17:01:18 #startmeeting ironic_qa 17:01:19 Meeting started Wed Feb 10 17:01:18 2016 UTC and is due to finish in 60 minutes. The chair is jlvillal. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:01:20 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:01:22 The meeting name has been set to 'ironic_qa' 17:01:31 <[1]cdearborn> o/ 17:01:41 Sorry I'm late 17:01:53 o/ 17:01:55 o/ 17:02:05 #topic Announcements 17:02:08 o/ 17:02:13 \o 17:02:16 o/ 17:02:40 I don't have any announcements. 17:02:51 Does anyone else have anything to announce? 17:03:03 mjturek1, maurosr - want to plug your ironic session? 17:03:23 sure, if it's not frowned upon 17:03:52 we have a ironic QA session for summit up for voting 17:03:53 Sure. I guess :) 17:04:21 #info mjturek1 maurosr Have an Ironic QA related session up for voting 17:04:24 link? 17:04:29 o/ 17:04:33 grabbing it now 17:04:35 well I'd encourage everyone to vote for their favorite talks 17:04:39 #link https://www.openstack.org/summit/austin-2016/vote-for-speakers/presentation/7334 17:04:43 Thanks 17:04:47 thanks krtaylor 17:04:49 Any other announcements? 17:04:59 Okay moving on.... 17:05:04 #topic Grenade testing 17:05:37 #info jlvillal again got involved with other work and didn't have time this past week. Things are looking much better for this week. 17:05:45 Sorry about that folks. 17:06:13 But I think my work in bringing up a bunch of servers is ramping down. So I should have more cycles this week. 17:06:42 #topic Announcements 17:06:56 #info No meeting next week, as we have the virtual mid-cycle 17:07:00 TinyIPA merged 17:07:04 :) 17:07:11 nice 17:07:14 sambetts: o/^5 17:07:16 #info TinyIPA merged! 17:07:33 #topic Functional testing 17:07:37 hey jlvillal, maurosr and I have an quick update on functional testing 17:07:45 mjturek1: Great! 17:07:54 We started some initial investigations on functional testing, and we're thinking the nova code base would be a better basis than glance. 17:07:57 Feel free to #info things 17:08:08 Specifically, the use of fixtures to create the services seems to be the right approach, though we are totally up for suggestions. 17:08:22 Sounds reasonable to me. From when I was looking at the fixtures. 17:08:28 very cool 17:08:38 We've started adding some code for creating fixtures for the ironic services. Though we haven't published it anywhere yet (it's not tested/functional yet) 17:08:41 The Nova stuff sort of confused me as it was an onion of layers it seemed. 17:09:01 lol, yeah it took some time to understand what was going on there 17:09:05 One class led to another class to another class. Etc.. From what I remember. 17:09:16 yep very true 17:09:16 My head started hurting... :) 17:09:20 hahaha 17:09:27 jlvillal: we're curious though about what ironic team decided to belong or not to functional testing 17:09:49 maurosr: Not quite sure I understand that question. 17:10:26 I think in general we would like to have functional testing to exercise the code more and hopefully improve the quality of the code. 17:10:37 Not sure if that answers the question maurosr 17:10:55 Oh, I think you are asking what should functional testing test? 17:11:05 Is that correct? 17:11:07 jlvillal: it does, so I think that includes several areas, like api, db, conductor and such 17:11:15 yeah, exactly 17:11:31 hm, well initially functional moving out of tempest was a place to move tests that were to big or slow for gate, but we have morphed that a bit to be a more simple way for ironic devs to test functionality as well 17:11:41 So my thought for functional testing is to have it as isolated as we can. As few external dependencies as possible. 17:11:48 ++ 17:11:54 We already have tempest testing for testing it as part of the bigger project. 17:11:54 o/ 17:12:15 So functional testing would be ironic-api and ironic-conductor and any bare essentials to make those work. 17:12:31 jlvillal: so do you agree that most of tempest's api testing could be moved to functional? 