| *** karimb has quit IRC | 01:30 | |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Complete user interface rewrite https://review.openstack.org/438118 | 01:40 |
|---|---|---|
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Complete user interface rewrite https://review.openstack.org/438118 | 01:49 |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Complete user interface rewrite https://review.openstack.org/438118 | 02:24 |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Complete user interface rewrite https://review.openstack.org/438118 | 02:28 |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Complete user interface rewrite https://review.openstack.org/438118 | 02:37 |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Complete user interface rewrite https://review.openstack.org/438118 | 02:46 |
| openstackgerrit | Merged openstack/ara master: Complete user interface rewrite https://review.openstack.org/438118 | 02:55 |
| *** themurph has quit IRC | 10:37 | |
| *** themurph has joined #ara | 14:21 | |
| *** themurph has quit IRC | 14:27 | |
| *** themurph has joined #ara | 14:53 | |
| *** tbielawa has joined #ara | 15:38 | |
| *** karimb has joined #ara | 18:05 | |
| *** tbielawa is now known as tbielawa|lunch | 18:33 | |
| -dmsimard- 0.12.0 which includes proper WSGI support as well as the UI rewrite was just tagged and released to pypi | 18:46 | |
| dmsimard | Release notes: https://github.com/openstack/ara/releases/tag/0.12.0 | 18:47 |
| *** dmsimard changes topic to "ARA - Ansible Run Analysis: Making playbook runs easier to visualize, understand and troubleshoot | Git: https://github.com/openstack/ara - Docs: https://ara.readthedocs.io/en/latest/ - Latest version: 0.12.0: https://github.com/openstack/ara/releases/tag/0.12.0 | This channel is logged: http://eavesdrop.openstack.org/irclogs/%23ara/" | 18:47 | |
| *** dmsimard changes topic to "ARA - Ansible Run Analysis: Making playbook runs easier to visualize, understand and troubleshoot | Git: https://github.com/openstack/ara - Docs: https://ara.readthedocs.io/en/latest/ - Latest version: 0.12.0: https://github.com/openstack/ara/releases/tag/0.12.0 | This channel is logged: http://eavesdrop.openstack.org/irclogs/%23ara/" | 18:47 | |
| *** tbielawa|lunch is now known as tbielawa | 19:34 | |
| tbielawa | ooooooo | 20:12 |
| tbielawa | new ara you say | 20:12 |
| dmsimard | yessir | 20:14 |
| dmsimard | tbielawa: some sample reports in my email here http://lists.openstack.org/pipermail/openstack-dev/2017-March/113181.html | 20:14 |
| tbielawa | hm | 20:16 |
| tbielawa | example report is confusing me at first. | 20:16 |
| tbielawa | clicking the playbook path brings a modal, heh. | 20:17 |
| * tbielawa keeps poking around | 20:17 | |
| * tbielawa looking for single page run report view | 20:19 | |
| dmsimard | tbielawa: what do you mean by single page run report view ? | 20:22 |
| tbielawa | dmsimard: .11 i could click a playbook on the runs list and it brought me to a new page that had all that functionality we see now, but on its own page. | 20:22 |
| tbielawa | dmsimard: let me rephrase | 20:22 |
| tbielawa | dmsimard: how does one provide a direct link to a playbook analysis now? | 20:22 |
| tbielawa | a single one, like you might want to see after your CI runs $jobs | 20:23 |
| tbielawa | report for jenkins #1337 [ARA Analysis Link] | 20:23 |
| dmsimard | tbielawa: how were you generating those links in the first place ? | 20:24 |
| dmsimard | are you using ara in a central/aggregated fashion and then linking to the specific report of the job ? | 20:24 |
| tbielawa | dmsimard: in .11 i open up http://127.0.0.1:9009/ and click the playbook name/path under the "Playbook" column | 20:24 |
| tbielawa | this is me running locally. what i was trying to convey w/ my example is that being able to link to a specific report from anywhere is valuable | 20:25 |
| dmsimard | Right, so there is a new parameter which is not exposed everywhere yet | 20:25 |
| dmsimard | http://ara.readthedocs.io/en/latest/configuration.html#ara-playbook-override | 20:26 |
