QA/Meetings/20120319

Attendees

• adamw (147)
• kparal (37)
• maxamillion (29)
• tflink (17)
• brunowolff (12)
• elad661 (6)
• zodbot (6)
• rdieter (3)
• jskladan (2)
• kalev (2)
• bodhi_zazen (2)
• pschindl (2)
• pjones (1)
• jreznik (1)

Agenda

• Previous meeting follow-up
• Fedora 17 Beta status
• Test Day report
• Upcoming QA events
• AutoQA update
• Open floor

Previous meeting follow-up

• adamw to co-ordinate with pjones on grub2 update: done
• j_dulaney to ensure announcements are sent for 2012-03-13 and 2012-03-14 test days: 2012-03-13 done, looks like 2012-03-14 was not done

Fedora 17 Beta status

Turned into a blocker review meeting: see summary for details

Test Day report

• Looks like the USB 3.0 Test Day misfired, but Desktop i18n and GNOME Shell went off well with good participation and bug filing

Upcoming QA events

• Freeze is tomorrow, RC is scheduled for Thursday, we have lots of blocker resolvin' to do
• Sugar test day set for Thursday, preparation does not look complete
• maxamillion will look at possibilities for kde/xfce test days

AutoQA update

• AutoQA 0.8 released and deployed, kparal blogged about it
• resultsdb also now in production

