17:00:38 #startmeeting F15Alpha Blocker Review 17:00:38 Meeting started Fri Feb 18 17:00:38 2011 UTC. The chair is jlaska. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:38 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:00:46 yo 17:00:47 #meetingname f15-alpha-bug-review 17:00:47 The meeting name has been set to 'f15-alpha-bug-review' 17:00:49 yo! 17:00:52 #topic Roll call .. 17:01:06 .fasinfo heffer 17:01:09 heffer: User: heffer, Name: Felix Kaechele, email: felix@fetzig.org, Creation: 2008-07-04, IRC Nick: heffer, Timezone: Europe/Berlin, Locale: en, Extension: 5109666, GPG key ID: 0xB30D8F7C, Status: active 17:01:10 * rbergeron is here 17:01:13 heffer: Approved Groups: cla_fedora cla_done packager fedorabugs ambassadors cvsl10n 17:01:28 * adamw decided to show up for a few minutes this week 17:01:28 * tflink is here 17:01:31 you lucky, lucky people 17:01:38 uh oh :) 17:01:41 * tflink is in awe 17:01:42 * Viking-Ice pops in.. 17:01:46 :) 17:01:50 * nirik is lurking around. Feel free to ping me if there's anything I can help with. 17:02:01 nirik: will do 17:02:04 * mclasen colurks 17:02:14 heyo 17:02:14 * rbergeron bows to adamw 17:02:24 * bcl waves 17:02:31 rbergeron: ah, the correct form of address 17:02:38 i shall let you live 17:02:41 * jlaska waits another minute to get started 17:03:58 we have someone from rel-eng, I think dgilmore is around 17:04:13 #topic Introduction 17:04:23 Under 1min reminder of why we are here ... 17:04:26 #link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting 17:04:49 We are here to determine whether the proposed bugs meet the Alpha criteria, and to ensure that accepted blockers are making progress 17:04:55 #link https://fedoraproject.org/wiki/Fedora_15_Alpha_Release_Criteria 17:05:02 I have 4 queries we can work out of today 17:05:17 Accepted blockers, proposed blockers, accepted NTH and proposed NTH 17:05:22 any preference where to start? 17:05:27 * jlaska leaning towards blockers 17:05:30 +1 17:05:31 that order looks fine 17:05:32 alphabetically 17:05:36 any volunteers to help with updating bz's 17:05:37 maybe proposed ahead of accepted 17:05:45 clumens: okay... gimme 10 minutes to update my queries! :P 17:05:59 * rbergeron can help with updating bz's 17:06:05 adamw: you want proposed blocker+nth ... then accepted... 17:06:06 when i'm told what to do, i can surely help update bzs 17:06:12 rbergeron: heffer thanks 17:06:33 Just another reminder ... *yesterday* was the Alpha RC compose date 17:06:41 jlaska: proposed blocker, accepted blocker, proposed nth, accepted nth 17:06:49 any further delays may likely introduce delays in releasing the Alpha 17:06:50 jlaska: that way we can discuss all actual accepted blockers together... 17:07:24 * dgilmore is here 17:07:28 * jsmith-busy is here (lurking) 17:07:32 * brunowolff arrives a bit late 17:07:40 hey all 17:07:42 * bcl waves to brunowolff 17:07:46 okay, starting with proposed blockers 17:08:04 #info proposed blockers - http://bit.ly/f15-alpha-blocker-proposed 17:08:13 #info accepted blockers - http://bit.ly/f15-alpha-blocker-accepted 17:08:29 going alphabetically as suggested by clumens ... 17:08:33 (by component) 17:08:43 #topic https://bugzilla.redhat.com/show_bug.cgi?id=663294 17:08:44 Bug 663294: medium, low, ---, bcl, ASSIGNED, SystemError: error return without exception set 17:09:06 I added this for review earlier today 17:09:14 seems this has been happening on and off for a while 17:09:24 I built a custom live image with the latest kernel to explore how the live installs looked 17:09:28 and I keep hitting this bug 17:09:46 If this is blocking all live installs, I think the Alpha criteria may apply ... 17:09:50 " The installer must boot (if appropriate) and run on all primary architectures from default live image, DVD, and boot.iso install media" 17:09:57 I say aye. Now that more are seeing it. 17:10:19 we don't have a lot of time against the live installer, because of several previous issues tha are now resolved 17:10:23 what exactly is the nature of this bug and who does it affect 17:10:28 * rbergeron thinks wwoods needs a beefymiracle cloak for irc 17:10:30 mean only custom builder 17:10:39 Viking-Ice: that remains to be seen 17:10:40 rbergeron: haaah 17:10:51 I'd like to see if others are seeing this on *official* live images 17:11:06 if the live-installer works I would think this was an NTH 17:11:25 at best 17:11:35 Viking-Ice: that's the point, this is preventing the live install from working 17:11:42 what are the chances that this bug is any better on official live images? 17:12:02 Viking-Ice: I have seen it on nightlies, I need to try the latest to see if it is more consistent for me. 17:12:11 well, shall we? 17:12:11 ...is this thing on? 17:12:11 let's not, and say we did 17:12:12 clumens: do you think we're screwed up or they are? 17:12:38 zoiks, i think we just had a mini-netsplit 17:12:50 that's right scooby 17:13:08 * bcl grabs jlaska's coat tails 17:13:12 bumpy freenode ride :P 17:13:17 so #topic ... 17:13:27 ENOTENOUGHINFO :-/ 17:13:31 right :( 17:13:45 better to be safe than sorry +1 it's a blocker 17:13:54 +1 because: it makes live install fail. 17:14:01 we see how rc1 turns out 17:14:04 +0.75 17:14:06 +1 live install should work imo 17:14:10 Viking-Ice: if it's a blocker and we don't fix it, we don't spin rc1 17:14:17 mostly +1 17:14:21 * dgilmore is on the fence. 17:14:31 the number of people hitting it is worryingly high for a tc 17:14:32 adamw: alpha rc1 17:14:38 not a whole lot of people actually run tcs 17:14:45 fixing it may be tricky. When I last looked at it, it was inconsistent and might be a race of some kind. 17:14:47 Viking-Ice: yes - we only start spinning RCs if the blocker list is empty 17:14:53 Viking-Ice: we dont make rc1 until we think we have all blocker bugs fixed 17:15:02 we need the new kernel in stable, and then we can spin up live images to test this 17:15:03 I c 17:15:16 jlaska: i tried a live image with the latest kernel last night 17:15:18 jlaska: the kernel will land today 17:15:20 jlaska: live install failed 17:15:25 jlaska: don't think it was this bug, though 17:15:34 #info Need more information to determine whether the problem is widespread 17:15:52 it hit something at partitioning 17:16:19 proposed #agreed 663294 - tentatively accepted as a blocker as this appears to impact live installs. Needs more testing to determine scope of problem. 17:16:36 jlaska: i think for now we keep it potential. and see if we can get some more info 17:16:42 adamw: did you hit 678607? 17:17:03 jlaska: looks familiar 17:17:13 dgilmore: you propose not accepting, but leaving on the proposed blocker list? 17:17:28 jlaska: yep 17:17:43 proposed #agreed 663294 - Needs more testing to determine scope of problem. Leave on proposed blocker list pending more information 17:17:47 ack/nack/patch? 17:17:51 ack 17:17:52 ack 17:17:53 ack 17:17:54 fine 17:17:59 yup 17:18:02 ack 17:18:08 #agreed 663294 - Needs more testing to determine scope of problem. Leave on proposed blocker list pending more information 17:18:20 I'll revisit when there is a nightly with the new kernel. 