| *** themurph has quit IRC | 00:21 | |
| boxrick1 | dmsimard: I have finally upgraded to the new version, seems that Chrome support seems a little broken and in the docs "Set up the Apache virtual host at /etc/apache/sites-available/ara.conf:" -> http://ara.readthedocs.io/en/latest/webserver.html should be /etc/apache2* | 01:55 |
|---|---|---|
| boxrick1 | Also did you ever add in the ability to disable tooltips? An option to default directly to the reports page would be handy! | 01:57 |
| *** karimb has joined #ara | 08:09 | |
| *** openstackgerrit has quit IRC | 08:33 | |
| *** permalac has quit IRC | 08:55 | |
| *** permalac has joined #ara | 09:00 | |
| *** permalac has quit IRC | 09:06 | |
| *** permalac has joined #ara | 09:09 | |
| *** cuongnv has joined #ara | 09:15 | |
| *** cuongnv has quit IRC | 10:38 | |
| dmsimard | boxrick1: hey thanks for spotting the typo | 11:45 |
| dmsimard | boxrick1: yes, there will eventually be a way to disable the tooltips. | 11:46 |
| boxrick1 | Ok cool :) | 11:47 |
| boxrick1 | With each load, it seems to pull in all the data for that page and each individual playbook run. | 11:50 |
| boxrick1 | In my case, this is quite massive. Even if paginated down | 11:50 |
| boxrick1 | A full page load is around 50 seconds. | 11:52 |
| boxrick1 | From SSDs and a fast hexacore Xeon server. | 11:52 |
| dmsimard | boxrick1: the home page ? Yeah, someone told me that. There's probably a way to fix that. We're doing dumb "select count(*) from <table>" | 11:58 |
| boxrick1 | If I have a poke through the code at some point and push some patches, is that the sort of thing you mind? | 11:59 |
| dmsimard | boxrick1: go for it, the more contributors the better | 12:01 |
| dmsimard | boxrick1: keep in mind that the project doesn't use the github workflow so it's a bit more involved but there are many benefits | 12:02 |
| dmsimard | There's a doc to get started: http://ara.readthedocs.io/en/latest/contributing.html | 12:02 |
| dmsimard | boxrick1: from quick googling around, it seems like a known issue that the way SQLAlchemy counts records is slow, there's some faster ways which are easy to implement so I'll try that firsr | 12:04 |
| dmsimard | first* | 12:04 |
| boxrick1 | I wonder if it simply just needs a different take on it | 12:06 |
| boxrick1 | And only request the records on click ? | 12:06 |
| *** dougbtv has joined #ara | 12:08 | |
| dmsimard | Yeah, there's different ways to approach the problem.. I'll at least improve the performance first, it's like a 5 line fix | 12:10 |
| boxrick1 | If you want a guinea pig let me know. | 12:15 |
| dmsimard | boxrick1: there will be a 0.12.4 this week, I'll cut a release candidate before | 12:17 |
| boxrick1 | Ok cool . | 12:18 |
| *** themurph has joined #ara | 12:35 | |
| leifmadsen | dmsimard: thanks again for all the help last week -- got ARA working nicely with TOAD | 13:20 |
| leifmadsen | (again :)) | 13:20 |
| dmsimard | neat | 13:20 |
| dmsimard | Does anyone run ARA in https ? This user mentions there's some links in http causing issues but I can't seem to be able to reproduce it. https://storyboard.openstack.org/#!/story/2000947 | 13:22 |
| dmsimard | mnaser: ^ he says he uses jenkins html publisher, did you have to hack anything in particular ? | 13:22 |
| mnaser | dmsimard we dont run jenkins over ssl but i think i know whats going on here | 13:23 |
| mnaser | let me see | 13:23 |
| dmsimard | It's sort of weird because as far as I know, ARA only uses relative links from the root and doesn't specify the protocol .. but I might have missed something | 13:26 |
| mnaser | looks like its happening in the modals | 13:27 |
| mnaser | so <a href="#" data-toggle="modal" data-target="#file_modal" data-load="35d04aaa-3674-47b5-8f95-ec810bef9064">/tmp/tmp.l5pXVGn45T/ansible/post-deploy.yml</a> | 13:27 |
