15:01:02 #startmeeting Fedora QA meeting 15:01:02 Meeting started Mon Aug 8 15:01:02 2011 UTC. The chair is tflink. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:01:02 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:01:10 #meetingname fedora-qa 15:01:10 The meeting name has been set to 'fedora-qa' 15:01:19 #topic roll call 15:01:33 alrighty, let's get started 15:01:34 * nirik is lurking 15:01:51 * pschindl is here 15:02:01 * jlaska is here 15:02:03 * athmane is here 15:02:05 * kparal here 15:02:13 * mkrizek is present 15:02:23 hello nirik, pschindl, jlaska, athmane, kparal, mkrizek 15:02:46 hi 15:03:37 let's wait another couple of minutes for people to join and then get to business 15:04:26 * rbergeron is lurking in a double-booked-meeting type of way :\ 15:05:08 ooh, that sounds like fun :) 15:05:30 * Viking-Ice sneaks in late... 15:05:39 alrighty, lets get started. I'm new at this, so let me know if I'm not doing something right :) 15:05:49 #topic Previous Meeting Followup 15:05:52 tflink: meetbot is very forgiving :) 15:06:01 * adamw throws peanuts from the front row 15:06:07 SEND IN THE CLOWNS! 15:06:23 * tflink ducks and is thankful adamw isn't throwing tomatoes yet 15:06:50 unless I'm missing something, I don't see anything to follow up with from last meeting that isn't already on the agenda 15:07:13 yeah 15:07:21 * Viking-Ice is armed with lettuce heads and laughs at people with tomatoes 15:07:41 * tflink is worried 15:07:59 the vegetable artillery is out in force 15:08:02 #topic F16 Alpha Status 15:08:17 #link http://rbergero.fedorapeople.org/schedules/f-16/f-16-quality-tasks.html 15:08:55 #link F16 Alpha RC1 went out on Friday 15:09:21 * kparal made sure today we don't starve on reported bugs 15:09:31 gee, thanks ;) 15:09:33 #link http://lists.fedoraproject.org/pipermail/test/2011-August/101689.html 15:09:45 kparal: bugs are only happy if they come with criteria :) 15:09:54 hrm, that wasn't supposed to be a link 15:09:57 #undo 15:09:57 Removing item from minutes: 15:09:58 we need to make another rc from the looks of it 15:10:00 #unco 15:10:02 #undo 15:10:02 Removing item from minutes: 15:10:11 #info F16 Alpha RC1 went out on Friday 15:10:15 #link http://lists.fedoraproject.org/pipermail/test/2011-August/101689.html 15:10:16 so as an overview it looks like test coverage on rc1 install is fairly good 15:10:21 base and desktop need doing 15:10:33 and it's looking better than tc1 but clearly seems like we're gonna have to go to rc2 15:10:46 did we get the new anaconda and NM in RC1? 15:10:56 dennis says the new anaconda was in there 15:11:03 and I think NM should be as i listed it on the ticket 15:11:23 * rbergeron notes she is getting ready to ship out the mail for wednesday's go/no-go meeting, FYI. 15:11:26 yup I encountered bug 728700 and dgilmore mentioned some new file conflicts 15:11:36 an updated patch for the NM anaconda fix just landed on anaconda-devel@ 15:12:11 rhe entered a pass for "QA:Testcase_install_repository_HTTP/FTP_graphical" so that would suggest networking is better in rc1 15:12:12 ^---> but this is for NTH 727951 15:12:13 * nirik wonders if we could get abiword in as a NTH... 15:12:21 adamw: rc1 had anaconda and nm updates 15:12:22 nirik: sure. file a bug 15:12:44 https://bugzilla.redhat.com/show_bug.cgi?id=716005 might apply. 15:13:14 jlaska: is that for the NM bug where interfaces are listed multiple times? 15:13:19 tflink: you got it 15:13:26 http://kojipkgs.fedoraproject.org/mash/bleed/ is the side repo. where i put the packages we need to pull in 15:13:31 well, if anaconda team does a build in time we'll pull it into rc2, i guess. 