Open floor

N/A

Action items

• adamw to bug j_dulaney to bug Sugar test day folks

IRC Log

adamw	#startmeeting Fedora QA meeting	15:03
zodbot	Meeting started Mon Mar 19 15:03:15 2012 UTC. The chair is adamw. Information about MeetBot at http://wiki.debian.org/MeetBot.	15:03
zodbot	Useful Commands: #action #agreed #halp #info #idea #link #topic.	15:03
adamw	#meetingname fedora-qa	15:03
zodbot	The meeting name has been set to 'fedora-qa'	15:03
adamw	#topic roll call	15:03
* elad661 is here		15:03
* brunowolff is here		15:03
adamw	sorry i'm a bit late	15:03
adamw	who's around for happy qa fun times?	15:03
* pschindl is here		15:03
tflink	adamw: not acceptable!	15:03
* tflink is present		15:03
* kparal here		15:04
* jskladan is in		15:04
adamw	.fire adamw	15:05
zodbot	adamw fires adamw	15:05
adamw	back to bed for me!	15:05
adamw	#topic previous meeting follow-up	15:05
* maxamillion is here		15:05
adamw	#info "adamw to co-ordinate with pjones on grub2 update" - we got that done, f16 has the core.img size regression fixed, f17 has a shiny new grub2 2.0-beta2 which is currently still pretending to be 1.99.	15:06
pjones	as does rawhide.	15:06
adamw	yup.	15:07
adamw	#info "j_dulaney to ensure announcements are sent for 2012-03-13 and 2012-03-14 test days" - that happened too.	15:07
adamw	yay efficiency!	15:07
adamw	so on to...	15:08
adamw	#topic fedora 17 Beta status	15:08
adamw	not looking so shiny here	15:08
* maxamillion fears for the state of this topic		15:08
adamw	the accepted blockers are mostly not terrible, but the new anaconda has, 'ow you say, broken ze world.	15:09
adamw	we've got five proposed blockers, so let's just review them quickly	15:09
adamw	#topic fedora 17 Beta status - http://bugzilla.redhat.com/show_bug.cgi?id=804454	15:09
adamw	"Fedora-17-Beta-TC2-i686-Live-KDE.iso is oversized"	15:10
adamw	sounds like a clear blocker, criterion "The network installation image, DVD image, and live images for release-blocking desktops must meet current size requirements"	15:10
tflink	pretty clear blocker	15:10
adamw	anyone -1 on this?	15:10
adamw	#agreed 804454 is a clear blocker, criterion "The network installation image, DVD image, and live images for release-blocking desktops must meet current size requirements"	15:11
brunowolff	+1 blocker	15:11
adamw	(trying to get through this fast, so shortcutting the protocol a bit)	15:12
adamw	#topic fedora 17 Beta status - http://bugzilla.redhat.com/show_bug.cgi?id=804446	15:12
maxamillion	I thought the bugzappers meeting was for the +1/-1 of blockers	15:12
maxamillion	?*	15:12
adamw	maxamillion: we do a regular blocker review on friday	15:12
adamw	but we often do this in qa meetings close to releases too	15:12
adamw	in the interests of...not waiting till friday	15:13
adamw	there are no bugzappers meetings at present.	15:13
maxamillion	ah, ok ... I've been irregular in my meeting attendance for too long and just don't know better :X	15:13
brunowolff	We only have a week right?	15:13
adamw	brunowolff: yup.	15:13
adamw	"cannot fresh install from DVD"	15:13
maxamillion	ouch	15:13
adamw	so this is the one for TC2 apparently not picking up packages from the install media	15:13
tflink	also sounds like a pretty clear blocker	15:13
maxamillion	tflink: +1	15:13
kparal	+1 blocker. btw do we have criteria?	15:14
adamw	alpha criterion "When booting from a DVD ISO image, the installer must be able to use the DVD local package source options"	15:14
elad661	+1 blocker.	15:14
kparal	great	15:14
adamw	kparal: apparently we thought of that!	15:14
adamw	#agreed 804446 is a clear blocker under Alpha criterion "When booting from a DVD ISO image, the installer must be able to use the DVD local package source options"	15:14
kparal	there are some oracles in our midst	15:14
adamw	#topic fedora 17 Beta status - https://bugzilla.redhat.com/show_bug.cgi?id=804483	15:15
adamw	also sounds like a nice clear blocker, criterion "The installer must be able to complete an installation using the text, graphical and VNC installation interfaces"	15:15
adamw	on the bright side, at least none of these bugs is borderline...	15:15
kparal	what does that mean_	15:16
maxamillion	adamw: that's true	15:16
adamw	none of them are those bugs where we spend half an hour arguing about whether they're blockers :)	15:16
brunowolff	We don't need to waste time arguing about status.	15:16
kparal	yep	15:16
maxamillion	its all pretty cut and dry	15:16
kparal	+1 blocker	15:16
tflink	+1 blocker	15:16
maxamillion	+1 blocker	15:16
* jskladan is with the crowd		15:16