| mnaser | i suspect that is where it poops out | 13:27 |
| *** openstackgerrit has joined #ara | 13:28 | |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Change http URLs links to https https://review.openstack.org/450219 | 13:28 |
| dmsimard | mnaser: I found ^ but it can,t be related | 13:28 |
| mnaser | yeah most likely not | 13:28 |
| mnaser | im triyig to track down the data-load stuff | 13:29 |
| dmsimard | mnaser: https://github.com/openstack/ara/blob/master/ara/templates/report_list.html#L418-L429 | 13:29 |
| leifmadsen | I planned to front ARA with traefik and https on the front with letsencrypt instead of managing multiple certs | 13:29 |
| leifmadsen | (that doesn't help solve this problem :)) | 13:30 |
| leifmadsen | just that I hadn't run into that yet | 13:30 |
| dmsimard | er no | 13:30 |
| dmsimard | mnaser: yeah what I linked is where the modals are handled | 13:31 |
| mnaser | ok so it looks the iframe src gets replaced | 13:31 |
| dmsimard | mnaser: yeah, it's a "clever" way to dynamically load the right iframe in the modal on display | 13:32 |
| dmsimard | the alternative being writing all the modals and all the iframes and throwing them at the browser | 13:32 |
| dmsimard | but perhaps there's an issue with that particular bit | 13:33 |
| mnaser | dmsimard any reason why you dont just pull the content directly in and avoid the iframe? | 13:34 |
| dmsimard | mnaser: it doesn't scale, it needs to be loaded dynamically | 13:34 |
| mnaser | yeah still load dynamically | 13:35 |
| mnaser | it'll do an ajax request, get the content, put it inside the modal | 13:35 |
| mnaser | so the same thing you're doing now, but instead of an iframe, you can do $(this).find('contaienr-div').html(content); | 13:35 |
| mnaser | and it might eliminate that weird issue as well at the same time, i guess | 13:36 |
| mnaser | what could be useful if that bug report included the source code of the playbook reports page (the one where they click the facts from) | 13:36 |
| dmsimard | mnaser: I'm not sure I follow how we can do anything else than what we're doing now without writing and loading things at page load time | 13:36 |
| mnaser | dmsimard so instead of having the iframe load the page, you download the contents (with an ajax request) and put it inside the html div (when the modal gets shown) | 13:38 |
| dmsimard | I suck at ajax stuff, probably part of why I'm having a hard time understanding :( | 13:39 |
| mnaser | its okay no worries | 13:40 |
| dmsimard | I'll try and ask for more details in the bug | 13:40 |
| mnaser | im working on a little propopsal | 13:40 |
| mnaser | i asked for details in the bug | 13:40 |
| dmsimard | oh | 13:40 |
| mnaser | and ill do this | 13:40 |
| mnaser | (getting rid of iframes that is) | 13:40 |
| dmsimard | hey if the end result is the same, it works and resolves this issue I'm down | 13:41 |
| mnaser | hm ok this is not as easy as i thought it would be | 13:47 |
| mnaser | because some things you load with a bookmark | 13:47 |
| mnaser | and that cant easily be as replicated :< | 13:47 |
| mnaser | so how do we make storyboard send us notifications.. | 13:48 |
| logan- | I'm the big reporter. Out of town today but I've been digging on it some. Is it possible that the auth cookie isn't being passed to Jenkins on the iframe load? I noticed that chrome shows a stalled/cancelled request to https:// and then it seems to attempt http:// as a fallback | 13:49 |
| logan- | Anyway based on what I have found now feel free to let it sit a while and I'll see if I can sort out the root cause this week | 13:50 |
| mnaser | logan- it could be some weird jenkins stuff to prevent csrf | 13:50 |
| mnaser | it probably blocks iframes or something like that | 13:50 |