15:13:36 and given there are some new proposed anaconda blockers, that might end up landing in Alpha still 15:13:39 #link https://fedoraproject.org/wiki/Current_Release_Blockers#Proposed_Blockers 15:13:45 so i guess we should be going through the blockers filed so far and voting on those 15:13:57 get the base and desktop testing done 15:14:08 and then start pulling together a list of updates for rc2 15:14:12 adamw: I assume that you mean right now, no? 15:14:15 * rbergeron nods 15:14:16 tflink: right 15:14:22 adamw: heads up, that fedora-logos update needs an accompanying generic-logos update included 15:14:24 tflink: do it async via comments 15:14:29 jlaska: ok 15:14:39 adamw: discussing that in #fedora-releng 15:14:46 yay to depcheck for finding that :) 15:14:57 tflink: kparal: jskladan ^^^ 15:15:04 good to hear :) 15:15:08 * kparal yays 15:15:15 alrighy, time for some proposed blockers 15:15:22 Depcheck html results found fedora-logos problem - http://autoqa.fedoraproject.org/results/150310-autotest/10.5.124.164/depcheck/results/fedora-logos-16.0.0-.html 15:15:38 #topic https://bugzilla.redhat.com/show_bug.cgi?id=692433 15:15:42 #info Cannot connect to Gnome keyring daemon dialog becomes annoying 15:15:51 jlaska: that's not strictly a blocker issue, of course, since generic-logos isn't on the dvd. 15:15:58 (is it?) 15:16:20 fyi that bug is against F15 15:16:20 adamw: it is not, you are correct 15:16:34 see later comment 15:16:38 comment #6 from rhe 15:16:40 adamw: we'll have to think about this for future updates when autoqa blocks bad deps ... but for now, we're clear 15:16:44 Viking-Ice: true but rhe just reproduced it on F16 15:16:47 that explains why it's proposed as a blocker 15:17:09 ugh 15:17:30 well, this seems pretty cut'n'dry for existing Alpha criteria 15:17:32 The only thing I can think of for blocker criteria on this one is that it impedes testing 15:17:52 tflink: it's the same criterion we accepted the NM bug under 15:17:52 #info May impact Alpha criteria "The installer must be able to report failures to Bugzilla, with appropriate information included " 15:17:54 nvm 15:17:56 it prevents submission of bugs 15:18:02 I missed rhe's last comment 15:18:07 good news is there's a fix already, by the looks of it... 15:18:07 oh that criteria 15:18:49 proposed #agreed 692433 - AcceptedBlocker - The installer must be able to report failures to Bugzilla, with appropriate information included" 15:18:55 ack 15:18:57 nack 15:19:11 this criteria is failirue in it self if you are not network connected 15:19:21 Viking-Ice: not true 15:19:29 ack 15:19:37 Viking-Ice: it prompts and allows the user to activate the network via nm-c-editor 15:19:44 (or at least that's the intended use) 15:19:49 and if no network present 15:19:52 et all 15:19:54 Viking-Ice: it's worded that 'the installer must be capable of' 15:19:58 then install a network adapter :) 15:20:00 Viking-Ice: not 'all users must always be able to' 15:20:15 Viking-Ice: i.e., _as long as you provide it with the appropriate conditions_, the installer should be able to handle reporting bugs via the network 15:20:23 obviously if there's no network, that's not an infringement of this criterion :) 15:20:49 ok, we're at 3+ and 1-. Any other votes? 