| *** hwoarang has quit IRC | 02:10 | |
| *** hwoarang has joined #ara | 05:14 | |
| *** hwoarang has quit IRC | 06:23 | |
| *** hwoarang has joined #ara | 06:23 | |
| *** hwoarang has joined #ara | 06:23 | |
| *** openstackgerrit has quit IRC | 06:49 | |
| *** gvincent has joined #ara | 09:51 | |
| *** hwoarang has quit IRC | 09:53 | |
| *** hwoarang has joined #ara | 09:54 | |
| *** hwoarang has joined #ara | 09:54 | |
| *** hwoarang has quit IRC | 09:56 | |
| *** hwoarang has joined #ara | 09:57 | |
| *** hwoarang has quit IRC | 09:57 | |
| *** hwoarang has joined #ara | 09:57 | |
| *** gvincent has quit IRC | 10:21 | |
| tristanC | prometheanfire: wouldn't https://review.openstack.org/#/c/574669/ help? | 11:18 |
|---|---|---|
| *** bcoca has joined #ara | 12:37 | |
| *** bcoca has joined #ara | 12:37 | |
| *** themurph has joined #ara | 13:26 | |
| dmsimard | prometheanfire: hey, I'm not sure exactly to what extent I'll be able to help | 13:34 |
| dmsimard | prometheanfire: I'm trying to put as little maintenance in ARA 0.x as possible so I can spend that time getting 1.0 out the door | 13:35 |
| dmsimard | if bumping flask from 0.x to 1.x means there's a bunch of work involved I might not be able to do it | 13:35 |
| *** tbielawa has joined #ara | 13:44 | |
| *** themurph has quit IRC | 13:50 | |
| *** guilhermesp has joined #ara | 14:23 | |
| *** tbielawa is now known as tbielawa|stand | 14:28 | |
| *** themurph has joined #ara | 14:33 | |
| *** openstackgerrit has joined #ara | 14:37 | |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Stop supporting Ansible =<2.3 in config compatibility layer https://review.openstack.org/575240 | 14:37 |
| *** tbielawa|stand is now known as tbielawa | 14:45 | |
| openstackgerrit | Merged openstack/ara master: Improve self-healing of missing playbook file references https://review.openstack.org/578202 | 14:45 |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Bump requirements and update gate jobs for latest version of Ansible https://review.openstack.org/579891 | 14:46 |
| dmsimard | bcoca: still waiting for that slack invite :p | 14:53 |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Avoid join and yield per and just iterate https://review.openstack.org/579706 | 14:56 |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Drop the rest of the usage of YIELD_PER https://review.openstack.org/579710 | 14:56 |
| ara-slack | <dmsimard> @harlowja ^ rebased your patches, the first one was based on an ancient parent for some reason | 14:57 |
| bcoca | dmsimard: i canot do those anymore | 14:59 |
| bcoca | permissions were changed | 14:59 |
| ara-slack | <dmsimard> @dcreno @logan.attwood ^ Josh has sent patches which should improve the behavior when dealing with a large amount of tasks/hosts. Let me know if you want to test/try them out ahead of the release | 15:00 |
| ara-slack | <logan.attwood> ty | 15:00 |
| bcoca | dmsimard: but pushed up the chain ... lets see what happens | 15:01 |
| dmsimard | bcoca: sure, thanks :D | 15:01 |
| dmsimard | bcoca: submitted a talk to ansiblefest about ara, hope it gets selected :) | 15:01 |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Add a trailing slash to file links in the file tab https://review.openstack.org/574739 | 15:03 |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Add a trailing slash to file links in the file tab https://review.openstack.org/574739 | 15:07 |
| prometheanfire | tristanC: nice | 15:11 |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Bump requirements and update gate jobs for latest version of Ansible https://review.openstack.org/579891 | 15:24 |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Bump requirements and update gate jobs for latest version of Ansible https://review.openstack.org/579891 | 15:26 |