17:18:24 do we have enough to get more testing going on this? 17:18:30 bcl: should be later today 17:18:32 or do we wait until tomorrow for a new .iso ? 17:18:55 #info New nightly live .iso's expected later today 17:18:59 new .iso, just to make sure it isn't caused by a local build problem/old repo/cache/etc. 17:19:02 jlaska: i pushed the proposed kernel to stable last night. it will be in todays compose 17:19:17 will that link up with the nightly schedule? 17:19:22 dgilmore: anything we can do to speed up those live iso creation? 17:19:39 jlaska: teh hold up is the branched compose process 17:19:54 dgilmore: oh that's what you were mentioning in #anaconda 17:19:54 once thats done the livecd creation is actually pretty quick 17:20:10 jlaska: part of it yeah 17:20:10 hrmm, is this another blocker that needs trackign? 17:20:13 You can ask nirik to expedite things once the branch compose finishes. 17:20:25 brunowolff: oh right, thanks 17:20:30 brunowolff: nirik or I will kick it of asap 17:20:38 dgilmore: perfect, thanks 17:20:38 * nirik notes that doing the nightlies in koji means they can land ~30min after branched. ;) 17:20:58 #action dgilmore or nirik - kick off live image builds once branched compose completes 17:21:10 ooh shiny. 17:21:18 moving on ... 17:21:21 #topic https://bugzilla.redhat.com/show_bug.cgi?id=677131 17:21:22 Bug 677131: unspecified, unspecified, ---, clumens, MODIFIED, How to skip update disk prompt 17:22:02 For this issue, the installer prompts the user to select which disk to load an updates.img from 17:22:10 ... even when not requesting updates 17:22:21 the workaround is easy, click "cancel" 17:22:43 * dgilmore is inclined to push this to beta 17:22:45 i'd say for alpha it'd be more a nth 17:23:00 Only if the fix is pretty safe. 17:23:15 nice to have for alpha, but i think it needs fixed at beta time 17:23:15 I was NTH on this as well, but wanted to bring it here for feedbcak 17:23:16 yeah, this "only" affects unattended installs 17:23:20 i'm fine with waiting for beta 17:23:31 yeah, definitely not blocker, on the fence between nth and leave it alone 17:23:32 so if we are spinning anaconda for Alpha, we might take this ... otherwise post-Alpha ? 17:23:44 +1 for Beta blocker, NTH otherwise 17:23:54 +1 to what jsmith-busy proposed 17:24:04 jlaska: sounds good 17:24:05 NTH if the fix looks safe. 17:24:06 proposed #agreed 677131 - Accepted as Alpha NTH, rejected as Alpha blocker 17:24:09 ack/nack/patch 17:24:11 ack 17:24:13 ack 17:24:14 ack 17:24:16 ack 17:24:18 ack 17:24:21 note ... I've tested the updates.img that clumens provided on this ... so it seems safe 17:24:37 #agreed 677131 - Accepted as Alpha NTH, rejected as Alpha blocker 17:24:41 my response to every anaconda bug here is going to be "i'm fine with waiting for beta", incidentally 17:24:53 clumens: perfect 17:24:54 clumens: :-) 17:25:05 * dgilmore is also fine with waiting till beta for the anaconda bugs 17:25:06 clumens: f16 beta? :P 17:25:08 anyone want to update the bz ... otherwise, we can do the updates post-meeting 17:25:08 unless there's something new i haven't heard of 17:25:19 #topic https://bugzilla.redhat.com/show_bug.cgi?id=673238 17:25:20 Bug 673238: high, high, ---, anaconda-maint-list, MODIFIED, anaconda creates incorrect udev 70-persistent-net.rules entry for network device 17:25:32 * rbergeron is updating bz's 17:25:42 rbergeron: thank you :) 17:25:45 rbergeron: if you'd go ahead i could take a look at how it's done 17:25:48 I think mdomsch added this issue yesterday ... 17:25:54 "Moving to blocker, as feature is intended to be enabled by default, but isn't 17:25:57 w/o this comps change." 17:26:01 heffer: sure, though I'm still learning myself :) 17:26:09 so i think that the fix here im ok with is making sure biosdevname is installed 17:26:33 I don't think features get a free pass as blockers, do they? 17:26:49 * jlaska needs a feature criteria refresher 17:27:14 well, the goal is to have something we can test the features with 17:27:30 jlaska: if notting did fix up comps then biosdevname will be installed on systems 17:27:46 since we pull a fresh comps checkout each compose 17:27:46 dgilmore: is that the proposed fix ... updated comps? @core group? 17:27:58 man, 17:28:02 so far we've not taken feature issues as blockers 17:28:03 midair collision with adamw ;) 17:28:09 jlaska: i think there is 2 issues here 17:28:24 on the theory that the fallback for features is 'drop the feature' not 'delay the release' 17:28:26 one is that biosdevname is used at install time, but wasnt available for run time 17:28:38 i've always said this is something to discuss more fully but we never have yet 17:28:43 the second is that upstream says anaconda shouldnt write out the udev rule 17:29:05 the 2nd sounds like something we don't have a patch for yet 17:29:10 will this stop users from establishing a working network connection.. .If yes it's a blocker if no it's not 17:29:12 jlaska: right 17:29:19 Viking-Ice: i like that take on it 17:29:23 Viking-Ice: nice 17:29:26 look at the practical consequence 17:29:29 yup 17:29:57 is there a workaround if they can't establish a network connection? 17:30:01 I haven't seen any problems with TC1 or TC2 and current biosdevname that would impact the ability to install, get networking, and get updates 17:30:14 Viking-Ice: i dont believe it does its just a weirdness internally 17:30:48 then it's simply it's not a blocker weirdness can be fixed in an update 17:30:57 jlaska: we have a fix for the udev rules from notting 17:30:57 s/simple/simply 17:30:58 i think what happens is that it just wouldn't pick up the config in sysconfig that is generated by anaconda 17:31:10 bcl: oh okay 17:31:37 bcl: which is in f15-alpha-branch 17:31:44 It is in 15.19-1 17:31:58 bcl: so we have both fixes already then 17:31:59 I think they want to move that rules stuff into udev (not anaconda)? 17:32:07 or did I read that wrong 17:32:18 we have 15.20-1 now right 17:32:29 anaconda is pulling in biosdevname to its environment, though there's no RPM 17:32:33 Requires: on it. Until notting's comps fix lands, the installed system won't 17:32:36 have biosdevname automatically installed. Hence anaconda wrote out 17:32:38 persistent-net rules that couldn't be used during subsequent boots. 17:32:41 IMHO, anaconda shouldn't be writing out persistent-net rules, but let udev do 17:32:44 that. 17:32:46 - mdomsch 17:32:50 jlaska: comps fix will be in rc1 17:33:00 if approved, right? 17:33:04 or it's already fixed 17:33:12 jlaska: and the udev fix is in already 17:33:17 if already fixed ... is there anything to discuss? 17:33:18 so its already fixed 17:33:45 sure sounds like. 17:34:10 proposed #agreed 673238 - Likely not an Alpha blocker, but comps+udev fixes are already included and will be present in RC1 17:34:17 ack/nack/patch 17:34:22 ack 17:34:23 ack 17:34:23 ack 17:34:24 ack 17:34:25 ack 17:34:30 #agreed 673238 - Likely not an Alpha blocker, but comps+udev fixes are already included and will be present in RC1 17:34:39 okay that sucked, hopefully these others go faster ... 