15:21:02 I can file another one if we block this one based on I could not save the traceback to a stick 15:21:14 but had to choose "logger" 15:21:23 Viking-Ice: you can, but that's not explicitly covered by the above criteria 15:21:33 Viking-Ice: we can discuss that after this list of bugs 15:21:55 #agreed 692433 - AcceptedBlocker - The installer must be able to report failures to Bugzilla, with appropriate information included" 15:22:05 #topic https://bugzilla.redhat.com/show_bug.cgi?id=728700 15:22:13 #info Unhandled exeption occured Livecd RC1 spin 15:22:27 oops, looks like I should move the Blocks to the other bug it got DUP'd against 15:22:55 I'm not able to login to the desktop to test this issue yet, but if this is impacting all live installs, then yeah it meets Alpha criteria 15:23:16 #info May impact Alpha criteria "The installer must boot (if appropriate) and run on all primary architectures from default live image, DVD, and boot.iso install media " 15:23:31 #link https://bugzilla.redhat.com/show_bug.cgi?id=720070 - AttributeError: 'Iso9660FS' object has no attribute 'labelType' 15:24:14 jlaska, happens on the dvd as well same issue 15:24:17 "You must be the first to try a liveusb installation with anaconda's new bootloader code on x86. I'm working on a patch." 15:24:27 implies that it happens to all live usb installs 15:24:39 oh, so it's specific to setting the keyboard to Icelandic? 15:24:40 I used usb not plastic 15:24:48 jlaska: looks like no 15:25:39 jlaska: from the error type, it looks like something's assuming it's running from an optical disc-type-thing 15:25:39 proposed #agreed - 720070 - AcceptedBlocker - The installer must boot (if appropriate) and run on all primary architectures from default live image, DVD, and boot.iso install media 15:26:07 jlaska: so i expect this happens if you install via usb, whether writing a live image to usb or the dvd iso to usb 15:26:11 so it's a problem when running the installer from a USB ... okay 15:26:30 adamw: err, do you mean installing to a USB device? 15:26:37 rather than installing _from_ a USB device? 15:26:41 no, i think from one. 15:27:02 ah okay 15:27:28 i'd vote provisional +1 given that understanding: we can confirm with dlehman in the bug and revisit if we're wrong? 15:27:42 we don't differentiate between the device type used when installing from ISO media 15:27:51 Viking-Ice: btw...your USB stick isn't made of plastic? :) 15:27:54 yeah, it just says live images 15:28:11 but yeah, I'd be provisional +1 and we can consider adjusting the criteria as needed 15:28:20 we've been a bit fussy about 'supporting' install from dvd-written-to-usb before, but i think we've been pretty clear that live-written-to-usb is supported. 15:28:34 brb, call of nature 15:29:14 ack on the proposal 15:29:34 ok, I've got +2 on the proposal 15:29:37 ack 15:29:48 #agreed - 720070 - AcceptedBlocker - The installer must boot (if appropriate) and run on all primary architectures from default live image, DVD, and boot.iso install media 15:29:52 and we can reevaluate should more info from dlehman turn up a more precise failure condition 15:29:57 s/condition/scenario/ 15:30:00 #topic https://bugzilla.redhat.com/show_bug.cgi?id=728868 15:30:09 #info anaconda could not parse metalink during the net installation 15:30:26 This is now fixed, dgilmore added a MirrorManager redirect so that path=fedora-16-ALpha is not recognized 15:30:46 however, code will be needed to ensure this bug doesn't come up every milestone 15:30:47 ok so close? 15:31:01 well, it doesn't sound like it was quite fixed yet 15:31:04 Viking-Ice: no ... I'd suggest we can move this to Beta since the original problem has been resolved 15:31:07 it was band-aided 15:31:09 but future updates are still needed 15:31:17 so NTH for Alpha, or propsoed Beta maybe? 