install designate_dashboard in venv of designate_container and mention if in the local_config.py file of horizon container? | 05:59 |
| cloudnull | yes | 06:02 |
| cloudnull | once you activate it | 06:02 |
| cloudnull | you can do pip install git+https://github.com/openstack/designate-dashboard@stable/liberty | 06:02 |
| cloudnull | which will install it . | 06:03 |
| cloudnull | then all you should have to do is step 2 from -- https://github.com/openstack/designate-dashboard/tree/stable/liberty#howto | 06:03 |
| cloudnull | prometheanfire: ah. | 06:07 |
| *** izaakk has quit IRC | 06:17 | |
| *** fawadkhaliq has joined #openstack-ansible | 06:18 | |
| *** woodard has joined #openstack-ansible | 06:19 | |
| *** raddaoui_ has quit IRC | 06:22 | |
| *** raddaoui has quit IRC | 06:22 | |
| *** woodard has quit IRC | 06:24 | |
| *** mariusv has quit IRC | 06:25 | |
| *** phiche has joined #openstack-ansible | 06:25 | |
| *** mariusv has joined #openstack-ansible | 06:32 | |
| *** mariusv is now known as Guest98039 | 06:32 | |
| *** javeriak_ has joined #openstack-ansible | 06:35 | |
| *** javeriak has quit IRC | 06:35 | |
| *** javeriak has joined #openstack-ansible | 06:38 | |
| *** javeriak_ has quit IRC | 06:41 | |
| swati | Kevin_ is it necessary to include venv files in all openstack components? | 06:45 |
| swati | for Liberty? | 06:45 |
| swati | as initially, I had included a component in Kilo without venv ....n it was working.....but I guess some change like this has been made in Liberty release...can u pls confirm? | 06:46 |
| *** asettle has joined #openstack-ansible | 06:51 | |
| *** Guest98039 has quit IRC | 06:53 | |
| *** marius- has joined #openstack-ansible | 06:53 | |
| *** marius- has quit IRC | 06:54 | |
| *** severion has quit IRC | 06:54 | |
| *** severion has joined #openstack-ansible | 06:55 | |
| *** asettle has quit IRC | 06:56 | |
| *** phiche has quit IRC | 06:57 | |
| *** phiche has joined #openstack-ansible | 07:07 | |
| *** fawadkhaliq has quit IRC | 07:11 | |
| *** cemmason has joined #openstack-ansible | 07:16 | |
| *** javeriak_ has joined #openstack-ansible | 07:19 | |
| *** cemmason has quit IRC | 07:19 | |
| *** cemmason1 has joined #openstack-ansible | 07:19 | |
| *** woodard has joined #openstack-ansible | 07:20 | |
| *** javeriak has quit IRC | 07:21 | |
| *** cemmason has joined #openstack-ansible | 07:22 | |
| *** cemmason1 has quit IRC | 07:22 | |
| *** cemmason1 has joined #openstack-ansible | 07:25 | |
| *** woodard has quit IRC | 07:25 | |
| *** cemmason has quit IRC | 07:27 | |
| *** cemmason1 has quit IRC | 07:32 | |
| *** mrda has quit IRC | 07:32 | |
| *** cemmason has joined #openstack-ansible | 07:32 | |
| *** cemmason1 has joined #openstack-ansible | 07:34 | |
| *** cemmason has quit IRC | 07:34 | |
| *** swati has quit IRC | 07:38 | |
| *** Bjoern_ has quit IRC | 07:42 | |
| *** mikelk has joined #openstack-ansible | 07:43 | |
| *** asettle has joined #openstack-ansible | 07:47 | |
| *** asettle has quit IRC | 07:52 | |
| *** tomoe_ has joined #openstack-ansible | 07:54 | |
| *** admin0 has joined #openstack-ansible | 07:56 | |
| *** admin0 has quit IRC | 07:56 | |
| *** javeriak has joined #openstack-ansible | 07:58 | |
| *** pcaruana has joined #openstack-ansible | 07:58 | |
| *** admin0 has joined #openstack-ansible | 07:59 | |
| *** javeriak_ has quit IRC | 08:02 | |
| *** javeriak has quit IRC | 08:04 | |
| *** javeriak has joined #openstack-ansible | 08:07 | |
| *** pcaruana is now known as pcaruana|afk| | 08:08 | |
| *** pcaruana|afk| is now known as pcaruana | 08:08 | |
| *** asettle has joined #openstack-ansible | 08:10 | |
| *** Oku_OS has quit IRC | 08:13 | |
| *** javeriak has quit IRC | 08:13 | |
| *** asettle has quit IRC | 08:15 | |
| *** mariusv has joined #openstack-ansible | 08:20 | |
| *** woodard has joined #openstack-ansible | 08:22 | |
| *** woodard has quit IRC | 08:26 | |
| *** javeriak has joined #openstack-ansible | 08:29 | |
| *** subscope has joined #openstack-ansible | 08:32 | |
| *** subscope has quit IRC | 08:36 | |
| *** swati has joined #openstack-ansible | 08:38 | |
| swati | Hi...I am getting error "openstack-ansible failed to find required executable rabbitmqctl"....any idea? | 08:43 |
| *** subscope has joined #openstack-ansible | 08:47 | |
| *** swati has quit IRC | 08:48 | |
| *** javeriak has quit IRC | 08:56 | |
| *** javeriak has joined #openstack-ansible | 08:59 | |
| *** evrardjp has joined #openstack-ansible | 09:06 | |
| *** mrda has joined #openstack-ansible | 09:16 | |
| openstackgerrit | Javeria Khan proposed openstack/openstack-ansible: Update PLUMgrid Appendix Doc paths https://review.openstack.org/283465 | 09:16 |
| *** subscope has quit IRC | 09:20 | |
| *** woodard has joined #openstack-ansible | 09:23 | |
| *** woodard has quit IRC | 09:27 | |
| *** eford has joined #openstack-ansible | 09:28 | |
| *** dford has quit IRC | 09:31 | |
| *** phalmos has quit IRC | 09:33 | |
| *** asettle has joined #openstack-ansible | 09:34 | |
| odyssey4me | Bofu2U did you find the cause of the instance<->volume traffic going over br-mgmt instead of br-storage ? | 09:36 |
| *** eford has quit IRC | 09:38 | |
| *** eford has joined #openstack-ansible | 09:39 | |
| *** asettle has quit IRC | 09:42 | |
| *** asettle has joined #openstack-ansible | 09:42 | |
| *** subscope has joined #openstack-ansible | 09:43 | |
| *** mikelk has quit IRC | 09:45 | |
| *** mikelk has joined #openstack-ansible | 09:45 | |
| *** elo has quit IRC | 09:49 | |
| odyssey4me | Bofu2U bgmccollum tiagogomes it seems that you all may have hit the same bug at the same time, as http://paste.openstack.org/show/487757/ and https://review.openstack.org/283234 relate to the same problem, although they are different solutions | 09:50 |
| openstackgerrit | Michael Davies proposed openstack/openstack-ansible-ironic: Fix tests for os_ironic https://review.openstack.org/283384 | 09:52 |
| tiagogomes | Hmm, I can't see how they are related. | 09:53 |
| odyssey4me | rohanp ping? | 09:54 |
| tiagogomes | But coffee didn't kick in yet | 09:54 |
| odyssey4me | tiagogomes so is mine, so I might be completely off :) | 09:55 |
| odyssey4me | mrda I was going to aubmit an updated tox/test config for the ironic repo today... good to see you're on it :) | 09:56 |
| *** subscope has quit IRC | 09:57 | |
| *** cemmason1 has quit IRC | 09:57 | |
| mrda | well, I'm about to give up, so feel free to continue while I sleep. | 09:57 |
| odyssey4me | mrda you may wish to take a look at https://github.com/openstack/openstack-ansible-galera_server for the updated tox config, and updated test configuration... | 09:57 |
| tiagogomes | odyssey4me one is for neutron, another for cinder | 09:57 |
| odyssey4me | ah ok, I'll take it over :) | 09:57 |
| mrda | Feel free to push a new revision to https://review.openstack.org/#/c/283384 | 09:57 |
| *** cemmason has joined #openstack-ansible | 09:57 | |
| mrda | ...which is just to get things running | 09:57 |
| odyssey4me | tiagogomes sure, but I think they related to the same problem - finding the right IP to push traffic over, right? | 09:58 |
| tiagogomes | ok same problem, but different bugs :) | 09:59 |
| * mrda heads off for the night, especially now odyssey4me can beat tox into submission on my behalf :) | 09:59 | |
| odyssey4me | tiagogomes but if you check my comments, I think they related to the same mechanism which originated from the same patch | 10:00 |
| *** cemmason1 has joined #openstack-ansible | 10:02 | |
| *** subscope has joined #openstack-ansible | 10:02 | |
| *** cemmason has quit IRC | 10:02 | |
| odyssey4me | tiagogomes ah, I see that the Neutron local_ip fact determination was done at a different time. See this for the Nova one: https://github.com/openstack/openstack-ansible/commit/c506dda61d303735189abbee7c70d972b3096a40 | 10:03 |
| *** eford has quit IRC | 10:03 | |
| *** eford has joined #openstack-ansible | 10:03 | |
| *** fawadkhaliq has joined #openstack-ansible | 10:03 | |
| odyssey4me | javeriak good morning :) I had hoped you'd be able to make the mid cycle either in person or remotely, but alas it didn't happen :( | 10:05 |
| *** cemmason has joined #openstack-ansible | 10:05 | |
| odyssey4me | javeriak you may wish to be rid of https://github.com/plumgrid/plumgrid-ansible/blob/master/plumgrid_playbooks/ansible.cfg#L28 as it causes connectivity issues | 10:06 |
| *** cemmason1 has quit IRC | 10:06 | |
| odyssey4me | javeriak see https://github.com/openstack/openstack-ansible/commit/c7ce0982ff0c5e357fe957c9f929a9bf71deed7c#diff-c65aaeaf20dfc7d75fca73c229cb9e85 | 10:06 |
| *** javeriak has quit IRC | 10:08 | |
| *** cemmason has quit IRC | 10:09 | |
| *** asettle has quit IRC | 10:09 | |
| *** cemmason has joined #openstack-ansible | 10:09 | |
| *** asettle has joined #openstack-ansible | 10:09 | |
| odyssey4me | tiagogomes bgmccollum tiagogomes I've added more comments to https://review.openstack.org/283234 to clarify. Hopefully that helps with further digging. | 10:11 |
| tiagogomes | Seeing this: "_overlay_network: "{{ container_networks.tunnel_address|default({}) }}" , I grepped for tunnel_address but didn't find anything useful | 10:11 |
| *** cemmason has quit IRC | 10:13 | |
| *** cemmason1 has joined #openstack-ansible | 10:13 | |
| *** swati has joined #openstack-ansible | 10:13 | |
| *** fawadkhaliq has quit IRC | 10:14 | |
| swati | Hi....I am facing a similar error for Designate in Liberty : https://bugs.launchpad.net/openstack-ansible/+bug/1527809 | 10:14 |
| openstack | Launchpad bug 1527809 in openstack-ansible "repo_build - create role based venv role in liberty branch fails" [Low,Fix released] - Assigned to Kevin Carter (kevin-carter) | 10:14 |
| *** fawadkhaliq has joined #openstack-ansible | 10:14 | |
| *** cemmason has joined #openstack-ansible | 10:16 | |
| odyssey4me | Hi swati - that's a pretty old bug. What's the issue you're facing? | 10:17 |
| *** cemmason1 has quit IRC | 10:17 | |
| *** electrofelix has joined #openstack-ansible | 10:18 | |
| *** cemmason1 has joined #openstack-ansible | 10:20 | |
| *** cemmason has quit IRC | 10:20 | |
| odyssey4me | swati with regards to the venvs, you would not have to implement venvs for a new role - that could be added later... I have answered your question several times before | 10:20 |
| odyssey4me | I'm not quite sure why you keep going back to that. Your role was close to being in a ready state! | 10:21 |
| odyssey4me | swati I asked before whether you'd be happy to pair up with someone to help you finalise the role? | 10:21 |
| odyssey4me | I haven't yet received an answer, and would appreciate it if you could answer. | 10:22 |
| swati | failed: [infr002-clddev_repo_container-32598725] => (item={'key': 'os_designate', 'value': ['MySQL-python', 'designate', 'httplib2', 'pycrypto', 'python-designateclient', 'python-keystoneclient', 'python-memcached', 'virtualenv', 'virtualenv-tools', 'warlock']}) => {"attempts": 5, "cmd": "/tmp/openstack-venv-builder/venvs/designate/bin/pip install --timeout 120 --find-links /var/www/repo/os-releases/12.0.5 --no-index --log | 10:22 |
| swati | This is the error I am getting while running setup-infrastructure.ymkl. | 10:22 |
| swati | I have added vens in playbook only now..for liberty release.. | 10:23 |
| *** woodard has joined #openstack-ansible | 10:23 | |
| odyssey4me | swati ok, so that's telling you that there's an issue with building the venv... so the question is which of the packages did it actually fail on | 10:23 |
| odyssey4me | do you have a -vv log that you can pastebin ? | 10:24 |
| odyssey4me | and you'll only need to run the repo-build playbook | 10:24 |
| *** cemmason1 has quit IRC | 10:24 | |
| *** cemmason has joined #openstack-ansible | 10:24 | |
| swati | it is failing while running repo_build ( create role based venv) | 10:25 |
| odyssey4me | swati are you still pushing all your work to https://github.com/sharmaswati6/designate_files ? | 10:25 |
| swati | k..i ll paste the logs.. | 10:25 |
| odyssey4me | swati what time zone do you work in? is it morning/afternoon/evening for you now? | 10:26 |
| swati | no..i am not..i ll push the changes when the playbook will run without any error.. | 10:26 |
| *** woodard has quit IRC | 10:28 | |
| odyssey4me | swati ah, I thought it'd be easier to help diagnose if I could see where your work currently is | 10:28 |
| *** cemmason1 has joined #openstack-ansible | 10:29 | |
| swati | the logs for repo_venv_builder.log are showing the names of the packages which have been successfully installed. There is no error logged . | 10:29 |
| *** cemmason has quit IRC | 10:29 | |
| swati | I will try to update the files as soon as possible. | 10:30 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Store openstack_host_sysstat_enabled as lower case value for sysstat.default.j2 https://review.openstack.org/283337 | 10:31 |
| *** cemmason has joined #openstack-ansible | 10:33 | |
| *** cemmason1 has quit IRC | 10:34 | |
| *** cemmason has quit IRC | 10:38 | |
| *** cemmason has joined #openstack-ansible | 10:39 | |
| *** eford has quit IRC | 10:41 | |
| *** cemmason1 has joined #openstack-ansible | 10:42 | |