17:34:53 #topic https://bugzilla.redhat.com/show_bug.cgi?id=678070 17:34:54 Bug 678070: urgent, unspecified, ---, mgracik, MODIFIED, problem with an non ascii character in username 17:35:13 looks like we need to test the fix for this 17:35:23 not a blocker anyway 17:35:27 Fixed in firstboot-1.117-1 17:35:28 well 17:35:32 #info Fixed in firstboot-1.117-1 17:35:37 character set issues are a bit of a grey area 17:35:46 users can change the name later on since this only affected firstboot 17:35:57 s/on/ 17:36:00 i'd +1 nth for this, anyway 17:36:05 * dgilmore knows firsthand that not propperly supporting utf8 sucks 17:36:12 "When booting a system installed without a graphical environment, or when using a correct configuration setting to cause an installed system to boot in non-graphical mode, the system should boot to a state where it is possible to log in through at least one of the default virtual consoles " 17:36:19 so it depends how many would be impacted by this issue 17:36:31 well, that criterion doesn't have an 'in most cases 17:36:33 ' rider 17:36:34 right 17:36:35 hence the greynes 17:36:44 but anyway 17:36:49 i think we should just call it nth and take the fix 17:36:50 are there workarounds for this? 17:36:50 so you could get by just fine 17:36:51 makes things easier 17:37:01 workaround: don't use non-US-ascii characters in your usernam 17:37:02 e 17:37:03 adamw: true ... an update is already available 17:37:10 jlaska: workaround is obvious dont use non ascii characters in username 17:37:11 +1 for adamw 17:37:15 of course, that's sucky for some people 17:37:19 * dgilmore thinks its NTH 17:37:38 proposed #agreed 678070 - accepted as NTH for Alpha. Test feedback against updated firstboot needed to ensure this lands in Alpha 17:37:43 ack 17:37:47 ack 17:37:48 ack 17:37:49 ack 17:37:51 ack 17:37:52 * adamw loves glossing over things! 17:38:13 :) 17:38:16 #agreed 678070 - accepted as NTH for Alpha. Test feedback against updated firstboot needed to ensure this lands in Alpha 17:38:28 #topic https://bugzilla.redhat.com/show_bug.cgi?id=677797 17:38:30 Bug 677797: medium, medium, ---, bruno, MODIFIED, Broken dependency: hugin-base-2010.4.0-4.fc15.x86_64 requires libpano13.so.2()(64bit) 17:38:36 now i have to figure out a way to get a working install so i can test the fix. =) 17:38:51 "There must be no file conflicts (cases where the files in some packages conflict but the packages have explicit Conflicts: tags are acceptable) or unresolved package dependencies during a media-based (CD/DVD) install " 17:39:12 this may already be in stable 17:39:14 right, obvious blocker 17:39:16 no not yet 17:39:30 http://koji.fedoraproject.org/koji/buildinfo?buildID=228678 is marked dist-f15 17:39:49 do we need a bodhi update, or did this sneak in before branch? 17:40:12 i think it was before branching. 17:40:51 a repoquery -i libpano13 shows the fixed build in repo 'fedora', for me. 17:41:01 jlaska: we snuck it in right around branching 17:41:08 proposed #agreed 677797 - accepted as Alpha blocker - fix already available now in stable 17:41:15 ack 17:41:16 ack 17:41:19 adamw: dgilmore perfect 17:41:20 ack 17:41:21 ack 17:41:22 ack 17:41:26 #agreed 677797 - accepted as Alpha blocker - fix already available now in stable 17:41:28 i can do a 'yum install hugin', it pulls all three packages from repo 'fedora', they work 17:41:42 adamw: libpano13 one of them? 17:41:45 yes 17:41:53 if so ... we can move this to VERIFIED + CLOSED 17:41:59 jlaska: yeah 17:42:11 this came about due to developer lazyness 17:42:11 #info Appears fix is resolved, will CLOSE out bug 17:42:25 a few bugs now for bcl ... 17:42:27 #topic https://bugzilla.redhat.com/show_bug.cgi?id=678609 17:42:29 Bug 678609: unspecified, unspecified, ---, bcl, NEW, livecd-iso-to-disk started reporting problems for media checks 17:42:29 yay! 17:42:47 brunowolff added this earlier today I believe 17:42:49 not a blocker, and not a broblem for me. 17:42:54 This might be a NTH, or maybe something we could doc in common bugs. 17:43:23 i haven't hit this either 17:43:28 except with images which were actually bad, heh 17:43:38 I first noticed this a couple of days ago during som testing and thought it was a one off problem, but it was recurring last night and today. 17:44:06 I re-tested F14 and F14 .iso's with 16.1 this morning and all pass. 17:44:09 The image testing works despite the error, so I suspect the error message is wrong. 17:44:10 I say not a blocker and not a nth just a regular bug if not an actual corrupted image 17:44:14 err, F13 and F14 17:44:20 oh, i only have 15.4 17:44:30 bcl: what have you been testing with? 17:44:34 ah, i see, bm 17:44:35 nm 17:44:40 brunowolff: did you check this with another image 17:44:40 this is something we could reasonably fix as a post-Alpha update 17:44:57 I checked several images, including some that used to not get that warning. 17:45:03 or does this impact official images created by rel-eng too? 17:45:21 well if the error is in the tool it worries me less 17:45:25 I don't think it was a change in livecd-tools that caused the problem. I am not sure what did though. 17:45:28 since most people aren't going to be using the f15 build of the tool anyway 17:45:54 I think the main issue is that people testing live images will be confused. 17:45:59 most people aren't running f15 yet, and those who are probably don't want to write an alpha live usb stick from it =) 17:46:08 we just use checkisomd5 to do this 17:46:30 i expect by far the most common case will be that people write their usb stick from f14 17:46:33 I use live images for my spin testing, it's a lot faster and checper the using optical media. 17:46:36 * dgilmore thinks not a blocker 17:46:50 +1 17:46:50 proposed #agreed 678609 - not accepted as Alpha blocker or NTH. Candidate for Common_F15_Bugs or post-Alpha update 17:46:54 ack/nack/patch? 17:46:57 ack 17:46:58 ack 17:46:59 ack 17:46:59 ack 17:47:01 ack 17:47:01 ack 17:47:07 #agreed 678609 - not accepted as Alpha blocker or NTH. Candidate for Common_F15_Bugs or post-Alpha update 17:47:24 brunowolff: gimme or adamw a holler if you want to get this on Common_F15_Bugs 17:47:36 #topic https://bugzilla.redhat.com/show_bug.cgi?id=676904 17:47:37 Bug 676904: high, medium, ---, bruno, NEW, osmin.img is sometimes not a valid squashfs on livecd images 17:47:52 This one should be fixed. 17:48:08 I didn't have this issue witwh the live image I built this morning 17:48:13 I haven't seen the issue with the sub-bug reapear either. 17:48:14 using the latest kernel 17:48:31 yeah, i confirm jlaska's test 17:48:35 I'd say this is accepted as a blocker, but appears to be fixed 17:48:36 ? 17:48:38 i tested two xfce images which only differed in the kernel 17:48:44 i hit the bug with the old kernel, didn't with the new one 17:48:45 +1 17:48:53 adamw, jlaska: can one of you give me a link to your live image you used to reproduce 678607? 17:49:10 dlehman: i haven't put it anywhere public yet 17:49:16 dlehman: jlaska can probably get it to you faster than i can 17:49:26 proposed #agreed 676904 - accepted as Alpha Blocker. Appears to be resolved with latest F15 stable kernel 17:49:31 dlehman: I'll give you access ... 17:49:38 ack 17:49:39 ack 17:49:39 ack 17:49:40 ack 17:49:58 ack 17:50:28 #agreed 676904 - accepted as Alpha Blocker. Appears to be resolved with latest F15 stable kernel 17:50:39 one for mclasen ... 