QA/Meetings/20130916

Attendees

• tflink (111)
• Viking-Ice (32)
• kparal (31)
• pwhalen (12)
• dgilmore (10)
• jreznik (10)
• satellit_e (9)
• zodbot (6)
• satellit (6)
• jskladan (4)
• pschindl (3)
• roshi (1)

Agenda

• Followup from previous meeting
• Fedora 20 Planning/Status
• Open floor

Followup from previous meeting

• tflink to pester handsome_pirate about cloud testing coordination and/or talk to cloud folks about testing and eventual location of images - tflink followed up to some degree, but the status of cloud images and testing did not seem satisfactory, further action required

Fedora 20 Planning/Status

F20 Alpha RC2 Testing Status

• All known issues were being tracked via blocker bugs

Arm Hardware Status

• kparal raised concerns about a lack of links to ARM-specific documentation for certain tests

Open floor

• testcase_stats page tracks how often each validation test case has been run (and with what result), and when it was last run: re-generated every hour
• An inconsistency was identified between requiring the images to boot from physical media at Alpha, but only requiring them to be appropriately sized at Beta: tflink would propose criteria changes

Action items

• tflink to make sure some cloud testing is getting done
• tflink to file blocker bug about DOA i386 AMI
• pschindl to send out karma/testing request for not-yet-stable updates in RC2
• tflink to propose moving 'boot from optical media' criterion from alpha to beta

