14:59:54 #startmeeting Fedora QA Meeting 14:59:54 Meeting started Mon May 2 14:59:54 2011 UTC. The chair is jlaska. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:59:54 Useful Commands: #action #agreed #halp #info #idea #link #topic. 14:59:58 #meetingname fedora-qa 14:59:58 The meeting name has been set to 'fedora-qa' 15:00:03 #topic Roll Call 15:00:20 * kparal first 15:00:20 Happy Monday all ... anyone lurking for the QA meeting? 15:00:23 * brunowolff is here 15:00:56 Hi kparal and brunowolff 15:00:59 * tflink is here 15:01:03 * jlaska tips hat to tflink 15:01:30 * jlaska surprised to see vhumpa 15:01:55 will wait a few more minutes ... 15:02:26 robatino: are you available for the meeting? 15:02:33 * robatino here 15:02:41 yay! 15:03:06 * jsmith-busy lurks 15:03:32 okay, well ... let's get moving. I'm not sure if adamw will be with us today, and I'm sure jskladan is lurking 15:03:41 yo 15:03:47 oooh, i'm a ghost! 15:03:58 cool, already you are here :) 15:04:02 welcome back 15:04:12 #topic Previous meeting follow-up 15:04:17 jlaska: jskladan regrets but can't attend 15:04:31 kparal: okay, thanks. Hope all is well 15:04:43 We didn't gather for a meeting last week ... so I don't have any action items 15:04:53 but there were a few items from the Friday blocker review 15:05:17 #info adamw reached out to rdieter for status of 684846 - COMPLETED 15:05:44 and there is a new selinux-policy available for testing, score! 15:06:06 #info jlaska - reach out to mgracik for input on 696320 - COMPLETED - https://admin.fedoraproject.org/updates/firstboot-1.118-1.fc15 15:06:18 #info jlaska - add https://admin.fedoraproject.org/updates/imsettings-1.2.2-1.fc15 to F15-TC1 rel-eng ticket - COMPLETED - more on this later 15:06:31 is that the same selinux-policy that's failing depcheck right now? 15:06:34 #info jlaska - test 700085 using an updates.img prior to Final TC1 - COMPLETED - just tsted ... looks good 15:06:44 tflink: sounds like you're more up to speed on that 15:07:02 That's all I have on my list ... any other outstanding items from last week? 15:07:46 alright, moving on ... 15:07:59 #topic F-15-Final-TC1 status 15:08:20 According to the schedule, Final TC1 should be arriving sometime today 15:08:21 #link http://rbergero.fedorapeople.org/schedules/f-15/f-15-quality-tasks.html 15:08:43 I didn't see any Final rel-eng tickets yet, so I created 2 so we can track any bodhi updates that need to be included 15:08:50 #link https://fedorahosted.org/rel-eng/ticket/4685 15:08:59 and ... 15:09:00 #link https://fedorahosted.org/rel-eng/ticket/4686 15:09:11 cool 15:09:18 adamw: I *think* I got all the items we noted from last Friday .. but lemme know if I missed any 15:09:45 if these tickets get closed or dup'd by any autogenerated ones, let's just make sure the list of updates is fwd'd 15:10:24 I still need to catch up with clumens for a anaconda-15.30-1 build, but I know he's already started the process on his end 15:10:38 the blocker list seems to have cleaned up well since Friday 15:10:40 #link https://fedoraproject.org/wiki/Current_Release_Blockers 15:10:44 bah! ... Re: Roll Call - /me is here 15:10:53 Hi maxa :) 15:10:59 you're late, stand in the corner 15:11:08 we have 2 proposed items ... any objections if we attempt to knock those out here (*quickly*)? 15:11:16 #link https://bugzilla.redhat.com/show_bug.cgi?id=700967 15:11:22 ^^^ seems pretty straight forward 15:12:21 This appears to impact the final criteria "All applications listed under the Applications menu or category must start successfully" 15:12:36 and this is for KDE ... which does qualify as a blocker I believe 15:12:51 sure 15:12:51 any comments/objections? 