15:00:01 #startmeeting Fedora QA 15:00:01 Meeting started Mon Aug 2 15:00:01 2010 UTC. The chair is jlaska. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:01 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:07 meetingname fedora-qa 15:00:10 #meetingname fedora-qa 15:00:10 The meeting name has been set to 'fedora-qa' 15:00:20 #topic Gathering in the lobby 15:00:29 Will wait a few minutes for folks to join 15:00:32 * kparal joins on time 15:00:38 * fenris02 waves 15:00:46 kparal: fenris02 vaschenb hi there :) 15:00:55 * vaschenb uff, 1 minute delay :-) 15:00:57 g'morning 15:01:12 * jlaska notes ... adamw won't be around today ... today is Civic day in Canada 15:01:34 if we see adamw respond in the meeting, we get to shush him ... he's supposed to be taking the day off 15:01:35 * j_dulaney is a pirate today 15:01:49 j_dulaney: pirates welcome 15:01:59 sweet 15:02:08 vaschenb: adjusting to life back in Brno? 15:02:49 jlaska: everything is so fast and every drink is alcohol-free :-D 15:02:57 jlaska: I can tell you vaschenb now lacks a lot of wine :) 15:03:07 haha 15:03:12 well, glad to have you back 15:03:20 kparal: psst :-) 15:03:55 I think wwoods is lurking ... we may need to lure him out with some treats 15:04:28 .bacon 15:04:28 I love bacon, you love bacon, WE ALL LOVE BACON! 15:04:49 that should do it, as long as it's soy bacon 15:04:54 okay ... let's get started 15:05:01 #topic Previous meeting follow-up 15:05:08 * wwoods appears in a puff of vegetarian bacon 15:05:14 eeew 15:05:20 :) 15:05:37 first, kparal, many thanks for wiki-zing the minutes from last week 15:05:56 I have an item for adamw and I, but I'll save that for when he's back 15:06:01 only other item ... 15:06:03 jlaska: ah, I really wasn't good at it 15:06:09 #info j_dulaney discussed possible proventesters from his OS class 15:06:26 j_dulaney: any updates on this topic? 15:06:33 Classes start up late August 15:06:36 kparal: funny, I think the same thing about my notes each week too! 15:07:12 I want to say around the 22nd 15:07:38 okay ... anything you'd need in prep for classes? 15:07:48 feel free to raise in the meetings here, or on the list 15:08:15 Not sure just yet, still trying to work that out with Mr. Edwards. 15:08:30 Like adamw mentioned last time, we've had some experience with organizing QA events in a university setting ... but always looking for more opportunities 15:08:47 I'm thinking you'll get a flood of proventester requests 15:09:32 Of course, this is also assuming that enough people sign up for the class, which hasn't always happened. 15:09:55 okay, as we get closer, we may want to come up with some ideas to quickly process the flood 15:10:13 Righteo. I'll keep up on this end 15:10:22 j_dulaney: cool, let us know how we can help :) 15:10:33 alright ... a few quick #topic updates ... 15:10:35 Roger 15:11:02 #topic F-14 Alpha Test Compose -- it's dead Jim 15:11:42 is there a new anaconda image coming rsn? 15:11:46 #info F-14 Alpha test compose CD and DVD images are available for test (see https://fedorahosted.org/rel-eng/ticket/3856) 15:11:50 I still have not been able to succeed in yum updating to it 15:12:03 dead Jim? 15:12:10 fenris02: there is a new anaconda already in koji ... I'll need ot catch up with anaconda-devel to see if an even newer version is expected 15:12:12 Star Trek 15:12:17 meaning it's completely broken? 15:12:26 #info Current F-14-Alpha blockers - https://bugzilla.redhat.com/showdependencytree.cgi?id=611990&hide_resolved=1 15:12:28 kparal, yes. py fail on bootup 15:12:40 kparal: yeah ... there's a ugly remnant of the python-2.7 migration .. 15:12:41 fenris02: alright. little less work for me then :) 15:12:49 .bug 620274 15:12:50 jlaska: Bug 620274 pyconfig-32.h is missing, anaconda fails - https://bugzilla.redhat.com/show_bug.cgi?id=620274 15:13:16 I'll try an updates.img after the meeting (although this might not be updates.img-able) 15:13:21 I posted a similar bug, let me see if I can find it 15:13:32 will post to the mailing list if that helps get tests further along 15:15:04 so stay tuned ... we'll likely have new media for testing fairly soon 15:15:06 .bug 620082 15:15:10 j_dulaney: Bug 620082 Python dependance broken. - https://bugzilla.redhat.com/show_bug.cgi?id=620082 15:15:19 #info Testing currently blocked bug 620274 15:16:12 j_dulaney: your bug is different, but similar to what newgle posted on the install matrix (I think) ... 15:16:38 j_dulaney: still more fall-out from python-2.7 15:16:46 so stay tuned gang ... 15:16:46 Indeed 15:17:02 #topic Accepting Test Day Proposals 15:17:18 As adamw noted during open-discussion last week ... 15:17:21 #info Adamw notes that we're looking very light on test days for this cycle, so if anyone has any ideas, please pitch in 15:17:33 Hurry (rhe) gets the prize for the first F-14 test day on the schedule 15:17:59 #info Installer translation/keyboard test day scheduled (see http://fedorahosted.org/fedora-qa/ticket/112) 15:18:30 Only thing I wanted to point out here was that we have a few proposals for events 15:18:41 #info Current proposed test days -- https://fedorahosted.org/fedora-qa/query?status=new&status=assigned&status=reopened&component=Test+Day&milestone=Fedora+14&order=priority 15:19:00 But if you have a good idea for a test day, or want to host one ... please submit an idea 15:19:04 #link https://fedoraproject.org/wiki/QA/Test_Days/Create 15:19:18 alright ... flying through the agenda today 15:19:34 #topic AutoQA package acceptance update 15:20:09 Quite a bit of movement in recent days ... wwoods, can you walk through recent updates and next steps? 