IRC Log

tflink	#startmeeting Fedora QA meeting	15:01
zodbot	Meeting started Mon Sep 16 15:01:25 2013 UTC. The chair is tflink. Information about MeetBot at http://wiki.debian.org/MeetBot.	15:01
zodbot	Useful Commands: #action #agreed #halp #info #idea #link #topic.	15:01
tflink	#meetingname fedora-qa	15:01
zodbot	The meeting name has been set to 'fedora-qa'	15:01
tflink	#topic Roll Call	15:01
* satellit listening		15:01
* jskladan lurks		15:01
* tflink tries to figure out what he did with the agenda		15:01
* roshi here		15:02
* kparal is a bit late again, but that shouldn't surprise you		15:02
* jreznik is here, but will have to leave in a short for an hour		15:03
tflink	ah, saved it on the wrong page	15:03
tflink	#link https://fedoraproject.org/wiki/QA/Meetings/20130916	15:03
* pschindl is here		15:04
* tflink waits another minute or so for folks to show up		15:05
tflink	ok, it's been 5 minutes - anyone not here yet will miss out on all the fun	15:06
* pwhalen is here		15:06
tflink	#topic previous meeting followup	15:06
tflink	the only thing I'm seeing is:	15:06
tflink	#info	15:06
tflink	#info tflink to pester handsome_pirate about cloud testing coordination and/or talk to cloud folks about testing and eventual location of images	15:06
tflink	hrm	15:06
tflink	#undo	15:06
zodbot	Removing item from minutes: <MeetBot.items.Info object at 0x29dfb3d0>	15:06
tflink	#undo	15:06
zodbot	Removing item from minutes: <MeetBot.items.Info object at 0x28f4c950>	15:06
tflink	#info tflink to pester handsome_pirate about cloud testing coordination and/or talk to cloud folks about testing and eventual location of images	15:07
* Viking-Ice joins in 50%		15:07
tflink	I think I pestered him about it, but I can't remember the outcome	15:07
kparal	well, repeat the procedure :)	15:08
tflink	yeah, I just found my irc logs and am realizing that I didn't get quite all of the information	15:08
tflink	but most of it :)	15:08
tflink	not sure there's much else to say on this	15:09
jreznik	what's missing? if you got most of it	15:09
tflink	eventual location, but I think that's also been figured out	15:09
tflink	theoretically, the cloud sig is helping to test the cloud images	15:09
jreznik	and from email, mattdm: This still doesn't boot. (We're working on it.) - from Sep 13	15:09
tflink	yeah, the i386 AMI hasn't been booting for a while	15:10
tflink	like I said, theoretically	15:10
tflink	location is determined by releng, I think - new Images dir	15:11
tflink	testing ... need to ping rbergeron about possibly getting people access to a fedora amazon account	15:11
tflink	but that's something to deal with outside the QA meeting	15:12
tflink	#action tflink to make sure some cloud testing is getting done	15:13
tflink	anything else here"	15:13
tflink	?	15:13
jreznik	and also how to track i386 ami better	15:13
Viking-Ice	well is that not best handled by the cloud sig	15:14
tflink	Viking-Ice: if they were actually doing the testing, sure	15:14
tflink	jreznik: not sure I understand?	15:14
tflink	i386 ami was just never tested	15:15
Viking-Ice	tflink, they are the ones that should be doing the testing	15:15
Viking-Ice	we should just help them prepare for it	15:15
tflink	ok, moving on	15:15
jreznik	tflink: I mean, we know it does not work, it's potential blocker, you did not know where/how to report it and it's not on the blocker bugs list (so could be easily missed, hard to get udpates...)	15:15
tflink	jreznik: good point. it's still not blocking release	15:16
tflink	#action tflink to file blocker bug about DOA i386 AMI	15:16
tflink	Viking-Ice: I'm not disagreeing with you but the other option is DOA cloud images	15:17
tflink	#topic F20 Alpha RC2 Testing Status	15:17
jreznik	tflink: yep and you know how I hate last minute surprises when someone brings it up :)	15:17
tflink	jreznik: meh, it adds excitement. what good would a boring, predictable release be :-D	15:18
jreznik	:D	15:18
Viking-Ice	but back to topic where are we at with regards to the alpha blocker bugs and testing	15:18
tflink	any huge issues from RC2 testing thus far?	15:18
tflink	other than the current blocker issues	15:18
satellit	I am having a problem using anaconda liveinst to install to an external USB HD....	15:19
tflink	hopefully, we'll have a new anaconda build today for #1007387	15:19
kparal	I reported just a few cosmetic issues, I haven't seen any other big issue	15:19
Viking-Ice	what about 1006113	15:19
jreznik	btw. for current blocking issues - anaconda guys did some mess in the list by cloning blocker bugs - fedora stuff should be cleaned from that bugs	15:20
kparal	tflink: will we get RC3 once that happens?	15:20
tflink	Viking-Ice: it needs to be closed	15:20
Viking-Ice	ah yes	15:20
tflink	I forgot to secretarialize after go/no-no	15:20
tflink	kparal: yep	15:20
tflink	jreznik: i thought that got cleaned up post-clone	15:20
Viking-Ice	what's our karma status on the bugs and their fixes	15:20
tflink	there are a few builds missing karma for stable	15:21
jreznik	tflink: seems like it wasn't	15:21
tflink	pschindl: were you planning to do a F20 alpha karma request?	15:22
tflink	I see several updates that should be stable once they get enough karma	15:22
pschindl	I was. But haven't done it yet.	15:22
Viking-Ice	pschindl, lure in karma via badges "Earn your badges test and provide karma" yata yata	15:23
tflink	#action pschindl to send out karma/testing request for not-yet-stable updates in RC2	15:23