| *** cemmason has quit IRC | 10:42 | |
| *** cemmason1 has quit IRC | 10:44 | |
| *** cemmason has joined #openstack-ansible | 10:44 | |
| *** cemmason1 has joined #openstack-ansible | 10:46 | |
| *** cemmason has quit IRC | 10:46 | |
| openstackgerrit | Merged openstack/openstack-ansible-openstack_hosts: Only update apt cache if necessary https://review.openstack.org/280235 | 10:46 |
| *** cemmason1 has quit IRC | 10:48 | |
| *** cemmason has joined #openstack-ansible | 10:49 | |
| *** cemmason has quit IRC | 10:51 | |
| *** cemmason has joined #openstack-ansible | 10:52 | |
| *** Oku_OS has joined #openstack-ansible | 10:52 | |
| *** cemmason1 has joined #openstack-ansible | 10:55 | |
| *** cemmason has quit IRC | 10:55 | |
| *** cemmason has joined #openstack-ansible | 10:59 | |
| *** cemmason1 has quit IRC | 11:00 | |
| *** cemmason has quit IRC | 11:02 | |
| *** cemmason has joined #openstack-ansible | 11:02 | |
| *** fawadkhaliq has quit IRC | 11:03 | |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible: [WIP] Automate setting of glance_api_version https://review.openstack.org/283508 | 11:03 |
| *** cemmason1 has joined #openstack-ansible | 11:06 | |
| *** cemmason has quit IRC | 11:06 | |
| *** subscope has quit IRC | 11:21 | |
| *** woodard has joined #openstack-ansible | 11:24 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ironic: Fix tests for os_ironic https://review.openstack.org/283384 | 11:28 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ironic: Fix tests for os_ironic https://review.openstack.org/283384 | 11:28 |
| *** woodard has quit IRC | 11:29 | |
| *** jaypipes has joined #openstack-ansible | 11:32 | |
| mattt | odyssey4me: so how should we proceed with https://review.openstack.org/#/c/283039/ ? shall we pin more stuff ? | 11:34 |
| *** javeriak has joined #openstack-ansible | 11:35 | |
| javeriak | hey odyssey4me, yea sorry to have missed it :(; things got super busy at work; havent even been able to be much active here as well | 11:37 |
| javeriak | odyssey4me thanks for that tip; will take it out | 11:38 |
| odyssey4me | mattt good question, have you had another chat in that regard in #pypa or with sigmavirus24_awa about that? | 11:38 |
| mattt | odyssey4me: no, it was your suggestion :) | 11:39 |
| mattt | odyssey4me: we can speculate all day but i guess the question is do we want to unblock kilo or not? | 11:39 |
| *** mancdaz has joined #openstack-ansible | 11:42 | |
| odyssey4me | mattt sure, but I think that this issue applies to more than just kilo | 11:42 |
| mattt | odyssey4me: the wheel issue affects everything indirectly | 11:43 |
| mattt | odyssey4me: it only breaks kilo build, but logan- was having issues with additional packages he was building on liberty | 11:43 |
| odyssey4me | so whatever we do, I'd like it to fit wherever it needs to go | 11:43 |
| odyssey4me | mattt kevin also hit issues which relate a little to this: https://review.openstack.org/279683 | 11:43 |
| odyssey4me | cloudnull ^ | 11:44 |
| mattt | any plans to use pip 8 in master? | 11:45 |
| mattt | feels a bit wrong to pin it there | 11:45 |
| mattt | (wheel, that is) | 11:45 |
| odyssey4me | yeah, I'm kinda thinking that perhaps master should move its pin to >8,<9 | 11:51 |
| odyssey4me | I think we should discuss this in the Thu meeting. | 11:51 |
| * odyssey4me heads over to add it to the agenda. | 11:51 | |
| mattt | odyssey4me: i WIP'd https://review.openstack.org/#/c/283039 until we can discuss it further | 11:52 |
| *** mancdaz has quit IRC | 11:56 | |
| *** subscope has joined #openstack-ansible | 11:58 | |
| *** mancdaz has joined #openstack-ansible | 11:58 | |
| odyssey4me | mattt kicking of a query in #pypa-dev, maybe we can resolve this sooner | 12:00 |
| mattt | odyssey4me: what precisely did you ask? | 12:00 |
| odyssey4me | if we're in a situation where we need to pin pip to a major version (eg: >7,<8) then what companion packages should also be pinned? it seems that setuptools and wheel are closely related and should be on that list | 12:01 |
| mattt | k | 12:01 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ironic: Fix tests for os_ironic https://review.openstack.org/283384 | 12:02 |
| mattt | odyssey4me: #pypa is quite active, could ask in there if no response in -dev | 12:04 |
| odyssey4me | thanks mattt - I've posed the same question there | 12:05 |
| odyssey4me | I know that pypa-dev becomes active once the dev's are around, which is mostly US timezones as far as I've seen | 12:06 |
| openstackgerrit | Merged openstack/openstack-ansible: Removing unused galera_lb_address var from galera playbook https://review.openstack.org/283227 | 12:18 |
| odyssey4me | mattt did you try using wheel v0.27 / 0.28 instead of just 0.26? how did you determine the cap? | 12:19 |
| odyssey4me | mattt / hughsaunders are any RPC builds working that are using the OSA 11.2.10 tag for rpc-repo? | 12:22 |
| odyssey4me | ie is a repo-clone working, rather than a repo-build? | 12:23 |
| odyssey4me | actually, we have a WIP review in for that from mancdaz - let's retry that | 12:23 |
| *** javeriak has quit IRC | 12:23 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: [WIP] [DO NOT MERGE] testing review 278343 https://review.openstack.org/278981 | 12:25 |
| odyssey4me | mattt according to the last test of https://review.openstack.org/278981 the current versions in http://rpc-repo.rackspace.com/os-releases/11.2.10/ may actually be fine - re-running a fresh test to confirm | 12:26 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ironic: Fix tests for os_ironic https://review.openstack.org/283384 | 12:29 |
| *** Nepoc has quit IRC | 12:34 | |
| mattt | odyssey4me: i went w/ what logan- suggested, and also based on the changelog on https://pypi.python.org/pypi/wheel | 12:35 |
| mattt | odyssey4me: the bulk of the changes were introduced in 0.27.0, so it seemed sensible to cap at the previous version | 12:35 |
| odyssey4me | yeah, that does seem sensible - although it seems that the more up to date wheel version may be ok, it might actually be setuptools that's an issue | 12:36 |
| odyssey4me | let's see how the test works out | 12:36 |
| mattt | odyssey4me: dstufft seemed familiar w/ the issue we were hitting, and he said either pin wheel or bump pip to >= 8 | 12:37 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Remove MySQL-python from Aodh pip requirements https://review.openstack.org/277339 | 12:41 |
| mgariepy | good morning everyone. | 12:47 |
| *** gparaskevas has joined #openstack-ansible | 12:51 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ironic: Fix tests for os_ironic https://review.openstack.org/283384 | 12:51 |
| bgmccollum | odyssey4me tiagogomes regarding cinder storage network...it doesn't even look like those facts are being used after being set. i think all that is needed is dropping the ` | default()` bits | 12:53 |
| odyssey4me | bgmccollum perhaps, I'd like cloudnull to weigh in as he did those patches and has more depth of understanding into the underlying purpose | 12:54 |
| bgmccollum | odyssey4me i understand now what its doing...its trying to discover a suitable storage network IP...which it looks to be doing correctly, but that fact isnt used anywhere after being discovered... | 12:55 |
| bgmccollum | also, the new role var allows splitting management type traffic (rabbit, db) from dataplane traffic (iscsi)... | 12:55 |
| bgmccollum | but yes, we'll wait for cloudnull to comment | 12:56 |
| odyssey4me | andymccr are you around? I'd like to try and fathom how mrda wanted a functional test in the gate to work... it looks like we're missing a few elements here | 12:59 |
| odyssey4me | (for the ironic role) | 12:59 |
| mattt | odyssey4me: he's afk at the minute | 13:00 |
| *** woodard has joined #openstack-ansible | 13:00 | |
| *** woodard has quit IRC | 13:01 | |
| *** woodard has joined #openstack-ansible | 13:02 | |
| *** johnmilton has joined #openstack-ansible | 13:03 | |
| *** subscope has quit IRC | 13:05 | |
| *** asettle has quit IRC | 13:07 | |
| *** shausy has quit IRC | 13:16 | |
| tiagogomes | is cloudnull in holiday or different timezone? | 13:21 |
| *** subscope has joined #openstack-ansible | 13:22 | |
| odyssey4me | tiagogomes he's from the US, so different time zone - he's been ill lately though, so I'm not sure if he'll be around today | 13:23 |
| tiagogomes | ok | 13:23 |
| odyssey4me | tiagogomes bgmccollum the best may be to figure out a solution which you think works in accordance with the intent, then we can continue discussion in review | 13:24 |
| *** jaypipes has quit IRC | 13:24 | |
| bgmccollum | odyssey4me adding my notes in review right now | 13:24 |
| *** jwitk0 has quit IRC | 13:29 | |
| bgmccollum | odyssey4me some clarification notes added | 13:30 |
| *** subscope has quit IRC | 13:32 | |
| openstackgerrit | Byron McCollum proposed openstack/openstack-ansible: Cinder should use the storage network for iscsi traffic if available https://review.openstack.org/283234 | 13:32 |
| *** cemmason has joined #openstack-ansible | 13:33 | |
| *** cemmason1 has quit IRC | 13:34 | |
| mattt | logan- mgariepy : added you both to https://review.openstack.org/#/c/283508/, can you have a peek when you're around? | 13:35 |
| *** cemmason has quit IRC | 13:37 | |
| *** cemmason has joined #openstack-ansible | 13:37 | |
| mgariepy | looking at it now ;) | 13:37 |
| *** Nepoc has joined #openstack-ansible | 13:40 | |
| *** cemmason1 has joined #openstack-ansible | 13:41 | |
| *** cemmason has quit IRC | 13:41 | |
| *** subscope has joined #openstack-ansible | 13:45 | |
| *** cemmason1 has quit IRC | 13:45 | |
| *** cemmason has joined #openstack-ansible | 13:45 | |
| mattt | mgariepy: tested on kilo, worked OK ... just waiting for this gate to pass | 13:48 |
| *** tiagogomes has quit IRC | 13:50 | |
| *** cemmason has quit IRC | 13:50 | |
| *** cemmason has joined #openstack-ansible | 13:50 | |
| *** klamath has joined #openstack-ansible | 13:51 | |
| *** automagically has left #openstack-ansible | 13:52 | |
| *** automagically has joined #openstack-ansible | 13:52 | |
| *** asettle has joined #openstack-ansible | 13:52 | |
| andymccr | odyssey4me: im not sure - i will follow up with mrda tonight. | 13:55 |
| *** subscope has quit IRC | 13:55 | |
| openstackgerrit | Travis Truman (automagically) proposed openstack/openstack-ansible: Clarify user config loading in dynamic inventory https://review.openstack.org/283313 | 13:56 |
| *** asettle has quit IRC | 13:56 | |
| *** cemmason has quit IRC | 13:56 | |
| *** cemmason has joined #openstack-ansible | 13:56 | |
| *** asettle has joined #openstack-ansible | 13:57 | |
| odyssey4me | andymccr I'm taking the path of what I hope to be least resistance which is to have the initial functional test be a standalone install. | 13:58 |
| odyssey4me | andymccr I expect thereafter that there should also be a distributed install using containers, but we can add that as a subsequent set. | 13:58 |
| andymccr | odyssey4me: ok fair enough - i think that is the path of least resistance, since i know that works | 13:58 |
| andymccr | im working on neutron integration rn | 13:58 |
| odyssey4me | andymccr I'm kinda thinking that it might be a good idea to just make the funcational test do nothing and pass, then the functional tests can be worked on as subsequent patches - otherwise this is going to end up being a frankenpatch | 13:59 |
| odyssey4me | that at least allows others to start pitching in as well to fix anything found | 14:00 |
| andymccr | thats probably sensible. the only other thing would've been to get everything in in such a way that its tested and working in 1 go. | 14:00 |
| odyssey4me | yeah, I think a fully functional test is going to take a few days considering the quirks of openstack-ci | 14:00 |
| *** cemmason1 has joined #openstack-ansible | 14:00 | |
| *** evrardjp has quit IRC | 14:00 | |
| odyssey4me | ok, let me get that going, then we can move on from there | 14:01 |
| *** cemmason has quit IRC | 14:01 | |
| *** pcaruana has quit IRC | 14:02 | |
| *** subscope has joined #openstack-ansible | 14:04 | |
| *** cemmason1 has quit IRC | 14:06 | |
| *** cemmason has joined #openstack-ansible | 14:06 | |
| *** evrardjp has joined #openstack-ansible | 14:07 | |
| *** b3rnard0_away is now known as b3rnard0 | 14:09 | |
| *** cemmason has quit IRC | 14:09 | |
| *** cemmason has joined #openstack-ansible | 14:09 | |
| logan- | morning | 14:10 |
| logan- | odyssey4me: when i opened the wheel bug, i chose 0.26.0 because of https://github.com/openstack/requirements/blob/stable/liberty/upper-constraints.txt#L334 .. I did not test the versions in between | 14:11 |
| odyssey4me | heh, I got so sucked into the multiple things I'm working on in parallel that I forgot about my coffee | 14:11 |
| odyssey4me | thanks goodness for my keep warm cafetiere :) | 14:13 |
| *** cemmason1 has joined #openstack-ansible | 14:13 | |
| *** cemmason has quit IRC | 14:13 | |
| *** cemmason has joined #openstack-ansible | 14:17 | |
| *** cemmason1 has quit IRC | 14:18 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ironic: Fix docs & linting tests https://review.openstack.org/283384 | 14:18 |