17:50:41 #topic https://bugzilla.redhat.com/show_bug.cgi?id=677842 17:50:42 Bug 677842: unspecified, unspecified, ---, pbrobinson, NEW, [abrt] mutter-2.91.6-4.fc15: ureg_src: Process /usr/bin/mutter was killed by signal 11 (SIGSEGV) 17:51:14 this one's pretty icky 17:51:28 hw spesific 17:51:52 not necessarily 17:52:23 note the possible dupe 17:52:26 I say it's a beta blocker 17:52:33 the other possible dupe mentioned from an nvidia tester 17:52:34 mclasen gone 17:52:40 and hongqing reported seeing it in their testing 17:52:46 without providing many details 17:53:30 see mail "Subject: Fedora 15 Alpha TC2 install testing report" 17:53:34 I have seen popups saying mutter crashed. 17:53:44 I was seeing this like crazy, and then it just started working 17:53:56 adamw: it was linked in one of the wiki results matrices 17:54:03 i was just about to check that 17:54:05 :) 17:54:28 I know that I haven't had much luck getting mutter to run well in any of the alpha images that I've used 17:54:33 due to a different issue users we getting the non-PAE kernel installed in rc1 and tc2 17:54:37 tc1 17:54:43 I'm getting the other variant ureg_src_undef as opposed to ureg_src 17:54:53 so his saying he switched to the non pae kernel doesnt make sense 17:56:02 i'd really like to have some damn clue what's going on with this bug 17:56:23 in general i wish we had someone who was actually in charge of figuring out what the hell to do with mutter crashes :/ 17:56:44 they need someone with clue to sort them out into graphics driver issues or mutter code issues or whatever 17:56:50 mclasen is offline now 17:56:57 anyone else from desktop? 17:57:01 adamw: mesa 17:57:05 Viking-Ice: right 17:57:51 I say it's not an alpha but a beta blocker ( if this is an alpha then so is this bug 677371 ) 17:57:52 Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=677371 unspecified, unspecified, ---, ajax, NEW, [abrt] mutter-2.91.6-3.fc15: ureg_src_undef: Process /usr/bin/mutter was killed by signal 11 (SIGSEGV) 17:58:05 i think we don't really have enough info 17:58:10 agreed 17:58:22 and I was hoping that fallback mode would kick in here too 17:58:29 point 17:58:29 but that's completely out of scope for this 17:59:07 proposed #agreed 677842 - not enough information to accept or reject, need input from desktop SIG. 17:59:10 ack/nack/patch? 17:59:11 ack 17:59:21 ack 17:59:24 ack 17:59:27 ack 17:59:28 we need an action item for someone to poke desktop (and possibly graphics) team 17:59:40 I pinged jrb this morning, and a moment ago 17:59:44 cool 17:59:47 ajax not around 17:59:57 ( mesa issue = 17:59:58 but wouldn't hurt to have another contact 17:59:59 ) 18:00:07 i think we break the airlied glass in case of ajax non-availability 18:00:15 #action desktop-sig - input needed to determine impact of 677842 18:00:34 anything else on this one? 18:00:36 and bug 677371 18:00:46 Viking-Ice: does yours have the same impact (prevent gnome3 from loading at all)? 18:00:53 yup crashes 18:00:57 then agreed 18:00:58 #action desktop-sig - input needed to determine impact of 677371 18:01:08 moving on ... 18:01:18 #topic https://bugzilla.redhat.com/show_bug.cgi?id=677790 18:01:20 Bug 677790: unspecified, unspecified, ---, rdieter, ON_QA, virtuoso-opensource-6.1.3 20110105 snapshot breaks nepomuk 18:01:39 not sure what criteria this impacts 18:02:04 nepomuk is pretty significant to kde functionality i believe 18:02:05 nth maybe allthou seems general bug 18:02:06 rdieter_work: ping? 18:02:13 adamw: hola 18:02:26 "For the purpose of Alpha-blocking status, please note that this also affects 18:02:26 Dolphin, our default file manager." 18:02:38 rdieter_work: we're on https://bugzilla.redhat.com/show_bug.cgi?id=677790 , what's the impact of this exactly? how bad does it break kde? 18:02:40 Bug 677790: unspecified, unspecified, ---, rdieter, ON_QA, virtuoso-opensource-6.1.3 20110105 snapshot breaks nepomuk 18:02:43 impact here is pretty bad, dolphin hangs, and is not functional 18:02:51 yeah, that's kinda bad 18:02:53 can you reach a browser 18:03:03 browser, probably 18:03:03 i think +1 nth 18:03:16 the alpha desktop requirements are met, i tested it 18:03:16 rdieter_work: can you update 18:03:24 Viking-Ice: see https://fedoraproject.org/wiki/Test_Results:Current_Desktop_Test 18:03:38 so i'm -1 blocker +1 nth 18:03:38 * Viking-Ice looks 18:03:47 Viking-Ice: updating should work 18:03:59 it does 18:04:13 ok +1 to nth -1 to a blocker 18:04:20 since the fixed build is available i think we should just go ahead and tag it and move on 18:05:09 proposed #agreed 677790 - RejectedBlocker AcceptedNTH - Fix available, will include with sufficient karma 18:05:12 ack/nack/patch 18:05:13 it has karma 18:05:14 ack 18:05:29 ack 18:05:33 ack 18:05:36 ack 18:06:04 #agreed 677790 - RejectedBlocker AcceptedNTH - Fix available, will include with sufficient karma 18:06:08 #undo 18:06:08 Removing item from minutes: 18:06:17 #agreed 677790 - RejectedBlocker AcceptedNTH - Fix available, will include since it has needed karma 18:06:46 last on the proposed blocker list 18:06:48 #topic https://bugzilla.redhat.com/show_bug.cgi?id=676827 18:06:49 Bug 676827: high, unspecified, ---, xgl-maint, NEW, keyboard with german layout doesn't work in gdm 18:07:57 not a blocker 18:08:06 workaround present and fix upstream 18:08:09 i'd definitely like this to be fixed 18:08:31 * mclasen hasn't been able to get hold any x dev yet 18:08:55 nth ? 18:09:22 it's arguable for blocker status as the workaround is pretty tricky and it does prevent you reaching a working desktop with some hardware 18:09:24 mclasen's back! 18:09:42 adamw: workaround is F2 ... do stuff ? 18:09:45 yeah 18:09:55 halfline: when you say this is fixed upstream...upstream X? 18:10:47 note he does not mention if this was an fresh install or an upgrade 18:11:37 and Icelandic keyboard layout works just fine 18:11:49 hum, that's a point 18:11:54 * adamw doesn't have a french keyboard handy 18:12:12 * dgilmore is guessing that its upgrades 18:12:23 fix is here: http://lists.x.org/archives/xorg-devel/2011-February/019377.html 18:12:25 * adamw still hoping halfline will appear :) 18:12:31 * dgilmore should get a spanish keyboard 18:12:53 I say NTH 18:12:53 Can we catch up with halfline, and come to conclusion? :) 18:12:57 * rbergeron cracks that whip of hers 18:13:22 dgilmore: that's what I suspect is the case that this is an upgrade issue 18:13:38 if it doesn't hit fresh boots / installs i care much less. 18:13:58 so if fresh install, blocker. If upgrade, post-alpha ? 18:14:09 I say not a blocker et all 18:14:26 Viking-Ice: you think the workaround is reasonable? 18:15:03 #info workaround posted at https://bugzilla.redhat.com/show_bug.cgi?id=676827#c5 18:15:05 Bug 676827: high, unspecified, ---, xgl-maint, NEW, keyboard with german layout doesn't work in gdm 18:15:10 jlaska: yup 18:15:25 let's get the info on whether it hits fresh installs first 18:15:33 okay 18:15:50 this is alpha + he does not mention if he tried logging in and changing the keyboard layout and those changes stuck after reboot 18:16:13 proposed #agreed 676827 - Unable to accept or reject. Need more information from reporter (fresh install or upgrade) 18:16:19 he can reach the internet and he can update 18:16:38 ack 18:16:43 * rbergeron +1's to jlaska. 