Viking-Ice	hm we actually should be monitoring if karma brings in more people	15:23
Viking-Ice	mean badges	15:23
pschindl	Viking-Ice: thanks for the idea for the mail :)	15:24
* tflink wonders if it would be worth requesting/making badges for providing karma on blockers		15:24
kparal	tflink: would be hard to query for and maintain	15:24
Viking-Ice	tflink, dont think so	15:24
tflink	yeah, something for another day at least	15:25
Viking-Ice	have filed for a zombie badges for testing in the test matrix	15:25
tflink	ok, anything else on testing status?	15:25
tflink	#topic ARM Status	15:26
pwhalen	tflink, arm bugs look resolved, testing the new anaconda today which should clear up our initial-setup blocker	15:26
satellit	soas will start if lightdm is added to .ks and -gdm is included in it	15:26
tflink	it looks like things are working better than they were	15:26
tflink	satellit: bug filed?	15:26
pwhalen	vexpress not booting was resolved with an update to qemu	15:26
satellit	pbrobinson is aware but away atm	15:27
tflink	satellit: is there a bug filed?	15:27
tflink	if there's no bug, it's not supposed to change	15:27
Viking-Ice	no bug no fix ;)	15:27
satellit	no I will do so	15:27
tflink	cool, thanks	15:27
tflink	I'm hoping RC3 will be the last for f20 alpha, so I'd suggest doing it soon	15:28
satellit	I did a test remix with this fix here: http://wiki.sugarlabs.org/go/Fedora_20#SoaS_86_64-dm_.28remix.29	15:28
tflink	pwhalen: sounds like arm is in good shape for f20 alpha, though?	15:28
kparal	I'm not sure why there are white fields in the general tests in installation matrix for ARM, when the only test case available speaks about dd-ing images to a SD card	15:28
pwhalen	tflink, I'll confirm anaconda, but yes it is. The fix was very small for our setup issue	15:28
pwhalen	https://admin.fedoraproject.org/updates/FEDORA-2013-16735/anaconda-20.16-1.fc20	15:29
tflink	kparal: not sure I understand your question	15:29
pwhalen	looks like someone left karma but didnt test the right version, can that be removed or we just need to get additional karma	15:29
tflink	the lack of arm-specific stuff in the test cases?	15:29
kparal	tflink: how do you run anaconda on ARM? I don't see any test case covering that	15:29
tflink	yeah, not sure on that one myself	15:30
kparal	I simply don't see any instructions	15:30
tflink	i didn't think anaconda actually ran on arm	15:30
tflink	outside of initial-setup	15:30
pwhalen	kparal, pxe boot on highbank	15:30
kparal	pwhalen: does it work with qemu as well?	15:30
pwhalen	kparal, not that I'm aware of	15:31
kparal	hmm, ok. maybe we should provide this info somehow in the matrix	15:31
kparal	I saw empty cells for ARM, but I didn't know how to fill them	15:32
tflink	they look mostly filled for RC2	15:32
kparal	as long as pwhalen takes care of it, I don't mind... :)	15:32
pwhalen	kparal, which. I think i missed - QA:Testcase_Kickstart_Nfs_Server_Path_Ks_Cfg , but all else should be done	15:33
kparal	pwhalen: for example https://fedoraproject.org/wiki/QA:Testcase_Anaconda_autopart_install has ARM field enabled	15:33
kparal	but I guess we can't put ARM-specific instructions into every test case	15:33
kparal	we need to have a general guide at the top of the matrix or something like that	15:34
* jreznik has to leave now for a short, will read the log		15:34
tflink	but some more links to ARM info might be useful?	15:34
kparal	definitely, but I wouldn't duplicate it into every test case	15:34
pwhalen	kparal, I dont see any adjustments needed, boot the installer by any means	15:34
kparal	something like "ARM testing" info box at the top of the page, perhaps?	15:35
tflink	#info concerns about a lack of links to arm documentation were raised	15:35
pwhalen	tflink, lack of links to arm documentation? that testcase seems pretty straightforward	15:35
kparal	pwhalen: well, simple people like me don't realize that it means pxeboot (as the only method) :) I was looking for some real installation images, and they are not there, of course	15:36
tflink	kparal: can you think of a better way to phrase the #info above?	15:37
tflink	#chair kparal	15:37
zodbot	Current chairs: kparal tflink	15:37
tflink	pwhalen: I've not run through it before, to be honest	15:38
kparal	#info People with no or little experience with ARM might wonder where ARM installation images are. We might consider adding a brief info to the installation matrix wiki page.	15:38
tflink	ok, anything else here?	15:39
* tflink assumes not		15:40
tflink	#info from earlier discussion, there are no currently known major issues with F20 alpha	15:40
tflink	which brings us to ...	15:40
tflink	#topic Open Floor	15:40
tflink	I have one thing for open floor if nobody else has anything?	15:40
kparal	I think we should mention that thanks to jskladan we have a simple way of seeing which test cases haven't been tested for a long time	15:41
kparal	http://testdays.qa.fedoraproject.org/testcase_stats/	15:41
kparal	it's regenerated every hour or so	15:41
* jskladan blushes		15:41
satellit_e	https://bugzilla.redhat.com/show_bug.cgi?id=1008569	15:41
satellit_e	not sure how to mark it for FE	15:41
tflink	#link http://testdays.qa.fedoraproject.org/testcase_stats/	15:42
tflink	satellit: mark it as blocking 980650	15:42
satellit_e	ok	15:42
Viking-Ice	+1 FE	15:42
tflink	#info test case run statistics are regenerated ~ every hour and available at the url above	15:43