| odyssey4me | andymccr ^ let's see how well that works | 14:19 |
| *** Bjoern_ has joined #openstack-ansible | 14:20 | |
| *** Bjoern_ is now known as Bjoern_zZzZzZzZ | 14:20 | |
| *** weshay has joined #openstack-ansible | 14:20 | |
| *** kencjohnston has joined #openstack-ansible | 14:24 | |
| *** kencjohnston has quit IRC | 14:24 | |
| *** cemmason has quit IRC | 14:25 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ironic: Fix docs & linting tests https://review.openstack.org/283384 | 14:25 |
| *** cemmason has joined #openstack-ansible | 14:25 | |
| *** kencjohnston has joined #openstack-ansible | 14:26 | |
| *** kencjohnston has quit IRC | 14:26 | |
| *** cemmason has quit IRC | 14:31 | |
| *** cemmason1 has joined #openstack-ansible | 14:31 | |
| *** jaypipes has joined #openstack-ansible | 14:32 | |
| *** Bjoern_zZzZzZzZ is now known as Bjoern_ | 14:33 | |
| *** cemmason1 has quit IRC | 14:34 | |
| *** cemmason has joined #openstack-ansible | 14:34 | |
| odyssey4me | lbragstad you've been working with cloudnull on https://github.com/os-cloud/openstack-ansible-os_keystone - I'm taking some of the concepts from there for some other work and have noticed that https://github.com/os-cloud/openstack-ansible-os_keystone/blob/master/tasks/keystone_install_apt.yml#L68-L81 has no corresponding list in the defaults... is that intentional? | 14:35 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ironic: Ironic: Fix docs & linting tests https://review.openstack.org/283384 | 14:36 |
| *** Mudpuppy has joined #openstack-ansible | 14:38 | |
| *** Mudpuppy has quit IRC | 14:38 | |
| *** Mudpuppy has joined #openstack-ansible | 14:38 | |
| *** izaakk has joined #openstack-ansible | 14:43 | |
| *** dmsimard is now known as dmsimard|afk | 14:48 | |
| *** ShannonM has joined #openstack-ansible | 14:50 | |
| odyssey4me | andymccr mrda alright, that gets a gate pass - https://review.openstack.org/283384 - please reviews and provide thoughts | 14:53 |
| *** kencjohnston has joined #openstack-ansible | 14:54 | |
| *** kencjohnston has quit IRC | 14:54 | |
| *** KLevenstein has joined #openstack-ansible | 14:54 | |
| *** sdake has joined #openstack-ansible | 14:55 | |
| openstackgerrit | Byron McCollum proposed openstack/openstack-ansible: Cinder should use the storage network for iscsi traffic if available https://review.openstack.org/283234 | 14:56 |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible: [WIP] Automate setting of glance_api_version https://review.openstack.org/283508 | 14:57 |
| *** kencjohnston has joined #openstack-ansible | 14:57 | |
| *** raddaoui has joined #openstack-ansible | 14:58 | |
| *** raddaoui_ has joined #openstack-ansible | 14:58 | |
| *** evrardjp has quit IRC | 14:58 | |
| *** sigmavirus24_awa is now known as sigmavirus24 | 15:01 | |
| *** raddaoui__ has joined #openstack-ansible | 15:02 | |
| sigmavirus24 | odyssey4me: mattt no one told me about that problem with the wheel package | 15:03 |
| mattt | sigmavirus24: i only saw it yesterday | 15:03 |
| mattt | but there has been an osa bug open about it for some time | 15:04 |
| automagically | TIL about inventory-manage.py. Thanks to whomever wrote it. Very useful little tool | 15:04 |
| odyssey4me | :) morning automagically | 15:04 |
| automagically | morning | 15:04 |
| *** jthorne has joined #openstack-ansible | 15:04 | |
| *** elopez_ has joined #openstack-ansible | 15:04 | |
| mattt | logan-: wonder if i should do this instead: cinder_glance_api_version: "{{ glance_enable_v2_api|bool | ternary('2','1') }}" | 15:05 |
| openstackgerrit | Merged openstack/openstack-ansible: Remove MySQL-python from Aodh pip requirements https://review.openstack.org/277339 | 15:05 |
| odyssey4me | wow, that finally merged | 15:06 |
| odyssey4me | I have no idea why it kept failing. | 15:06 |
| mattt | odyssey4me: i was starting to wonder about that one | 15:06 |
| logan- | mattt: not sure.. that is a lot more impactful than only changing the default cinder behavior to accommodate rbd compatibility.. but also don't know if tehre would be a reason not to use glance v2 | 15:07 |
| *** elopez has quit IRC | 15:07 | |
| *** raddaoui_ has quit IRC | 15:08 | |
| *** raddaoui has quit IRC | 15:08 | |
| odyssey4me | mattt sigmavirus24 logan- so, interestingly - https://review.openstack.org/278981 does a clone from http://rpc-repo.rackspace.com/os-releases/11.2.10/ which includes setuptools 20.0 and wheel 0.29 and it works just fine, at least from a gate test standpoint... so it would seem to me that the issue is not necessarily the wheel version... unless I'm missing something? | 15:08 |
| logan- | the issue for me was 100% wheel, i was building the broken package manually using pip wheel and the only thing i did to fix it was downgrade wheel... let me see what version of setuptools is on the repo | 15:09 |
| mattt | odyssey4me: dunno, it failed for me when i tried doing a repo-build | 15:10 |
| mattt | or rather, it did the repo-build but then another playbook failed due to being unable to install the right wheels | 15:10 |
| odyssey4me | mattt yeah, repo-build is using newer versions of both packages | 15:10 |
| mattt | i downgraded wheel as logan- suggested and it worked | 15:10 |
| *** raddaoui__ has quit IRC | 15:10 | |
| *** subscope has quit IRC | 15:11 | |
| mattt | odyssey4me: so whatever is in rpc-repo isn't what you'd actually get if you did a repo-build :) | 15:11 |
| odyssey4me | sigmavirus24 back to the question at hand - if we are putting constraints of >7,<8 around pip, which packages should be constrained with pip (and with what constraints)? it seems that setuptools and wheel are rather tightly connected | 15:11 |
| odyssey4me | mattt yes, but what's in rpc-repo is a result of a repo-build at a point in time... so it adds data to the problem at hand | 15:12 |
| mattt | yeah so all those wheels were built with wheel 0.26.0 | 15:13 |
| mattt | which is why they're installable | 15:13 |
| *** raddaoui has joined #openstack-ansible | 15:13 | |
| logan- | yep thats my guess | 15:13 |
| *** raddaoui_ has joined #openstack-ansible | 15:13 | |
| logan- | it is placing 0.29.0 in the repo but using 0.26.0 to build | 15:13 |
| *** raddaoui__ has joined #openstack-ansible | 15:13 | |
| odyssey4me | ah, so that's entirely plausible | 15:14 |
| logan- | i am guessing rpc-repo is not building from a fresh repo container every time? | 15:14 |
| odyssey4me | the rpc-repo build server is long lived | 15:14 |
| odyssey4me | nope - it's an instance which remains alive and has done so for a very long time | 15:14 |
| *** swati has quit IRC | 15:15 | |
| logan- | gotcha. yeah i think it is just making sure wheel=present not wheel=latest so this would only affect new installs after 0.29.0 dropped | 15:15 |
| logan- | or rather... new repo container builds after 0.29.0 dropped | 15:16 |
| *** mgoddard__ has joined #openstack-ansible | 15:16 | |
| odyssey4me | alright, so I can totally accept that - but then I am still concerned that we may need to also pin setuptools | 15:17 |
| odyssey4me | can we validate or invalidate that? | 15:17 |
| mattt | odyssey4me: well we don't explicitly install setuptools | 15:17 |
| *** raddaoui_ has quit IRC | 15:18 | |
| *** raddaoui has quit IRC | 15:18 | |
| mattt | so presumably installing wheel will pull in the right version of setuptools? | 15:18 |
| odyssey4me | I suppose the gate test is working with mattt's patch, so it would seem that no setuptools cap is required | 15:18 |
| *** weezS has joined #openstack-ansible | 15:18 | |
| *** raddaoui__ has quit IRC | 15:18 | |
| *** raddaoui has joined #openstack-ansible | 15:18 | |
| *** raddaoui_ has joined #openstack-ansible | 15:18 | |
| *** pcaruana has joined #openstack-ansible | 15:18 | |
| logan- | yeah it does not look like repo does a pip install of setuptools, my container just uses the os setuptools it looks like | 15:18 |
| odyssey4me | logan- mattt so that would mean that https://review.openstack.org/283039 is applicable to the liberty & master branches too, right? | 15:19 |
| odyssey4me | hmm, but why are those two branches working? | 15:20 |
| *** mgoddard_ has quit IRC | 15:20 | |
| odyssey4me | I take it that venvs are making a difference there? | 15:20 |
| mattt | odyssey4me: upper constraints i think | 15:20 |
| mattt | odyssey4me: the package i was having issues w/ is cffi | 15:20 |
| *** pcaruana has quit IRC | 15:20 | |
| odyssey4me | ah, that makes sense | 15:20 |
| mattt | and that version is actually lower in liberty/master iirc | 15:20 |
| *** raddaoui__ has joined #openstack-ansible | 15:21 | |
| odyssey4me | so it's an issue that will come when we next do a sha update | 15:21 |
| *** pcaruana has joined #openstack-ansible | 15:21 | |
| mattt | odyssey4me: yeah i believe it would need to be addressed everywhere | 15:21 |
| logan- | maybe osa should follow openstack/requirements upper constraints for repo pip install? | 15:21 |
| *** pcaruana|afk| has joined #openstack-ansible | 15:21 | |
| logan- | that would make this an issue in master only | 15:21 |
| mattt | logan-: side note, my repo server has another setuptools installed under /usr/local/lib | 15:21 |
| logan- | where you are talking about moving the pip pin | 15:21 |
| odyssey4me | logan- the upper constraints are only available from liberty onwards | 15:21 |
| mattt | but i also have the ubuntu package installed | 15:21 |
| odyssey4me | logan- kilo doesn't have that option as it still follows the only whack-a-mole upper cap model | 15:22 |
| *** cemmason has quit IRC | 15:22 | |
| odyssey4me | logan- so yeah, we do use uppser constraints already - which is why master & liberty don't have an issue | 15:22 |
| odyssey4me | so my suggestion is as follows - for liberty we also ensure that wheel is constrained to a specific version, and for master we shift the pip pin to >8,<9 | 15:23 |
| *** raddaoui__ has quit IRC | 15:23 | |
| *** raddaoui__ has joined #openstack-ansible | 15:24 | |
| logan- | it doesn't seem like the upper constraints are being used in liberty for the repo-install, only repo-build | 15:24 |
| odyssey4me | logan- ah, you're quite correct - that is a good point | 15:24 |
| odyssey4me | considering that the local repo is not available, we'd have to devise a way of doing that | 15:25 |
| odyssey4me | ie pull the appropriate sha's upper constraints from the upstream repo and make use of those constraints when doing the repo-install | 15:26 |
| sigmavirus24 | odyssey4me: so setuptools has no dependency on wheel | 15:26 |
| sigmavirus24 | wheel is something that hooks into setuptools | 15:26 |
| *** admin0 has quit IRC | 15:26 | |
| sigmavirus24 | I would think we should figure out what changed, how, etc. and file a bug https://bitbucket.org/pypa/wheel/issues?status=new&status=open | 15:27 |
| mattt | sigmavirus24: well we figured that all out already :) | 15:28 |
| sigmavirus24 | You should be able to use moderately recent version of setuptools, any version of pip, and any version of wheel that doesn't cause the problem :P | 15:28 |
| sigmavirus24 | setuptools/pip/wheel are pretty loosely coupled in reality | 15:28 |
| sigmavirus24 | even if it doesn't appear that way on the surface | 15:28 |
| odyssey4me | lol | 15:28 |
| *** pcaruana|afk| has quit IRC | 15:29 | |
| odyssey4me | so I think we can more closely define the problem now I guess? | 15:30 |
| odyssey4me | cffi-1.5.2 built by wheel 0.29.0 is not installable by pip 7.x | 15:31 |
| odyssey4me | is that right, mattt? | 15:31 |
| mattt | odyssey4me: that is precisely what i put in the commit message :) | 15:31 |
| hughsaunders | M04R specifics required mattt | 15:32 |
| mattt | wonder if i can put my #pypa transcript in a commit message :) | 15:32 |
| odyssey4me | mattt yep, that's where I got those details from - now I have a far better understanding of all the bits around it | 15:32 |
| mattt | odyssey4me: sweet | 15:33 |
| odyssey4me | and yeah, a link to an eavesdrop log or something would be amazing | 15:33 |
| mattt | perhaps we need to look at a venv for the repo server | 15:33 |
| mattt | which gets sucked in from an external source | 15:33 |
| odyssey4me | sigmavirus24 wheel 0.26 created a wheel called cffi-1.5.2-cp27-none-linux_x86_64.whl and wheel 0.29 created one called cffi-1.5.2-cp27-cp27mu-linux_x86_64.whl - what's the extra bit in the middle signify? | 15:34 |
| andrei_ | hey guys! Is there any way to _not_ copy ceph keyrings from monitor nodes? In current playbook https://github.com/openstack/openstack-ansible/blob/master/playbooks/roles/ceph_client/tasks/ceph_auth.yml#L25-L77 where is connection toward monitors and ceph auth get. I want to create users on ceph, copy them somewhere so OA can read them. | 15:36 |
| *** spotz_zzz is now known as spotz | 15:36 | |
| *** mgugino has joined #openstack-ansible | 15:36 | |
| *** pcaruana has quit IRC | 15:37 | |
| *** mgoddard__ has quit IRC | 15:37 | |
| sigmavirus24 | odyssey4me: it's part of the platform identifier but I'm honestly not sure exactly what it means. I suspect it has to do with the fact that it's identifying C-extensions and that it adds that information to the identifier for pip/other tools | 15:38 |
