QA/Meetings/20131202

Attendees

• adamw (86)
• jreznik_ (26)
• Viking-Ice (19)
• jreznik (14)
• pjones (7)
• zodbot (5)
• kparal (5)
• mkolman (4)
• pwhalen (4)
• roshi (3)
• tflink (2)
• nirik (1)
• brunowolff (1)
• mkrizek (1)
• satellit (1)
• pschindl (1)
• viking-ice (0)

Agenda

• Previous meeting follow-up
• Fedora 20 Final status
• Open floor

Previous meeting follow-up

• adamw to draft a new test case and matrix row for validating cloud image checksums - adamw didn't get around to this yet, put back on the list
• adamw to make sure the desktop tests are completed for TC2 - they were covered between TC2 and TC3, many thanks to roshi, satellit and mkrizek for chipping in
• jreznik to check in with KDE team if they're on top of the two proposed KDE blockers or if help would be appreciated - they had been working hard on them but still not with clear resolutions at time of meeting, jreznik would continue to monitor
• cmurf to send a quick email to test-announce asking folks to test and karma critpath updates ahead of freeze tomorrow - this was done

Fedora 20 Final status

• Test coverage for TC2+TC3 was looking fairly good, we were still missing some of the more difficult-to-test cases like SCSI, iSCSI, hardware RAID and kickstart deployment
• Current blocker bug load was large, but work was progressing on most bugs; we still had a shot at on-time release

Action items

• adamw to draft a new test case and matrix row for validating cloud image checksums