adamw	#agreed 804483 is a clear blocker per Alpha criterion "The installer must be able to complete an installation using the text, graphical and VNC installation interfaces"	15:16
adamw	#topic fedora 17 Beta status - http://bugzilla.redhat.com/show_bug.cgi?id=785729	15:17
adamw	well hey, here's a borderline one! sigh.	15:18
adamw	kay's rationale for blocker status is in the first comment	15:18
adamw	is argyllcms in a default install package set, does anyone know?	15:18
* kparal checks		15:19
adamw	it's not directly in comps as default	15:19
adamw	just looking at dependencies	15:19
kparal	not installed	15:19
adamw	okay, if it's not in a default install, i'm -1	15:20
elad661	-1 as well	15:20
brunowolff	I am not sure we want to block for hypothetical problems, on the other hand I'd like to see this fixed early so we don't find out about real problems too late.	15:20
adamw	brunowolff: if it was installed by default i'd be on the same track as you, but if it's not i'm rather less worried	15:20
tflink	-1 blocker	15:20
maxamillion	-1 blocker	15:20
brunowolff	-1 blocker (yeah I didn't see the not in default until after I finished typing)	15:21
pschindl	-1 blocker	15:21
adamw	whee	15:21
adamw	#agreed 785729 is not a blocker as it's not present in the default install from DVD, network install or any release-blocking live spin	15:22
adamw	(i just checked the lives.)	15:22
adamw	and finally...	15:22
adamw	#topic fedora 17 Beta status - https://bugzilla.redhat.com/show_bug.cgi?id=804451	15:22
adamw	"zenity depends on webkitgtk3"	15:23
adamw	sounds odd, but this is the cause of kde live being oversize	15:23
adamw	and it's not directly blocking f17beta, it's just via the dep on the 'tracking bug' kevin created	15:23
adamw	by policy we review 'indirect blokcers' too, i'm happy to say +1 to this	15:23
brunowolff	I can't see this as a blocker, even though it would fix the oversize issue.	15:23
adamw	brunowolff: well, if we vote it not a blocker, that means it has to not block the tracker bug any more	15:24
brunowolff	There may be other avenues to getting the KDE spin size down.	15:24
elad661	There is a fix there, I think it should be a blocker.	15:24
elad661	"zenity-3.2.0-4.fc17 has been submitted as an update for Fedora 17."	15:24
elad661	we just need to push this to stable quickly.	15:24
tflink	we're sure that nothing uses the webkitgtk3 stuff?	15:25
adamw	i only just found out about this five minutes ago, so not personally, no. but i trust kk to know what he's talking about when it comes to the kde spin.	15:25
brunowolff	Some optional stuff was turned on that hadn't been before. So turning it off should be OK.	15:25
rdieter	I can confirm that webkitgtk3 was added as a new dep to zenity only very recently, and it is the main contribution to the kde spin growing by 10mb since alpha	15:26
kparal	so what's the result? --html and --url options have been removed?	15:26
adamw	i think the update doesn't turn the optional stuff off, it just removes the dep on the thing that makes the optional stuff work.	15:26
brunowolff	Given they are fixing this, I am not worried about blocker status. Lets just call it one and move on.	15:26
adamw	but it'd probably be good to check with mclasen that he's okay with the fix.	15:26
adamw	rdieter: do you know if you or kk has contacted the desktop team?	15:26
rdieter	mclasen is the one submitted the new build	15:26
rdieter	so I assume he's ok with it. :)	15:27
adamw	ah :)	15:27
adamw	i guessed it was kk fixing it manually	15:27
adamw	okay, so this seems all alright	15:27
tflink	yep, I was assuming that it was fixed manually as well	15:28
adamw	propose #agreed 804451 is accepted as a blocker as the avenue being taken to reduce KDE spin size	15:28
tflink	I'm OK with blocker on this	15:28
adamw	tflink: er, i menat something else. =)	15:28
adamw	unilaterally! that's it.	15:28
adamw	i knew it had a lly in it.	15:28
tflink	eh, as long as we both knew what we meant :)	15:29
adamw	anyone still have concerns?	15:29
kparal	if we accept this as a blocker, don't we say this is the road they must take to take down the size?	15:30
kparal	maybe they could have done it otherwise	15:30
kparal	other way	15:30
adamw	we can always change things.	15:31
adamw	bugs can be un-accepted.	15:31
tflink	NTH would also accomplish the purpose but would make things a bit messy in bz	15:31
kparal	ok then	15:31
kparal	in this case it doesn't probably matter so much, the fix is already on its way	15:32
tflink	also true but you're probably right about blocker/nth	15:32
adamw	#agreed 804451 is accepted as a (dependent) blocker as it's the root cause of the KDE live being oversize	15:32
adamw	okay, that's all the proposed blockers; at least now we know what we're dealing with	15:33
