| *** karimb has quit IRC | 00:03 | |
| *** permalac has quit IRC | 09:12 | |
| *** permalac has joined #ara | 09:14 | |
| *** karimb has joined #ara | 09:43 | |
| *** karimb has quit IRC | 09:43 | |
| *** permalac has quit IRC | 09:45 | |
| *** permalac has joined #ara | 09:47 | |
| *** karimb has joined #ara | 09:52 | |
| *** karimb has quit IRC | 10:31 | |
| *** karimb has joined #ara | 10:50 | |
| *** karimb has quit IRC | 11:12 | |
| *** karimb has joined #ara | 11:27 | |
| *** karimb has quit IRC | 11:28 | |
| *** karimb has joined #ara | 11:30 | |
| *** karimb has quit IRC | 12:16 | |
| *** karimb has joined #ara | 12:19 | |
| *** karimb has quit IRC | 12:39 | |
| *** karimb has joined #ara | 13:20 | |
| *** jrist has quit IRC | 14:56 | |
| *** jrist has joined #ara | 15:10 | |
| *** karimb has quit IRC | 15:44 | |
| *** karimb has joined #ara | 15:52 | |
| *** jrist has quit IRC | 16:10 | |
| *** karimb has quit IRC | 16:15 | |
| *** jrist has joined #ara | 16:22 | |
| *** karimb has joined #ara | 16:27 | |
| *** karimb has quit IRC | 17:33 | |
| *** karimb has joined #ara | 17:36 | |
| *** karimb has quit IRC | 17:46 | |
| berendt | 19:03:28 OperationalError: (sqlite3.OperationalError) table data already exists [SQL: u'\nCREATE TABLE data (\n\tid VARCHAR(36) NOT NULL, \n\tplaybook_id VARCHAR(36), \n\t"key" VARCHAR(255), \n\tvalue BLOB, \n\tPRIMARY KEY (id), \n\tFOREIGN KEY(playbook_id) REFERENCES playbooks (id) ON DELETE RESTRICT, \n\tUNIQUE (playbook_id, "key")\n)\n\n'] | 19:04 |
|---|---|---|
| berendt | i got this issue when trying to initiate a new sqlite database with current release | 19:04 |
| berendt | is this issue already known? | 19:04 |
| berendt | it is related to the ara_record data table: INFO [alembic.runtime.migration] Running upgrade da9459a1f71c -> e8e78fd08bf2, ara_record data | 19:10 |
| berendt | the upgrade is called twice | 19:10 |
| berendt | that's interesting, i set database = sqlite:///ara.sqlite to generate the file in the current directory, when i do not set this parameter everything is working like expected | 19:13 |
| berendt | absolute path (sqlite:////tmp/ara.sqlite) is also working like expected | 19:13 |
| dmsimard | berendt: Haven't seen that happen | 19:20 |
| dmsimard | berendt: can you let me know if you figure out how to reproduce the issue ? | 19:20 |
| berendt | give me a second | 19:20 |
| *** karimb has joined #ara | 19:26 | |
| berendt | dmsimard: https://gist.github.com/berendt/5a661fdf0711009eb60ccfd72edab441 | 19:30 |
| berendt | according to http://docs.sqlalchemy.org/en/latest/core/engines.html#sqlite this connection url should be possible | 19:31 |
| *** karimb has quit IRC | 19:45 | |
| dmsimard | berendt: hm, but current working directory relative to what ? | 20:35 |
| dmsimard | berendt: to where you are running ansible? to where ara is installed? | 20:35 |
| berendt | dmsimard: i think sqlite:///ara.sqlite means: use ara.sqlite in cwd | 20:35 |
| berendt | ara/ansible installed inside a venv | 20:35 |
| dmsimard | berendt: behavior for that is not clear to me, I mean I undertand that if you have a .py file in cwd and you execute it from there it'll properly use it in cwd | 20:36 |
| dmsimard | but when used from ansible/ara /me shrugs | 20:36 |
| berendt | as a workaround i set the ara_database env var in my jenkins job to an absolute path pointing to a dir inside the workspace | 20:37 |
| dmsimard | okay | 20:38 |
| berendt | i thought that it is possible to say: create the sqlite file in the directory where i am calling ansible-playbook from | 20:38 |
| berendt | this way it would be possible to set the parameter inside ansible.cfg without using the env vars | 20:38 |
| dmsimard | You could probably create your ansible.cfg dynamically ? | 20:39 |
| berendt | yes, this would be an other way to workaround this issue | 20:39 |
| dmsimard | sort of like how it's shown here http://ara.readthedocs.io/en/latest/configuration.html#using-ansible-cfg | 20:39 |
| berendt | i try to recude as many dynamic generated files as possible | 20:40 |
| berendt | an other issue: 20:27:07 UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 139: ordinal not in range(128) | 20:40 |
| berendt | i used some german characters inside a playbook | 20:40 |
| berendt | got this error after calling ara generate | 20:40 |