IRC Log

adamw	#startmeeting Fedora QA meeting	16:00
zodbot	Meeting started Mon Dec 2 16:00:29 2013 UTC. The chair is adamw. Information about MeetBot at http://wiki.debian.org/MeetBot.	16:00
zodbot	Useful Commands: #action #agreed #halp #info #idea #link #topic.	16:00
adamw	#meetingname fedora-qa	16:00
zodbot	The meeting name has been set to 'fedora-qa'	16:00
adamw	ahoyhoy, folks	16:00
* tflink is present		16:01
kparal	is it roll call?	16:01
kparal	topic is missing	16:01
adamw	#topic Roll call	16:01
adamw	DON'T KNOW WHAT YOU'RE TALKING ABOUT	16:01
* pschindl is here		16:01
* pwhalen is here		16:01
* satellit listening		16:01
* mkrizek is here		16:01
* nirik is lurking in the back		16:01
* kparal is here, but can't attend the blocker bug meeting in an hour. going to theatre. you now, culture.		16:02
* roshi is here		16:02
* jreznik is mostly here, needs three more minutes		16:02
adamw	.fire kparal the only culture Fedora QA folks are allowed is 'to the suburbs' (oh snap! inside joke!)	16:02
zodbot	adamw fires kparal the only culture Fedora QA folks are allowed is 'to the suburbs' (oh snap! inside joke!)	16:02
kparal	fire me again for not getting that joke	16:03
adamw	kparal: https://fedoraproject.org/wiki/QA:Testcase_audio_basic	16:03
adamw	kparal: that's the example sound track. i know the first three seconds of it...pretty well.	16:04
tflink	yeah, have heard that song a few times :)	16:04
kparal	oh right, I thought we had stallman's free software song in that test case. should be replaced	16:04
adamw	! excellent plan.	16:04
adamw	send a pathc	16:04
adamw	also a patch	16:04
brunowolff	I'm around if you need to ask about spin-kickstarts.	16:04
adamw	brunowolff: thanks	16:05
adamw	#topic Previous meeting follow-up	16:05
adamw	brunowolff: that'll probably come up in the blocker meeting, which will be in #fedora-blocker-review, just so you know	16:05
adamw	#info "adamw to draft a new test case and matrix row for validating cloud image checksums" - I totally didn't get around to doing this. too busy playing with tablets. my bad	16:05
adamw	#action adamw to draft a new test case and matrix row for validating cloud image checksums	16:06
adamw	#info "adamw to make sure the desktop tests are completed for TC2" - they're covered between TC2 and TC3, many thanks to roshi satellit and mkrizek for chipping in	16:07
adamw	and pwhalen for arm	16:07
* Viking-Ice is here		16:08
adamw	"jreznik to check in with KDE team if they're on top of the two proposed KDE blockers or if help would be appreciated" - jreznik, back yet?	16:08
adamw	ahoy viking	16:08
jreznik_	adamw: I'm here!	16:10
jreznik_	the whole team works on arm one but without any success	16:10
adamw	zoiks	16:10
adamw	Fedora 20: ARM's A Primary Arch (As Long As You're Not Super Keen On Clicking On Things)	16:11
jreznik_	one idea is to workaround the bug by moving that offending widgets to the desktop	16:11
adamw	eeexcellent, it's hack time	16:11
jreznik_	advanced panel test case would still fail in some way but at least basic usecase could be ok	16:11
jreznik_	and seems like mouse events are broken on intel too, just not such horribly	16:12
jreznik_	so plasma does something in a really bad manner	16:12
jreznik_	and it's a question how you would like this hack, making it better, not fixed with a lot of stuff still broken...	16:13
jreznik_	for the rest of kde bugs - it's more up to your testing with kdm based image, how does it look?	16:13
adamw	sounds likes classic fedora to me	16:13
adamw	seemed fine, I ran the whole desktop_login test case	16:14
adamw	of course it'll hit TC4 and explode on impact, that's just a given	16:14
roshi	yay	16:14
jreznik_	adamw: what I don't get - upstream aims on arm based tablets, they are aware of this bug for a long time and it's still there... :(	16:14
adamw	jreznik: yeah, seems weird	16:15
adamw	maybe they figure it's not super important. if they have some discussion of it somewhere where they reached that conclusion, it'd be useful to have the reference, so we can use it to justify our inevitable fudge	16:15
jreznik_	I can ask guys, they talked to sebas	16:16
jreznik_	it's *still* arm... but at least one desktop should be somehow usable... if not primary one	16:17
adamw	yeah, maybe we should've made xfce primary after all. quick change? :P	16:17
pwhalen	the desktop is still very usable, more an inconvenience. xfce works well.. it was our original blocker desktop	16:18