15:20:46 well, we've been working on some patches to lay groundwork we need for advanced testing 15:21:40 kparal wrote some patches that rework how tests handle their arguments and how they can label themselves as requiring certain capabilities 15:21:49 like what arch they want to run on or whether they need virt capabilities 15:22:03 or distro 15:22:07 right 15:22:22 there's a friendly wiki doc for that ... 15:22:33 #link https://fedoraproject.org/wiki/Managing_autotest_labels 15:22:37 it lets the tests use various labels that we've defined in autotest for the test hosts 15:23:15 which is awesome. there was also a patch from jskladan that moved all the tests to using a common base class 15:23:30 which lets us define some default behaviors for autoqa tests, like sending mail and/or reporting to a future resultdb 15:23:36 without having to add code to every test 15:23:40 which is also awesome 15:23:56 * j_dulaney likes less work 15:24:07 and (IIRC) vaschenb wrote a 'helloworld' test that we can use to be sure that tests are running properly 15:24:20 the changes are not really documented on our wiki yet 15:24:27 that is a nice improvement ... I was getting worried with maintaining all that results mail code in each test 15:24:29 which has been really useful in testing all the (fairly invasive) changes we've made here 15:25:03 so - huge thanks to kparal, jskladan, and vaschenb for all their work 15:25:13 there's more than this but those are the hilights for the moment 15:25:24 When can we expect some Wiki love so that us newbs can get in on the fun? 15:25:40 so the next steps are to finish testing, merge the code into the master branch, and write some docs on the new changes 15:25:52 j_dulaney: sure. in the meantime just look at the doc/ directory 15:26:02 (that is new there) 15:26:23 kparal: Roger 15:26:39 once the new code is deployed I'm going to resume work on getting depcheck up and running as promised 15:26:55 jskladan mentioned he was looking into updating https://fedoraproject.org/wiki/Writing_AutoQA_Tests once those patches all land in master 15:27:08 I think I saw some test code kicking around - using rpmfluff(?) to generate RPMs with certain dependency structures that we can use as tests cases 15:27:46 * kparal has a question 15:27:46 so that'll probably be the next major target - using various resources (like https://fedoraproject.org/wiki/Updates_Lessons) to design test cases for depcheck 15:27:54 to make sure it's handling everything as expected 15:27:54 I was going to work on that, but my lack of python knowledge got in the way 15:28:28 * j_dulaney is working on learning Python 15:28:30 wwoods: what's the current status of post-bodhi-update hook? 15:28:38 there are a lot of other important things in the works - they might be able to be done in parallel with that effort, if we have enough free time. but we'll see. 15:28:48 kparal: should be merged and functional, I believe? 15:29:07 if it's not, we should fix that. it has a couple of known shortcomings but should be good enough for what we want to do 15:29:22 and now that we can have multihook tests we can wire up rpmlint/rpmdiff there, if we like 15:29:30 wwoods: well I thought it's not finished yet. I'll look tomorrow, now I'm remotely connected 15:30:15 kparal: I'll check and make sure it's there - I haven't gotten to testing it yet 15:30:26 vaschenb: have you had a chance to start thinking about designing depcheck unittests? 15:30:32 it will be tested and confirmed working before we push new code to production 15:30:37 wwoods: but everything should be already in master, right? 15:31:01 jlaska: not yet, now i'm working on upgradepath test 15:31:21 jlaska: but some skeleton I already made, but I need more specific information... 15:31:29 vaschenb: oh right, forgot about that test (https://fedorahosted.org/autoqa/ticket/123) 15:32:09 kparal: I thought it was 15:32:10 jlaska: I think vaschenb will be our upgrade path guy nowadays, so I can look at rpmfluff instead 15:32:36 kparal: hooks/post-bodhi-update exists in master, anyway 15:33:07 the code at least exists - whether or not it Does The Right Thing is unsure 15:33:08 kparal: vaschenb: exciting, sounds like a plan 15:33:11 jlaska: so upgradepath test is not my business now? 15:33:28 vaschenb: I believe it is :) 15:33:35 kparal: me too :-) 15:33:40 vaschenb: correction, sounds like you have the upgrade path, and kparal was going to take depcheck unit tests 15:33:46 did I get that right? 15:33:53 yep 15:33:58 jlaska: ok, now I understand, uff :-) 15:34:05 roger ... okay ... back to wwoods 15:34:17 * jlaska 15:34:35 Ooh, squirrel 15:34:52 hey, I'm loosing you 15:35:20 wwoods: anything else you wanted to include today? 15:36:05 that's all I had in mind for the near future - there's more, of course, but we'll have another meeting before we get to any of that 15:37:38 roger, alright ... thanks for the update gang 15:37:45 #topic Open Discussion - 15:37:56 Any open discussion items this week? 15:38:20 if not ... we'll close this puppy out and let everyone get back to rockin! 15:39:15 * j_dulaney notes that he already has Oasis going 15:40:56 Looks like we have a buggout in progress. 15:41:04 alrighty, let's close it out 15:41:10 thanks gang ... talk to you all online :) 15:41:17 as usual, I'll send the minutes to the list 15:41:25 #endmeeting