| dmsimard | hmm | 20:56 |
| dmsimard | I've seen that happen before on breaking whitespace characters | 20:56 |
| dmsimard | I have an idea on how to fix it, hang on. | 20:58 |
| *** openstackgerrit has joined #ara | 21:32 | |
| openstackgerrit | David Moreau Simard proposed openstack/ara: Ensure we're dealing with decoded utf-8 strings for file formatting https://review.openstack.org/425932 | 21:32 |
| dmsimard | berendt: that should do it ^ | 21:33 |
| dmsimard | I was able to reproduce the issue with a breaking whitespace character, applied the fix and then tried again and it worked | 21:33 |
| berendt | can you test with ä, ö, ü | 21:33 |
| dmsimard | berendt: sure, hang on | 21:34 |
| dmsimard | berendt: funny story, it's damn near impossible to test encoding related issue in py.test unit tests :/ | 21:35 |
| dmsimard | oh, I just had an idea on how to test it .. I'll try after | 21:35 |
| dmsimard | berendt: yeah those characters are fine | 21:36 |
| berendt | nice, thanks :) | 21:37 |
| dmsimard | berendt: there's one last thing I want to merge and I'll cut a release with that bugfix included | 21:38 |
| berendt | it is not urgent, i simply removed all special chars for the moment | 21:38 |
| berendt | it is possible to write ae instead of ä | 21:38 |
| dmsimard | bah, there's a couple bugfixes due for release | 21:39 |
| openstackgerrit | David Moreau Simard proposed openstack/ara: Ensure we're dealing with decoded utf-8 strings for file formatting https://review.openstack.org/425932 | 21:40 |
| openstackgerrit | David Moreau Simard proposed openstack/ara: Ensure we're dealing with decoded utf-8 strings for file formatting https://review.openstack.org/425932 | 21:41 |
| dmsimard | berendt: there ^ unit testing for encoding issue sucks but I added it to an integration test. | 21:41 |
| dmsimard | thanks for poking me about it, I've been meaning to fix it for a while | 21:42 |
| dmsimard | My keyboard layout has me typing breaking whitespace characters all the time :( | 21:43 |
| openstackgerrit | David Moreau Simard proposed openstack/ara: Ensure we're dealing with decoded utf-8 strings for file formatting https://review.openstack.org/425932 | 21:44 |
| berendt | dmsimard: do you know/use molecule? | 21:46 |
| dmsimard | berendt: I've heard a lot about it but I haven't had the opportunity of using it yet | 21:47 |
| dmsimard | I happen to already have tooling to re-run things in ephemeral environments so I need another excuse to spend time with it :) | 21:49 |
| berendt | dmsimard: https://github.com/metacloud/molecule/issues/728 | 21:52 |
| berendt | ;) | 21:52 |
| dmsimard | hah | 21:52 |
| dmsimard | thanks for the advertising :) | 21:52 |
| berendt | i use molecule for the test of my ansible roles | 21:52 |
| berendt | i think custom integration of ara by using a custom ansible cfg is already possible | 21:53 |
| berendt | but would be nice to have a native integration | 21:53 |
| dmsimard | I need to revive my patch to add it in Zuul natively :) | 21:53 |
| berendt | is it possible to only output relative paths in the visualisation? e.g. playbook.yml instead of /var/jenkins_home/workspace/aio-reset-node/playbook.yml | 21:57 |
| dmsimard | berendt: not currently | 22:03 |
| dmsimard | berendt: I think I had an idea around that, I forgot | 22:04 |
| dmsimard | berendt: ah there you got https://storyboard.openstack.org/#!/story/2000817 | 22:04 |
| dmsimard | I'll take it into account in the upcoming UI refactor | 22:04 |
| dmsimard | mnaser: fyi I know you use molecule too https://github.com/metacloud/molecule/issues/728 :) | 22:08 |
| mnaser | oh we already do this dmsimard | 22:09 |
| mnaser | i'll find time for a comment | 22:09 |
| dmsimard | I think it's about adding it natively or something to that effect | 22:09 |
| * dmsimard needs to find time to try molecule one day | 22:09 | |
| mnaser | dmsimard the time you spend to get it running will result in many hours saved | 22:09 |
| mnaser | tho i think molecule is becoming *kinda* useless with molecule v2 | 22:10 |
| mnaser | it pretty much becomes a scaffolding tool and uses the openstack modules to create servers etc | 22:10 |
| *** themurph has quit IRC | 23:20 | |
| *** themurph has joined #ara | 23:28 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!