| *** mgoddard has joined #openstack-ansible | 15:38 | |
| hughsaunders | apparently the m signifies --wtih-pymalloc and u --with-wide-unicode | 15:39 |
| hughsaunders | according to https://www.python.org/dev/peps/pep-3149/ | 15:39 |
| mattt | odyssey4me: so i put that wheel review through on kilo, since that is the most pressing branch ... i'll remove WIP and manually cherry-pick to liberty and create a separate review for the repo_server role in master | 15:41 |
| cloudnull | o/ all | 15:42 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ironic: Ironic: Fix docs & linting tests https://review.openstack.org/283384 | 15:42 |
| odyssey4me | thanks mattt | 15:44 |
| cloudnull | odyssey4me: from the keystone role keystone_developer_apt_packages is in the vars | 15:45 |
| cloudnull | https://github.com/os-cloud/openstack-ansible-os_keystone/blob/master/vars/ubuntu-14.04.yml | 15:45 |
| stevelle | andrei_: doesn't look like we have support for that right now. | 15:45 |
| odyssey4me | lol cloudnull I should have thought to look there :p | 15:45 |
| cloudnull | i followed the other patterns for package installs | 15:45 |
| Bofu2U | morn | 15:46 |
| *** pcaruana has joined #openstack-ansible | 15:49 | |
| palendae | raddaoui: I didn't get to testing https://review.openstack.org/#/c/282067/ yesterday, but spinning up an AIO to play with it now | 15:49 |
| raddaoui__ | okay thanks palendae let me know if you need any explanation on it | 15:53 |
| cloudnull | odyssey4me: what patches did you need me to review? | 15:53 |
| cloudnull | reffering back to your convo with bgmccollum | 15:54 |
| spotz | o/ | 15:55 |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible: Pin wheel version in repo_server role https://review.openstack.org/283657 | 15:57 |
| izaakk | o/ | 15:57 |
| *** gparaskevas has quit IRC | 15:58 | |
| cloudnull | also mattt odyssey4me what issues were you seeing with https://review.openstack.org/#/c/279683/ | 15:59 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Resolve packaging problems in virtualenv(s) https://review.openstack.org/279683 | 16:00 |
| *** mgoddard_ has joined #openstack-ansible | 16:01 | |
| *** phalmos has joined #openstack-ansible | 16:01 | |
| *** raddaoui__ has quit IRC | 16:01 | |
| *** raddaoui__ has joined #openstack-ansible | 16:01 | |
| *** mgoddard has quit IRC | 16:04 | |
| rromans | . | 16:04 |
| *** raddaoui__ has quit IRC | 16:05 | |
| *** raddaoui has quit IRC | 16:05 | |
| *** raddaoui_ has quit IRC | 16:06 | |
| *** raddaoui has joined #openstack-ansible | 16:06 | |
| *** raddaoui_ has joined #openstack-ansible | 16:06 | |
| *** raddaoui has quit IRC | 16:06 | |
| *** raddaoui_ has quit IRC | 16:06 | |
| *** raddaoui_ has joined #openstack-ansible | 16:07 | |
| *** raddaoui has joined #openstack-ansible | 16:07 | |
| *** severion has quit IRC | 16:10 | |
| *** kysse_ has joined #openstack-ansible | 16:10 | |
| *** tiagogomes has joined #openstack-ansible | 16:12 | |
| *** subscope has joined #openstack-ansible | 16:14 | |
| *** baker has joined #openstack-ansible | 16:14 | |
| tiagogomes | A question: is master already targeting Mitaka? And if so, is the development being made on a Ubuntu 16.04 image? | 16:14 |
| *** nn_brian has joined #openstack-ansible | 16:14 | |
| palendae | tiagogomes: Yes, no | 16:15 |
| *** phiche has quit IRC | 16:15 | |
| palendae | tiagogomes: The release windows are super tight, and there was the indpendent role repository work going into mitaka, too | 16:15 |
| *** phalmos has quit IRC | 16:16 | |
| tiagogomes | thanks palendae | 16:18 |
| mattt | spotz: argh :P | 16:18 |
| *** KLevenstein_ has joined #openstack-ansible | 16:19 | |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible: Pin wheel version in repo_server role https://review.openstack.org/283657 | 16:19 |
| openstackgerrit | Matt Thompson proposed openstack/openstack-ansible: Pin wheel version in repo_server role https://review.openstack.org/283039 | 16:20 |
| spotz | mattt That's why you guys keep me around:) | 16:20 |
| mattt | spotz: nah, good catch, thanks ! :) | 16:21 |
| nn_brian | Hey, folks. Executing my first openstack-ansible install. Attempting AIO on an Ubuntu 14.04 VM on KVM. The run-playbooks.sh step fails with dozens of errors like the following: | 16:21 |
| mattt | i need a vim spell checker :) | 16:21 |
| nn_brian | <172.29.238.182> REMOTE_MODULE pip name=pycrypto state=present extra_args='' | 16:21 |
| nn_brian | <172.29.238.182> EXEC ssh -C -vvv -o ControlMaster=auto -o ControlPersist=60s -o ControlPath="/root/.ansible/cp/ansible-ssh-%h-%p-%r" -o StrictHostKeyChecking=no -o KbdInteractiveAuthenticati | 16:21 |
| nn_brian | on=no -o PreferredAuthentications=gssapi-with-mic,gssapi-keyex,hostbased,publickey -o PasswordAuthentication=no -o ConnectTimeout=120 172.29.238.182 /bin/sh -c 'LANG=en_US.UTF-8 LC_CTYPE=en_U | 16:21 |
| nn_brian | S.UTF-8 /usr/bin/python' | 16:21 |
| nn_brian | Result from run 1 is: {'msg': "stdout: Ignoring indexes: https://pypi.python.org/simple\nCollecting pycrypto\n\n:stderr: Retrying (Retry(total=4, connect=None, read=None, redirect=None)) afte | 16:21 |
| nn_brian | r connection broken by 'ProtocolError('Connection aborted.', error(111, 'Connection refused'))': /os-releases/11.2.10/\n Retrying (Retry(total=3, connect=None, read=None, redirect=None)) afte | 16:21 |
| nn_brian | r connection broken by 'ProtocolError('Connection aborted.', error(111, 'Connection refused'))': /os-releases/11.2.10/\n Retrying (Retry(total=2, connect=None, read=None, redirect=None)) afte | 16:21 |
| nn_brian | r connection broken by 'ProtocolError('Connection aborted.', error(111, 'Connection refused'))': /os-releases/11.2.10/\n Retrying (Retry(total=1, connect=None, read=None, redirect=None)) afte | 16:21 |
| nn_brian | r connection broken by 'ProtocolError('Connection aborted.', error(111, 'Connection refused'))': /os-releases/11.2.10/\n Retrying (Retry(total=0, connect=None, read=None, redirect=None)) afte | 16:21 |
| nn_brian | r connection broken by 'ProtocolError('Connection aborted.', error(111, 'Connection refused'))': /os-releases/11.2.10/\n Could not find a version that satisfies the requirement pycrypto (from | 16:21 |
| nn_brian | versions: )\nNo matching distribution found for pycrypto\n", 'failed': True, 'attempts': 1, 'cmd': '/usr/local/bin/pip install pycrypto'} | 16:21 |
| nn_brian | Not sure where to look for root causes. Any ideas? | 16:22 |
| *** KLevenstein has quit IRC | 16:22 | |
| *** KLevenstein_ is now known as KLevenstein | 16:22 | |
| spotz | mattt but then I won't have a job! | 16:23 |
| *** phiche has joined #openstack-ansible | 16:25 | |
| *** subscope has quit IRC | 16:27 | |
| nn_brian | The output I submitted, above, is the result of setting the -vvvv flag on the ansible calls in the script. The error is not limited to pycrypto. A number of other packages also elicit the complaint that a matching distro can't be found. I've tried installing the packages by hand in advance, but still I see these errors. | 16:30 |
| odyssey4me | nn_brian convention when working via IRC would be to paste logs into pastebin rather than in the channel | 16:30 |
| odyssey4me | then you just link the pastebin link | 16:31 |
| nn_brian | Oops. Thanks, odyssey4me. I'll figure out how to do that and repost. | 16:33 |
| odyssey4me | nn_brian this is an AIO, correct? did you complete the boostrap-ansible then bootstrap-aio steps without a problem? | 16:33 |
| *** retreved has joined #openstack-ansible | 16:33 | |
| odyssey4me | nn_brian also, does your AIO have more than one network interface? is it virtualbox/vmware/a public cloud/what exactly? | 16:34 |
| nn_brian | odyssey4me yes, both bootstrap steps completed without error. The VM is running on KVM on a CentOS server in my lab. No public access. Checking on the number of interfaces... | 16:36 |
| odyssey4me | nn_brian it will need public access - it needs to fetch packages from pypi, keys from key servers, etc | 16:37 |
| nn_brian | odyssey4me It has one virtual NIC. It is configured for NAT and can get out. What I meant by "no public access" is that there isn't an easy way for someone on the outside to get in. | 16:38 |
| nn_brian | odyssey4me I can, for example, do a command line pip install. No problems there. | 16:39 |
| odyssey4me | nn_brian ah, that's no problem - assuming you don't want to get to Horizon and are prepared to do all testing from inside the environment | 16:39 |
| odyssey4me | alright, that's a start - now it'd be useful to know which branch/tag you cloned and where it actually failed the install process | 16:40 |
| nn_brian | odyssey4me after the clone I did a git checkout kilo | 16:40 |
| odyssey4me | run_playbooks runs through a sequence of playbooks - it failed on one of them... it'd be good to know which one it failed in | 16:41 |
| cloudnull | bgmccollum: you around ? | 16:41 |
| nn_brian | odyssey4me let me check | 16:41 |
| cloudnull | on https://review.openstack.org/#/c/283234/ | 16:41 |
| cloudnull | i think all we need is https://review.openstack.org/#/c/283234/1/playbooks/os-cinder-install.yml | 16:41 |
| cloudnull | and the rest is done for us | 16:41 |
| odyssey4me | nn_brian ah ok, in that case it's a known issue being dealt with by https://review.openstack.org/283039 | 16:41 |
| odyssey4me | nn_brian is there a particular reason you're going for kilo rather than something later? | 16:42 |
| cloudnull | seems there was a regression/typo when the storage address bits were addressed a while ago | 16:42 |
| nn_brian | odyssey4me no. I started with kilo because I assumed stability. What would you recommend as a starting place? This is for Nuage integration work with RackSpace... | 16:43 |
| *** subscope has joined #openstack-ansible | 16:45 | |
| mattt | cloudnull: sorry, just saw your message ... not seeing any issues w/ https://review.openstack.org/#/c/279683/, but i think odyssey4me mentioned it as potentially being related to https://review.openstack.org/#/c/283039/ | 16:45 |
| odyssey4me | nn_brian kilo is stable for production, but the production deployment mechanism does not use run_playbooks (which is for development/testing purposes) | 16:45 |
| nn_brian | odyssey4me ah. I didn't know that. | 16:46 |
| odyssey4me | nn_brian kilo, though, is already in security-only fixes in OpenStack, and close to it in OpenStack-Ansible... so Liberty would be a better bet for stability | 16:46 |
| cloudnull | seems like those patches should go hand in hand. | 16:46 |
| odyssey4me | nn_brian especially when you're doing an eval for production purposes | 16:47 |
| cloudnull | virtualenv is pulling in pip8 by default for new venvs | 16:47 |
| cloudnull | so pinning wheel makes sense | 16:47 |
| nn_brian | odyssey4me so I should clean up, switch to liberty, and execute run-playbooks. Cool. Thanks. | 16:48 |
| *** javeriak has joined #openstack-ansible | 16:48 | |
| odyssey4me | nn_brian yep, there is a teardown script in scripts that'll do a rather destructive teardown | 16:48 |
| nn_brian | odyssey4me very good. thanks for the help. And I'll use pastebin next time. :-) | 16:49 |
| odyssey4me | nn_brian considering that you're building an AIO with no tweaks - I'd suggest just tearing down the whole VM and restarting from scratch - otherwise you'll be missing config bits | 16:49 |
| odyssey4me | nn_brian no problem :) happy to help and advise - looking forward to further interaction :) | 16:49 |
| nn_brian | odyssey4me I have a clone of just the OS bits. I can recreate the VM quickly and start with the clone, bootstraps, and all that. | 16:50 |
| nn_brian | odyssey4m many thanks! | 16:50 |
| odyssey4me | nn_brian yeah, just don't forget to checkout to liberty or the most recent liberty tags (12.x) | 16:50 |
| *** chhavi has joined #openstack-ansible | 16:51 | |
| nn_brian | odyssey4me yeah, i saw that in the quick-start guide. I'll grab the latest 12.x tag | 16:51 |
| tiagogomes | cloudnull, my deployment is failing in the "set local_ip fact (is_metal)" task, on os-neutron-install.yml. The error is that _overlay_network does not have a bridge defined. I can't see how _overlay_network would get any meanful value, as 'tunnel_address' does not seem to be defined anywhere | 16:54 |
| *** mgoddard has joined #openstack-ansible | 16:55 | |
| *** mgoddard_ has quit IRC | 16:55 | |
| bgmccollum | cloudnull | 16:56 |
| cloudnull | bgmccollum: ohai . | 16:56 |
| bgmccollum | cloudnull i was trying to also separate management and storage traffic...hence the separate cinder_iscsi_ip_address role variable... | 16:57 |
| *** javeriak_ has joined #openstack-ansible | 16:57 | |
| bgmccollum | db, rabbit, api over mgmt, iscsi over storage | 16:57 |
| *** javeriak has quit IRC | 16:57 | |
| cloudnull | bgmccollum: all we should have to do is that one change to set the myip | 16:58 |
| cloudnull | https://review.openstack.org/#/c/283234 | 16:58 |
| cloudnull | i updated that | 16:58 |
| cloudnull | sorry | 16:58 |
| cloudnull | i can revert it | 16:58 |
| cloudnull | but thats what we had set in juno/icehouse. | 16:59 |
