| *** etienne has quit IRC | 02:48 | |
| *** jrist has quit IRC | 04:54 | |
| *** jrist has joined #ara | 05:07 | |
| *** gvincent has joined #ara | 07:58 | |
| *** herald85 has joined #ara | 08:32 | |
| *** logan- has quit IRC | 09:21 | |
| *** themroc has joined #ara | 10:06 | |
| *** sshnaidm|afk is now known as sshnaidm | 10:31 | |
| *** openstackgerrit has quit IRC | 11:05 | |
| ara-slack | jeansebastien.guyot: Hi ! | 13:16 |
|---|---|---|
| ara-slack | jeansebastien.guyot: ara-manage-db show give me : | 13:19 |
| ara-slack | Rev : <xxxxx> (head) | 13:19 |
| ara-slack | Parent : <xxxxx> | 13:19 |
| ara-slack | Path : </usr/lib/python2.7/site-packages/ara/db/versions/005_ansible_metadata.py | 13:19 |
| ara-slack | ansible_metadata | 13:19 |
| ara-slack | revision ID : <xxxxxxx> | 13:19 |
| ara-slack | Revises : <xxxxxx> | 13:19 |
| ara-slack | Create Date : 2017-05-02 13:34:41.150156 | 13:19 |
| ara-slack | nothing wrong in this ... | 13:19 |
| ara-slack | jeansebastien.guyot: have you a limit from playbook number ? | 13:42 |
| ara-slack | actually, for exemple i have more 125000 rows in tasks | 13:42 |
| ara-slack | dmsimard: No, there isn't any hard limits but the more playbooks there are, the slower things are likely to get | 14:09 |
| ara-slack | jeansebastien.guyot: ok ... | 14:36 |
| ara-slack | I have test 16.0.2 version from scratch, it's OK ... but i want my old data .... | 14:36 |
| ara-slack | between 0.14.6 and 0.15.1 you have add SQLALCHEMY ... Do you think that the problem could come from? | 14:36 |
| ara-slack | dmsimard: sqlalchemy was always in ara, that's how flask talks to the database | 14:39 |
| ara-slack | dmsimard: perhaps the sql migration failed to be applied, timed out or was interrupted | 14:39 |
| ara-slack | dmsimard: The migrations are all here: https://github.com/openstack/ara/tree/master/ara/db/versions | 14:40 |
| ara-slack | dmsimard: Have you tried downgrading ? In fact, what have you tried ? | 14:40 |
| ara-slack | jeansebastien.guyot: I just try from 0.14.6 to execute pip install ara --upgrade | 14:56 |
| ara-slack | jeansebastien.guyot: before for old version this was ok with this | 14:57 |
| ara-slack | dmsimard: ok, do yo have a backup or your database ? | 15:11 |
| ara-slack | jeansebastien.guyot: actually not ... | 15:14 |
| ara-slack | jeansebastien.guyot: i compare schema of two database (by schemaspy), its the same ... | 15:15 |
| ara-slack | jeansebastien.guyot: seem same | 15:15 |
| ara-slack | dmsimard: Ok, I would at least dump/backup your database at this point | 15:18 |
| ara-slack | jeansebastien.guyot: I will try to make a dump of the base in 0.14.6 and to import it in the base 0.16.2 without the creation of table ... | 15:19 |
| ara-slack | dmsimard: ok, sorry that I can't be more helpful .. let me know if you end up figuring what was the issue | 15:20 |
| ara-slack | jeansebastien.guyot: no problem :slightly_smiling_face: | 15:20 |
| dmsimard | apollo13: so I learned that apparently namespaced packages are directly installable by pip | 15:40 |
| apollo13 | yes, what did you think? | 15:40 |
| dmsimard | well, right now we have "ara", "ara-server", "ara-plugins" and "ara-clients" | 15:40 |
| dmsimard | but we could have "ara", "ara.server", "ara.plugins", "ara.clients" | 15:41 |
| apollo13 | is that a valid pypi package name? | 15:41 |
| dmsimard | it's not | 15:42 |
| dmsimard | I mean | 15:42 |
| dmsimard | the package is still ara | 15:42 |
| dmsimard | but it'd only install the namespaced one ? I don't know.. I found that somewhere on stack overflow | 15:42 |
| dmsimard | I would not really have the opportunity to test this without breaking anything, perhaps we could try the test pypi instance | 15:43 |
| dmsimard | https://packaging.python.org/guides/packaging-namespace-packages/ doesn't mention the "pip install" syntax | 15:45 |
| dmsimard | this is where I found it https://stackoverflow.com/questions/45324189/python-install-sub-package-from-package | 15:47 |
| dmsimard | I thought it would have been interesting even if just to keep "pip install ara" for people who aren't interested in the separated packages | 15:50 |
| dmsimard | but then the people who care could do pip install ara.server, for example | 15:50 |
| *** sshnaidm has quit IRC | 16:10 | |
| *** sshnaidm has joined #ara | 16:11 | |
| ara-slack | nj.chandu: Right now I have multiple jobs that run in parallel on the same host. And I am using ARA to generate static report at the end of the job. What’s happening is all the playbook executions are recorded and displayed in the static report. Is there a way to show only that specific job’s playbook report when multiple jobs are running in the same host in parallel? Thanks in advance | 16:17 |