adamw	RC is due tomorrow but i doubt we'll make that, with the blocker list we have; it's conceivable, though	15:33
adamw	anyone have any specific concerns to raise about the beta?	15:33
maxamillion	would make today one hell of a bug smashing sprint	15:33
adamw	even though tc2 is obviously bad we should still be able to confirm several of the accepted blockers are fixed with it, and set those to VERIFIED	15:34
adamw	so please do take a look through that list, everyone	15:34
adamw	i'll fire out a 'blocker bug status' email today	15:34
adamw	hi andre	15:36
adamw	okay, well, guess we'll move on then...	15:37
adamw	#topic Test Day report	15:37
adamw	we had another packed test day schedule last week, three events	15:37
adamw	j_dulaney's not around, anyone else have news from any of the test days?	15:37
adamw	from what I can see, the USB 3.0 one didn't really get off the ground, the GNOME 3.0 one looks to have gone off well, as does desktop i18n	15:38
adamw	lots of bugs filed from desktop i18n	15:39
* adamw looks around nervously, wonders if the rapture happened while he wasn't looking		15:40
maxamillion	sorry, distracted by other bits of the internet :X	15:40
tflink	adamw: its moving from east to west and hasn't gotten to you yet :)	15:41
adamw	NOT ACCEPTABLE	15:41
adamw	.fire maxamillion	15:41
zodbot	adamw fires maxamillion	15:41
* maxamillion puts on his dunce cap and sits in the corner in shame		15:41
adamw	#info looks like the USB 3.0 test day misfired, but desktop i18n and GNOME Shell went off well with good participation and bug filing	15:42
adamw	#topic Upcoming QA Events	15:42
adamw	this week, in 'stuff adamw got wrong'...	15:42
adamw	looks like RC is scheduled for Thursday, not Tuesday. you'd think I'd know this stuff, right?	15:42
* adamw vows to switch to diet crack		15:42
tflink	Tuesday, Thursday ... same thing	15:43
adamw	so many of the letters are the same!	15:43
adamw	#info freeze is tomorrow, RC is scheduled for Thursday, we have lots of blocker resolvin' to do	15:43
adamw	we also have the Sugar desktop test day coming up Thursday	15:43
adamw	that looks like it still needs work, the test day page is still mostly template	15:44
adamw	#action adamw to bug j_dulaney to bug Sugar test day folks	15:44
maxamillion	adamw: diet crack!?!? blasphemy, you need more a more wholesome high but with less calories ... get Lite ... >.>	15:44
adamw	#info Sugar test day set for Thursday as well	15:44
adamw	maxamillion: crack lite just doesn't smoke smoothly, you know? you can't ever forget you're smoking second-rate crack	15:45
maxamillion	adamw: fair criticism	15:45
adamw	that's everything on my list, anyone have exciting plans for the week?	15:45
maxamillion	oh, are there XFCE and KDE test days scheduled yet?	15:46
maxamillion	I don't see any on the list .... in all reality the XFCE spin should be pretty solid since not a whole lot in XFCE land has changed but testing is still good, I know KDE went to 4.8.x so that likely needs some run throughs .... or are those just included as part of the Desktop validation testing and don't need dedicated test days anymore?	15:47
adamw	maxamillion: neither.	15:47
maxamillion	rgr	15:47
adamw	maxamillion: nothing 'needs' a test day	15:47
adamw	we don't have a set schedule; they're entirely optional	15:47
adamw	it just comes down to whether anyone wants to run one or not	15:47
adamw	no-one proposed an xfce test day or a kde test day, so...we don't have them.	15:48
adamw	i don't think we've ever done a kde test day, actually	15:48
maxamillion	yeah, come to think of it maybe we haven't	15:48
kalev	on a different note, I noticed that the GNOME live cd failed the last best test compose and tracked it down	15:48
maxamillion	alright, well .... I'll look and see if I can attend two of the open time slots to run both a KDE and XFCE test day and if so I'll propose slots on the schedule	15:49
kalev	should be fixed by https://admin.fedoraproject.org/updates/FEDORA-2012-4138 , maybe it could get some karma?	15:49
adamw	kalev: oh right, should have put that in beta status	15:49
adamw	we should really have a blocker bug for that	15:49
adamw	#info OUT-OF-TOPIC-BAND: kalev tracked down cause of TC2 desktop spin failing to compose, it is fixed by https://admin.fedoraproject.org/updates/FEDORA-2012-4138	15:50
adamw	maxamillion: the open slots are very late, but there are non-thursdays available of course	15:50
maxamillion	adamw: yeah, true .... I'll see what I can do, if not I'll just test and file bugs the best I can	15:51
maxamillion	TEST ALL THE THINGS!	15:51
maxamillion	>.>	15:51