15:12:57 +1 Blocker for that 15:12:59 +1 15:13:14 * maxamillion stands in the corner in shame 15:13:27 +1 15:13:37 (from the corner) 15:13:37 #agreed 700967 - AcceptedBlocker - Impacts basic application functionality for s-c-lang when run by KDE 15:13:52 adamw: I'll note in the bz post-meeting 15:13:55 ok 15:13:59 #link https://bugzilla.redhat.com/show_bug.cgi?id=696278 15:14:07 I don't feel any closer to understanding this issue 15:14:24 I've got a preupgrade test inprogress, so we'll see 15:14:32 but still no idea what the root cause is here 15:14:49 adamw: any brilliant ideas on your end? 15:14:57 +1 for 700967 as well (I was reading it while you guys were voting) 15:15:00 reading 15:15:09 brunowolff: noted, thank you 15:15:15 oh, that one 15:15:18 yeah :( 15:15:26 well no, i had no real idea before and i don't now 15:15:35 it's worth noting that we only guarantee upgrades from _clean install_ 15:15:53 "The installer must be able to successfully complete an upgrade installation from a clean, fully updated default installation (from any official install medium) of the previous stable Fedora release, either via preupgrade or by booting to the installer manually. " 15:16:14 we don't really guarantee any upgrade from a 'real' working system; it's just too hard to do 15:16:21 definitely 15:16:33 I asked for a slew of additional logs from their systems 15:16:38 so if neither of them hit this on a clean install that could be a -1 15:16:40 maybe that'll help us figure out the paths taken 15:16:58 and we'll get some good preupgrade testing this week against TC1 ... that'll be informative 15:17:04 we'd still want to look at getting it fixed, of course, but it might not be a blocker 15:17:05 yep 15:17:10 It sounds like it could still be NTH if many people might experience this on upgrade. 15:17:18 brunowolff: sure, could be 15:17:30 so ... shall we keep this as is until we can further pinpoint _a_ cause? 15:17:43 it sounds like it's still in triage to me ... nothing tangible for a maintainer yet 15:17:48 ack/nak/patch? 15:18:01 I think the statement might also mention something about "upgrade not guaranteed in a situation where third party repositories are enabled and any software installed from outside of the official Fedora repositories is not guaranteed to work post-upgrade" 15:18:22 15:18:23 It would be nice to keep an eye on it in blocker discussions until we have a better idea of the cause. 15:18:36 brunowolff: right ... I don't think we need to Reject it yet 15:18:39 jlaska: ack 15:18:45 jlaska: ack 15:18:59 so I guess the decision is ot keep i on the proposed list for now ... continue monitoring (and hopefully attempting to recreate the failure) 15:18:59 maxamillion: i'd worry about the precedent of describing a negative case in the criteria 15:19:05 maxamillion: it could be a slippery slope:) 15:19:15 writing down all the things that aren't covered would take a while... 15:19:50 #agreed 696278 - Continue traiging - Still unclear on the exact cause, and the exact failure. Requested additional logs from reporters, preupgrade is planned for testing against TC1 15:19:57 adamw: good point 15:20:04 okay, thanks ... this concludes our mini blocker review 15:20:19 I know kparal and vhumpa wanted to experience taste :) 15:20:32 adamw: I just worry more about unrealistic expectations of users ... but I do agree that the documentation would quickly get overly verbose 15:20:43 so ... stay tuned to those rel-eng tickets for when ISO's arrive 15:21:03 jlaska: thank you :) 15:21:04 robatino: will you be doing the wiki updates and announcements, or will that be Tao Wu? 15:21:13 kparal: trust me, my pleasure!! :D 15:21:17 jlaska: afaik it'll be me 15:21:46 robatino: okay ... drop me a line if I can help in any way. You've clearly done it before and I think qualify as an expert here :) 15:22:13 If no other thoughts on F15-Final-TC1 ... we'll move on 15:22:23 i'll be fully spelling out "test compose" and "release candidate" in the announcements since there were complaints in the devel list 15:22:45 robatino: great thinking 15:22:58 robatino: heh, sure. I hadn't realized those weren't generally accepted terms. Yeah +good_idea :) 15:23:14 Note, TC1 will mark the start of a hectic 2+ weeks 15:23:27 I know everyone will do there best to stay on top of bugs, and provide feedback in a timely manner 15:23:43 if anything feels stuck or is missing information ... don't hesitate to make noise 15:24:00 #topic AutoQA update 15:24:30 kparal, also known as the most interesting man in the world ... take it away 15:24:47 not me, that's our mascot 15:25:00 http://troll.me/images/the-most-interesting-man-in-the-world/i-dont-always-test-my-code-but-when-i-do-i-do-it-in-production.jpg.pagespeed.ce.rM9YUkz0Br.jpg 15:25:06 alright 15:25:16 Here's a short update from the AutoQA world. 