| logan- | Yeah I had to disable CSP just to get ARA loading | 13:50 |
| logan- | None of the fonts would load with Jenkins default CSP header | 13:51 |
| mnaser | logan- https://wiki.jenkins-ci.org/display/JENKINS/XFrame+Filter+Plugin | 13:51 |
| mnaser | ithink this might be it | 13:51 |
| mnaser | i think jenkins has really locked down defaults | 13:51 |
| mnaser | logan- could be related as well - http://jenkins-ci.361315.n4.nabble.com/iframes-no-longer-displayed-after-Jnekins-upgrade-1-557-td4697705.html | 13:53 |
| logan- | Thanks. I'll give it a try when I'm back at the pc | 13:54 |
| dmsimard | mnaser: you click the star at the top of the story (to subscrbe only to that story) or the star at the top of the project (to subscribe to the whole project) | 14:02 |
| dmsimard | mnaser: you also have to make sure in your personal preferences you have email notifications enabled | 14:02 |
| mnaser | dmsimard awesome thanks, i'll have to figure this out as openstack stuff moves to it | 14:02 |
| boxrick1 | When using the new layout ( 0.12 vs 0.11 ) you can sadly no longer share direct pages as links, and when working with a failure it lives in two separate windows ( the role and the failure ) where it used to be all visible on one. | 14:02 |
| *** tbielawa has joined #ara | 14:03 | |
| dmsimard | boxrick1: yeah, so the new UI is great but we not without some sacrifices | 14:04 |
| dmsimard | s/we// | 14:04 |
| dmsimard | boxrick1: for "permalinkability" I created an issue here https://storyboard.openstack.org/#!/story/2000904 | 14:05 |
| dmsimard | mnaser, logan-: if you figure this one out, happy to include some docs on how to do it if it's a jenkins implementation issue | 14:07 |
| dmsimard | It'd be part of the different use case documentation (i.e, https://storyboard.openstack.org/#!/story/2000860 ) | 14:07 |
| mnaser | yeah i think its probably a really common use case | 14:07 |
| *** tbielawa is now known as tbielawa|mtg | 14:10 | |
| leifmadsen | sounds like a lot of us seem to be using ARA + Jenkins together | 14:16 |
| leifmadsen | I wonder if there are some "best practices" in this area that we could document together | 14:16 |
| dmsimard | There's probably different approaches.. I tend to leave artifacts out of jenkins, especially html/js files since they're such a pain to view from Jenkins due to all the security things | 14:17 |
| dmsimard | But I can see how people might want everything centralized from inside jenkins so we should definitely document some hints and gotchas | 14:19 |
| dmsimard | russellb: o/ fancy seeing you here, hello sir | 14:56 |
| russellb | :) | 14:56 |
| russellb | leifmadsen was showing me ara the other day, pretty cool. | 14:57 |
| dmsimard | I ought to send him some stickers, he's been a great advocate :) | 14:57 |
| leifmadsen | :D | 14:58 |
| leifmadsen | dmsimard: just send me the SVG and I can get some printed out at Roberts Signs and Awnings | 14:59 |
| leifmadsen | saves you the shipping | 14:59 |
| dmsimard | leifmadsen: it costs an envelope and a stamp, I should have just enough left to feed my kids | 14:59 |
| leifmadsen | lol | 15:00 |
| leifmadsen | russellb: did you even work last week, or did I just get you on a bunch of demos? :) | 15:02 |
| russellb | i was connected to irc all week | 15:02 |
| leifmadsen | lol | 15:02 |
| leifmadsen | aen. | 15:02 |
| leifmadsen | amen* | 15:02 |
| * dmsimard is connected to IRC 24/24 | 15:02 | |
| leifmadsen | you mean 24/7? :) | 15:03 |
| leifmadsen | or are you only connected for just over 3 weeks of every month? | 15:04 |
| leifmadsen | then you ragequit for a week | 15:04 |
| leifmadsen | all my meetings are done for the day, and now I have no idea what to do lol | 15:04 |
| dmsimard | ah, 24/24 doesn't translate very well to english indeed | 15:05 |