18:16:49 ack 18:16:52 ack 18:17:01 -1 still think this is not an alpha bug 18:17:03 the workaround doesn't seem horrible to me for an Alpha, but yeah, let's get more information 18:17:15 it's not like he does not have a working keyboard 18:17:17 i'm here 18:17:22 but not in the german layout 18:17:36 mclasen is closer to the situation than me 18:17:43 i don't have anything to add 18:17:55 mclasen: do you know whether this affects fresh boots? 18:18:02 no, I don't 18:18:23 we don't know how the bogus info was getting added to /etc/sysconfig/keyboard 18:18:26 Viking-Ice: it's all about how many people this would hit ... we just want to make an informed decision before moving on 18:18:43 okay 18:18:48 so we'll wait for reporters to test 18:19:00 adamw, can you put a call out on test@? 18:19:03 jlaska: I think for alpha we should draw the line with a working keyboard vs non working keyboard 18:19:05 (if it's worthy of that) 18:19:28 Viking-Ice: we can propose updated criteria here ... and follow-up on them post meeting 18:19:31 jlaska: i put out a call on the bug 18:19:37 if keyboard works on US layout it's not an alpha 18:19:41 adamw: thanks i justa saw it 18:19:47 wow, I can't type. 18:20:05 Viking-Ice: this will be a longer discussion probably ... no objections, but think we should raise this on test@ or next QA meeting 18:20:17 agreed 18:20:24 #agreed 676827 - Unable to accept or reject. Need more information from reporter (fresh install or upgrade) 18:20:42 rbergeron: i thought you'd just joined the cartoon mafia. 18:20:50 #info possible improvements to the alpha criteria to clarify US vs non-US keymaps 18:20:58 okay,that's it for Proposed Blockers 18:21:01 now to Accepted blockers 18:21:13 #topic https://bugzilla.redhat.com/show_bug.cgi?id=672265 18:21:14 Bug 672265: high, medium, ---, bruno, CLOSED RAWHIDE, "Install to harddrive" fails with "not a live image" 18:21:22 awesome, nothing else to discuss here :) 18:21:23 it's closed. yay. move on. 18:21:44 woot 18:21:44 #agreed fixed and included in F15Alpha 18:21:51 #topic https://bugzilla.redhat.com/show_bug.cgi?id=676032 18:21:53 Bug 676032: unspecified, unspecified, ---, anaconda-maint-list, VERIFIED, Import Error: No module named iutil 18:22:04 Note ... this, and the next 3 are in VERIFIED 18:22:04 same 18:22:15 we can skip, or confirm the bodhi update has what it needs? 18:22:42 jlaska: i think we just need to close bugs 18:22:43 #agreed confirmed fixed and included in anaconda-15.20-1 (TC2) 18:22:52 yeah, if they aren't linked to bodhi 18:22:57 jlaska: the fix was in pre branch/bodhi 18:23:13 perfect 18:23:18 we can close this sucker 18:23:19 #topic https://bugzilla.redhat.com/show_bug.cgi?id=673907 18:23:20 Bug 673907: unspecified, unspecified, ---, mgracik, VERIFIED, Install from DVD fails: /var/run/dbus/system_bus_socket is missing 18:23:25 same deal here ... 18:23:40 #agreed confirmed fixed in lorax-0.4-1, used to build TC2 18:23:52 jlaska: wrong version of lorax 18:24:07 I don't see a lorax bodhi update ... does that matter? 18:24:23 lorax-0.3-3.fc15 18:24:25 dgilmore: sorry, I was going by the fixed_in bz field 18:24:27 it was pre branch 18:24:40 * jlaska updates fixed_in and closes 18:24:41 jlaska: np. 0.4 is not built 18:25:04 #topic https://bugzilla.redhat.com/show_bug.cgi?id=674978 18:25:05 Bug 674978: high, unspecified, ---, rstrode, VERIFIED, GDM in autologon error loop 18:25:11 same deal ... confirmed fix by many testers 18:25:17 thanks halfline for the fix :) 18:25:30 "The package needs to get pushed to stable before we can close it." -adamw 18:25:51 #link https://admin.fedoraproject.org/updates/gdm-2.91.6-9.fc15 18:26:13 jlaska: gdm-2.91.6-8.fc15 is queued up to go to stable 18:26:13 the previous release had enough karma - https://admin.fedoraproject.org/updates/gdm-2.91.6-8.fc15 18:26:17 okay 18:26:17 dgilmore: thx 18:26:26 bodhi should then close it out for us 18:26:34 jlaska: -9 is not yet pushed to testing 18:26:38 #topic https://bugzilla.redhat.com/show_bug.cgi?id=670379 18:26:39 Bug 670379: medium, low, ---, mgracik, VERIFIED, Unknown lvalue 'ValidNoProcess' in section 'Service'. 18:26:59 this is fixed 18:26:59 confirmed, but awaiting additional karma 18:27:01 #link https://admin.fedoraproject.org/updates/firstboot-1.116-1.fc15 18:27:10 wait ... nevermind 18:27:15 dgilmore took care of this -- This update has been submitted for stable by ausil. 18:27:23 so looks to be in good shape 18:27:34 jlaska: just need to push it 18:27:38 #agreed fix confirmed and bodhi update submitted for stable 18:27:47 alright ... now on to proposed NTH bugs 18:28:11 #topic Proposed Blocker review 18:28:15 #link http://bit.ly/f15-alpha-nth-proposed 18:28:20 #link http://bit.ly/f15-alpha-nth-accepted 18:28:37 sorting by component name, we start off with anaconda ... 18:28:42 #topic https://bugzilla.redhat.com/show_bug.cgi?id=678095 18:28:44 Bug 678095: unspecified, unspecified, ---, clumens, MODIFIED, If no display mode is provided, kickstart installs default to cmdline 18:29:13 the subject pretty much captures it 18:29:28 an annoyance for kickstart, I proposed that we pull this in if we respin anaconda for any reason 18:29:33 sure. 18:29:37 * rbergeron nods 18:29:44 fine by me 18:30:08 which I guess would make this NTH 18:30:38 proposed #agreed 678095 - AcceptedNTH, will take tested fix if anaconda respins for the Alpha 18:30:41 ack/nack/patch 18:30:58 ack 18:31:03 ack 18:31:09 * dgilmore thinks the behaviour is correct 18:31:12 nak 18:31:28 dgilmore: it's a change in behavior from all previous releases 18:31:30 the default display mode is graphical. 18:31:35 according to the kickstart docs 18:31:59 jlaska: hrrm its the behaviour i personally want 18:32:09 heh 18:32:22 lol 18:32:27 I think we'd need another bug to change the default behavior then :) 18:32:55 i vaguely remeber discussion about it 18:33:07 clumens: how do you want to handle this? 18:33:40 jlaska: well, i didn't mean to change the behavior, so it seems reasonable to patch it back to how it used to be 18:33:53 we encourage graphical mode anyway 18:34:05 is this the correct behavior or the wrong one 18:34:13 but used by many 18:35:11 the behavior as documented and as existed in many previous releases was that no mode in kickstart meant you got graphical 18:35:31 that does it for me 18:35:50 any talks on changing that expectation can happen outside this meeting 18:35:54 dgilmore: you okay with that? 18:36:27 clumens: are you +0 on these? 