| *** weshay|ruck is now known as weshay | 15:29 | |
| *** sshnaidm is now known as sshnaidm|rover | 15:29 | |
| *** prometheanfire has left #ara | 15:42 | |
| ara-slack | <dcreno> File uploaded https://ara-community.slack.com/files/U9PSY5M7Y/FBJ3L05TN/-.txt / https://slack-files.com/T6VAB05L7-FBJ3L05TN-c5c81456bd - <@U6U5DKJ59>, interested but currently battling “Internal Server Error” due to this… | 16:04 |
| ara-slack | <dmsimard> @dcreno your error is fixed by another commit which hasn't landed yet :( Let me land the stuff that is in flight and get a proper beta release out. I'll let you know when it's good to go. Probably tomorrow. | 16:08 |
| ara-slack | <dcreno> that would be great. | 16:08 |
| *** jparrill_ has quit IRC | 17:04 | |
| *** jparrill_ has joined #ara | 17:04 | |
| *** tbielawa is now known as tbielawa|lunch | 17:10 | |
| *** jparrill_ has quit IRC | 17:15 | |
| weshay | howdy.. trying the adhoc example for ara | 17:23 |
| weshay | ansible localhost -m ara_record \ | 17:23 |
| weshay | -a "playbook=uuuu-iiii-dddd-0000 key=logs value={{ lookup('file', '/var/log/ansible.log') }}" | 17:23 |
| weshay | nvrmind for now :) | 17:24 |
| *** jparrill_ has joined #ara | 17:24 | |
| dmsimard | weshay: ok :p | 17:40 |
| weshay | :) | 17:40 |
| weshay | dmsimard, so after a ansible ara_read that appears to have worked.. I don't have any playbooks in the sqllite db | 17:58 |
| weshay | :( | 17:58 |
| weshay | http://paste.openstack.org/show/724911/ is a partial paste | 17:59 |
| * weshay tries to get more data | 17:59 | |
| dmsimard | weshay: "playbook=deploy_steps_playbook.yaml" is not right | 18:00 |
| dmsimard | weshay: you need the actual playbook uuid | 18:00 |
| weshay | OH | 18:00 |
| dmsimard | this can be retrieved in different ways | 18:00 |
| dmsimard | I already have a gist for that I think, hang on | 18:00 |
| dmsimard | weshay: so I have this but this isn't what you want -- though it does show you that running an ara_record (or ara_read) task within a playbook provides the playbook uuid in the return output (registered var) | 18:02 |
| dmsimard | https://gist.github.com/dmsimard/0335667e69539c1dd22a169cbd43e616 | 18:02 |
| weshay | this is all post ansible run btw | 18:02 |
| * weshay looks | 18:02 | |
| weshay | essentially post processing http://logs.openstack.org/49/575749/6/check/tripleo-ci-centos-7-containers-multinode/fef5b0c/logs/undercloud/var/lib/mistral/ee6e8d26-ac5d-40bb-96b4-b43d4d89cd79/ | 18:03 |
| dmsimard | weshay: there's basically two ways of retrieving the playbook uuid | 18:03 |
| dmsimard | weshay: 1) running an ara_record task within a playbook and looking at the playbook_id of the registered var | 18:03 |
| dmsimard | 2) doing something like "ara playbook list" and retrieving the uuid from there | 18:03 |
| weshay | ya. ran that | 18:03 |
| weshay | nothing returned | 18:03 |
| dmsimard | weshay: where is the database file located ? | 18:04 |
| dmsimard | weshay: if you're not exporting a database path, by default it'll be in ~/.ara/ansible.sqlite | 18:04 |
| dmsimard | in a job you might have $WORKSPACE/ansible.sqlite or whatever, for example | 18:04 |
| weshay | ya.. that is where it's at | 18:04 |
| weshay | ok.. | 18:05 |
| weshay | I have a few jobs that have ara configured for the overcloud | 18:05 |
| weshay | it appeared as though I could just take the ansible.log and the playbook and generate this data | 18:05 |
| weshay | I think that was my mistake | 18:05 |
| dmsimard | weshay: you can, that's more or less what ara already does in it's own integration tests | 18:06 |
| dmsimard | weshay: 1) https://github.com/openstack/ara/blob/master/run_tests.sh#L129-L130 | 18:06 |
| dmsimard | 2) https://github.com/openstack/ara/blob/master/run_tests.sh#L160 | 18:06 |