| leifmadsen | :D | 15:05 |
| *** tbielawa|mtg is now known as tbielawa | 15:12 | |
| tbielawa | lol | 16:13 |
| openstackgerrit | Merged openstack/ara master: Use Ansible's CallbackBase._dump_results to dump results https://review.openstack.org/449827 | 16:46 |
| openstackgerrit | Merged openstack/ara master: Change http URLs links to https https://review.openstack.org/450219 | 16:46 |
| *** harlowja has quit IRC | 17:18 | |
| *** harlowja has joined #ara | 17:24 | |
| *** tbielawa is now known as tbielawa|lunch | 18:36 | |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Fix typo for apache config in WSGI docs for Debian-like distros https://review.openstack.org/450422 | 19:08 |
| *** tbielawa|lunch is now known as tbielawa | 19:34 | |
| openstackgerrit | Merged openstack/ara master: Fix typo for apache config in WSGI docs for Debian-like distros https://review.openstack.org/450422 | 20:07 |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: WIP: Improve count performance https://review.openstack.org/450449 | 20:29 |
| harlowja | dmsimard qq | 20:40 |
| harlowja | if say we have multiple related ansible playbook runs | 20:40 |
| harlowja | is there anyway to have ara correlate them and show them all as 'one megabook' | 20:40 |
| harlowja | like by passing an id around | 20:40 |
| harlowja | or call it a 'tome id' | 20:40 |
| harlowja | lol | 20:40 |
| harlowja | or bookshelf | 20:41 |
| harlowja | lol | 20:41 |
| dmsimard | harlowja: megabook | 20:41 |
| dmsimard | it has a nice ring to it | 20:41 |
| harlowja | lol | 20:42 |
| * dmsimard opens upstream ansible issue | 20:42 | |
| harlowja | ya, we have distinct activations from python -> different books | 20:42 |
| dmsimard | harlowja: so, it depends on you're running ara | 20:42 |
| dmsimard | harlowja: how are the playbooks related ? | 20:42 |
| harlowja | 'deploy openstack' | 20:42 |
| dmsimard | from the same ci job or something ? | 20:42 |
| harlowja | same kind of goal, but with sub books to do each part | 20:42 |
| dmsimard | or just the same playbook file for example | 20:42 |
| harlowja | like deploy openstack.book is to big | 20:43 |
| harlowja | so we've broken it up | 20:43 |
| harlowja | (as u can imagine) | 20:43 |
| harlowja | lol | 20:43 |
| dmsimard | harlowja: is it a short lived relation ? :D | 20:44 |
| dmsimard | harlowja: like, for example, OSA has like 35 playbook runs in each of their CI job and they're somewhat grouped together I guess ? | 20:44 |
| harlowja | ya | 20:44 |
| dmsimard | I'm lazy to find a job but it looks like this http://ara-demo.dmsimard.com/index.html | 20:45 |
| harlowja | ya, hmmm | 20:45 |
| dmsimard | harlowja: are you using a centralized setup ? HOW DO YOU USE ARA | 20:46 |
| dmsimard | tell me | 20:46 |
| harlowja | :-P | 20:46 |
| harlowja | we have a chat operator that is going to kick off all this stuff (when requested) | 20:46 |
| harlowja | so that chatop container now has ara running inside of it as well | 20:47 |
| harlowja | chatop is in python so that it can do more complicated things (in code) | 20:47 |
| dmsimard | so it's a long-lived/persistent database ? | 20:47 |
| harlowja | i'd like it to be | 20:47 |
| dmsimard | i.e, it's not discarded like for a CI job/context | 20:47 |
| harlowja | right | 20:47 |
| dmsimard | ok, that makes things a bit more complicated if you want to logically group things | 20:47 |
| dmsimard | because you see, there's no feature to logically group things | 20:47 |
| harlowja | :) | 20:48 |
| harlowja | except to shove it all under ansible and sort of get a grouping via it | 20:48 |
| harlowja | that requires everything to be programmed up in ansible though :-/ | 20:48 |
| dmsimard | So there's *something* | 20:48 |