18:36:53 jlaska: i'm fine with holding off until beta 18:37:09 id prefer us to wait till beta 18:37:17 i guess im outvoted here 18:37:19 so -1 NTH from you guys? 18:37:48 I dont see why this cant wait for beta 18:37:53 it can 18:38:02 I'm proposing it for NTH 18:38:09 so if we respin anaconda, we'd take this tested fix too 18:38:12 but we can vote against that 18:38:21 current proposed #agreed 678095 - AcceptedNTH, will take tested fix if anaconda respins for the Alpha 18:38:22 at this point id rather not change anaconda unless ist a omg noone can install bug 18:38:24 ack/nak/patch 18:38:35 sure 18:38:45 proposed #agreed 678095 - RejectedBlocker RejectedNTH, post-alpha 18:38:47 +/-0, i don't really care :P 18:38:55 too much time on this :) 18:38:55 if the fix is there I say ack 18:39:09 it doesn't really seem important enough to risk a change, so i guess small +1 to second proposal 18:39:43 #agreed 678095 - RejectedBlocker, RejectedNTH, document Common_F15_Bugs and accept post-Alpha 18:39:58 #topic https://bugzilla.redhat.com/show_bug.cgi?id=677773 18:39:59 Bug 677773: unspecified, unspecified, ---, clumens, MODIFIED, F-15-Alpha DVD install (TC2) prompts for networking at repo step 18:40:02 dgilmore: yeah, me neither. let's get the alpha out and then when we have a little breathing room, i'll rebuild for all these minor annoyances. 18:40:31 similar type of issue 18:40:38 non-blocking annoyance bug 18:40:46 you are prompted to enable networking on DVD installs 18:40:51 you can click cancel (twice) and proceed with the install 18:40:58 I think that's an acceptable workaround 18:41:03 but was hoping we could include if we rebuilt 18:41:08 but in light of previous bug ... 18:41:14 proposed #agreed 678095 - RejectedBlocker, RejectedNTH, document Common_F15_Bugs and accept post-Alpha 18:41:21 ack 18:41:25 ack 18:41:35 ack 18:41:39 ack 18:41:46 perfect, thanks all 18:41:49 #agreed 678095 - RejectedBlocker, RejectedNTH, document Common_F15_Bugs and accept post-Alpha 18:41:59 #topic https://bugzilla.redhat.com/show_bug.cgi?id=677648 18:42:00 Bug 677648: unspecified, unspecified, ---, clumens, MODIFIED, Cannot get translation file 18:42:26 From the reporter ... 18:42:26 "After selecting e.g. German as installation language I get an 'Cannot get 18:42:30 translation file de.tr' on Alpha TC2." 18:42:56 Fix is available in the next anaconda 18:43:02 * jsmith-busy thinks that this is pretty important to have 18:43:13 I added this for review because I wasn't sure about impact ... all langs, just German 18:43:17 all langs 18:43:45 will this fix be in alpha? 18:43:47 * rbergeron agrees with jsmith-busy, particularly if it's all langs 18:43:56 Viking-Ice: that's what we need to decide 18:44:09 +1 for blocker 18:44:14 -1 for blocker 18:44:20 (as there's no workaround) 18:44:21 we got a working language english 18:44:25 incidentally, this will only affect the loader. 18:44:35 aah, not graphical installer? 18:44:57 it sure shouldn't. and if it does, people can pass "nopass" as a cmdline argument and get a graphical language prompt. 18:44:58 Viking-Ice: we have to define that new criteria before we can start using it 18:45:13 Hmmmmn... 18:45:18 I'm less concerned about it now 18:45:22 * jsmith-busy is too 18:45:33 (Might be useful to put those details in the bug report) 18:45:39 wait, what does this affect exactly? 18:45:50 adamw: the early text portion of the installer (aka loader) 18:46:29 * dgilmore thinks this is a document Common_F15_Bugs issue 18:46:40 and what happens? 18:46:50 clumens: when the failure happens, is it recoverable? 18:46:52 you get the early part in english and rest in your language? 18:46:53 no matter what language you pick, you get english 18:47:13 i've not tested that the graphical part is translated properly, but it'd be easy to check. 18:47:14 hence my -1 18:47:21 but if you pass in "nopass" you can select a language other than english? 18:47:27 at graphical time, yes 18:47:40 assuming it's not being translated right. someone should check that out real quick. 18:48:02 checking ... 18:49:43 confirmed 18:49:46 graphical portion is in german 18:49:57 a+ 18:50:07 * Viking-Ice never gets an a+ 18:50:16 proposed #agreed 677648 - RejectedBlocker, RejectedNTH, Common_F15_Bugs - issue only affects translation of loader (not graphical installer) 18:50:38 confirmation - http://jlaska.fedorapeople.org/screenshots/Screenshot.png 18:50:56 +1 18:51:02 ack 18:51:05 ack 18:51:22 #agreed 677648 - RejectedBlocker, RejectedNTH, Common_F15_Bugs - issue only affects translation of loader (not graphical installer) 18:51:34 #topic https://bugzilla.redhat.com/show_bug.cgi?id=676942 18:51:35 Bug 676942: unspecified, unspecified, ---, msivak, MODIFIED, Text mode missing dependencies dialog out uses single column 18:51:50 if the text-mode installer hits any package dependencies, the error dialog looks wonky 18:52:17 the original problem the user encountered has been fixed (package dependency) 18:52:21 so I don't know why I raised this 18:52:29 I'm -1 to Blocker and NTH 18:52:32 ack 18:52:54 I'm with jlaska on this 18:53:19 jlaska: :) im with you on it 18:53:20 proposed #agreed 676942 - RejectedBlocker, RejectedNTH - a mal-formed package dep dialog in text-mode doesn't meet criteria 18:53:24 #agreed 676942 - RejectedBlocker, RejectedNTH - a mal-formed package dep dialog in text-mode doesn't meet criteria 18:53:33 ack 18:53:35 well, I should have waited, but I assumed you all agreed 18:53:43 * jlaska waits a bit in case -1 18:53:56 (DON'T) SHIP IT. 18:54:22 :) 18:54:31 #topic https://bugzilla.redhat.com/show_bug.cgi?id=676940 18:54:32 Bug 676940: unspecified, unspecified, ---, clumens, MODIFIED, %include checked before %pre runs, or not found 18:54:57 folks: I have to jet for another meeting. The last bug i update will be 676942, but will check logs if nobody else can update bz's from hereforth. 18:55:14 clumens: is this already fixed in 15.20-1? 18:55:23 rbergeron: thank you for handling the bz updates!! 18:55:35 jlaska: no, 15.21 18:55:41 okay 18:55:50 I don't remember why I proposed this for NTH now 18:55:52 yay fail! 18:55:56 it would be fixed in 15.20.1 if we decided to rebuild for alpha 18:56:02 k 18:56:24 jlaska: cobbler install properly why you suggested it 18:56:25 oh, cause it was going to impact cobbler installs 18:56:29 Viking-Ice: yeah 18:56:30 which it looks like we are desperately trying to avoid 18:56:35 haha 18:56:45 I don't feel strong about this one anymore 18:56:52 so unless anyone else does ... 18:57:17 clumens: is this updates.img-able? 18:57:20 no 18:58:19 * adamw doesn't feel strongl 18:58:20 y 18:58:27 probably worth commonbugsing if we don't fix it, though 18:58:27 proposed #agreed 676940 - RejectedNTH, will resolve post-alpha 18:58:39 proposed #agreed 676940 - RejectedNTH, CommonBugs, will resolve post-alpha 18:58:41 if you propose for commonbugs please add a note explaining the impact is cobbler 18:58:57 so i know what to write for commonbugs if i wind up doing it =) 18:59:13 and we know you will if I'm lazy! 18:59:27 I'm accepting this and moving on 18:59:29 I ack on rj-cbs 18:59:41 #agreed 676940 - RejectedNTH, CommonBugs, will resolve post-alpha 18:59:44 Viking-Ice: nice :) 19:00:05 probably should be rj-nth-cbs 19:00:08 hehe 19:01:02 #topic https://bugzilla.redhat.com/show_bug.cgi?id=678510 19:01:03 Bug 678510: urgent, unspecified, ---, matt_domsch, NEW, biosdevname segfaults when BIOS advertizes a zero sized PIRQ routing table 19:01:09 3 biosdevname bugs 19:01:40 "Proposing F15 Alpha blocker." -Narendra 19:01:51 no idea on this one 19:01:56 what's the impact of biosdevname segfaulting? 19:01:58 would it kill install? 