| cloudnull | tiagogomes: tunnel_address is set within the openstack_user_config | 16:59 |
| bgmccollum | looking | 16:59 |
| tiagogomes | cloudnull I don't use tunnels for tenant networks, only VLANs | 17:00 |
| cloudnull | ah. | 17:01 |
| *** mikelk has quit IRC | 17:05 | |
| bgmccollum | cloudnull so, i was under the assumption that things like db, rabbit, internal api would use the br-mgmt...and just the storage type traffic would use the storage network. they way it is now, either everything is storage network, or everything is management network. | 17:05 |
| bgmccollum | look again at patch set 4 | 17:06 |
| odyssey4me | bgmccollum if that is the case, then I'd prefer your model - mgmt related traffic should be isolated onto the appropriate network, and data on its own - it's important for security and performance to isolate them properly | 17:07 |
| odyssey4me | mhayden ^ your thoughts? | 17:07 |
| *** asettle has quit IRC | 17:08 | |
| odyssey4me | mhayden also, I thought you may be keen to look into a PoC for implementing something like https://www.hastexo.com/blogs/florian/2016/02/21/containers-just-because-everyone-else/ suggests ? | 17:09 |
| * mhayden ganders | 17:09 | |
| mhayden | ah, overlayfs? | 17:09 |
| odyssey4me | it would have to be an optional feature, not enabled by default (at least initially while it's being battle tested), but if implemented properly it could probably be a massive optimisation | 17:10 |
| mhayden | it might be easier to get to systemd-nspawn at some point instead | 17:10 |
| *** nn_brian has quit IRC | 17:11 | |
| *** phalmos has joined #openstack-ansible | 17:11 | |
| odyssey4me | automagically that may be of interest to you guys too | 17:13 |
| cloudnull | bgmccollum: I see what your getting at | 17:14 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Set the storage address for cinder https://review.openstack.org/283234 | 17:15 |
| cloudnull | bgmccollum: ^ | 17:18 |
| cloudnull | tiagogomes: it seems we've made the assumption that an overlay network will be there. | 17:19 |
| bgmccollum | Yeah one issue...since we're using multi-backend style configs, you have to specify the iscsi_ip_address inside each backend section | 17:19 |
| cloudnull | really ? | 17:19 |
| bgmccollum | yeah its stupid | 17:19 |
| cloudnull | we didnt have to do that in Juno ? | 17:19 |
| *** severion has joined #openstack-ansible | 17:20 | |
| bgmccollum | http://lists.openstack.org/pipermail/openstack-operators/2013-August/003410.html | 17:20 |
| cloudnull | bummer | 17:21 |
| bgmccollum | also, i tried limiting it to just the LVM iSCSI driver...but i dont know if thats futile, or if there is any harm to set for every backend type | 17:21 |
| bgmccollum | like, are there other drivers that use the option, and do we need to maintain a list, or just fire and forget | 17:22 |
| cloudnull | so that needs to be set in the openstack_user_config | 17:22 |
| cloudnull | for the various drivers. | 17:22 |
| bgmccollum | will the variable "resolve" before the set fact tasks that discover `cinder_storage_address`? | 17:22 |
| bgmccollum | templated variable precedence in ansible is hard to wrap my head around sometimes | 17:23 |
| *** nn_brian has joined #openstack-ansible | 17:26 | |
| automagically | odyssey4me: Thanks for the pointer | 17:27 |
| *** sdake_ has joined #openstack-ansible | 17:28 | |
| *** GheRivero has joined #openstack-ansible | 17:28 | |
| cloudnull | bgmccollum: yes we can set it here https://github.com/openstack/openstack-ansible/blob/master/etc/openstack_deploy/openstack_user_config.yml.aio#L114 | 17:29 |
| *** phalmos has quit IRC | 17:30 | |
| *** sdake has quit IRC | 17:30 | |
| *** Nepoc has quit IRC | 17:30 | |
| bgmccollum | ok, but set it to what? isn't this the input to generate the inventory? can you even reference a yet to be defined variable here? | 17:31 |
| bgmccollum | "{{ cinder_storage_address }}" ? | 17:31 |
| odyssey4me | bgmccollum so the one option is for a deployer to set it themselves, and we take all the guessing out of the playbooks/tasks... this is simple, but not particularly friendly | 17:32 |
| *** sdake_ is now known as sdake | 17:33 | |
| odyssey4me | another option is to try and derive it - this is friendly, but causes a frankenmess of code | 17:33 |
| *** subscope has quit IRC | 17:33 | |
| * stevelle votes for explicit deployer choice | 17:34 | |
| odyssey4me | I'm actually leaning towards taking the code-based guessing out and implementing documentation, and perhaps show by example in the AIO, to show that if you want the traffic segregation then you need to change your user config | 17:34 |
| automagically | Seems like the most desirable case would be for traffic segregation | 17:35 |
| bgmccollum | so...to actually make it work, are you saying you have to specify the as yet unknown storage IP address in your use config...or is there a was to reference the storage address after the dynamic inventory has created it... | 17:35 |
| odyssey4me | if we do anything to simplify this, then this is another case where an improvement to the dynamic inventory could be used | 17:35 |
| automagically | And as much as possible, I’d expect OSA to be opinionated in that regard and do the best practice thing | 17:36 |
| stevelle | ^ or at least document and make it somewhat easy | 17:36 |
| automagically | Arriving at a clean implementation seems to be the sticky bit | 17:36 |
| bgmccollum | how do you set the iscsi_ip_address user config if you dont know it yet because it hasn't been assigned | 17:36 |
| odyssey4me | yeah, but what I'm seeing at the moment is a set of tasks which don't quite reach the desired result and are complex and hard to follow | 17:37 |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Set the storage address for cinder https://review.openstack.org/283234 | 17:37 |
| bgmccollum | although, cinder-volume LVM iSCSI runs on metal now by default...but the issue still stands | 17:37 |
| cloudnull | bgmccollum: ^ | 17:37 |
| *** Nepoc has joined #openstack-ansible | 17:38 | |
| bgmccollum | i was unsure if something like that would work... | 17:38 |
| odyssey4me | bgmccollum Using Cinder backed by LVM is, according to most everyone I talk to in OpenStack development, supposed to be for reference only - not production | 17:38 |
| odyssey4me | but that's a side issue anyway | 17:39 |
| bgmccollum | well, this all started because Bofu2U was confused why his really expensive storage network wasn't being used at all, and i recall this being fixed in icehouse...so dug in. | 17:40 |
| bgmccollum | cloudnull patch set 8 looks good...ill queue up a build with that and test... | 17:42 |
| odyssey4me | bgmccollum yeah, and thanks for doing so - I'm just expressing a desire for a fresh look at how we do things sometimes | 17:43 |
| *** admin0 has joined #openstack-ansible | 17:44 | |
| *** admin0 has quit IRC | 17:49 | |
| bgmccollum | understood | 17:49 |
| bgmccollum | mo betta | 17:49 |
| bgmccollum | noms | 17:49 |
| *** admin0 has joined #openstack-ansible | 17:50 | |
| *** tiagogomes has quit IRC | 17:55 | |
| *** chhavi has quit IRC | 18:02 | |
| *** chhavi has joined #openstack-ansible | 18:03 | |
| Bofu2U | yeah it makes me want to cry | 18:05 |
| Bofu2U | though I'm going to get one of these reattached to see if it worked | 18:05 |
| odyssey4me | andymccr mrda I have achieved convergance for Ironic in standlone mode.... it doesn't work, but the plays execute - which is a start | 18:09 |
| odyssey4me | lemme get a WIP patch up | 18:10 |
| cloudnull | im going to sleep a bit more, still feeling terribad. will try to be back later. | 18:12 |
| odyssey4me | thanks for checking in cloudnull - get well soon! | 18:13 |
| cloudnull | np. ill try to do more later today | 18:13 |
| cloudnull | bgmccollum Bofu2U: let me know how the storage thingamies go | 18:14 |
| cloudnull | tiagogomez i think we need to revisit the nova play to make it so that we're not assuming a tenant network isnt always there. | 18:14 |
| cloudnull | thats all i got. | 18:14 |
| sigmavirus24 | odyssey4me: I think if you're worried about pip breaking things then your concerns are misplaced. What you should do is black list the known bad versions of pip8 because those bugs you linked in #pypa-dev are fixed appropriately | 18:17 |
| sigmavirus24 | so blacklist 8.0.0, 8.0.1, 8.0.2, (I think 8.0.3 works fine iirc) | 18:17 |
| odyssey4me | sigmavirus24 sure, that is an option - except what's to prevent the same thing happening again | 18:18 |
| sigmavirus24 | But trying to cap pip and setuptools and wheel and whatever else you can imagine is going to breed pain, suffering, and a need in the future to remove those | 18:18 |
| odyssey4me | the whole point of constraints is to be preventative in a stable environment | 18:18 |
| sigmavirus24 | odyssey4me: except the only problem that led to the capping has been fixed (and it won't be reverted until we can get downstreams on board with a lot of packaging changes for their distros - in other words, we have several years before we'll get to a state where that change can be reintroduced) | 18:19 |
| sigmavirus24 | Further, wheel being uncapped is only problematic because we're on an old unsupported version of pip. It's a version distros are not supporting and neither is the PyPA | 18:19 |
| sigmavirus24 | So being preventative of possible issues in a stable environment is fine so long as we're committed to supporting unsupported versions in this project | 18:20 |
| sigmavirus24 | (note this is part of why OpenStack is using constraints and not upper caps at the gate) | 18:20 |
| *** elo has joined #openstack-ansible | 18:30 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ironic: [WIP] Ironic: Add standalone convergance test https://review.openstack.org/283742 | 18:31 |
| *** phalmos has joined #openstack-ansible | 18:32 | |
| odyssey4me | sigmavirus24 so I think we'd be ok with sticking with an older version of pip & wheel, assuming it works for the lifetime of the stable branch in question | 18:33 |
| odyssey4me | it's a question of risk and the management of that risk | 18:33 |
| odyssey4me | if we feel that the risk warrants staying at a cap, then we do that | 18:34 |
| odyssey4me | if we don't, then we uncap it and deal with that | 18:34 |
| sigmavirus24 | odyssey4me: except upstream is releasing things (like glance_store) that are being used on stable branches and tested with newer versions of pip and wheel meaning that anything we run into with our capped versions is going to hurt us | 18:35 |
| odyssey4me | but the theory is that if you deploy a tagged kilo branch at any time in the lifetime of the branch's support life cycle, you should end up with a functional result that it equal to the day the tag was pushed | 18:35 |
| odyssey4me | yeah, this is why this is specifically hurting us in the kilo branch where there are no upper constraints to rely on | 18:36 |
| sigmavirus24 | odyssey4me: speaking for openstack that isn't entirely true the way we do deploys because you'd have to coordinate upper-constraints and everything else | 18:36 |
| odyssey4me | in other branches we should better implement the use of upper constraints for all elements | 18:36 |
| *** phiche1 has joined #openstack-ansible | 18:37 | |
| *** phiche has quit IRC | 18:37 | |
| *** KLevenstein has quit IRC | 18:39 | |
| *** nn_brian has quit IRC | 18:41 | |
| *** javeriak_ has quit IRC | 18:43 | |
| odyssey4me | hmm, so there's a thing - all the devstack vm's have the current uppser constraints file on them and that file is used for the build | 18:46 |
| *** chhavi has quit IRC | 18:47 | |
| odyssey4me | I would guess that this means that the current master branch upper constraints are being used when testing stable/kilo patches | 18:47 |
| *** javeriak has joined #openstack-ansible | 18:48 | |
| odyssey4me | if that is true, then all we'd need to do is find a way to grab that file and consume it for the repo build process in kilo | 18:49 |
| odyssey4me | that said, this would be a partial reimplementation of what's already been done with the repo build process do-over | 18:49 |
| *** admin0 has quit IRC | 18:50 | |
| odyssey4me | stevelle FYI rechecking https://review.openstack.org/278037 doesn't give you much as it only affects the sources on the host - it's partnered with https://review.openstack.org/278033 which affects the containers | 18:51 |
| odyssey4me | stevelle once those are through I can continue work on https://review.openstack.org/279633 which is where the real optimisation starts to happen | 18:51 |
| stevelle | odyssey4me: Yeah, just doing some data collection and I wanted 3 samples for the hosts. | 18:52 |
| odyssey4me | stevelle sure - what's the data looking like so far? | 18:52 |
| stevelle | so far 1:08-1:10 are dominant timings | 18:52 |
| stevelle | outliers over 1:20 | 18:52 |
| odyssey4me | all the same provider? | 18:52 |
| stevelle | thats for all providers so far, | 18:53 |
| *** asettle has joined #openstack-ansible | 18:53 | |
| stevelle | need more samples (10+) to say anything meaningful for specific providers | 18:53 |