| harlowja | which doesn't feel that great for certain things (like 'deploy openstack' i don't think i can full hand it over to ansible without ansible being fully turing complete, lol) | 20:49 |
| dmsimard | harlowja: so there's https://storyboard.openstack.org/#!/story/2000920 but that's not going to help you right now | 20:50 |
| dmsimard | I had documented something somewhere, trying to find it | 20:50 |
| harlowja | ya, intersing | 20:50 |
| harlowja | just being able to pass in a 'tag_id' ('megabook_id'?) then having that located later would do it (sort of, lol) | 20:50 |
| dmsimard | harlowja: see the second bullet point here https://storyboard.openstack.org/#!/story/2000904 | 20:50 |
| dmsimard | ARA_PLAYBOOK_OVERRIDE -> http://ara.readthedocs.io/en/latest/configuration.html#ara-playbook-override | 20:51 |
| harlowja | hmmm | 20:51 |
| dmsimard | you could pseudo-code: ara playbook list -f value |grep openstackbook |awk '{print $1}' and then do something with that | 20:52 |
| dmsimard | but it's pretty ugly | 20:52 |
| dmsimard | what you probably need is really that first story I linked | 20:53 |
| harlowja | megabook | 20:53 |
| harlowja | lol | 20:53 |
| dmsimard | well, more generically put, "logical grouping/search of playbooks" | 20:53 |
| dmsimard | harlowja: what'd you like to "search" by ? playbook path ? something else ? | 20:54 |
| harlowja | that'd be a start | 20:54 |
| harlowja | or "deployment on XYZ date"? | 20:54 |
| dmsimard | so the challenge is that I'd like, as much as possible, to keep the existing 100% feature parity between the hosted application and the statically generated one | 20:55 |
| dmsimard | I'll try and think of something, but not on my short term todo for now | 20:56 |
| harlowja | k | 20:57 |
| dmsimard | another user said he might want to take a stab at it | 20:59 |
| dmsimard | I forget who it was | 20:59 |
| harlowja | also, another question, if i use the embedded version of ara, do i need to do anything to get the static content to show up | 21:00 |
| harlowja | seems like its not showing up when pip install ara and then using `/ara-manage runserver -h 0.0.0.0 -p 8090` | 21:01 |
| dmsimard | harlowja: what is not showing up ? | 21:02 |
| dmsimard | harlowja: the playbook runs ? | 21:02 |
| harlowja | http://imgur.com/a/rNRGN | 21:03 |
| harlowja | so the main logo i think? | 21:03 |
| harlowja | but maybe something else is up? | 21:03 |
| harlowja | lol | 21:03 |
| dmsimard | harlowja: huh, that's weird | 21:03 |
| dmsimard | those are SVGs, can you see them if you open them up individually ? what browser/OS ? | 21:04 |
| dmsimard | you gotta see the parrots, that's like the most important part of the UI | 21:04 |
| harlowja | firefox mac | 21:04 |
| harlowja | we have ara running out of a docker container | 21:04 |
| harlowja | though i don't think that matters | 21:04 |
| dmsimard | harlowja: hmmm, WFM on Firefox 52 on fedora | 21:05 |
| harlowja | http://imgur.com/a/yCBRj same on chrome | 21:05 |
| harlowja | so wonder if its just not finding static | 21:05 |
| dmsimard | harlowja: you would not have css/javascript if it couldn't find statifcf | 21:06 |
| dmsimard | harlowja: do you see the logos on http://ara-demo.dmsimard.com/index.html ? | 21:06 |
| harlowja | ya | 21:06 |
| harlowja | why u got the logos | 21:06 |
| harlowja | lol | 21:06 |
| dmsimard | why do you not have the logos* | 21:06 |
| harlowja | lol | 21:06 |
| dmsimard | if you open up the developer console thing | 21:07 |
| dmsimard | do you get a 404 or something ? | 21:07 |
| harlowja | let's see | 21:07 |
| harlowja | http://imgur.com/a/FFwCl | 21:08 |
| harlowja | weird | 21:08 |
| harlowja | feels like the content-type is wrong | 21:08 |
| harlowja | and maybe its confused | 21:08 |
| dmsimard | yeah that's probably supposed to be image/svg+xml | 21:09 |