| tbielawa | front page: https://gyazo.com/eb41bcf35dfe1930743e09be134945dd then click that top playbook link and it brings you to https://gyazo.com/fe54df67ce6569498f8191c53dbb0d14 | 20:26 |
| tbielawa | a unique page | 20:26 |
| tbielawa | linkable | 20:26 |
| * tbielawa looks @ doclink | 20:26 | |
| dmsimard | ARA_PLAYBOOK_OVERRIDE will load the web app with only the specified (comma delimited) list of playbooks | 20:26 |
| tbielawa | hm | 20:26 |
| tbielawa | I see | 20:27 |
| tbielawa | soooooooooo, from this I can know that directly linking to one of the playbooks available through the server view is no longer a feature? | 20:28 |
| tbielawa | rather, one must generate html w/ a parameter for a playbook? | 20:28 |
| dmsimard | ultimately it's a parameter recognized by the web app | 20:28 |
| dmsimard | so you can use it if you run the local dev server and it'll work too | 20:28 |
| dmsimard | But yes, I recognize I had to do some tradeoffs with "permalinkability" -- that said, maybe I could expose that override parameter over a GET param in the URL or something. Need to think a little bit about it. | 20:29 |
| tbielawa | dmsimard: and to be expressly clear, in 0.12 I can not directly link to a playbook report unless I use gen html with a playbook ID and that parameter | 20:29 |
| dmsimard | tbielawa: not currently, no. | 20:29 |
| tbielawa | dmsimard: thank you for clearing that up. I understand it better now :) | 20:29 |
| tbielawa | i like the things you did implement though that aren't related to that. Like we tlaked about before, the sorting is a much desired feature i've wanted :) | 20:30 |
| dmsimard | tbielawa: The problem with "permalink" ability that was all over the place before (for playbooks, for hosts, for tasks, for task results, for files, everything) is that you could easily end up with thousands of file to render when generating static pages -- this was not just slow but it was heavy | 20:30 |
| tbielawa | dmsimard: oh man | 20:30 |
| tbielawa | i think i wound up with almost 200MiB of html pages when I generated HTML before for a single run of the openshift installer :) | 20:31 |
| dmsimard | I explained some of the rationale behind the UI changes in the changelog https://github.com/openstack/ara/releases/tag/0.12.0 | 20:31 |
| dmsimard | and I'll dig a little bit deeper in an upcoming blog post | 20:31 |
| * tbielawa click | 20:31 | |
| tbielawa | LOL | 20:32 |
| tbielawa | I see you noticed that too in the report. the openstack-ansible one anyway. we have about as many tasks and results in our runs. that long time there | 20:32 |
| tbielawa | heh | 20:32 |
| tbielawa | and those are your gzipped sizes@! | 20:32 |
| *** rteague has joined #ara | 20:32 | |
| dmsimard | So that one has 1500something tasks | 20:33 |
| dmsimard | and 1700 or so results | 20:33 |
| dmsimard | but they have this one series of playbooks that has over 7500 results | 20:33 |
| tbielawa | yeah, my last playbook run (10 minutes ago) says 807 OK, 46 Changed, 0 failed, and 1421 skipped | 20:33 |
| dmsimard | now, I'm not overoptimizing for their use case but I struck some decent middle ground | 20:33 |
| tbielawa | that's a lotta results | 20:33 |
| tbielawa | yeah I think this is a good step forward. Not sure if I'll update to this release yet, I really like the direct-link feature too much to change right now. I can more easily compare run results if i just open a report in a different tab | 20:34 |
| tbielawa | we're doing actual analysis of our overhaul/refactoring right now so having that available is pretty important to me. | 20:34 |
| dmsimard | I think you'll like not only the fact that the tables are sortable now but you can also search things | 20:35 |
| tbielawa | ctrl+f? heh | 20:35 |
| dmsimard | you can search hosts based on some limited amount of facts or tasks based on their results | 20:35 |
| dmsimard | before this UI revamp, it wasn't possible to see an aggregated list of host results | 20:35 |
| dmsimard | you could only see all the tasks and then only the task results for a specific host | 20:35 |
| dmsimard | now they're all available in that single table | 20:36 |
