*** harlowja has quit IRC | 00:58 | |
*** dougbtv__ has joined #ara | 02:15 | |
*** dougbtv_ has quit IRC | 02:15 | |
*** bcoca has quit IRC | 02:17 | |
*** dcreno has quit IRC | 02:49 | |
*** dcreno has joined #ara | 02:59 | |
*** dcreno has quit IRC | 03:03 | |
*** dcreno has joined #ara | 03:11 | |
*** dcreno has quit IRC | 03:15 | |
*** dcreno has joined #ara | 03:22 | |
*** dcreno has quit IRC | 03:26 | |
*** dcreno has joined #ara | 03:28 | |
*** dcreno has quit IRC | 03:32 | |
*** larsks has quit IRC | 03:34 | |
*** larsks has joined #ara | 03:35 | |
*** dougbtv__ has quit IRC | 03:44 | |
*** rvgate has quit IRC | 03:44 | |
*** jmccrory has quit IRC | 03:44 | |
*** berendt_ has quit IRC | 03:44 | |
*** zxiiro has quit IRC | 03:44 | |
*** tristanC has quit IRC | 03:44 | |
*** harlowja has joined #ara | 04:17 | |
*** harlowja has quit IRC | 04:20 | |
*** sshnaidm_pto has joined #ara | 06:35 | |
*** gvincent has quit IRC | 07:09 | |
*** gvincent has joined #ara | 07:11 | |
*** dougbtv__ has joined #ara | 07:47 | |
*** rvgate has joined #ara | 07:47 | |
*** jmccrory has joined #ara | 07:47 | |
*** berendt_ has joined #ara | 07:47 | |
*** tristanC has joined #ara | 07:47 | |
*** zxiiro has joined #ara | 07:47 | |
*** rvgate has quit IRC | 07:49 | |
*** rvgate has joined #ara | 07:50 | |
*** jparrill has joined #ara | 08:19 | |
*** sshnaidm_pto has quit IRC | 08:29 | |
*** sshnaidm_pto has joined #ara | 08:30 | |
*** sshnaidm_pto has quit IRC | 08:36 | |
*** sshnaidm_pto has joined #ara | 08:51 | |
hughsaunders | dmsimard: I see you already talked to odyssey4me about this, and the static thing is not long term. I'll give up trying to get that to work with swift. I really should have remembered from last time that this doesn't work and isn't easy to fix.. | 09:05 |
---|---|---|
*** jparrill has quit IRC | 09:20 | |
*** sshnaidm_pto has quit IRC | 09:20 | |
hughsaunders | Turns out swift/cloudfiles can redirect / --> /index.html. By setting X-Container-Meta-Web-Index: index.html. I was already setting that header but on a request to the wrong endpoint #doh. | 09:59 |
openstackgerrit | Artem Goncharov proposed openstack/ara master: repair apache mod_wsgi configuration https://review.openstack.org/571699 | 11:10 |
*** dougbtv__ is now known as dougbtv | 11:11 | |
*** dcreno has joined #ara | 11:25 | |
*** sshnaidm_pto has joined #ara | 11:26 | |
*** dcreno has quit IRC | 11:26 | |
openstackgerrit | Artem Goncharov proposed openstack/ara master: repair apache mod_wsgi configuration https://review.openstack.org/571699 | 11:46 |
*** jlozadad has joined #ara | 12:06 | |
*** gtema has joined #ara | 12:24 | |
*** dcreno has joined #ara | 12:46 | |
*** dcreno has quit IRC | 12:50 | |
*** dcreno has joined #ara | 12:50 | |
*** ara-slack1 has quit IRC | 13:04 | |
*** ara-slack has joined #ara | 13:04 | |
*** ara-slack has joined #ara | 13:04 | |
*** ChanServ sets mode: +v ara-slack | 13:04 | |
*** tbielawa has joined #ara | 13:10 | |
*** gvincent has quit IRC | 13:13 | |
*** gvincent has joined #ara | 13:24 | |
*** themurph has joined #ara | 13:33 | |
dmsimard | hughsaunders: hey o/ sorry I missed you | 14:39 |
dmsimard | yeah I discussed this with odyssey4me yesterday | 14:39 |
hughsaunders | dmsimard: no problem, it now works :DD | 14:39 |
dmsimard | I did mention that when I tried out s3, I had to do the same index thing -- they had a setting for it | 14:40 |
hughsaunders | yeah | 14:40 |
dmsimard | good thing it's there for cloudfiles too | 14:40 |
dmsimard | I wonder if it's worth documenting | 14:40 |
hughsaunders | I thought was enabling the index thing, but I was hitting the wrong endpoint with my request.. | 14:40 |
hughsaunders | fixed that and now all is good | 14:40 |
hughsaunders | though we'll have to rethink if the static generator is going away | 14:40 |
dmsimard | hughsaunders: yeah, it's a bit early to tell if 1.0 will provide something like that, it depends if there is something in the django ecosystem that's able to do it fairly easily (and efficiently) | 14:41 |
dmsimard | The need is to be able to easily upload your reports and browse them without complicated dependencies, we'll make that happen | 14:42 |
dmsimard | Not sure how yet -- the sqlite middleware approach will likely be ported | 14:42 |
hughsaunders | https://github.com/datadesk/django-bakery | 14:44 |
hughsaunders | dmsimard: out of interest why the move to django? | 14:46 |
dmsimard | hughsaunders: I've put a ton of work in 1.0 based on the current backend with flask by using flask-restful: https://github.com/openstack/ara/tree/feature/1.0 | 14:47 |