19:02:02 looks like it was fixed, no? 19:02:07 the pkg is in updates-testing 19:02:15 http://koji.fedoraproject.org/koji/buildinfo?buildID=229252 19:02:19 appaers to be fixed in 0.3.7 which is already built 19:02:20 tflink: yeah, the q is whether we take it 19:02:37 mdomsch isn't around 19:02:41 if it helps short-cut discussion, i think 675048 should definitely be at least nth and make us take the fix 19:02:50 I'm on taking the fix 19:02:52 i'm fine with taking the other two fixes at the same time 19:03:00 yup me too 19:03:01 want to jump forward to that issue? 19:03:13 well ... 19:03:17 semantics 19:03:23 im ok with it being nth 19:03:24 let's accept this as NTH 19:03:28 ill likely pull it in 19:03:39 since there was behaviour change in the latest version 19:04:02 let's just +1 all three and take the fix 19:04:18 ack 19:04:35 ack 19:04:36 ack 19:04:53 yay 19:05:12 proposed #agreed 678510 - AcceptedNTH, updated koji build available 19:05:14 #agreed 678510 - AcceptedNTH, updated koji build available 19:05:20 #agreed 675048 - AcceptedNTH, updated koji build available 19:05:26 #agreed 662905 - AcceptedNTH, updated koji build available 19:05:39 that felt good 19:05:40 #topic https://bugzilla.redhat.com/show_bug.cgi?id=656737 19:05:42 Bug 656737: medium, high, ---, nb, MODIFIED, Link to Release Notes using legacy URL 19:05:55 * dgilmore says push to beta 19:05:59 "This is nice-to-have for f15-alpha per docs team." 19:06:05 it still works as is 19:06:13 it seems to redirect fine 19:06:15 yeah 19:06:19 beta it is? 19:06:34 do we know when the symlink is going away? 19:06:40 * nb wonders why not take it? 19:06:46 it's pretty safe to take it since it's just a bookmarks package, but i don't really care either way. 19:06:48 all the diff is is just 1 line in the default bookmarks 19:06:50 let's toss a coin and move on! 19:07:10 tflink, as soon as we can, the change only affects new installations 19:07:15 proposed #agreed 656737 - AcceptedNTH, will take fix into Alpha 19:07:16 so it'll be a while 19:07:23 * jlaska tries to shake it up 19:07:28 jlaska: is there anything else with my name on it? 19:07:29 ack/nak/patch? 19:08:04 clumens: just a check-in to make sure the acceptedNTH bug 671922 is good to go. But since we aren't respinning, I think we're good 19:08:05 Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=671922 unspecified, unspecified, ---, dlehman, VERIFIED, ValueError: isys.umount() can only umount by mount point 19:08:11 ack 19:08:20 ack 19:08:39 jlaska: they're moving cubes around here and it's a hell of a lot of racket, so i'm going to relocate to somewhere quieter as soon as my part is done here. 19:08:41 neutral 19:09:08 dgilmore: ? 19:09:21 jlaska: im pretty neutral 19:09:30 im leaning towards least change right now 19:09:37 #agreed 656737 - AcceptedNTH, will take fix into Alpha 19:09:43 but i really really want alpha out on time 19:09:57 +1 to that :) 19:10:07 dont we always ;) 19:10:25 #topic https://bugzilla.redhat.com/show_bug.cgi?id=675652 19:10:26 Bug 675652: high, unspecified, ---, rstrode, MODIFIED, Can't start alternative (non-default) DE with gdm-2.91.6-2.fc15 19:10:31 this ought to be a good one 19:10:59 jlaska: i maintain my stance on not rebuilding. 19:11:10 clumens: roger 19:11:30 halfline: adamw: what's the scoop here? 19:11:38 er, unclear =) 19:11:44 well, halfline knows better than me 19:11:49 There are two things breaking xfce: 19:11:49 1) They were shipping an old version of gdm that was getting hit by bug 674978 19:11:51 Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=674978 high, unspecified, ---, rstrode, VERIFIED, GDM in autologon error loop 19:11:52 2) after upgrading gdm to a recent version, they started hitting a crasher in 19:11:55 gnome-session 19:11:57 so #1 is solved 19:12:02 gdm-2.91.6-8.fc15 was not working for me 19:12:06 yes. but that's not what I hit in testing, as i used a newer gdm. 19:12:13 ill gladly take gdm-2.91.6-9.fc15 if things are working 19:12:24 i still don't understand why launching xfce should involve gnome-session at all. 19:12:32 so how does a crasher in gnome-session break xfce? 19:12:41 gdm uses gnome-session 19:13:10 okay. so, is that actually fixed? 19:13:14 afaict -9 adds back the session chooser 19:13:19 but that doesn't fix this bug, right? 19:13:23 cwickert had to go 19:13:26 er 19:13:30 it fixes the #topic bug 19:13:34 so we weren't able to debug the gnome-session crash he was hitting 19:13:37 but not the actual important bug, which is xfce session not running 19:13:40 okay 19:13:49 i think i hit the same issue in my test 19:13:53 so i'll have a look at it later 19:13:56 cool 19:14:04 practically speaking, we switched spin-kickstarts to use lxdm last night 19:14:05 im going to install gdm-2.91.6-9.fc15 now and see what i get 19:14:07 (for xfce spin) 19:14:19 dgilmore: really you need to build an xfce live spin with that gdm and see what happens 19:14:21 but i expect it would fail 19:14:27 adamw: so that works around this bug? 19:14:30 it should, yes 19:14:39 it's not been very heavily tested, but it worked for me, mostly 19:14:59 once the nightly is built we can confirm the switch worked and it boots 19:15:00 * jlaska rereads the NTH criteria .. 19:15:00 bugs which constitute infringements of the desktop-related Fedora_Release_Criteria as applied to non-default desktops 19:15:03 bugs which result in a system being unable to reach a graphical environment 19:15:07 significant installer bugs which do not meet the criteria to be blocker bugs 19:15:09 but this specific bug should probably be marked not nth anyway 19:15:15 as it's not really the problem 19:15:34 * Viking-Ice takes on this whole issue is that each *DE should have it's own corresponding *DM not crossbreed with other *DM 19:15:38 so, proposal - this is -1 nth, i and cwickert and halfline will look into the gnome-session crash and review teh situation 19:15:49 +1 19:15:52 +1 19:15:59 +1 19:16:37 There was a patch posted to us a different dm for lxde or xfce because of this. I don't know that it is being used for the spin yet. 19:16:42 +1 19:16:46 #agreed 675652 - RejectedNTH. adamw, cwickert and halfline will investigate gnome-session crash for futher action 19:16:56 brunowolff: will be in todays nightlys 19:17:03 #topic https://bugzilla.redhat.com/show_bug.cgi?id=654762 19:17:04 Bug 654762: medium, low, ---, rstrode, NEW, plymouth does not start due to "Address already in use" error 19:17:07 brunowolff: we kicked it through last night. lxde already uses lxdm. 19:17:15 adamw: is the live spin with lxdm available somewhere? 19:17:24 I added this since I *think* this means we get an empty /var/log/boot.log 19:17:30 tflink: today's nightly will do. my personal test spin, no, my upload sucks. 