| odyssey4me | I don't see it making a big difference for an AIO. I think it'll be a nice difference when you combine the two, and when you have more hosts. | 18:54 |
| stevelle | docs changes are fast but don't tell us much :) | 18:54 |
| stevelle | yeah, betting the containers change will help a little bit more | 18:54 |
| *** kencjohnston has quit IRC | 18:55 | |
| odyssey4me | the apt cache update dependent on whether the cache has expired seems to have helped just a little bit | 18:55 |
| *** subscope has joined #openstack-ansible | 18:56 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Gate: Use apt sources configured on the host https://review.openstack.org/279633 | 18:57 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Gate: Use apt sources configured on the host https://review.openstack.org/279633 | 18:58 |
| bgmccollum | re caps...why cant *everything* be capped after the last successful build before being tagged...then caps removed after tagging for dev work...this is environment rot because of uncapped constraints. | 18:58 |
| odyssey4me | bgmccollum because then we'd need to have a task (figure out the current caps required and put together a patch) for someone to do before a tag, we'd have to ensure that everything else is blocked from merging into the branch, then merge the patch, then tag, then undo all the above | 18:59 |
| odyssey4me | upper constraints was built to reduce that overhead, but unfortunately is not in kilo | 19:00 |
| bgmccollum | ok so its just a problem for kilo ATM | 19:00 |
| odyssey4me | it's only a problem for kilo when doing a repo-build, which is not what is used for production (that uses a repo-clone from rpc-repo) | 19:01 |
| odyssey4me | so it's very specifically an issue for gating and development | 19:01 |
| odyssey4me | and anyone who's using run-playbooks, and I hope to goodness that no-one is using that in production! | 19:02 |
| odyssey4me | stevelle I'm not sure if you've seen https://review.openstack.org/276766 - but it is another very small optimisation | 19:06 |
| stevelle | hadn't pulled that one up yet odyssey4me | 19:06 |
| *** electrofelix has quit IRC | 19:07 | |
| *** admin0 has joined #openstack-ansible | 19:07 | |
| *** subscope has quit IRC | 19:08 | |
| admin0 | can we DVR using the ansible playbooks ? | 19:12 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Remove dependency on python2_lxc git source https://review.openstack.org/276766 | 19:13 |
| odyssey4me | admin0 not at this stage - DVR is not supported by Neutron when using the Linuxbridge Agent | 19:14 |
| admin0 | ok | 19:14 |
| admin0 | is it on the roadmap ? | 19:14 |
| odyssey4me | admin0 there are some guys in the project working on adding OVS support, and then DVR on top of that | 19:14 |
| admin0 | ok | 19:14 |
| odyssey4me | admin0 you can see their progress in the etherpad linked from the meeting agenda | 19:15 |
| admin0 | aha :) | 19:15 |
| admin0 | \o/ | 19:15 |
| odyssey4me | sigmavirus24 we're going to discuss the pip capping, etc in the weekly meeting on thu - can you make it to help with the discussion? I'd like us to be in a position to make informed choices. :) | 19:17 |
| sigmavirus24 | odyssey4me: I'll add it back to my calendar I think I have a different meeting at the same time though | 19:19 |
| sigmavirus24 | Just ping me anyway | 19:19 |
| odyssey4me | sigmavirus24 I'll ping you when the topic comes up. Thanks! | 19:20 |
| sigmavirus24 | Any idea why "{# keystone_service_internaluri_v3 #}" would end up in a config file when we were using {{ keystone_service_internaluri_v3 }}? | 19:20 |
| sigmavirus24 | did we kill that var? | 19:21 |
| *** neerbeer has joined #openstack-ansible | 19:22 | |
| *** javeriak_ has joined #openstack-ansible | 19:22 | |
| odyssey4me | sigmavirus24 that var should only be used inside the keystone role as I recall | 19:22 |
| sigmavirus24 | odyssey4me: aha, this was working a week ago | 19:23 |
| odyssey4me | liberty+ the internaluri var is v3 anyway | 19:23 |
| meteorfox | if have a standalone-swift with 5 nodes, and they are configured as 1 swift_host per zone. If I change the zone values on the swift.yml and re-run os-swift-install playbook, will it work? or do I have to start from scratch? Also same question for the repl_ip? I'm trying to reconfigure it to match another lab which has 1 zone per Rack, so I can configure | 19:23 |
| meteorfox | them in a multi-region configuration | 19:23 |
| sigmavirus24 | Guess anyone developing third party roles shouldn't try to depend on any vars :+1: | 19:23 |
| odyssey4me | sigmavirus24 yep - no such var any more - if it worked previously, you must be using something horribly old: https://github.com/openstack/openstack-ansible/blob/master/playbooks/inventory/group_vars/hosts.yml#L190-L196 | 19:24 |
| *** javeriak has quit IRC | 19:24 | |
| sigmavirus24 | Might be two weeks old but was definitely working last I used this role | 19:24 |
| odyssey4me | odd, as this change was implemented for the Liberty release - so it's been there for quite some time | 19:26 |
| *** Nepoc has quit IRC | 19:26 | |
| *** javeriak has joined #openstack-ansible | 19:26 | |
| stevelle | sigmavirus24: sounds like a topic we need to establish a set of expectations on (dependence on vars) | 19:27 |
| *** Nepoc has joined #openstack-ansible | 19:27 | |
| sigmavirus24 | stevelle: yeah or we just document what ones other people can expect to work | 19:27 |
| sigmavirus24 | the latter would be nicer honestly | 19:27 |
| *** javeriak_ has quit IRC | 19:27 | |
| stevelle | +1 ^ | 19:27 |
| stevelle | we will need to establish an api contract of sorts if we expect to make IRR work, let alone allow 3rd party roles to build on osa reliably | 19:28 |
| palendae | You mean like a public API? | 19:28 |
| stevelle | jinx palendae | 19:28 |
| palendae | mrda was having a lot of headaches building ironic last night | 19:29 |
| odyssey4me | yeah, we started a discussion around approximately that topic at the mid cycle - but never quite finished it | 19:29 |
| palendae | No, and I'm not sure how many in the greater ansible community are approaching it, either | 19:29 |
| palendae | The only project I know of that's making roles on this scale is debops | 19:29 |
| odyssey4me | loads of people are making many roles, but most roles are only applied to small numbers of servers | 19:30 |
| *** Nepoc has quit IRC | 19:30 | |
| stevelle | I think a small working group of a couple folks can probably start to enumerate the kinds of things we know we will need (things like identity endpoints, db and mq endpionts, etc) | 19:30 |
| *** KLevenstein has joined #openstack-ansible | 19:30 | |
| *** asettle has quit IRC | 19:30 | |
| odyssey4me | we have scale complexities in many directions in the project | 19:30 |
| stevelle | and it can grow from that starting point | 19:30 |
| palendae | odyssey4me: True, but are they public like this? | 19:30 |
| palendae | My sense is that most ansible stuff is being written for private use | 19:31 |
| odyssey4me | stevelle yeah, I'm in favour of a small working group being put together to put together a starting doc which can then be put into a review | 19:31 |
| odyssey4me | palendae lots is in the public domain too, but not too many of those are reusable by other environments | 19:31 |
| odyssey4me | and certainly not very reusable at scale | 19:32 |
| palendae | Yeah | 19:32 |
| odyssey4me | most of the role do simple things like setup Apache - they don't care too much about more complex changes thereafter.... when you deal with small number of servers, you don't care that your config is rewritten every time you execute a playbook because you only ever do that once in a while | 19:33 |
| palendae | Right | 19:34 |
| odyssey4me | we're trying to achieve lofty goals like idempotence, maximum uptime, and stable interfaces into roles and to implement playbook entry points that go through a deprecation cycle | 19:35 |
| odyssey4me | most roles I've seen don't even bother tagging or considering a 'stable' policy | 19:35 |
| odyssey4me | they're either roles that never change, or they're roles used in very tight ecosystems | 19:36 |
| stevelle | we be pioneerin | 19:37 |
| stevelle | ...lunch time | 19:37 |
| automagically | The var dependency and namespacing thing is a must | 19:39 |
| automagically | The problem is that you end up with lots of assign this var to this role namespaced var all over the place | 19:39 |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Remove dependency on python2_lxc git source https://review.openstack.org/276766 | 19:42 |
| *** eil397 has joined #openstack-ansible | 19:43 | |
| *** Nepoc has joined #openstack-ansible | 19:44 | |
| *** phiche1 has quit IRC | 19:46 | |
| *** phiche has joined #openstack-ansible | 19:46 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible-ironic: [WIP] Ironic: Add standalone convergance test https://review.openstack.org/283742 | 19:47 |
| odyssey4me | automagically yeah, so it falls to us to figure out a pattern which works for us | 19:47 |
| odyssey4me | something that is more predictable, easy to understand and easy to also troubleshoot when precedence gets hairy | 19:48 |
| automagically | Completely agree, well worth a blueprint/spec writeup | 19:50 |
| *** fawadkhaliq has joined #openstack-ansible | 19:52 | |
| *** javeriak_ has joined #openstack-ansible | 19:53 | |
| *** javeriak has quit IRC | 19:56 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: [WIP] Add support for upgrading from Kilo to Liberty https://review.openstack.org/272182 | 19:57 |
| odyssey4me | palendae would it be possible to break that upgrade patch down into some foundational elements (scaffolding for docs and initial script that does nothing) in order to allow focused patches on top of it that deal with specific elements? | 19:59 |
| odyssey4me | that makes it easier to evaluate and review items in isolation | 19:59 |
| palendae | odyssey4me: I'm thinking about it; I really hate the idea of a commit for a scaffold, but if that's what people want I can do it | 20:00 |
| palendae | I'm almost to the point of letting it go for review | 20:00 |
| palendae | In which case I'll break it up | 20:00 |
| odyssey4me | ah ok, in that case keep at it | 20:00 |
| odyssey4me | I'd just rather not have patches being held back because people are waiting to see if you do it | 20:00 |
| palendae | I'm testing my last changes, then going to do a release note once those are done | 20:00 |
| palendae | Right - my intention with it now is to stop adding new things | 20:01 |
| palendae | If there's problems with what's there I'll fix it | 20:01 |
| openstackgerrit | Merged openstack/openstack-ansible: Docs: IP range values should be consistently formatted https://review.openstack.org/283237 | 20:01 |
| palendae | But getting it merged and unblocking stuff is the priority | 20:01 |
| openstackgerrit | Merged openstack/openstack-ansible: Fix bootstrap-ansible.sh invocation directory https://review.openstack.org/278523 | 20:01 |
| palendae | Got a little carried away with it, admittedly | 20:01 |
| odyssey4me | palendae cool - it doesn't have to be perfect, just having it work on a prestine AIO is certainly good enough as a starting point | 20:01 |
| palendae | Right | 20:01 |
| *** javeriak has joined #openstack-ansible | 20:01 | |
| palendae | That's what I'm trying now | 20:01 |
| palendae | I'm hoping my latest changes also negate the need for RUN_TASKS, but I'm not gonna make a big deal of that point in this round | 20:02 |
| *** javeriak_ has quit IRC | 20:02 | |
| odyssey4me | I'm hoping that we can implement the scaffolding in Mitaka asap too, so that we can also require upgrade tasks (along with docs and release notes) where necessary for new patches coming in | 20:04 |
| palendae | Annnd of course as soon as I get it on an AIO, I see dumb typose | 20:04 |
| palendae | Yeah, I was considering doing some Mitaka work in parallel | 20:04 |
| palendae | But I'd rather not take all of that on myself | 20:04 |
| odyssey4me | that way the upgrade burden is far less when we start actually testing master branch upgrades | 20:04 |
| palendae | Yep | 20:05 |
| palendae | That was my hope for liberty, too | 20:05 |
| palendae | But alas, did not happen | 20:05 |
| *** Bjoern_ has quit IRC | 20:05 | |
| openstackgerrit | Jesse Pretorius (odyssey4me) proposed openstack/openstack-ansible: Check local and upstream repo for pip before using pypi https://review.openstack.org/277905 | 20:06 |
| palendae | Upgrade gating would be awesome | 20:06 |
| palendae | But that's a task in and of itself | 20:06 |
| *** Nepoc has quit IRC | 20:12 | |
| *** Nepoc has joined #openstack-ansible | 20:12 | |
| palendae | odyssey4me: I have made what I believe is my last code change that I know is necessary. Want me to get a release note in before I remove the WIP tags? | 20:14 |
| odyssey4me | palendae please - the release note should add a 'feature' not (or set of feature notes) and refer to the URL for the documentation... no need to retype all the docs stuff | 20:15 |
| odyssey4me | release notes should just be a short reference summary with links for further details | 20:15 |