jreznik_	pwhalen: define "very"	16:18
adamw	#info lots of firepower has been aimed at the KDE / ARM plasma systray widget bug - https://bugzilla.redhat.com/show_bug.cgi?id=969524 - without success so far. we'll keep working on it, but may need to fudge it for release	16:19
pwhalen	besides the convenience of the plasma widgets, everything else seems to be okay.	16:19
jreznik_	and if we could be able to make it at least a bit better, it would be win	16:20
adamw	okay, sounds like we're on top of it.	16:21
adamw	"cmurf to send a quick email to test-announce asking folks to test and karma critpath updates ahead of freeze tomorrow"	16:21
adamw	well, we don't have a cmurf, but i believe he did that	16:21
adamw	#info "cmurf to send a quick email to test-announce asking folks to test and karma critpath updates ahead of freeze tomorrow" - this was done: https://lists.fedoraproject.org/pipermail/test-announce/2013-November/000821.html	16:22
Viking-Ice	pwhalen, hmm was that decided by fesco as an DE could block the ARM release	16:23
adamw	Viking-Ice: we decided it at Flock, you were there	16:23
Viking-Ice	I'm lucky if I can remember what happened last week	16:24
jreznik_	Viking-Ice: lucky man, I have troubles to remember what happend day before sometimes!	16:25
adamw	Viking-Ice: hehe :)	16:25
adamw	#topic Fedora 20 Final status	16:25
Viking-Ice	since fesco has started butting into our processes ( like they are doing with the EOL process ) it's harder to keep taps of where things got decided	16:26
adamw	...where am i? who are all you people? how do I work this machine?	16:26
Viking-Ice	jreznik_, that does just happen to me after serious drinking the night before	16:27
jreznik_	for other F20 bugs - I was able to get status for	16:27
Viking-Ice	where you wake up in strange places wearing nothing but mankini	16:27
jreznik_	other two heisenbugs are systemd ones, especially that lvm on bios raid	16:27
adamw	#info test coverage for TC2+TC3 is looking fairly good, we're still missing some of the usual suspects on the install matrix (annoying cases like SCSI, iSCSI, hardware RAID and those icky kickstart deployment ones)	16:27
jreznik_	systemd guys seems to be clueless for that one	16:27
adamw	jreznik_: the instantiated service one?	16:28
jreznik_	yep	16:28
Viking-Ice	lvm issue most likely I would say	16:28
Viking-Ice	race condition	16:28
jreznik_	Viking-Ice: it's race and they are trying to help peter but no luck so far	16:28
jreznik_	for journald - mschmidt was on PTO today but also seems like the proposed fixes are not enough :(	16:29
jreznik_	for grubby bug, pjones says it's rfe, not a blocker as it's not supported configuration	16:30
Viking-Ice	grubby has plenty of bugs	16:31
pjones	jreznik_: right - the bug is specifically about /boot on a subvolume, which isn't something anaconda will allow you to do.	16:31
pjones	Viking-Ice: wild categorizations of grubby notwithstanding, the question is if this bug is a blocker.	16:31
jreznik_	so this is my status for currently accepted blockers I was able to collect so far	16:32
jreznik_	for the proposed, I can share it at blocker review meeting	16:33
Viking-Ice	pjones, 864198 got accepted as a blocker hence it is	16:34
adamw	pjones: i thought you can do it in custom partitioning, but autopart doesn't do it.	16:34
* adamw checks the bug report		16:34
jreznik_	from non bugs related side - we are getting pretty ready	16:34
Viking-Ice	all docs an all workarounds documented in release notes ;)	16:35
jreznik_	Viking-Ice: well, the maintainer is always one party to decide what should be blocker and not	16:35
pjones	adamw: the bug specifically states that during installation he puts /boot on ext4	16:35
adamw	https://bugzilla.redhat.com/show_bug.cgi?id=864198#c9	16:35
adamw	"This bug is still alive, and anaconda will let the user create a layout as describe in comment 7"	16:36
adamw	so, the obvious alternative here if you want this to be an unsupported configuration is don't allow it in anaconda.	16:36
Viking-Ice	rifht	16:36
Viking-Ice	mean right	16:36
pjones	I really thought that was the case - let me check with dlehman, because he's the expert on that.	16:36
adamw	i can do a quick test easy enough	16:37
pjones	but in any case adding this to grubby is a fairly significant effort that's not going to happen at the last minute	16:37
adamw	okay	16:37
adamw	so basically the plan is: check if it's really allowed in anaconda, if so, patch it out	16:37