dmsimard | However, I was convinced to switch to django -- not so much for django itself but rather for django-rest-framework which is ridiculously good, maintained and developed. flask-restful is not very active. | 14:48 |
dmsimard | AWX/Tower uses django/django-rest-framework as well but that didn't really influence the decision | 14:49 |
dmsimard | You can see an early version of the django backend/model/API here: https://github.com/openstack/ara-server | 14:49 |
hughsaunders | dmsimard: might as well switch to vuejs while you're at it :p | 14:50 |
*** tbielawa is now known as tbielawa|brb | 14:51 | |
dmsimard | hughsaunders: I'm not 100% sure we've made up our mind. gvincent did bring up vuejs but we might end up using react because that's where patternfly (our css framework) is focusing I think. | 14:53 |
dmsimard | I suck at frontend though, but there's plenty of opportunities to contribute if you'd like :) | 14:53 |
gvincent | :D | 14:53 |
gvincent | as long as you do not ask me to do Angular | 14:54 |
gvincent | i'm fine | 14:54 |
dmsimard | I honestly do not have a personal preference in JS frameworks, so long as we pick something that's maintained and hopefully simple enough | 14:55 |
hughsaunders | Heh, I know nothing about frontend, but I'm doing a project in vue at the moment because it looked the least daunting of the current options! | 14:55 |
dmsimard | hughsaunders: exactly my feelings :p | 14:55 |
dmsimard | hughsaunders: so anyway, the general idea with 1.0 is that the core will be very focused: A database model and an API. You'll then have things leveraging that API, whether it's the callbacks, the web UI, etc. If someone wants to leverage the API to generate a static report of sorts I'm not going to stop them | 14:59 |
dmsimard | The API will open up a lot of interesting opportunities I think. Programmatically retrieving failed playbooks/tasks for monitoring/alerting, using metrics, etc. | 15:00 |
*** Bakey has joined #ara | 15:10 | |
hughsaunders | dmsimard: sounds interesting, so we could standup one api instance somewhere and point all our playbook runs at it? | 15:12 |
hughsaunders | Would need a way of grouping playbook executions - eg all playbooks associated with a Jenkins build for example. | 15:13 |
dmsimard | hughsaunders: yup, there will be a feature to logically group playbooks | 15:17 |
dmsimard | hughsaunders: in the context of zuul, this could be used to define the "pre", "run" and "post" playbooks, for example.. or "setup", "install", "configure", etc. it'll be generic. | 15:17 |
hughsaunders | dmsimard: bonus points for a generalised tag system | 15:18 |
dmsimard | that's the intent, yes | 15:18 |
dmsimard | right now everything ties back to a playbook in the model but we'd like to have a layer on top to arbitrarily group playbooks together. | 15:19 |
hughsaunders | Yass, sounds great. Thanks for all your work on ARA :) | 15:19 |
dmsimard | people could use this to group different playbooks into a single report or, for example, have a generic tag and people interested in that tag can follow that or something | 15:20 |
dmsimard | i.e, "openstack rdo ci" and then you just keep on piling things into that tag so you know where to find those reports | 15:20 |
dmsimard | this would be very generic so people can do what they can | 15:20 |
dmsimard | I haven't found exactly how tags will be passed -- i.e, how to identify that a playbook is part of a tag | 15:21 |
dmsimard | an env var has been suggested but I find that a bit clunky | 15:22 |
hughsaunders | Nice thing about an env var, is that they are easy to inject without modifying stuff - so CI environments could inject tags. | 15:26 |
dmsimard | yeah it's probably okay to have it as an option but perhaps not the *only* option | 15:27 |
dmsimard | I've been thinking about a simple ansible var | 15:27 |
dmsimard | could be in a vars file, or just -e "ara_tag=something" | 15:28 |
Bakey | I think that's the most flexible | 15:30 |
Bakey | let's you set a tag at runtime | 15:30 |
Bakey | s/let's/lets | 15:30 |
*** tbielawa|brb is now known as tbielawa | 15:32 | |
dmsimard | right, ara would pick up whatever tag was the last one used during that playbook | 15:32 |
dmsimard | so you could tag things dynamically if needed | 15:32 |
dmsimard | i.e, block/rescue -> set_fact: ara_tag=rescued | 15:32 |
dmsimard | I suppose we could/should support multiple tags | 15:37 |
ara-slack | <dbpiv> when we start ARA via `ara-manage runserver`, is there a way force debug on? I am getting more errors and it would be helpful to see what the error is? Currently all I know is that I am getting 500 errors…. | 15:56 |