| harlowja | hmmm | 21:09 |
| dmsimard | stackoverflow to the rescue ? http://stackoverflow.com/questions/11494793/how-do-i-serve-a-static-svg-file-with-flask | 21:11 |
| dmsimard | jesus those people | 21:11 |
| dmsimard | A easy (but hacky) way is to add a new route just for svgs | 21:11 |
| harlowja | weird | 21:12 |
| harlowja | is that normal? | 21:12 |
| dmsimard | I don't know why it wouldn't be served properly out of the box | 21:13 |
| harlowja | ya, whacky | 21:13 |
| dmsimard | harlowja: you said it's installed side by side your chat bot ? can you do a pip freeze ? | 21:13 |
| harlowja | ya | 21:13 |
| * dmsimard coughs | 21:18 | |
| harlowja | coughs back | 21:19 |
| harlowja | my manager came by | 21:19 |
| harlowja | lol | 21:19 |
| harlowja | https://gist.github.com/harlowja/4967950e9e081aac10a0f95816374930 | 21:19 |
| harlowja | 'daddy' is the bot | 21:19 |
| harlowja | lol | 21:19 |
| harlowja | maybe just need a newer flask or osmething | 21:20 |
| harlowja | didn't expect flask to have weird problems with svg | 21:20 |
| dmsimard | oh ansible 2.2.2 is out | 21:20 |
| dmsimard | TIL | 21:20 |
| dmsimard | 2.1.5 too | 21:20 |
| harlowja | i wonder how it gets served for u | 21:21 |
| dmsimard | harlowja: so for the record, a fresh virtualenv http://paste.openstack.org/raw/604383/ | 21:23 |
| harlowja | ya, i was trying the following to | 21:23 |
| harlowja | >>> import mimetypes | 21:23 |
| harlowja | >>> mimetypes.types_map['.svg'] | 21:23 |
| harlowja | Traceback (most recent call last): | 21:23 |
| harlowja | File "<stdin>", line 1, in <module> | 21:23 |
| harlowja | KeyError: '.svg' | 21:23 |
| dmsimard | only difference I see is the version of jinja | 21:23 |
| dmsimard | but that's really be related | 21:23 |
| dmsimard | I expected maybe something different with werkzeug | 21:23 |
| harlowja | on 3.4python | 21:24 |
| harlowja | >>> import mimetypes | 21:24 |
| harlowja | >>> mimetypes.types_map['.svg'] | 21:24 |
| harlowja | 'image/svg+xml' | 21:24 |
| harlowja | evil python | 21:24 |
| harlowja | lol | 21:24 |
| dmsimard | is mimetypes stdlib ? | 21:25 |
| *** tbielawa has quit IRC | 21:25 | |
| harlowja | yes | 21:25 |
| harlowja | though i am noticing a diff | 21:25 |
| harlowja | cat /etc/mime.types exists on host centos system | 21:25 |
| harlowja | but not in container | 21:25 |
| harlowja | so i wonder if that gets installed | 21:25 |
| dmsimard | huh, maybe. | 21:26 |
| harlowja | rpm -q -f /etc/mime.types | 21:26 |
| harlowja | mailcap-2.1.41-2.el7.noarch | 21:26 |
| harlowja | weird | 21:26 |
| dmsimard | yeah, mailcap-2.1.41-2.el7.noarch : Helper application and MIME type associations for file types | 21:28 |
| dmsimard | mailcap doesn't exactly scream MIME TYPES to me | 21:28 |
| dmsimard | ¯\_(ツ)_/¯ | 21:28 |
| harlowja | lol | 21:28 |
| harlowja | ya, let me try shoving that into the docker image | 21:28 |
| harlowja | and see if svg starts showing up right | 21:28 |
| dmsimard | harlowja: hey completely unrelated, is simplejson better performance wise ? | 21:29 |
| dmsimard | iirc it was json < simplejson < ujson | 21:29 |
| harlowja | that's what i've heard, though i'm not sure it matters anymore | 21:29 |
| harlowja | pretty sure simplejson went into upstream json? | 21:29 |
| dmsimard | harlowja: fwiw, how to pronounce json: https://www.youtube.com/watch?v=wf-BqAjZb8M&t=1102 | 21:30 |
| harlowja | jlol | 21:31 |
| harlowja | lol | 21:31 |
| harlowja | like french | 21:31 |
| harlowja | jsooon | 21:31 |
| dmsimard | :D | 21:31 |
| harlowja | dmsimard what's typical of usage for ara_record? | 21:34 |
| harlowja | like just various logs? | 21:34 |
| harlowja | or other blogs | 21:35 |
| harlowja | *blobs | 21:35 |
| dmsimard | harlowja: I know there's someone who uses it on a registered command to get the sha1 git hash of his playbooks dir | 21:35 |