adamw	seems fine to me	16:37
pjones	(unless, you know, $mystery_coder shows up with some patches)	16:37
jreznik_	seems like	16:37
Viking-Ice	jreznik_, I would say more the input from developers is appreciated but the actual decision making if an bug is a blocker is not	16:38
adamw	Viking-Ice: per policy it's supposed to be a shared decision between qa, devel and releng	16:38
adamw	though lately it's mostly qa folks showing up to meetings	16:38
adamw	#chair roshi viking-ice	16:38
zodbot	Current chairs: adamw roshi viking-ice	16:38
Viking-Ice	well historically speaking it has more or less been that case	16:38
adamw	in the early days of meetings we did a better job at getting more people out...	16:39
adamw	i think they found they couldn't stand the four hours of sheer excitement ;)	16:39
adamw	eep. my 'quick test' produces an unrelated crasher bug. stand by	16:40
Viking-Ice	in anycase anyone who actually shows up can have a say in a vote unless he has directly related or responsible for what is being discussed and voted upon	16:40
adamw	so, https://bugzilla.redhat.com/show_bug.cgi?id=1036025 , fun. back to trying to check this.	16:42
adamw	pjones: yeah, custom part happily let me out with just a btrfs /boot, btrfs /, and swap.	16:43
pjones	ugh.	16:44
adamw	install is running now. no warnings, even.	16:44
adamw	okay, we know the status anyhow. two days to fix it, plenty of time :)	16:44
Viking-Ice	or we simply slip if necessary	16:45
adamw	sure	16:46
adamw	updated the bug	16:46
adamw	so, taking a look at the current blocker list...https://qa.fedoraproject.org/blockerbugs/milestone/20/final/buglist	16:46
mkolman	adamw: looks like a duplicate	16:46
mkolman	adamw: checking with vpodzime now	16:46
adamw	it's quite long, but some of them are bureaucracy bugs	16:46
jreznik	update for plasma arm one - guys are building fix right now	16:47
adamw	yay	16:47
jreznik	(like a real one, not a workaround but still nasty hack)	16:47
adamw	we've mostly covered the most tricky ones	16:47
* pwhalen is watching the build		16:47
adamw	we probably need to point some dcbw firepower at https://bugzilla.redhat.com/show_bug.cgi?id=1018317	16:48
mkolman	adamw: 1036025 yep is a dupe	16:48
adamw	ok	16:49
adamw	dupe it	16:49
jreznik	adamw: he should know how to fix it	16:49
mkolman	dshea already did :)	16:49
adamw	ah, cool	16:49
jreznik	jklimes added a comment there - it seems like impact *could* be limited to only particular use cases but...	16:50
adamw	jreznik: i just poked dcbw about it	16:50
adamw	#info current blocker bug load is large, but work is progressing on most bugs; we still have a shot at on-time release	16:50
jreznik	[15:55] <jklimes> danw, dcbw: jreznik pinged me to find out what happens, so that they could decide blocker/noblocker	16:50
jreznik	[15:55] <jklimes> danw: if you have a solution (in mind), go for that	16:50
jreznik	[15:55] <danw> it happens for some openvpn users, but not all, and it depends on how the openvpn server is configured, so there's no workaround on the client side	16:50
jreznik	[15:56] <danw> i don't think we have any clue what percentage of openvpn users is affected	16:50
jreznik	[15:56] <danw> and yeah, i think we know how to fix it	16:50
adamw	we can discuss the %age thing in blocker review meeting	16:51
jreznik	hmm, from the last two comments it looks like it's again unsupported config	16:52
adamw	as far as TC4 goes...we have enough stuff queued up to do it today, probably	16:52
adamw	sound ok to everyone?	16:52
adamw	i'm hoping the next build after that may be RC1	16:52
roshi	works for me	16:52
jreznik	ok, still systemd bugs are the hard ones, but we will see	16:53
Viking-Ice	so have we somehow magically moved this meeting to a blocker bug meeting	16:54
jreznik	well, topic is status :D	16:54
adamw	i'm trying to follow the agenda i wrote :P	16:54
adamw	we're about done with it now	16:54
adamw	5 minutes of open floor then blocker review meeting time, sound goo?	16:54
adamw	or good	16:54
Viking-Ice	yeah	16:54
jreznik	sure, goo	16:54
adamw	#topic open floor	16:58
adamw	sorry, forgot...anyone have anything?	16:58
adamw	alrighty	17:00
adamw	blocker review meeting should be starting up in #fedora-blocker-review shortly - see you folks voer there	17:00
* adamw sets quantum fuse		17:00
adamw	thanks for coming, folks!	17:00
adamw	#endmeeting	17:01

Generated by irclog2html.py 2.12.1 by Marius Gedminas - find it at mg.pov.lt!