19:17:44 but this should be easily fixed post-Alpha 19:17:45 adamw: OK, thanks 19:18:03 i guess i'd be +1 nth on this if we have a reasonably safe fix as it does make debugging a pain 19:18:20 right, but I'm not keen on fiddling with plymouth now 19:18:28 like you said, 'safe fix' 19:18:44 halfline: have a take on this? 19:18:51 fun ... I have /var/log/boot.log now 19:18:55 so something may have already fixed it 19:19:02 (could be in updates-testing ... not sure) 19:19:20 someone claimed that it was fixed in plymouth-0.8.4-0.20101119.2.fc15.x86_64 19:19:38 wait, I'm reading too quickly 19:19:44 hang on 19:19:53 the Address already in use error is completely benign 19:20:01 it's already in use because it was already running 19:20:09 halfline: aah okay ... so I was wrong 19:20:19 it won't prevent /var/log/boot.log from filling up 19:20:25 also mezcalero fixed systemd to not try to start it a second time 19:20:41 and i changed plymouth not to print the message anymore 19:20:58 jlaska: i think the boot.log thing was fixed in systemd 19:21:12 proposed #agreed 654762 - RejectedNTH. Error message is benign. Updated systemd resolves issue 19:21:15 so the key fix is systemd 19:21:20 is that one in dist-f15? 19:21:34 https://admin.fedoraproject.org/updates/systemd-18-1.fc15 19:22:07 * jlaska adding my karma feedback 19:22:43 ack/nak/patch? 19:23:00 ack 19:23:08 #info Updated systemd available - karma needed 19:23:15 ack 19:23:19 ack 19:23:36 note ... that systemd resolves a previously accepted blocker 676815 19:23:41 scratch that 19:23:46 accepted *Beta* blocker 19:23:53 so that's not required for Alpha 19:24:28 #agreed 654762 - RejectedNTH. Error message is benign. Updated post-alpha systemd resolves issue 19:24:52 Okay gang ... that's it for Blockers (proposed + accepted) and NTH (proposed) 19:24:59 I'm out of gas for going through the accepted NTH issues 19:25:06 we don't really need to 19:25:10 jlaska: cheers 19:25:11 agreed 19:25:12 just look at the list and test any available updates 19:25:12 good! 19:25:22 we can track nth fixes that should be pulled in the trac ticket 19:25:25 * dgilmore has lots of work to do to get a alpha rc composed 19:25:25 #topic Accepted NTH bugs 19:25:26 #link http://bit.ly/f15-alpha-nth-accepted 19:25:36 #topic Open Discussion 19:25:46 anything else we need to touch on? 19:25:51 not from me 19:25:52 otherwise, let's get down to bidness 19:26:06 are we spinning an rc> 19:26:07 ? 19:26:11 we still have unresolved open blockers, don't we? 19:26:30 yes 19:26:38 adamw: yes 19:26:39 so, we need to work on those 19:26:41 k 19:26:48 sorry for disturbing... any ideas about bug 678431 ? is this a blocker or nth? 19:26:50 Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=678431 unspecified, unspecified, ---, dennis, MODIFIED, Incorrect key for fedora-debug repo 19:27:04 debug repos aren't used in an update by default, so no 19:27:17 you only hit it if you have the yum plugin that auto-updates debuginfo packages installed i think 19:27:28 abrt? 19:27:30 well, all bug reporting tools are useless 19:27:40 both abrt and drkonqi 19:27:41 adamw: that effects all repos, the fix is built for it 19:27:43 ah 19:28:00 i put the gpg keys in the wrong order 19:28:09 let's take it then 19:28:11 +1 blocker 19:28:17 cool, thanks 19:28:20 ill +1 it 19:28:31 I'll ack to it 19:28:34 #agreed 678431 - accepted as a blocker bug 19:28:35 the change is one line on the config file 19:29:20 We have an update bug#676827 19:29:21 Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=676827 high, unspecified, ---, xgl-maint, NEW, keyboard with german layout doesn't work in gdm 19:29:28 it was *not* from a fresh install, but from a yum update 19:29:38 what does that change? 19:29:41 nothing much 19:29:48 we need to know _if it happens_ from a fresh install 19:29:54 jlaska: it's not an alpha blocker :) 19:29:55 if the reporter hasn't tested, we still don't know that 19:30:07 we just know that we don't know...it's a known unknown ;) 19:30:30 clocks ticking 19:30:35 not a blocker 19:30:54 Viking-Ice: do you have an opinion on this one? :D 19:31:40 * dgilmore thinks we cant know 19:31:47 but that it likely a upgrade only issue 19:31:56 and there is workarunds 19:32:02 and i cant type 19:32:04 I'm going to join Viking-Ice on this one 19:32:20 just so we aren't playing games with "if this is the last bug on the list" 19:32:35 I'm still saying that this is not an blocker since he has a working "keyboard" 19:32:42 imo ... if this was the last on the list, I'd be in favor of documenting the workaround 19:32:48 fine 19:32:51 i'll go with the majority 19:32:54 which means ... why shoudl we keep it on still? 19:33:04 well, votes ... I think it was just Viking-Ice & I 19:33:25 * dgilmore is with Viking-Ice 19:33:49 proposed #agreed 663294 remove from F15Alpha, acceptable workaround exists, *suspect* this only affects updates (not install) 19:34:25 * dgilmore needs to run to physio in a couple of minutes 19:34:37 s/updates/upgrades right ? 19:34:38 new proposed blocker showed up 19:34:40 ack 19:34:55 #agreed 663294 remove from F15Alpha, acceptable workaround exists, *suspect* this only affects upgrades (not install) 19:35:01 #topic https://bugzilla.redhat.com/show_bug.cgi?id=678644 19:35:03 Bug 678644: unspecified, unspecified, ---, pmatilai, NEW, yum reinstall hangs 19:35:08 adamw: that one? 19:35:11 -1 blocker, yum reinstall is not part of alpha criteria 19:35:12 yes 19:35:19 not a blocker 19:35:24 I didn't even know about that feature 19:35:29 -1 blocker 19:36:00 -1 blocker 19:36:05 -1 19:36:08 #agreed 678644 - RejectedBlocker, yum reinstall not part of Alpha release criteria 19:36:25 okay ... #endmeeting fuse set to 2 mins 19:36:31 #topic Open Discussion 19:36:52 so action is with anaconda team to clear up live install blockers and desktop team to look at clutter bugs? 19:36:58 from what i can see 19:37:00 adamw: thanks for clarifying 19:37:14 jlaska: er, that live install issue we hit isn't set as blocker 19:37:17 jlaska: should we? 19:37:31 it may be after rbergeron had to leave, so we have some bz updates to make 19:37:45 i'm not sure we reviewed it as a blocker at all 19:37:51 we just mentioned it in side discussion 19:37:58 #chair adamw 19:37:58 Current chairs: adamw jlaska 19:38:04 what was the url again? i lost it 19:38:16 https://fedoraproject.org/wiki/User:Jlaska#Links 19:38:46 no, to the bug 19:39:16 oh 678607 19:39:29 #topic https://bugzilla.redhat.com/show_bug.cgi?id=678607 19:39:31 Bug 678607: medium, medium, ---, anaconda-maint-list, NEW, AttributeError: 'NoneType' object has no attribute 'format' 19:39:42 propose this as blocker - jlaska and i both hit it testing live install with the fixed kernel 19:39:45 ah, thank you 19:39:50 +1 19:39:55 +1 19:40:02 jlaska: what was your hardware for the test btw? vm? bare metal? what disk? 19:40:05 one more thing lurking behind the untested doors of live install 19:40:07 vm 19:40:12 disk size? 19:40:22 15G, 10G, 2G, 2G 19:40:36 k 19:40:42 mine's just a 10G i told it to wipe 19:40:46 so it hits different disk configs 19:41:46 #agreed 678607 - AcceptedBlocker, appears to block the limited Live testing performed so far 19:42:09 anything else? 19:42:11 Well I got to run 19:42:20 Viking-Ice: thanks for helping us through ... cya 19:42:26 like wize 19:42:30 calling #endmeeting in 1 min 19:42:33 happy brewing this weekend 19:42:40 :D 19:43:27 endmeeting in 10 seconds 19:43:31 that's a dupe of 672030 I think 19:44:20 looks like it 19:44:21 looks like 19:44:26 okay ... let's close this puppy out 19:44:31 we can do the DUP game out of meeting 19:44:32 thanks all! 19:44:41 I'll send minutes to the list(s) 19:45:11 #endmeeting