| palendae | Ok. I have no docs stuff right now; it's certainly need but also a big patchset >.< | 20:15 |
| palendae | needed* | 20:15 |
| *** subscope has joined #openstack-ansible | 20:16 | |
| odyssey4me | now you see why I suggested the scaffolding :) | 20:16 |
| palendae | Yeah, I get it | 20:16 |
| odyssey4me | if it's in a working proof of concept state, then it may be easier to break up now | 20:16 |
| palendae | Yeah, I'll upload this last change so it's not just on my computer and go back and do that | 20:17 |
| palendae | Not sure if that'll abandon https://review.openstack.org/#/c/272652/2 or not | 20:17 |
| palendae | I could even start with that one | 20:17 |
| odyssey4me | thanks palendae, I know this is tough work and I'm happy to see it taken on - I hope that we're learning enough through each iteration in order to provide feedback into the development process | 20:18 |
| palendae | Yeah; I have some notes on the specific things, so I'll try to get the commit messages focused around that | 20:18 |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: [WIP] Add support for upgrading from Kilo to Liberty https://review.openstack.org/272182 | 20:19 |
| palendae | Ok, abandoned with a note, but I'll start into breaking that up this afternoon. That may break https://review.openstack.org/#/c/272652/ then | 20:21 |
| *** javeriak has quit IRC | 20:21 | |
| *** javeriak has joined #openstack-ansible | 20:24 | |
| *** javeriak_ has joined #openstack-ansible | 20:26 | |
| odyssey4me | mrda please feel free to continue work on https://review.openstack.org/283742 when you come online, or start putting together an alternative, or just comment on the existing review - I'm working a little blind on it but I think it's important to have a convergence and functional test (in the end) for the standalone and distributed code paths). | 20:27 |
| odyssey4me | I simply picked the standalone code path because I figured it'd be a simpler start. | 20:27 |
| odyssey4me | I'm going offline now. afk until morning. | 20:28 |
| spotz | Night odyssey4me | 20:29 |
| *** javeriak has quit IRC | 20:29 | |
| *** flwang has quit IRC | 20:31 | |
| *** javeriak_ has quit IRC | 20:31 | |
| *** javeriak has joined #openstack-ansible | 20:35 | |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Add condition to local IP for overlay net https://review.openstack.org/273793 | 20:50 |
| *** javeriak_ has joined #openstack-ansible | 20:50 | |
| mrda | odyssey4me: thanks for your work on this today! Greatly appreciated - there were some assumptions about environments (linters, functional etc) that didn't click until late yesterday. Thanks for your help! | 20:50 |
| sigmavirus24 | spotz: someone said you were interested in/working on barbican? Is that true? | 20:54 |
| *** javeriak has quit IRC | 20:54 | |
| *** dmsimard|afk is now known as dmsimard | 20:57 | |
| *** flwang has joined #openstack-ansible | 20:58 | |
| flwang | hi guys, i have a question about this doc http://docs.openstack.org/developer/openstack-ansible/developer-docs/quickstart-aio.html | 21:00 |
| flwang | now i'm working on ansible zaqar and i'm wondering the dev/test steps | 21:00 |
| flwang | after download the openstack-ansible code, then i ran $ scripts/bootstrap-ansible.sh | 21:01 |
| flwang | so for the next step, do i have to run $ scripts/bootstrap-aio.sh ? can i just run $ cd /opt/openstack-ansible/playbooks $ openstack-ansible os-zaqar-install.yml | 21:01 |
| sigmavirus24 | flwang: bootstrap-aio.sh is a good idea but it depends on how os-zaqar-install works :) | 21:02 |
| bgmccollum | bootstrap-aio.sh only if you're actually setting up an AIO | 21:03 |
| *** leifmadsen has quit IRC | 21:04 | |
| bgmccollum | and as long as you have an appropriate env.d and conf.d files for your zaqar role...then it should work... | 21:04 |
| bgmccollum | "should" ™ | 21:05 |
| *** leifmadsen has joined #openstack-ansible | 21:05 | |
| sigmavirus24 | And user vars and other stuff | 21:06 |
| sigmavirus24 | But yeah, I'm too lazy to be certain everything's correct :) | 21:06 |
| cloudnull | flwang: i | 21:07 |
| cloudnull | sorry. | 21:07 |
| cloudnull | flwang: I'd just run the gate-check-commit.sh script which will do all that for you, as you stand up a test env | 21:07 |
| openstackgerrit | Merged openstack/openstack-ansible-lxc_hosts: Set container apt sources to use a configured list of components https://review.openstack.org/278033 | 21:08 |
| *** Bofu2MBP has joined #openstack-ansible | 21:10 | |
| flwang | cool, thank you guys. | 21:10 |
| flwang | the next question | 21:10 |
| flwang | now i'm putting all the ansible zaqar code in a separate repo | 21:10 |
| flwang | so when I test the code, should i copy the code into openstack-ansible tree? or there is another good way? | 21:11 |
| sigmavirus24 | ansible-galaxy flwang | 21:11 |
| sigmavirus24 | assuming it's in a role structure | 21:11 |
| sigmavirus24 | flwang: something structured roughly like https://github.com/sigmavirus24/openstack-ansible-magnum will allow you to do `ansible-galaxy install https://github.com/sigmavirus24/openstack-ansible-barbican` and then put your playbook in `openstack-ansible/playbooks/` and run it from that directory with `openstack-ansible os-zaqar-install.yml` | 21:12 |
| flwang | sigmavirus24: cool, but the help usage of ansible-galaxy is not very detailed | 21:12 |
| flwang | and then put your playbook in `openstack-ansible/playbooks/` so for this step, i still need to manually copy the code, right? | 21:13 |
| *** neerbeer has quit IRC | 21:13 | |
| spotz | sigmavirus24 I used to be involved with the team, I sit over with them still | 21:14 |
| sigmavirus24 | spotz: someone mentioned you might be interested in hacking on an OSA role to deploy it | 21:15 |
| sigmavirus24 | flwang: yes you would need to manually copy your playbook | 21:15 |
| spotz | sigmavirus24 I did look at it one day when I was bored | 21:15 |
| flwang | sigmavirus24: got it, thanks a lot | 21:16 |
| *** javeriak has joined #openstack-ansible | 21:18 | |
| *** javeriak_ has quit IRC | 21:20 | |
| openstackgerrit | Kevin Carter proposed openstack/openstack-ansible: Set the storage address for cinder https://review.openstack.org/283234 | 21:20 |
| *** asettle has joined #openstack-ansible | 21:23 | |
| palendae | mrda: The tox stuff's making some more sense now? | 21:23 |
| mrda | my patch set doesn't fail now :) | 21:24 |
| palendae | Progress at least! | 21:26 |
| *** javeriak has quit IRC | 21:26 | |
| mrda | actually I lie, it is failing, but I understand it a little more. My goal today is to get them all passing in the gate | 21:27 |
| palendae | A worthy goal! | 21:27 |
| palendae | I'll probably try to not be on IRC at midnight this evening :p | 21:27 |
| palendae | But glad there's progress | 21:27 |
| mrda | I'm depending on you being there palendae! :-P | 21:28 |
| * mrda hopes my AU humour isn't taken incorrectly | 21:28 | |
| palendae | :p | 21:28 |
| palendae | In May I will be working on your timezone at least | 21:28 |
| mrda | oh good! | 21:28 |
| sigmavirus24 | palendae: that's actually happening then? | 21:29 |
| palendae | Off by an hour | 21:29 |
| * mrda hopes things are working by then :) | 21:29 | |
| sigmavirus24 | Congrats! | 21:29 |
| palendae | sigmavirus24: Me going to Japan? Yeah. | 21:29 |
| sigmavirus24 | last I had heard it was uncertain | 21:29 |
| palendae | May 12 to 26 | 21:29 |
| palendae | Ah, yeah | 21:29 |
| palendae | Mostly I was trying to decide on before or after summit | 21:29 |
| *** kencjohnston has joined #openstack-ansible | 21:33 | |
| *** raddaoui__ has joined #openstack-ansible | 21:37 | |
| *** sdake has quit IRC | 21:37 | |
| *** fawadkhaliq has quit IRC | 21:39 | |
| openstackgerrit | Robb Romans proposed openstack/openstack-ansible: Flatten deployment host section https://review.openstack.org/283814 | 21:40 |
| *** raddaoui_ has quit IRC | 21:41 | |
| *** raddaoui has quit IRC | 21:41 | |
| *** raddaoui has joined #openstack-ansible | 21:42 | |
| *** raddaoui_ has joined #openstack-ansible | 21:42 | |
| *** jthorne_ has joined #openstack-ansible | 21:42 | |
| *** jthorne has quit IRC | 21:42 | |
| *** raddaoui__ has quit IRC | 21:43 | |
| *** lane_kong has joined #openstack-ansible | 21:44 | |
| *** jthorne_ has quit IRC | 21:47 | |
| *** johnmilton has quit IRC | 21:47 | |
| *** jthorne has joined #openstack-ansible | 21:47 | |
| *** retreved has quit IRC | 21:54 | |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Add Kilo to Liberty upgrade scaffold https://review.openstack.org/283823 | 21:56 |
| stevelle | flwang: I hope you noticed I added you as a reviewer on https://review.openstack.org/#/c/280770/ | 22:04 |
| flwang | stevelle: yep, i saw that | 22:06 |
| flwang | stevelle: and i'm going to comment it based on my above questions | 22:06 |
| stevelle | I suppose I could add a little more detail on copying the playbook over manually, and likely some of the other steps | 22:06 |
| stevelle | perfect | 22:07 |
| stevelle | thanks for that | 22:07 |
| flwang | stevelle: it would be nice | 22:07 |
| *** pellaeon has quit IRC | 22:07 | |
| flwang | stevelle: i would like to know how to setup the test env if i only care about my project | 22:07 |
| stevelle | flwang: how do you mean if you only care about your project. | 22:09 |
| stevelle | I assume you want to test against a keystone + heat + ceilometer environment at least | 22:10 |
| openstackgerrit | Nolan Brubaker proposed openstack/openstack-ansible: Add Kilo to Liberty upgrade scaffold https://review.openstack.org/283823 | 22:11 |
| *** sigmavirus24 is now known as sigmavirus24_awa | 22:13 | |
| flwang | stevelle: for example | 22:19 |
| flwang | stevelle: zaqar only need keystone | 22:19 |
| flwang | stevelle: so in my dev/test, i would like to know, how to build a minimum test env with only keystone + zaqar | 22:20 |
| flwang | just like devstack | 22:20 |
| *** JBenson has joined #openstack-ansible | 22:20 | |
| flwang | i can config in localrc | 22:20 |
| stevelle | flwang: we don't do that sort of thing typically. That would be a degenerate form of openstack, and because our AIOs use venvs inside containers by default having cinder around doesn't interfere, though it does take longer to build a new stack | 22:24 |
| flwang | stevelle: yep, i see. | 22:25 |
| stevelle | but you can do it, after you run bootstrap-ansible and bootstrap-aio playbooks, go to /etc/openstack_deploy/conf.d/ and remove the files for services you want to omit. | 22:25 |
| stevelle | then you can run scripts/run-playbooks.sh to continue the AIO install | 22:25 |
| flwang | stevelle: so at the init stage, at least i need to run scripts/bootstrap-aio.sh one time, right? | 22:26 |
| flwang | and after that, i can only run my playbook, is it? | 22:26 |
| stevelle | flwang: that sounds right. By bootstrapping AIO one time, you lay down our configuration for OpenStack | 22:26 |
| flwang | stevelle: like openstack-ansible os-zaqar-install.yml | 22:26 |
| stevelle | after that you can just run playbooks | 22:26 |
| stevelle | right | 22:27 |
| flwang | stevelle: cool, i see | 22:27 |
| flwang | stevelle: i will give it a try today and bug you guys later :) | 22:27 |
| flwang | thank you very much | 22:27 |
| stevelle | flwang: just let us know what we can do to help | 22:30 |
| flwang | stevelle: sure, thanks | 22:30 |
| *** sdake has joined #openstack-ansible | 22:30 | |
| *** neerbeer has joined #openstack-ansible | 22:35 | |
| *** woodard_ has joined #openstack-ansible | 22:43 | |
| *** Mudpuppy has quit IRC | 22:44 | |
| *** spotz is now known as spotz_zzz | 22:45 | |
| *** raddaoui_ has quit IRC | 22:45 | |
| *** raddaoui has quit IRC | 22:45 | |
| *** woodard has quit IRC | 22:47 | |
| *** subscope has quit IRC | 22:48 | |
| *** phiche has quit IRC | 22:49 | |
| *** phiche has joined #openstack-ansible | 22:49 | |
| *** phiche has quit IRC | 22:50 | |
| *** asettle has quit IRC | 22:57 | |
| *** mancdaz has quit IRC | 22:58 | |
| *** persia has quit IRC | 22:58 | |
| *** lkoranda has quit IRC | 22:58 | |
| *** dstanek has quit IRC | 22:58 | |
| *** andymccr has quit IRC | 22:58 | |
| *** leifmadsen has quit IRC | 22:58 | |
| *** klamath has quit IRC | 22:58 | |
| *** mariusv has quit IRC | 22:58 | |
| *** MCoLo has quit IRC | 22:58 | |
| *** john51 has quit IRC | 22:58 | |
| *** timrc has quit IRC | 22:58 | |
| *** gtt116_ has quit IRC | 22:58 | |
| *** eil397 has quit IRC | 22:58 | |
| *** baker has quit IRC | 22:58 | |
| *** pcaruana has quit IRC | 22:58 | |
| *** grumpycatt has quit IRC | 22:58 | |
| *** _hanhart has quit IRC | 22:58 | |
| *** prometheanfire has quit IRC | 22:58 | |
| *** portante has quit IRC | 22:58 | |
| *** rohanp has quit IRC | 22:58 | |
| *** neerbeer has quit IRC | 22:58 | |
| *** flwang has quit IRC | 22:58 | |
| *** weshay has quit IRC | 22:58 | |
| *** mrda has quit IRC | 22:58 | |
| *** lbragstad_ has quit IRC | 22:58 | |
| *** dweaver has quit IRC | 22:58 | |
| *** markvoelker has quit IRC | 22:58 | |
| *** jwagner has quit IRC | 22:58 | |
| *** d34dh0r53 has quit IRC | 22:58 | |
| *** bgmccollum has quit IRC | 22:58 | |
| *** stevelle has quit IRC | 22:58 | |