| dmsimard | see in the records http://logs.openstack.org/40/575240/2/check/ara-integration-ubuntu-py35-2.5.2/2959e25/logs/build-playbook/ | 18:07 |
| weshay | ya.. k.. | 18:07 |
| weshay | so maybe my ara_record command was wrong too | 18:07 |
| dmsimard | weshay: well, aside from the playbook id being wrong, the rest looks okay | 18:09 |
| weshay | ya.. so I just need the id | 18:09 |
| *** tbielawa|lunch is now known as tbielawa | 18:09 | |
| weshay | w/o the original db | 18:10 |
| weshay | maybe I'll try this on another deployment | 18:10 |
| dmsimard | if you're doing an "ara playbook list" and nothing comes up (while you're expecting stuff to be there), it's likely because the database is actually somewhere else | 18:10 |
| weshay | assuming ara was configured originally though | 18:10 |
| weshay | if ara was not configured in ansible.cfg or $other .. I would not have it | 18:10 |
| dmsimard | right, but I thought all ooo/oooq jobs had it ? | 18:11 |
| weshay | dmsimard, so there is the ci playbooks.. and now the playbooks that tripleo uses to deploy the undercloud or overcloud | 18:11 |
| weshay | we have the undercloud working.. which is awesome | 18:11 |
| weshay | now working on the overcloud playbooks | 18:11 |
| dmsimard | weshay: so inception-ish | 18:11 |
| weshay | ya | 18:11 |
| weshay | :( | 18:11 |
| weshay | https://review.openstack.org/#/c/565077/12/tripleo_common/actions/ansible.py | 18:12 |
| dmsimard | http://codesearch.openstack.org/?q=callback_plugins&i=nope&files=&repos=tripleo-ci,tripleo-common,tripleo-common-tempest-plugin,tripleo-docs,tripleo-ha-utils,tripleo-heat-templates,tripleo-image-elements,tripleo-ipsec,tripleo-puppet-elements,tripleo-quickstart,tripleo-quickstart-extras,tripleo-repos,tripleo-specs,tripleo-ui,tripleo-upgrade,tripleo-validations | 18:12 |
| *** tbielawa is now known as tbielawa|brb | 18:13 | |
| weshay | there is data in the /tmp/deploy.sqlite db | 18:13 |
| dmsimard | weshay: ah so do an export ARA_DATABASE=sqlite:////tmp/deploy.sqlite before running your commands | 18:17 |
| weshay | ya.. did that | 18:17 |
| weshay | dmsimard, so I'm doing two things btw.. which may confuse you.. | 18:18 |
| weshay | there are the jobs where we are configuring ara to work | 18:18 |
| weshay | http://paste.openstack.org/show/724913/ | 18:18 |
| weshay | and you get that ^ | 18:18 |
| weshay | and an experiment I'm trying to just pull the ansible logs and post process on the logs on my local dev box | 18:18 |
| weshay | the local devbox is the one coming up w/ no playbooks | 18:19 |
| weshay | http://logs.openstack.org/62/578462/7/check/tripleo-ci-centos-7-containers-multinode/67f8589/logs/ara_oooq_oc/ | 18:19 |
| weshay | is CI ^ atm | 18:19 |
| weshay | and that is consistent on my local recreate w/ the paste | 18:19 |
| dmsimard | weshay: btw I added some comments to that review | 18:21 |
| weshay | data from that is http://logs.openstack.org/62/578462/7/check/tripleo-ci-centos-7-containers-multinode/67f8589/logs/undercloud/var/lib/mistral/070b7a05-f674-4997-bf68-3ccd8a8ce1d6/ | 18:21 |
| weshay | dmsimard, k | 18:21 |
| weshay | dmsimard, the other thing is.. w/ all the inner and outter inception bullshit | 18:23 |
| weshay | we were trying to use different dbs for the different levels to keep it clean | 18:23 |
| weshay | notting the various ara dirs http://logs.openstack.org/62/578462/7/check/tripleo-ci-centos-7-containers-multinode/67f8589/logs/ | 18:23 |
| weshay | ara_oooq_root is the undercloud install | 18:23 |
| dmsimard | weshay: you guys need to stop generating all these html reports and use the raw databases instead | 18:24 |
| weshay | to what end? | 18:24 |
| weshay | we're sending the data to grafana afaik w/ just the dbs | 18:25 |
| dmsimard | weshay: http://lists.openstack.org/pipermail/openstack-dev/2018-March/128902.html | 18:25 |