| tbielawa | i did enjoy that feature. i found it useful to open the 'CHANGED' results tab for a host to compare between versions. We're trying to get more idempotent so that was my biggest use case. | 20:36 |
| tbielawa | i suppose though I could always just open two tabs up now | 20:37 |
| tbielawa | and then go to each playbook thing and get similar results | 20:37 |
| tbielawa | I understand from the docs you linked how I can get similar functionality, just different looking :) | 20:37 |
| tbielawa | sad about the no direct-link stuff, but, ya know, change happens | 20:37 |
| tbielawa | and there's work-arounds | 20:38 |
| dmsimard | I hear you, for the time being you can pin to 0.11 while I figure something out | 20:38 |
| rteague | FWIW, I liked the last 10 runs view. I upgraded to .12 and after several minutes it never returned when clicking on Playbooks. I've got 770+ runs recorded. I guess it was taking a while to pull them all from the DB. | 20:44 |
| tbielawa | :D | 20:45 |
| dmsimard | rteague: interesting, are things not paginated ? Should only have those 10 latest playbooks in the first page | 20:45 |
| rteague | dmsimard: well, it never returned, so I didn't realize it should be paginated. I guess it could be something else. | 20:46 |
| dmsimard | rteague: yeah, for example here there's 14 playbooks in all http://logs.openstack.org/24/396324/9/check/gate-openstack-ansible-openstack-ansible-ceph-centos-7-nv/f811e1a/logs/ara/reports/index.html | 20:47 |
| rteague | I reverted back to .11 for now. I'll give .12 a go again to see if something else was up. | 20:47 |
| dmsimard | the extra 4 are on the second page | 20:47 |
| dmsimard | rteague: ack, let me know if you figure it out | 20:47 |
| *** karimb has quit IRC | 21:37 | |
| *** karimb has joined #ara | 21:38 | |
| tbielawa | dmsimard: here's a perfect example of what i'm using the single-page view for: https://github.com/openshift/openshift-ansible/pull/3517#issuecomment-283792819 | 21:48 |
| dmsimard | tbielawa: so use case is diffing ? | 21:49 |
| tbielawa | dmsimard: um. more like an illustration/example of what I was describing earlier. What I found valuable about the easy-to-get individual result views from 0.11. | 21:49 |
| dmsimard | fair, I get requests for being able to diff two playbook runs often which is why I asked :) | 21:50 |
| tbielawa | :D | 21:51 |
| tbielawa | oh | 21:51 |
| tbielawa | i never thought of that as a first-class feature | 21:51 |
| tbielawa | that person is on point! | 21:51 |
| tbielawa | persons | 21:51 |
| dmsimard | yeah, it'll be complicated to implement, I don't know how I will do it yet | 21:52 |
| tbielawa | dmsimard: curious, do you get excited (in a nerdy way) like I do about interesting problems like that? | 21:52 |
| dmsimard | but it's probably going to be the next big feature | 21:52 |
| dmsimard | tbielawa: haha, it depends .. I'm excited until it gets frustrating :p | 21:53 |
| dmsimard | I lose patience fast on some things | 21:53 |
| tbielawa | lol | 21:53 |
| dmsimard | like the recursive function that we need to show files properly in the new UI | 21:53 |
| dmsimard | guess what, 0.12 shipped without it | 21:53 |
| * tbielawa scratches chin | 21:54 | |
| tbielawa | dmsimard: the recursive function, right? it didn't ship? | 21:56 |
| dmsimard | yeah, I gave up and wrote a literal TODO in the UI | 21:56 |
| dmsimard | http://i.imgur.com/Cz2V9RK.png | 21:57 |
| tbielawa | i admit that when I saw that panel there I went 'ick', that needs to move. i only read the first three bullets. i'm glad I held my toungue | 21:59 |
| tbielawa | haha | 21:59 |
| tbielawa | i had no idea why it was shaped like that. haha. | 21:59 |
| dmsimard | tbielawa: I want to do something like that: https://www.patternfly.org/pattern-library/widgets/#bootstrap-tree-view | 22:00 |
| dmsimard | but it's not straightforward to build the structure it needs | 22:00 |
| tbielawa | slick | 22:00 |
| tbielawa | dmsimard: see ya later | 22:07 |
| *** tbielawa has quit IRC | 22:08 | |
| *** joa_ has joined #ara | 22:23 | |
| joa_ | Hi | 22:23 |
| dmsimard | o/ | 22:24 |
| *** joa_ has left #ara | 22:25 | |
| *** karimb has quit IRC | 23:59 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!