15:32:17 I'd vote for NTH for Alpha ... looks like clumens has a patch out for testing already 15:32:27 proposed #agreed AcceptedNTH, proposed Beta Blocker - the symptoms have been band-aided and patches have been proposed to fix this in anaconda 15:32:31 ack 15:32:35 ack 15:32:45 ack 15:32:52 #agreed AcceptedNTH, proposed Beta Blocker - the symptoms have been band-aided and patches have been proposed to fix this in anaconda 15:33:03 #topic https://bugzilla.redhat.com/show_bug.cgi?id=728891 15:33:12 #info rd.luks=0 is not removed when installing with encrypted root partition '/' 15:33:37 hum dont think encrypted partition fall under alpha criteria 15:33:44 they do 15:33:45 The installer must be able to complete an installation using the entire disk, existing free space, or existing Linux partitions methods, with or without encryption or LVM enabled 15:33:53 great yet another thing added 15:33:56 yeah, what tflink said 15:33:59 Viking-Ice: that's been there for ages 15:34:04 well, we haven't added this ... been there for some time 15:34:15 ever since the big criteria re-write in fact 15:34:16 proposed #agreed - 728891 - The installer must be able to complete an installation using the entire disk, existing free space, or existing Linux partitions methods, with or without encryption or LVM enabled 15:34:25 ack 15:34:26 the Alpha covers install storage scenarios you can hit at the *first* partition screen ... no detailed/custom partitioning 15:34:26 yeah, I thought this was added F15 sometime 15:34:36 ack 15:34:37 ack 15:34:47 #agreed - 728891 - The installer must be able to complete an installation using the entire disk, existing free space, or existing Linux partitions methods, with or without encryption or LVM enabled 15:34:58 #topic https://bugzilla.redhat.com/show_bug.cgi?id=728657 15:35:07 #info fileconflicts failure in 16-Alpha.RC1 DVDs 15:35:14 Viking-Ice: https://fedoraproject.org/wiki/Fedora_13_Alpha_Release_Criteria <--- read it and weep =) 15:35:33 the /usr/lib/python2.7/site-packages/report/io/NewtIO.py conflict is already tracked in another bz, right? 15:36:00 as a process issue...when this test fails we should probably have multiple bugs for each conflicting package set 15:36:05 yeah 15:36:06 a single bug is much harder to wrok with 15:36:07 proposed #agreed - 728891 - AcceptedBlocker - Fileconflicts are interfering with media composition 15:36:40 and it winds up assigned to someone who can't really fix it... 15:36:40 ack 15:36:48 ack 15:36:49 we need to split this out and assign each bug to an appropriate maintainer 15:37:15 ack 15:37:25 adamw: do I hear a volunteer? 15:37:32 #agreed - 728891 - AcceptedBlocker - Fileconflicts are interfering with media composition 15:37:38 tflink: sure 15:37:49 jlaska: let me know if you find another report for the report conflict, i don't see one 15:38:03 #action adamw to split up 728891 into separate bugs and assign them appropriately 15:38:11 #topic https://bugzilla.redhat.com/show_bug.cgi?id=728863 15:38:19 adamw: will do 15:38:20 #info Unable to get to gdm login when booting F16-Alpha-RC1 Live media 15:38:28 selinux issue I belive 15:38:35 hit that one work for me with selinux=0 15:38:36 is selinux turned on in the live image? 15:39:15 I didn't think it was 15:39:19 I would think so since the workaround for me was selinux=0 ;) 15:39:20 I'm not a big fan of that as a workaround for such a common use case 15:39:32 if the workaround was built into the isolinux.cfg file ... that'd be fine 15:39:34 jlaska: yes. selinux is always turned on. 15:39:38 there's a fix for this already 15:39:43 so all we need to do is pull it into rc2 15:39:48 proposed #agreed - 728863 - AcceptedBlocker - Following on from the previous criterion, after firstboot is completed and on subsequent boots, a system installed according to any of the above criteria (or the appropriate Beta or Final criteria, when applying this criterion to those releases) must boot to a working graphical environment without unintended user intervention 15:39:55 yeah we should always pull in the latest selinux-policy 15:39:55 adamw: huh, sorry ... I didn't realize that was the case for live images 15:40:02 tflink: ack 15:40:04 ack 15:40:39 ack 15:40:45 #agreed - 728863 - AcceptedBlocker - Following on from the previous criterion, after firstboot is completed and on subsequent boots, a system installed according to any of the above criteria (or the appropriate Beta or Final criteria, when applying this criterion to those releases) must boot to a working graphical environment without unintended user intervention 15:40:54 ok, that's all of the proposed blockers that I see 15:41:20 Viking-Ice: did you have one you wanted to proposed? 