| ara-slack | dmsimard: @nj.chandu what we are doing in the openstack community is that we have one database per job | 16:19 |
| ara-slack | dmsimard: the database is stored as part of the artifacts of the job | 16:19 |
| ara-slack | dmsimard: we do not actually generate the static HTML report because it does not scale very well | 16:20 |
| ara-slack | dmsimard: the approach is documented here: https://ara.readthedocs.io/en/latest/advanced.html | 16:20 |
| *** herald85 has quit IRC | 16:26 | |
| ara-slack | dmsimard: that's not how it works | 16:47 |
| ara-slack | dmsimard: those are meant to be used by apache | 16:47 |
| *** themroc has quit IRC | 16:48 | |
| ara-slack | dmsimard: an example workflow would be something like your job sets ARA_DATABASE to a unique location for the build (ex: $WORKSPACE/ansible.sqlite) | 16:48 |
| ara-slack | dmsimard: when you archive your logs and artifacts, you recover that database and put it inside an ara-report directory | 16:48 |
| ara-slack | dmsimard: and then apache, with that sqlite middleware, links that ara-report URL to the path of a database on disk | 16:49 |
| ara-slack | dmsimard: The limitation is that apache needs to be running on the same server where the sqlite files are hosted and the locations must match | 16:49 |
| dmsimard | in practice, it looks like this | 16:50 |
| dmsimard | http://logs.openstack.org/23/628523/1/check/ara-server-ansible-integration/7007e52/ara-report/ | 16:50 |
| dmsimard | the database file is available at http://logs.openstack.org/23/628523/1/check/ara-server-ansible-integration/7007e52/ara-report/ansible.sqlite | 16:50 |
| dmsimard | and it is located on the server at a location like /srv/logs/23/628523/1/check/ara-server-ansible-integration/7007e52/ara-report/ansible.sqlite | 16:51 |
| *** jrist is now known as jrist__ | 18:16 | |
| *** jrist has joined #ara | 18:17 | |
| *** jrist__ has quit IRC | 18:18 | |
| ara-slack | nj.chandu: This /Users/chandanjavaregowda/.ara/ansible.sqlite database doesn’t have the latest reports. Although I am able to view the latest report by ara generate html build/ and visiting the build/ dir | 18:35 |
| ara-slack | nj.chandu: I was wrong the reports are loading the HTML content but not the data. Upon execution of playbook this is the error message I’m getting ^ | 20:14 |
| *** openstackgerrit has joined #ara | 20:39 | |
| openstackgerrit | David Moreau Simard proposed openstack/ara-plugins master: Implement support for naming playbooks https://review.openstack.org/629927 | 20:39 |
| openstackgerrit | David Moreau Simard proposed openstack/ara-plugins master: Implement support for setting host aliases https://review.openstack.org/629928 | 20:39 |
| *** openstackgerrit has quit IRC | 20:50 | |
| ara-slack | dmsimard: @nj.chandu are you running multiple concurrent playbooks simultaneously ? The "database is locked" error might due to concurrent writes to the sqlite database | 21:36 |
| ara-slack | dmsimard: ansible.cfg does not expand the "~", that is from your shell (i.e bash) | 21:45 |
| ara-slack | dmsimard: if you need to rely on the ~, you can use env variables instead | 21:45 |
| ara-slack | dmsimard: FWIW, dir=~/.ara is redundant because that's the default | 21:45 |
| ara-slack | dmsimard: if you want to try a new database, you can do "export ARA_DATABASE=sqlite:////tmp/test.sqlite" | 21:46 |
| ara-slack | nj.chandu: both setting the env variable and providing absolute path didn’t help | 21:49 |
| ara-slack | nj.chandu: I see a message in ara.log not sure if that can be of any help2019-01-10 16:46:23,643 INFO ara.webapp.configure_db: Initializing new DB from scratch | 21:52 |
| ara-slack | nj.chandu: apologies setting env variable works | 21:54 |
| ara-slack | nj.chandu: I made a mistake in ls tmp/test.sqlite, it should have been ls /tmp/test.sqlite | 21:54 |
| ara-slack | nj.chandu: @dmsimard I have apache running, and the sqlite file. I have created new dir ara-report under /tmp, what should be the path to open the report with the browser? | 22:00 |
| *** zoredache has quit IRC | 22:08 | |
| *** zoredache has joined #ara | 22:10 | |
| *** openstackgerrit has joined #ara | 23:03 | |
| openstackgerrit | David Moreau Simard proposed openstack/ara-server master: Serialize the value provided by ara_record https://review.openstack.org/629946 | 23:03 |
| *** gvincent has quit IRC | 23:50 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!