adamw	desktops actually make a reasonable amount of sense as late test days	15:51
adamw	since there's a high chance bugs found will be nth or even blocker	15:51
adamw	so...if we were going to run anything on those days, kde/xfce make sense, i guess.	15:51
adamw	#info maxamillion will look at possibilities for kde/xfce test days	15:52
maxamillion	well my schedule is about to get really crazy so I'm not even sure I could run the test days but I wanted to bring the topic up for discussion either way	15:52
adamw	sure	15:52
adamw	j_dulaney is the official test day co-ordinator for this cycle, so i'm a bit more hands-off than usual	15:52
kparal	maxamillion: contact jreznik, he was interested in KDE test day as well	15:53
adamw	#topic AutoQA update	15:53
maxamillion	kparal: very cool, will do! thanks	15:53
adamw	so we have ze big news here, right kparal?	15:53
kparal	#info AutoQA 0.8 released and deployed	15:53
kparal	#link https://kparal.wordpress.com/2012/03/16/autoqa-0-8-released/	15:54
kparal	soooo, it's running!	15:54
adamw	IT'S ALIIIIIIIVE	15:54
adamw	#chair kparal tflink brunowolff	15:54
zodbot	Current chairs: adamw brunowolff kparal tflink	15:54
jreznik	kparal: it was more "I try to propose it tmrw for our kde sig meeting", but yeah, I think it would be nice to have a test day	15:54
* adamw throws chairs around		15:54
kparal	we have the resultsdb running as well. just look at http://autoqa.fedoraproject.org/	15:54
adamw	#info AutoQA 0.8 is released and deployed, resultsdb is in production	15:54
brunowolff	Hey I don't want to be thrown around.	15:55
adamw	kparal: awesome news! congratulations.	15:55
kparal	our automated installation test is also running on production now: http://autoqa.fedoraproject.org/resultsdb/frontend/search?type=Testcase&terms=rats_install	15:55
kparal	the last two days it caught some broken dependencies problem	15:55
adamw	#info rats_install is also in production	15:56
kparal	I already asked relevant people to push fixed packages	15:56
kparal	which brings me to this question	15:56
kparal	do you think we should do automated installation from stable updates only, or updates-testing as well?	15:56
kparal	the latter means it will be broken more often	15:56
adamw	yeah, good question	15:57
kparal	the former means it's more serious when it breaks	15:57
adamw	the best answer, obviously, would be 'both' :)	15:57
tflink	and the actual images tend to be a combination of the two during development	15:57
kparal	I think the netinst has updates-testing enabled, right?	15:57
kparal	OTOH DVD is composed mainly of stable repo	15:57
kparal	so, if you have any preference, let's hear it	15:58
adamw	can we do both?	15:58
adamw	does the test take a lot of time/resources?	15:59
kparal	it does, but we can do both	15:59
kparal	but it will require some tweaking	15:59
adamw	i think that would be the best, because there's obvious value to catching failures down both paths	15:59
kparal	ok, I'll try it	16:00
adamw	if we just have to pick one, i guess i'd vote for updates-testing as it's the default netinst case and meets the original 'israwhidebroken.com' intent	16:00
kparal	great, that's the current behavior	16:01
brunowolff	Note that rawhide currently doesn't inherit from updates-testing.	16:01
adamw	anyone else have a take on this?	16:01
adamw	brunowolff: sure, that's not the rawhide i meant :)	16:01
tflink	brunowolff: AFAIK, rats uses images from releng so it isn't actually using rawhide anyways	16:02
kparal	that name is somewhat obsolete, yes. maybe rawhide -> branched, rats -> bats? :)	16:02
adamw	heh	16:03
maxamillion	lol	16:03
adamw	anyhoo, we're getting over time, so..	16:03
adamw	anything else from autoqa side, kparal?	16:03
kparal	all from me	16:03
adamw	tflink?	16:03
maxamillion	bats are informally know as "just wing'd rats" so there could be some fun pun about flying rats coming out of branches .... >.>	16:03
tflink	nothing from me	16:03
maxamillion	I see awesome t-shirts in our future	16:04
kparal	:)	16:04
adamw	halloween party!	16:05
adamw	#topic open floor	16:05
adamw	anything else from anyone else at all?	16:05
bodhi_zazen	adamw: thanks for all your hard work on this =)	16:07
adamw	on running around telling everyone what to do? you're welcome!	16:07
adamw	sometimes they even listen.	16:07
bodhi_zazen	your posts to the mailing list as well	16:08
adamw	okay	16:09
adamw	guess that's it, folks	16:09
adamw	so if everyone can help out with blocker re-testing with tc2 that'd help a lot...	16:09
adamw	thanks everyone!	16:10
adamw	#endmeeting	16:11

Generated by irclog2html.py 2.8 by Marius Gedminas - find it at mg.pov.lt!