15:41:25 there are no new proposed NTH and I think we can review the accepted blockers outside the meeting 15:41:28 s/d?$/d?/ 15:41:32 ugh, nm 15:41:35 ? 15:41:36 * tflink waits a minute before moving on 15:41:57 Viking-Ice: you mentioned you had another bug you wanted to propose as a blocker, is that still the case? 15:42:35 it's equally stupid use case as the report to bugzilla as in save to disk for anaconda traceback 15:42:43 I rather want us to drop the other criteria 15:43:01 then file a new one that you cant save the traceback to stick 15:43:02 so no bugs to propose then? 15:43:07 * tflink notes that we have another proposed alpha blocker 15:43:17 which is 15:43:30 I was waiting to see if you had one to propose 15:43:31 jlaska, no bugs to propose 15:43:36 #topic https://bugzilla.redhat.com/show_bug.cgi?id=729053 15:43:47 #info traceback after first iteration of device discovery 15:44:18 that's an interesting one but I'm not sure it's alpha blocker material 15:44:52 mm...yeah 15:45:03 ack on blocker 15:45:17 it's kind of a breakage of several criteria, but depends on a specific condition: an apt-based install already on the system 15:45:40 so we're in the 'room for discretion' area here 15:45:49 hrmm, no criteria that this hits ... so this may come down to ... right, as adamw said 15:46:06 I'd support NTH and commonBugs with an updates.img worst case 15:46:16 * jlaska would even write the entry 15:46:28 so we've got +1 blocker and +1 NTH 15:46:28 oh let's block on "cant save to bugzilla" but not on installer craps it self on device discovery. 15:46:33 15:46:52 ack on blocker 15:47:18 allegedly this is supposed to be fixed in anaconda-16.15 15:47:31 perhaps confirmation from anaconda team is in order 15:47:32 right, but that's the post-alpha anaconda branch 15:47:45 we'd need this to be pulled into 16.14.x to get it into rc2 15:47:51 I guess I'm +1 NTH, -1 Alpha blocker, propose for beta for final blocker 15:47:56 we don't have criteria for this, so this comes down to a judgement call on the expected user impact 15:48:11 (or Viking-Ice [or anyone else] can propose criteria for review) 15:48:11 jlaska: correction: we *do* have a criterion, but it's a conditional infringemnet. 15:48:19 adamw: yes, better phrased, thank you 15:48:33 I'm +1 NTH, -1 Blocker 15:48:50 i'm probably +1 nth and +1 beta or final blocker...it's not terribly important, as since there's a fix already, if we make it NTH, we will get the fix for rc2 15:49:26 * maxamillion is *waaaaayyyy* late ... hopefully didn't miss the whole thing (apologies) 15:49:32 maxamillion: we're on blocker bugs atm 15:49:42 proposed #agreed - 729053 - RejectedBlocker, AcceptedNTH, proposed for F16 beta blocker - This only hits systems that already have an apt based installation, so not too many people for alpha. 15:49:51 adamw: awesome :) 15:50:04 ack 15:50:07 nack it's an blocker a strong blocker 15:50:09 maxamillion: you made it in time for the last one 15:50:15 tflink: lol ... bugger :/ 15:50:25 silly $dayjob getting in the way 15:50:28 ack 15:50:38 it looks like we're +1 alpha blocker, -3 alpha blocker, +3 AlphaNTH 15:51:45 it looks like the acks have it 15:51:54 #agreed - 729053 - RejectedBlocker, AcceptedNTH, proposed for F16 beta blocker - This only hits systems that already have an apt based installation, so not too many people for alpha. 