| *** Guest51435 has quit IRC | 22:58 | |
| *** eglute has quit IRC | 22:58 | |
| *** rackertom has quit IRC | 22:58 | |
| *** darrenc has quit IRC | 22:58 | |
| *** d9k has quit IRC | 22:58 | |
| *** logan- has quit IRC | 22:58 | |
| *** evilrob has quit IRC | 22:58 | |
| *** arif-ali has quit IRC | 22:58 | |
| *** severion has quit IRC | 22:58 | |
| *** izaakk has quit IRC | 22:58 | |
| *** jamielennox has quit IRC | 22:58 | |
| *** finchd has quit IRC | 22:58 | |
| *** daneyon_ has quit IRC | 22:58 | |
| *** kysse_ has quit IRC | 22:58 | |
| *** h1nch has quit IRC | 22:58 | |
| *** errr has quit IRC | 22:58 | |
| *** rETROpunK has quit IRC | 22:58 | |
| *** sigmavirus24_awa has quit IRC | 22:58 | |
| *** mattoliverau has quit IRC | 22:58 | |
| *** GheRivero has quit IRC | 22:58 | |
| *** maximov_ has quit IRC | 22:58 | |
| *** jmccrory has quit IRC | 22:58 | |
| *** dalees has quit IRC | 22:58 | |
| *** jroll has quit IRC | 22:58 | |
| *** Bofu2MBP has quit IRC | 22:58 | |
| *** phalmos has quit IRC | 22:58 | |
| *** itsuugo has quit IRC | 22:58 | |
| *** kysse has quit IRC | 22:58 | |
| *** rromans has quit IRC | 22:58 | |
| *** miguelgrinberg has quit IRC | 22:58 | |
| *** dmsimard has quit IRC | 22:58 | |
| *** richoid has quit IRC | 22:58 | |
| *** bogeyon18 has quit IRC | 22:58 | |
| *** mcarden has quit IRC | 22:58 | |
| *** bryan_att has quit IRC | 22:58 | |
| *** etoews has quit IRC | 22:58 | |
| *** agireud has quit IRC | 22:58 | |
| *** b3rnard0 has quit IRC | 22:58 | |
| *** gus has quit IRC | 22:58 | |
| *** cloudnull has quit IRC | 22:58 | |
| *** palendae has quit IRC | 22:58 | |
| *** andrei_ has quit IRC | 22:58 | |
| *** mfisch has quit IRC | 22:58 | |
| *** galstrom_zzz has quit IRC | 22:58 | |
| *** metral has quit IRC | 22:58 | |
| *** spotz_zzz has quit IRC | 22:58 | |
| *** JBenson has quit IRC | 22:58 | |
| *** admin0 has quit IRC | 22:58 | |
| *** elo has quit IRC | 22:58 | |
| *** mgugino has quit IRC | 22:58 | |
| *** weezS has quit IRC | 22:58 | |
| *** zhangjn has quit IRC | 22:58 | |
| *** toddnni has quit IRC | 22:58 | |
| *** yarkot has quit IRC | 22:58 | |
| *** meteorfox has quit IRC | 22:58 | |
| *** jduhamel has quit IRC | 22:58 | |
| *** xar- has quit IRC | 22:58 | |
| *** antonym has quit IRC | 22:58 | |
| *** coolj has quit IRC | 22:58 | |
| *** woodard_ has quit IRC | 22:58 | |
| *** jthorne has quit IRC | 22:58 | |
| *** KLevenstein has quit IRC | 22:58 | |
| *** mgoddard has quit IRC | 22:58 | |
| *** ShannonM has quit IRC | 22:58 | |
| *** Oku_OS has quit IRC | 22:58 | |
| *** krotscheck has quit IRC | 22:58 | |
| *** loquacities has quit IRC | 22:58 | |
| *** McMurlock has quit IRC | 22:58 | |
| *** mgariepy has quit IRC | 22:58 | |
| *** gfa has quit IRC | 22:58 | |
| *** sshen has quit IRC | 22:58 | |
| *** neillc has quit IRC | 22:58 | |
| *** gmmaha has quit IRC | 22:58 | |
| *** arbrandes has quit IRC | 22:58 | |
| *** hughsaunders has quit IRC | 22:58 | |
| *** admiralboom has quit IRC | 22:58 | |
| *** dolphm has quit IRC | 22:58 | |
| *** nwonknu has quit IRC | 22:58 | |
| *** kmARC has quit IRC | 22:58 | |
| *** jcannava has quit IRC | 22:58 | |
| *** bapalm has quit IRC | 22:58 | |
| *** ggillies has quit IRC | 22:58 | |
| *** Tridde has quit IRC | 22:58 | |
| *** alextricity25_ has quit IRC | 22:58 | |
| *** odyssey4me has quit IRC | 22:58 | |
| *** jasondotstar has quit IRC | 22:58 | |
| *** dolpher has quit IRC | 22:58 | |
| *** lane_kong has quit IRC | 22:58 | |
| *** serverascode has quit IRC | 22:58 | |
| *** mhayden has quit IRC | 22:58 | |
| *** xek has quit IRC | 22:58 | |
| *** e-vad has quit IRC | 22:58 | |
| *** lbragstad has quit IRC | 22:58 | |
| *** charz has quit IRC | 22:58 | |
| *** ravla has quit IRC | 22:58 | |
| *** git-harry has quit IRC | 22:58 | |
| *** ChanServ has quit IRC | 22:58 | |
| *** sdake is now known as jpeeler | 22:58 | |
| *** jpeeler is now known as 64MAAECDI | 23:00 | |
| *** git-harry has joined #openstack-ansible | 23:00 | |
| *** ravla has joined #openstack-ansible | 23:00 | |
| *** charz has joined #openstack-ansible | 23:00 | |
| *** lbragstad has joined #openstack-ansible | 23:00 | |
| *** e-vad has joined #openstack-ansible | 23:00 | |
| *** xek has joined #openstack-ansible | 23:00 | |
| *** mhayden has joined #openstack-ansible | 23:00 | |
| *** serverascode has joined #openstack-ansible | 23:00 | |
| *** dolpher has joined #openstack-ansible | 23:00 | |
| *** jasondotstar has joined #openstack-ansible | 23:00 | |
| *** odyssey4me has joined #openstack-ansible | 23:00 | |
| *** alextricity25_ has joined #openstack-ansible | 23:00 | |
| *** lane_kong has joined #openstack-ansible | 23:00 | |
| *** portante has joined #openstack-ansible | 23:00 | |
| *** prometheanfire has joined #openstack-ansible | 23:00 | |
| *** _hanhart has joined #openstack-ansible | 23:00 | |
| *** grumpycatt has joined #openstack-ansible | 23:00 | |
| *** pcaruana has joined #openstack-ansible | 23:00 | |
| *** baker has joined #openstack-ansible | 23:00 | |
| *** eil397 has joined #openstack-ansible | 23:00 | |
| *** arif-ali has joined #openstack-ansible | 23:00 | |
| *** evilrob has joined #openstack-ansible | 23:00 | |
| *** logan- has joined #openstack-ansible | 23:00 | |
| *** darrenc has joined #openstack-ansible | 23:00 | |
| *** d9k has joined #openstack-ansible | 23:00 | |
| *** rackertom has joined #openstack-ansible | 23:00 | |
| *** eglute has joined #openstack-ansible | 23:00 | |
| *** stevelle has joined #openstack-ansible | 23:00 | |
| *** Guest51435 has joined #openstack-ansible | 23:00 | |
| *** bgmccollum has joined #openstack-ansible | 23:00 | |
| *** d34dh0r53 has joined #openstack-ansible | 23:00 | |
| *** jwagner has joined #openstack-ansible | 23:00 | |
| *** markvoelker has joined #openstack-ansible | 23:00 | |
| *** dweaver has joined #openstack-ansible | 23:00 | |
| *** 32NAAC6RR has joined #openstack-ansible | 23:00 | |
| *** mrda has joined #openstack-ansible | 23:00 | |
| *** weshay has joined #openstack-ansible | 23:00 | |
| *** flwang has joined #openstack-ansible | 23:00 | |
| *** neerbeer has joined #openstack-ansible | 23:00 | |
| *** spotz_zzz has joined #openstack-ansible | 23:00 | |
| *** galstrom_zzz has joined #openstack-ansible | 23:00 | |
| *** metral has joined #openstack-ansible | 23:00 | |
| *** mfisch has joined #openstack-ansible | 23:00 | |
| *** andrei_ has joined #openstack-ansible | 23:00 | |
| *** cloudnull has joined #openstack-ansible | 23:00 | |
| *** gus has joined #openstack-ansible | 23:00 | |
| *** palendae has joined #openstack-ansible | 23:00 | |
| *** b3rnard0 has joined #openstack-ansible | 23:00 | |
| *** agireud has joined #openstack-ansible | 23:00 | |
| *** etoews has joined #openstack-ansible | 23:00 | |
| *** bryan_att has joined #openstack-ansible | 23:00 | |
| *** mcarden has joined #openstack-ansible | 23:00 | |
| *** bogeyon18 has joined #openstack-ansible | 23:00 | |
| *** 64MAAECDI is now known as sdake | 23:01 | |
| *** GheRivero has joined #openstack-ansible | 23:01 | |
| *** severion has joined #openstack-ansible | 23:01 | |
| *** kysse_ has joined #openstack-ansible | 23:01 | |
| *** izaakk has joined #openstack-ansible | 23:01 | |
| *** jamielennox has joined #openstack-ansible | 23:01 | |
| *** finchd has joined #openstack-ansible | 23:01 | |
| *** daneyon_ has joined #openstack-ansible | 23:01 | |
| *** errr has joined #openstack-ansible | 23:01 | |
| *** h1nch has joined #openstack-ansible | 23:01 | |
| *** sigmavirus24_awa has joined #openstack-ansible | 23:01 | |
| *** rETROpunK has joined #openstack-ansible | 23:01 | |
| *** mattoliverau has joined #openstack-ansible | 23:01 | |
| *** maximov_ has joined #openstack-ansible | 23:01 | |
| *** jmccrory has joined #openstack-ansible | 23:01 | |
| *** dalees has joined #openstack-ansible | 23:01 | |
| *** jroll has joined #openstack-ansible | 23:01 | |
| *** mancdaz has joined #openstack-ansible | 23:01 | |
| *** persia has joined #openstack-ansible | 23:01 | |
| *** lkoranda has joined #openstack-ansible | 23:01 | |
| *** andymccr has joined #openstack-ansible | 23:01 | |
| *** dstanek has joined #openstack-ansible | 23:01 | |
| *** leifmadsen has joined #openstack-ansible | 23:02 | |
| *** klamath has joined #openstack-ansible | 23:02 | |
| *** mariusv has joined #openstack-ansible | 23:02 | |
| *** john51 has joined #openstack-ansible | 23:02 | |
| *** MCoLo has joined #openstack-ansible | 23:02 | |
| *** timrc has joined #openstack-ansible | 23:02 | |
| *** gtt116_ has joined #openstack-ansible | 23:02 | |
| *** john51 has quit IRC | 23:02 | |
| *** woodard_ has joined #openstack-ansible | 23:03 | |
| *** jthorne has joined #openstack-ansible | 23:03 | |
| *** KLevenstein has joined #openstack-ansible | 23:03 | |
| *** mgoddard has joined #openstack-ansible | 23:03 | |
| *** ShannonM has joined #openstack-ansible | 23:03 | |
| *** Oku_OS has joined #openstack-ansible | 23:03 | |
| *** krotscheck has joined #openstack-ansible | 23:03 | |
| *** loquacities has joined #openstack-ansible | 23:03 | |
| *** McMurlock has joined #openstack-ansible | 23:03 | |
| *** mgariepy has joined #openstack-ansible | 23:03 | |
| *** gfa has joined #openstack-ansible | 23:03 | |
| *** sshen has joined #openstack-ansible | 23:03 | |
| *** neillc has joined #openstack-ansible | 23:03 | |
| *** gmmaha has joined #openstack-ansible | 23:03 | |
| *** arbrandes has joined #openstack-ansible | 23:03 | |
| *** hughsaunders has joined #openstack-ansible | 23:03 | |
| *** admiralboom has joined #openstack-ansible | 23:03 | |
| *** dolphm has joined #openstack-ansible | 23:03 | |
| *** nwonknu has joined #openstack-ansible | 23:03 | |
| *** kmARC has joined #openstack-ansible | 23:03 | |
| *** jcannava has joined #openstack-ansible | 23:03 | |
| *** bapalm has joined #openstack-ansible | 23:03 | |
| *** ggillies has joined #openstack-ansible | 23:03 | |
| *** Tridde has joined #openstack-ansible | 23:03 | |
| *** zhangjn has joined #openstack-ansible | 23:03 | |
| *** toddnni has joined #openstack-ansible | 23:03 | |
| *** yarkot has joined #openstack-ansible | 23:03 | |
| *** meteorfox has joined #openstack-ansible | 23:03 | |
| *** jduhamel has joined #openstack-ansible | 23:03 | |
| *** xar- has joined #openstack-ansible | 23:03 | |
| *** antonym has joined #openstack-ansible | 23:03 | |
| *** coolj has joined #openstack-ansible | 23:03 | |
| *** rohanp has joined #openstack-ansible | 23:03 | |
| *** john51 has joined #openstack-ansible | 23:03 | |
| *** JBenson has joined #openstack-ansible | 23:03 | |
| *** admin0 has joined #openstack-ansible | 23:03 | |
| *** elo has joined #openstack-ansible | 23:03 | |
| *** mgugino has joined #openstack-ansible | 23:03 | |
| *** weezS has joined #openstack-ansible | 23:03 | |
| *** kysse__ has joined #openstack-ansible | 23:03 | |
| *** john51 has quit IRC | 23:03 | |
| *** rromans_ has joined #openstack-ansible | 23:03 | |
| *** sigmavirus24_awa has quit IRC | 23:04 | |
| *** GheRivero has quit IRC | 23:04 | |
| *** sigmavirus24_awa has joined #openstack-ansible | 23:05 | |
| *** h1nch has quit IRC | 23:06 | |
| *** ChanServ has joined #openstack-ansible | 23:06 | |
| *** asimov.freenode.net sets mode: +o ChanServ | 23:06 | |
| *** h1nch has joined #openstack-ansible | 23:06 | |
| *** sigmavirus24_awa has quit IRC | 23:06 | |
| *** sigmavirus24_awa has joined #openstack-ansible | 23:06 | |
| *** Bofu2MBP has joined #openstack-ansible | 23:06 | |
| *** phalmos has joined #openstack-ansible | 23:06 | |
| *** itsuugo has joined #openstack-ansible | 23:06 | |
| *** kysse has joined #openstack-ansible | 23:06 | |
| *** rromans has joined #openstack-ansible | 23:06 | |
| *** miguelgrinberg has joined #openstack-ansible | 23:06 | |
| *** dmsimard has joined #openstack-ansible | 23:06 | |
| *** richoid has joined #openstack-ansible | 23:06 | |
| *** dmsimard1 has joined #openstack-ansible | 23:06 | |
| *** miguelgrinberg has quit IRC | 23:07 | |
| *** kysse has quit IRC | 23:07 | |
| *** rromans has quit IRC | 23:07 | |
| *** dmsimard has quit IRC | 23:07 | |
| *** dmsimard1 is now known as dmsimard | 23:07 | |
| *** john51 has joined #openstack-ansible | 23:07 | |
| *** kencjohnston has quit IRC | 23:07 | |
| *** miguelgrinberg has joined #openstack-ansible | 23:08 | |
| *** baker has left #openstack-ansible | 23:09 | |
| *** klamath has quit IRC | 23:09 | |
| *** phalmos has quit IRC | 23:12 | |
| openstackgerrit | Merged openstack/openstack-ansible: Check local and upstream repo for pip before using pypi https://review.openstack.org/277905 | 23:12 |
| *** darrenc is now known as darrenc_afk | 23:13 | |
| *** KLevenstein has quit IRC | 23:21 | |
| *** retreved has joined #openstack-ansible | 23:25 | |
| *** retreved has quit IRC | 23:27 | |
| *** rohanp_ has joined #openstack-ansible | 23:27 | |
| *** weezS has quit IRC | 23:28 | |
| *** GheRivero has joined #openstack-ansible | 23:31 | |
| *** john51_ has joined #openstack-ansible | 23:37 | |
| *** john51 has quit IRC | 23:41 | |
| *** ShannonM has quit IRC | 23:48 | |
| *** gmmaha has quit IRC | 23:57 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!