| harlowja | k | 21:35 |
| harlowja | makes sense | 21:35 |
| dmsimard | harlowja: another uses it to show what host the playbook is running from (play on localhost with gather_facts, record ansible_fqdn) | 21:35 |
| dmsimard | harlowja: since ara does not yet record vars/extra-vars/tags, you can also record interesting ones there | 21:36 |
| harlowja | right | 21:36 |
| harlowja | other question | 21:36 |
| harlowja | since we have ansilble running via a bot | 21:36 |
| dmsimard | Another example was if the logging/artifact location is known, put it there | 21:36 |
| harlowja | if we kill the `ansible-playbook` process to stop it, i assume ara wouldn't puke? | 21:36 |
| harlowja | like 'bot deploy openstack ....' | 21:37 |
| harlowja | then oh shit, didn't mean to do that, stop stop | 21:37 |
| harlowja | lol | 21:37 |
| dmsimard | harlowja: it would puke just a bit | 21:37 |
| dmsimard | but it would be good puke | 21:38 |
| dmsimard | sec | 21:38 |
| dmsimard | harlowja: this is actually tested in the gate: https://github.com/openstack/ara/blob/master/run_tests.sh#L83-L86 | 21:38 |
| dmsimard | see the second playbook: http://logs.openstack.org/49/450449/1/check/gate-ara-integration-latest-centos-7/2229b40/logs/build/reports/index.html | 21:39 |
| dmsimard | it's grey and it says it's incomplete | 21:39 |
| dmsimard | it's best effort but data is bound to be incoherent due to mission relationships | 21:39 |
| harlowja | right | 21:41 |
| dmsimard | harlowja: you see parrots yet ? | 21:59 |
| harlowja | deploying container | 21:59 |
| harlowja | yup, got it | 21:59 |
| harlowja | mailcap was it | 22:00 |
| harlowja | weird stuff | 22:00 |
| harlowja | lol | 22:00 |
| * harlowja would have never guessed this would of happened | 22:00 | |
| harlowja | lol | 22:00 |
| dmsimard | Should probably do a "troubleshooting" doc, kinda like FAQs but for problems | 22:00 |
| harlowja | ya, this one would be a good one | 22:01 |
| dmsimard | Q: I don't see parrots | 22:01 |
| harlowja | feel free to take part of https://gist.github.com/harlowja/1327d63b10ad5955bc6758b8712fb081 | 22:01 |
| dmsimard | A: You thought docker would solve all your problems ? Hah | 22:01 |
| harlowja | `RUN yum -y install mailcap` did the trick | 22:01 |
| harlowja | which i guess makes sense after the fact (sort of) | 22:04 |
| harlowja | lol | 22:04 |
| dmsimard | gotta give you credit, you found the solution | 22:05 |
| dmsimard | hurray | 22:05 |
| *** themurph has quit IRC | 22:05 | |
| *** themurph has joined #ara | 22:11 | |
| *** karimb has quit IRC | 22:21 | |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: WIP: Improve count performance https://review.openstack.org/450449 | 22:30 |
| openstackgerrit | David Moreau Simard proposed openstack/ara master: Improve count performance https://review.openstack.org/450449 | 22:59 |
| dmsimard | boxrick1: do you want to try that patch to see if it helps your performance issue ? https://review.openstack.org/#/c/450449/ | 23:00 |
| dmsimard | boxrick1: to test it: http://paste.openstack.org/raw/604391/ | 23:01 |
| boxrick1 | Sure, I'll have to give it a go tomorrow morning im on a different time zone =) midnight here! Cheers though look forwards to trying it | 23:11 |
| dmsimard | boxrick1: np, it's no miracles but will have to do pending a better solution | 23:30 |
| boxrick1 | I clocked in my page load at 50s today ( 100 plays on a page ) so will see how it handles after this | 23:32 |
| dmsimard | Yeah | 23:36 |
| dmsimard | That doesn't make sense :D | 23:36 |
| boxrick1 | if a normal 12.xx client is sending data to the mysql is that going to make much difference here? Or do I need to update everything | 23:42 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!