Viking-Ice	should not be any problem to the release blocking desktop adding those missing requirements to the soas ks file	15:43
tflink	jskladan: I thought you were supposed to change those dns entries	15:43
tflink	they were going to be changed, rather	15:44
tflink	Viking-Ice: sure, but we still need to go through the motions	15:44
jskladan	tflink: I'm not sure what do you mean right now, to be honest	15:44
jskladan	to be honest, I just spawned the regular cronjob there, since it's easily accessible public machine :)	15:45
tflink	jskladan: we can talk after the meeting, not sure what the resolution was	15:45
Viking-Ice	tflink, those here can just vote on the bug now	15:45
Viking-Ice	mean the fix for this should relatively straight forward and actually faster then us voting on it	15:46
pwhalen	+1 FE	15:46
tflink	satellit: you add AcceptedFreezeException after it's accepted, not at proposal time :)	15:46
tflink	+1 FE	15:47
satellit_e	oops sorry	15:47
tflink	#info 980650 accepted as FE for F20 alpha, +3 votes in meeting	15:47
tflink	ok	15:48
* satellit_e wish there was a dropdown with 980650 in bugzilla.....		15:48
satellit_e	thanks	15:48
tflink	satellit_e: you'll be able to propose through the blocker tracking app again once I get it updated after the meeting	15:48
tflink	the other thing I wanted to bring up was optical media	15:49
tflink	size limits are not blocker issues for alpha, but booting from optical media is	15:49
Viking-Ice	yeah we should remove it	15:50
tflink	how do we want to handle the fact that the DVD.iso won't fit on a DVD?	15:50
kparal	Viking-Ice: remove what?	15:50
tflink	Viking-Ice: not going to happen	15:50
kparal	"Release-blocking live and dedicated installer images must boot when written to optical media of an appropriate size (if applicable) and when written to a USB stick with at least one of the officially supported methods."	15:50
Viking-Ice	kparal, tflink why not	15:51
Viking-Ice	with the exception of the releng DVD the SIG's themselfs control the target size	15:51
tflink	Viking-Ice: people still use it	15:51
* satellit_e not possible to split DVD to #1 and #2 with command to switch to #2 in boot?		15:51
tflink	but we're talking about the DVD, not the spins	15:51
kparal	so if I wanted to be super sneaky, if you can't write the image to a medium, it doesn't violate the criterion	15:51
tflink	kparal: yeah, I was thinking something similar but figured it was worth discussion	15:52
Viking-Ice	tflink, people still have floppy drives and your point being	15:52
kparal	on the other hand, I could say that the ability to write it is a pre-requisite to testing this criterion and therefore it is a test blocker	15:52
dgilmore	tflink: we have to make it smaller	15:52
tflink	Viking-Ice: still have and still use are not the same thing	15:52
Viking-Ice	tflink, I'm pretty sure they try to use it once in a full moon ;)	15:52
Viking-Ice	we should really just stop shipping the dvd	15:52
kparal	we don't have the concept of TestBlocker in Fedora QA yet, but I think we should have it	15:52
dgilmore	Viking-Ice: thats a discussion for another day	15:53
tflink	kparal: i was thinking that booting from optical media could be moved to beta blocker, with the size limits	15:53
dgilmore	for today we will need to remove things to mak ethe dvd small enough to fit	15:53
dgilmore	tflink: always a option also	15:53
Viking-Ice	dgilmore, we should not be hitting sizing issue every god dam release cycle	15:53
tflink	or we move size limitations to alpha, but it's a little late to be mucking around with that	15:53
kparal	tflink: I'm in favor of that. but they should continue to boot in VM from .iso	15:53
tflink	yeah, I'm also OK with USB and iso in VM for alpha	15:53
satellit_e	and dd USB	15:54
dgilmore	Viking-Ice: blame the developers and packagers :)	15:54
kparal	tflink: size limits are Beta or Final?	15:54
tflink	#action tflink to propose moving 'boot from optical media' criterion from alpha to beta	15:54
dgilmore	kparal: Beta	15:54
tflink	beta	15:54
Viking-Ice	dgilmore, is it not the comps crew that's to blame here ;)	15:54
kparal	tflink: sounds good	15:54
Viking-Ice	tflink, just put an either or clause in the criteria	15:54
dgilmore	Viking-Ice: the comps maintainers are the packagers	15:55
tflink	Viking-Ice: either or?	15:55
Viking-Ice	dvd or usb	15:55
dgilmore	Viking-Ice: and I really dont think its comps to blame. but i could be wrong. software seems to be constantly growing deps	15:55
tflink	or just move the dvd criterion to beta	15:55
Viking-Ice	we can add a hard beta criteria but it's fine to have the criteria for alpha	15:55
Viking-Ice	( it's going to be even worse hitting size-ing issues in beta )	15:56
tflink	what is the point of having an optical media criterion for alpha if we have no size criterion?	15:56
kparal	I wouldn't split optical criteria between Alpha and Beta just because of DVD. it adds complexity	15:56
dgilmore	tflink: none	15:56
* satellit_e 2 DVD w/ different Desktops ?		15:56
dgilmore	lets move optical media to Beta with size	15:56
tflink	but we're running short on time, if there's nothing else, I propose we move conversation to #fedora-qa and/or test@	15:56
Viking-Ice	yeah sure	15:56
Viking-Ice	tflink, just move it	15:57
tflink	if there's nothing else ...	15:57
* tflink sets the fuse for (0,2] minutes		15:58
tflink	Thanks for coming, everyone!	15:58
tflink	#endmeeting	15:58

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