15:52:16 alrighty, that should actually be the last of the proposed blockers this time 15:52:30 let's see if we can get though the other things real quick 15:52:43 #topic EC2 Test Day Recap 15:53:04 #info we had 5 people record test results during the test day 15:53:22 #info one new bug was filed, 3 people reproduced an old issue on the instance-store AMIs 15:53:39 overall, I think it went pretty well 15:53:56 it would have been nice to have a few more people but maybe for the F16 EC2 test day 15:54:09 cool 15:54:13 are you going to do a recap mail? 15:54:39 yeah, it's on my list 15:54:57 thanks again to athmane for putting the EC2 test cases together 15:55:05 the httpd test script is really nice 15:55:30 moving on since we're short on time 15:55:32 go athmane! 15:55:36 #topic 08-11 Test Day slot 15:55:41 yeah, slick 15:56:02 I assume that this slot is still empty? 15:56:21 * tflink checks the wiki page 15:56:24 slots are flexible 15:56:28 #link http://fedoraproject.org/wiki/QA/Fedora_16_test_days 15:56:30 if we need more we add more 15:56:45 Viking-Ice: it's not about that, more about that we don't have a test day for this week and i wondered if anyone had any proposals 15:57:02 oh expected running out of test slots 15:58:03 adamw, is it not more up to maintainer approaching us then ourselves hosting test days for various stuff? 15:58:10 Viking-Ice: both ways work 15:58:24 i just wanted to note that the slot's empty and wonder if anyone had plans...if not, oh well, we'll move on 15:58:28 not like we're not busy enough already =) 15:58:30 do we want to do another gnome3 test day for F16 (probably not this week, though)? 15:58:47 tflink: there probably should be one, yeah... 15:58:48 rule of thumb dont test *DE until after alpha release 15:59:02 yeah, I wasn't proposing it for this week 15:59:03 tflink: should really be a to-do for me i guess 15:59:47 I can't really think of any good ideas off the top of my head 15:59:58 and I'm not hearing any right now, either 16:00:03 well, let's move on then! 16:00:07 yep 16:00:27 #info send any test day suggestions to the lists 16:00:38 #topic Open Discussion 16:00:48 well, that's all that I can think of for today, any other topics? 16:00:57 thanks to adamw for sending out the agenda 16:01:26 since we're already over time, I'm setting the fuse to #endmeeting at 2 minutes 16:01:48 not much 16:01:57 if people could help with the base and desktop validation matrices that'd be great 16:02:02 the base tests in particular are easy 16:02:08 for desktop i guess you'll need to use selinux=0 =) 16:02:17 #link https://kparal.wordpress.com/2011/08/05/autoqa-0-6-released/ 16:02:28 how could I have forgotten that? 16:02:28 #info yay ... autoqa-0.6.0 was released :) ... nice job autoqa-devel@ 16:02:48 kparal: feel like doing an update on autoqa? 16:02:50 it surprisingly works it seems 16:03:04 ah how much faith is in these words 16:03:16 tflink: I think it's pretty much covered in the announcement 16:03:39 let's not waste time by repeating it 16:03:51 ok, works for me 16:04:10 yup, sounds good 16:04:30 any other topics? 16:04:33 When is it appropriate to start thinking about autoqa-0.6.1 (or 0.7.0) 16:05:24 we can set up another planning meeting for later this week or early next week 16:05:35 cool, sounds good 16:05:43 I've got nothing else ... happy Alpha testing everyone :) 16:06:00 * tflink resets fuse for 3 minutes 16:07:52 OK, that's all for today. Thanks for coming everyone and have fun testing the Alpha!! 16:08:09 will send out minutes to the list 16:08:12 #endmeeting