*** dcreno has quit IRC | 15:57 | |
*** bcoca has joined #ara | 16:04 | |
*** jrist has quit IRC | 16:08 | |
*** harlowja has joined #ara | 16:11 | |
*** jrist has joined #ara | 16:25 | |
*** gtema has quit IRC | 16:29 | |
*** sshnaidm_pto has quit IRC | 16:44 | |
*** tbielawa is now known as tbielawa|lunch | 17:11 | |
*** harlowja has quit IRC | 17:22 | |
*** gvincent has quit IRC | 17:23 | |
hughsaunders | multiple tags ++ | 17:31 |
*** myoung is now known as myoung|bbl | 17:47 | |
*** dcreno has joined #ara | 17:54 | |
ara-slack | <dbpiv> Never mind I found it… I knew I had seen it somewhere. | 18:07 |
*** harlowja has joined #ara | 18:14 | |
*** tbielawa|lunch is now known as tbielawa | 18:31 | |
ara-slack | <dbpiv> and in case anyone stumbles on to this. It is simply `ara-manage runserver --debug` as illustrated on https://ara.readthedocs.io/en/latest/webserver.html | 18:41 |
*** openstackgerrit has quit IRC | 19:04 | |
*** tbielawa is now known as tbielawa|mtg | 19:19 | |
*** hwoarang has quit IRC | 19:29 | |
*** spiette has quit IRC | 19:30 | |
*** spiette has joined #ara | 19:30 | |
*** hwoarang has joined #ara | 19:30 | |
*** myoung|bbl is now known as myoung | 19:30 | |
*** hwoarang has quit IRC | 19:35 | |
*** hwoarang has joined #ara | 19:36 | |
*** openstackgerrit has joined #ara | 19:55 | |
openstackgerrit | David Moreau Simard proposed openstack/ara master: Add support for running the main WSGI application from a virtualenv https://review.openstack.org/567259 | 19:55 |
openstackgerrit | David Moreau Simard proposed openstack/ara master: Support specifying virtualenv path with or without activate_this.py https://review.openstack.org/567280 | 19:55 |
openstackgerrit | David Moreau Simard proposed openstack/ara master: WSGI: Move the application creation within the app method https://review.openstack.org/567871 | 19:55 |
openstackgerrit | David Moreau Simard proposed openstack/ansible-role-ara master: Default the ARA installation to a virtualenv https://review.openstack.org/567094 | 19:56 |
openstackgerrit | David Moreau Simard proposed openstack/ansible-role-ara master: Make sure the ARA systemd service is not running with apache https://review.openstack.org/567435 | 19:56 |
dmsimard | ssbarnea: still around ? | 20:06 |
openstackgerrit | David Moreau Simard proposed openstack/ara master: Allow junit reports task result overriding https://review.openstack.org/571430 | 20:08 |
dmsimard | ssbarnea: ^ avoided a nitpick on the commit message to point to upstream docs rather than infrared | 20:09 |
openstackgerrit | Merged openstack/ara master: Documented nginx config for static serving https://review.openstack.org/571220 | 20:17 |
*** dougbtv_ has joined #ara | 20:43 | |
*** dougbtv has quit IRC | 20:45 | |
*** tbielawa|mtg is now known as tbielawa|g0n3 | 20:50 | |
*** tbielawa|g0n3 has quit IRC | 20:51 | |
*** dcreno has quit IRC | 20:58 | |
*** dcreno has joined #ara | 21:00 | |
openstackgerrit | Merged openstack/ara master: Add support for running the main WSGI application from a virtualenv https://review.openstack.org/567259 | 21:03 |
*** dcreno has quit IRC | 21:05 | |
openstackgerrit | Merged openstack/ara master: Support specifying virtualenv path with or without activate_this.py https://review.openstack.org/567280 | 21:09 |
openstackgerrit | Merged openstack/ara master: WSGI: Move the application creation within the app method https://review.openstack.org/567871 | 21:11 |
*** themurph has quit IRC | 21:17 | |
*** dcreno has joined #ara | 21:25 | |
*** dcreno has quit IRC | 21:29 | |
openstackgerrit | David Moreau Simard proposed openstack/ansible-role-ara master: Add support for installing from source https://review.openstack.org/571831 | 21:32 |
*** dcreno has joined #ara | 21:32 | |
*** dcreno has quit IRC | 21:32 | |
openstackgerrit | David Moreau Simard proposed openstack/ansible-role-ara master: WIP: Zuul jobs experiments https://review.openstack.org/557933 | 21:32 |
*** dcreno has joined #ara | 21:32 | |
openstackgerrit | David Moreau Simard proposed openstack/ansible-role-ara master: WIP: Zuul jobs experiments https://review.openstack.org/557933 | 21:43 |
*** dcreno has quit IRC | 21:49 | |
*** dcreno has joined #ara | 21:50 | |
openstackgerrit | David Moreau Simard proposed openstack/ansible-role-ara master: WIP: Zuul jobs experiments https://review.openstack.org/557933 | 22:32 |
*** harlowja has quit IRC | 23:18 | |
*** Bakey has quit IRC | 23:57 |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!