15:25:20 kparal: greast poster *EVAR* 15:25:32 maxamillion: kudos to jlaska 15:25:37 #info As you might have noticed, we have released AutoQA 0.4.7 last week. 15:25:44 #link https://kparal.wordpress.com/2011/04/29/autoqa-0-4-7-released/ 15:25:56 * jlaska does a dance ... nice work team 15:26:00 We hope that we have finally fixed all problems in depcheck and upgradepath tests and they will produce correct results now. 15:26:23 In the meantime we have received some complaints about the number of emails AutoQA sends (through Bodhi comments) and the readability of the logs. 15:26:36 According to tflink's suggestion we have decided to concentrate on those complaints first. Therefore we have wiped all the previous plans for 0.5.0 and started from scratch. 15:26:44 #info The main topic of the next 0.5.0 release should be "making AutoQA outputs more maintainer-friendly". 15:26:55 That mainly includes sending less emails and improving contents of our logs. 15:27:14 sounds like a good idea 15:27:16 I trust those changes will be well received by devel@ 15:27:24 usual call: If anyone has some suggestions related to that, please send your ideas to autoqa-devel@lists.fedorahosted.org. 15:28:02 And that's all! (in a nutshell) 15:28:08 [ OK ] dist-f15 + dist-f15-updates 15:28:18 I love that update ... boy, I completely missed that idea a while ago 15:28:18 i guess you did a quick list based on the feedback received so far already? 15:29:12 adamw: quick list of what? we have filed some tickets in our Trac according to the feedback 15:29:18 and we discuss them on autoqa-devel 15:29:31 I think discussion of those is still underway? 15:29:37 yes 15:29:43 * kparal is just writing one reply 15:29:46 or ... there is a tentative meeting scheduled for this week? 15:29:51 s/or/and/ 15:30:12 kparal: quick list of ideas 15:30:15 yeah, we're waiting on scheduling feedback for that one 15:30:17 so the answer's yes :) 15:30:18 jlaska: we just agreed with tflink we could have conference call tentatively on wednesday. depends on jskladan 15:30:33 adamw: yes :) 15:31:06 we don't want to fail our maintainers, that's our user base :) 15:31:33 anything else exciting on the AutoQA front? 15:32:01 seems like autoqa-0.4.7 went off without a glitch ... nicely done all 15:32:05 I think that's all from this weel 15:32:08 *week 15:32:24 okay, many thanks kparal tflink vhumpa jskladan :) 15:32:27 jlaska: well, production testing seems to be working 15:32:34 and jlaska 15:32:56 so, the only other thing I had on the agenda was a quick schedule reminder ... 15:33:06 #topic Upcoming QA events 15:33:11 #info Monday, May 02 (TODAY) - Final TC1 testing starts 15:33:19 #link http://rbergero.fedorapeople.org/schedules/f-15/f-15-quality-tasks.html 15:33:25 #info Friday, May 06 - Final blocker review meeting 15:33:44 I won't be available for the last scheduled blocker meeting this Friday ... but I can still send out the announcement etc... 15:33:52 we can coordinate as we get closer to ensure someone will be around 15:34:09 if not ... kparal can host from a pub or something! :D 15:34:33 i'll be around 15:34:40 (i think) 15:34:58 heh, okay ... we can work out the details later in the week 15:35:03 #info Wednesday, May 11 - Final Candidate testing starts 15:35:07 #info Tuesday, May 17 - Go/NoGo meeting 15:35:18 so that's 2+ weeks until the go/no_go 15:35:21 that'll go *fast* 15:35:56 any thoughts/questions/haiku? 15:36:11 if not ... we'll move on to the open-mic portion of the show 15:36:12 prayers? 