| dmsimard | weshay: html reports don't scale and are wasteful in terms of performance and resources | 18:25 |
| dmsimard | so you could have like | 18:26 |
| dmsimard | <root>/ara-report <-- zuul report | 18:26 |
| dmsimard | <root>/logs/undercloud/ara-report <-- undercloud report | 18:26 |
| dmsimard | <root>/logs/overcloud/ara-report <-- overcloud report | 18:26 |
| dmsimard | you don't need to generate anything, just need to create that directory structure and put the "ansible.sqlite" database there | 18:27 |
| dmsimard | for example, with openstack-ansible | 18:28 |
| dmsimard | This is Zuul's report: http://logs.openstack.org/33/572433/2/check/openstack-ansible-deploy-aio_basekit-opensuse-423/7b48a37/ara-report/ | 18:28 |
| dmsimard | and this is the OSA report: http://logs.openstack.org/33/572433/2/check/openstack-ansible-deploy-aio_basekit-opensuse-423/7b48a37/logs/ara-report/ | 18:28 |
| dmsimard | and then, you can also download the databases directly if you like, you just suffix "ansible.sqlite" so http://logs.openstack.org/33/572433/2/check/openstack-ansible-deploy-aio_basekit-opensuse-423/7b48a37/logs/ara-report/ansible.sqlite works | 18:29 |
| weshay | dmsimard, so using more than one sqlite db is ok? | 18:29 |
| weshay | may flip that dir structure though.. | 18:30 |
| weshay | \/logs/ara_report/undercloud | 18:30 |
| weshay | \/logs/ara_report/overcloud | 18:30 |
| weshay | because you have the undercloud logs, then subnode-$1,$2,$3 | 18:30 |
| dmsimard | weshay: I don't see why not -- the only thing to keep in mind if you want to use what I just described is that it's important that the directory is called "ara-report" and that the database is called "ansible.sqlite" because that's how the middleware is set up to dynamically load the databases | 18:30 |
| weshay | k | 18:31 |
| dmsimard | weshay: It'd probably save a few minutes on your jobs if you did it that way, because then you don't waste time generating hundreds of html files, compressing them and then rsyncing them. It's just this one small sqlite file. | 18:31 |
| dmsimard | plus, you get to be able to download the raw databases for experimenting locally or whatever | 18:32 |
| weshay | hrm k | 18:32 |
| weshay | k k.. someone just knocked.. bbiab | 18:32 |
| dmsimard | weshay: kind of got sidetracked but I couldn't help but point out an obvious improvement opporunity :) | 18:33 |
| dmsimard | fwiw, I don't think what you're running into is an issue with ara, just need to properly point it to the right database and stuff. | 18:33 |
| *** tbielawa|brb is now known as tbielawa | 18:51 | |
| openstackgerrit | Merged openstack/ara master: Stop supporting Ansible =<2.3 in config compatibility layer https://review.openstack.org/575240 | 19:00 |
| openstackgerrit | Merged openstack/ara master: Add a trailing slash to file links in the file tab https://review.openstack.org/574739 | 19:04 |
| openstackgerrit | Merged openstack/ara master: Avoid join and yield per and just iterate https://review.openstack.org/579706 | 19:04 |
| *** jparrill_ has quit IRC | 19:11 | |
| *** jparrill has joined #ara | 19:12 | |
| openstackgerrit | Merged openstack/ara master: Bump requirements and update gate jobs for latest version of Ansible https://review.openstack.org/579891 | 19:36 |
| *** tbielawa is now known as tbielawa|g0n3 | 19:53 | |
| *** themurph has quit IRC | 21:26 | |
| openstackgerrit | Merged openstack/ara master: Drop the rest of the usage of YIELD_PER https://review.openstack.org/579710 | 21:49 |
| tristanC | dmsimard: weshay: we have integrated ara-report in sf-3.1, so we can copy the settings to logs.rdoproject and have it added to the rdo base job. | 23:40 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!