17:12:34 Maybe that would be RabbitMQ or maybe there is a way to use oslo.messaging without RabbitMQ 17:12:57 we probably want to keep only the scenario tests on ironic side, correct? 17:12:59 maurosr: Not sure about 'moved'. But maybe we could 'copy' a lot of it. If it works. 17:13:08 right 17:13:36 I don't want us to lessen what the tempest tests are doing. 17:13:48 ok, got it 17:13:54 Thanks 17:13:54 is this something we should bring up at midcycle (what functional tests should excercise)? 17:13:54 we can not remove the API testing from tempest 17:14:00 +1 17:14:32 devananda: I didn't meant to remove from tempest, but from ironic's tempest plugin 17:14:33 and in fact, with the move to our tempest plugin, we need to start being very mindful of how we change those things 17:14:47 functional testing was meant to be moved out of tempest testing for projects that had a huge slow gate, they were meant to be periodic, but we don't have that problem with gate being bloated 17:14:47 #info maurosr mjturek1 have started some initial investigations on functional testing, and are thinking the nova code base would be a better basis than glance. 17:15:20 yes, lets discuss at midcycle 17:15:26 tempest API tests are the basis of defcore. ironic is not part of the set of projects currently considered for defcore ... but if we ever hope to be, those API tests would be the basis 17:16:49 #info Tempest API tests should remain in tempest. Tempest API tests could be used to help in writing functional tests. But tests should not be "moved" from Tempest to functional. 17:17:08 Anything else on functional testing? 17:17:19 no! 17:17:23 :) 17:17:25 not at the moment 17:17:36 thanks for the info tagging jlvillal 17:17:36 And now onto my favorite part of the meeting 17:17:46 #topic 3rd Party CI (krtarylor) 17:18:03 krtaylor: All yours. Please #info things of note :) 17:18:07 hehheh, favorite? 17:18:15 * jlvillal doesn't have to do anything :) 17:18:42 nothing to report this week, I too have been swamped with downstream activity, but it lifted today 17:19:15 I am hoping to get help from mjturek1 and maurosr for some doc changes based on their summit talk proposal 17:19:27 but thats about it 17:19:44 krtaylor: Anything on emails sent or anything else? 17:19:54 Deadlines? etc... 17:20:10 I have not seen any replies, but they may have gone directly to thingee 17:20:37 krtaylor: There was an email sent out last week. Correct? 17:20:55 yes, thingee sent a detailed message, excellent 17:21:05 Have not receieved anything 17:21:21 ok, good to know 17:21:52 jlvillal: http://lists.openstack.org/pipermail/openstack-dev/2016-February/085722.html 17:21:57 krtaylor: Anything you want to #info before we move on? 17:22:30 I think we're good for now, back to you 17:22:35 Thanks! 17:22:43 #topic Opens / General QA 17:22:55 Anyone? Anyone? ... 17:23:16 krtaylor: how many vendor/drivers have not yet responded // put up a CI thing? 17:24:16 thingee: ^^^ maybe you know? 17:24:54 devananda, I still need to crawl through that and identify those that have not done all the steps, han't had the time 17:24:57 I'm still waiting for my lab team to put together my rack and then I'll be getting our (Cisco) CI up and running, I've created the wiki page with the accounts etc and just put down a status of In development 17:25:03 devananda: it should be all in that etherpad we've been communicating from the previous meeting 17:25:11 * thingee digs it up 17:25:40 https://etherpad.openstack.org/p/IronicCI 17:26:07 #link https://etherpad.openstack.org/p/IronicCI 17:26:42 Anything else? 17:26:56 thingee: thanks 17:27:08 Okay. Thanks a lot everyone for attending. 17:27:18 Have a good day/evening. 17:27:27 Remember no meeting next week due to the mid-cycle. 17:27:39 #endmeeting