15:36:13 :) 15:36:20 古池や蛙飛込む水の音ふるいけやかわずとびこむみずのおと 15:36:25 * vhumpa promises to read the log when he's done with evil JavaEE :-) 15:36:36 * kparal hopes it is a polite haiku 15:36:36 vhumpa: good luck with that beast! 15:36:48 * tflink shudders ... JavaEE 15:37:05 kparal: boy, you got me stumped on that one 15:37:41 kparal: my japanese is a bit rusty but it looks OK to me 15:38:02 ok, I admit, I copied that from http://en.wikipedia.org/wiki/Haiku#Examples 15:38:03 kparal: that's what SHE said 15:38:03 :) 15:38:11 adamw: LOL!! 15:38:29 good times in the land of Fedora QA :) 15:38:36 #topic Open Discussion - 15:38:43 I think we've moved on this this #topic already :) 15:38:50 * adamw realizes 'that's what she said' is weirdly difficult to render in japanese 15:38:58 any business not already discussed that we need to consider? 15:39:24 did we review the cloud test days? 15:39:24 robatino: Viking-Ice: anything items/concerns from you as we prepare for TC1? 15:39:36 not that I'm aware of 15:39:48 jlaska: session saving in upgrades 15:40:16 there's a fix for that coming down the pipe afaik 15:40:20 do we usually do the test day review or the group involved? 15:40:31 robatino: I saw your feedback in that bz ... I'll be sure to test that as well. While I did experience that random black background ... it seems to save the session and no longer hang on login anymore 15:40:39 tflink: well, we usually just mention it in a meeting. the people running the test day should do the recap, but i can poke them to make sure it's going to happen 15:40:55 K, I'll poke them but they're all at JBW this week 15:40:58 ok 15:41:05 i'm worried whether people who had it enabled before upgrade will have to struggle with it or it'll work immediately 15:41:07 it looks like boxgrinder went off well and EC2 was DOA? 15:41:09 #info action tflink to follow-up with cloud sig for test day recap 15:41:12 tflink: thanks! 15:41:27 jlaska: info or action? :) 15:41:32 adamw: doh, thanks 15:41:34 yeah, I wasn't able to get any of the EC2 images to work 15:41:34 #undo 15:41:34 Removing item from minutes: 15:41:39 #action tflink to follow-up with cloud sig for test day recap 15:41:58 there has already been an update to BG for some of the issues found during the test day, though 15:42:09 robatino: I'll do another upgrade test after the meeting. We can sync up in #fedora-qa afterwards? 15:42:23 i'll be there 15:42:35 re: cloud test day - there aren't a ton of results on the wiki ... but the channel seemed to be active all day long 15:42:44 so that seemed fairly positive 15:42:56 jlaska: yeah, the number of people wasn't a whole lot 15:42:57 tflink: nice. the testers were external too, right, not just people from the group? 15:43:16 not sure - might have been from the BG community 15:43:30 tflink: still counts =) 15:43:44 jlaska: as long as the results were useful and we get some fixes i'd call that a win 15:43:49 #info Robatino expressed concerns about gnome-session-saving support for users upgrading from F14->F15. Will focus additional testing and review results in #fedora-qa 15:44:11 adamw: there were people I didn't recognize, so if that's what you were asking - yeah 15:44:18 cool 15:44:30 adamw: agreed ... I mentioned to tflink, I was keen on just getting some experience with these components ... not so interested in sheer results 15:44:44 s/sheer/number of/ 15:45:03 alright ... any other discussion items? 15:45:35 #info tflink and adamw discussed the Cloud SIG test day, tflink will follow-up with the Cloud SIG for an event recap 15:45:56 alright ... if no other topics ... let's #endmeeting in 1 minute 15:46:19 something always feels satisfying about not having a 1+ hour QA meeting 15:46:30 we save that for the blocker reviews :D 15:46:50 10 seconds until #endmeeting ... 15:46:58 Thanks everyone for your time today! 15:47:04 I'll follow-up to the list with minutes 15:47:11 Good luck testing these next 2+ weeks 15:47:16 #endmeeting