14:00:21 <licquia> #startmeeting LSB Bug Triage 2014 May 2 14:00:21 <lsbbot> Meeting started Fri May 2 14:00:21 2014 UTC. The chair is licquia. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:21 <lsbbot> Useful Commands: #action #agreed #help #info #idea #link #topic. 14:00:29 <licquia> good morning 14:00:48 <mwichmann> morning 14:01:29 * mwichmann offers warning: he is in foul mood today, pardon snippy responses if any creep in 14:01:58 <licquia> #info lsb: 1 reopened, 4 new 14:02:12 <licquia> #info fhs: 26 new 14:02:16 <licquia> ouch 14:04:25 * licquia figures it probably isn't the overcast and rainy weather in new mexico :-) 14:04:41 * licquia gets started 14:04:51 <licquia> !lsbbug 3195 14:04:54 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3195 normal, P2, ---, mats, REOPENED , LSB packages with unowned directories uninstallable on some rpm implementations. 14:05:09 <mwichmann> not even a sinkhole to cheer me up 14:05:32 <mwichmann> (those are useful for shoving politicians into) 14:05:46 <licquia> heh 14:05:59 <mwichmann> election season, but that's not the source of any of my irritation 14:06:00 <licquia> no news on this bug; we know what to do, but it isn't done yet 14:06:22 * licquia observes that his town is becoming a city, with first city elections happening very soon 14:07:09 <licquia> main deal with this bug: get some of the vip packages released already 14:07:19 <mwichmann> we have a village -> town -> city progression, don't think too much changes though 14:07:42 <mwichmann> I don't know if any missed the change except the one in question 14:07:52 <mwichmann> might have 14:08:02 <licquia> a lot changes here; we will have a mayor, council structure changes, etc. 14:08:15 <mwichmann> ah 14:08:48 <mwichmann> can we just push out lsb-qm without doing a bunch of others? (wish I knew if others were /neeeded/) 14:09:05 <licquia> i was thinking tet and qm, but yeah 14:09:31 <mwichmann> is there something we can check? 14:09:37 <mwichmann> tet is pretty simple as well 14:10:00 <licquia> i think we've pretty much checked it in snapshots 14:10:13 * licquia checks what the current status in staging is 14:11:14 <mwichmann> right, I mean is there a test like "compare rpm -qlp released.rpm with rpm -qlp staging.rpm" type of test that could highlight where we've failed to get out a pkg with the ownership question sorted? 14:12:01 <mwichmann> without spending a lot of time that is 14:12:03 <licquia> didn't we ferret that out w.r.t. qm? there's a comment on the bug which shows that the pkg release # changed at the commit's fix, and we have the old one 14:12:48 <mwichmann> it's clear with qm yes 14:12:48 <licquia> i think the only mystery is how we missed this fix in the last vip update, which happened after this fix 14:12:59 <mwichmann> qm probably wasn't part of that 14:13:22 <mwichmann> sometimes I think the test driver pkgs are in a strange state 14:13:38 <mwichmann> there's tet, t2c and qm 14:13:53 <mwichmann> dependencies of others, don't change often, etc. 14:14:49 <licquia> also, all forked from upstream by this point, sometimes by necessity, sometimes not 14:15:14 <mwichmann> yes 14:15:33 <licquia> we may want to look at that at some point 14:15:34 <mwichmann> we're almost upstream for t2c... I don't know if ispras continued to develop that elsewhere 14:15:44 <mwichmann> I did want to uplift tet to 3.8 14:16:01 <mwichmann> but cooler heads prevailed, hard to make that a priority 14:16:02 <licquia> there's a new major version, as i recall 14:16:27 * licquia triggers builds of tet and qm for potential release 14:18:06 <licquia> we'll see what happens with that 14:19:06 <licquia> #info no progress on 3195; started release build for tet-harness and qmtest-harness 14:19:17 <mwichmann> you have s390 working now? 14:19:23 <licquia> !lsbbug 3970 14:19:25 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3970 normal, P2, ---, licquia, NEW , Designer Tests give Segmentation fault 14:19:38 <licquia> no; since this is a production build, it's using the production sdk 14:19:57 <licquia> i did fix one problem with s390 which held things up before 14:20:23 <licquia> likely those builds would have failed w/o that fix 14:21:30 <mwichmann> I know it's off topic, but what is wrong with it that doesn't affect s390x? 14:22:05 <licquia> at one point, there was the quirk that the deps on g++ were borked 14:22:09 <licquia> in fedora 20 14:22:42 <licquia> i.e. you could in theory install 64-bit g++ with 32-bit libstdc++-devel only, but the combination wouldn't actually work 14:23:02 <licquia> installing 64-bit libstdc++-devel was the fix 14:23:29 <licquia> now there's an issue with lsbdev-qt i'm still trying to figure out 14:23:35 <licquia> MAXNAMLEN 14:23:36 <mwichmann> s390 seemed sicker than that, couldn't install base LSB pkgs 14:23:53 <mwichmann> well, sdk 14:24:11 <licquia> right; that's because the sdk isn't completing its build 14:24:33 <mwichmann> but that's not affecting s390x? 14:24:59 <licquia> apparently not 14:25:26 <mwichmann> we can move on to the 4 opens, but not sure I have anything to contribute, no time to look at this stuff 14:25:28 <licquia> oops; spoke too soon 14:25:56 <licquia> sdk failed to build on s390x too 14:26:08 <licquia> (but it did build once, so not sure what's going on there) 14:26:46 <licquia> on this bug and the others, orc_fedo has been working with him 14:27:10 <licquia> apparently, there are issues with that person just copying files around in an attempt to try and get things working 14:27:25 <licquia> the thought is that they will clear up with a clean test run 14:28:20 <mwichmann> seem to be pkg dep issues from the output thaat got tagged to one of the bugs 14:29:28 <licquia> i recall you were able to run one of these just fine on f20 14:30:18 <mwichmann> I randd something 14:30:25 <mwichmann> grr 14:30:45 <mwichmann> having problems, two wireless keyboards too close to each other, interfering even though they're not supposed to 14:30:57 <licquia> heh, fun 14:31:14 * licquia still sticks to wired stuff whenever possible 14:31:25 <mwichmann> I'm going to file a lawsuit against logitech 14:31:36 <mwichmann> isn't that the way these things get resolved? 14:31:56 <licquia> too often 14:31:58 <mwichmann> ... if you saw my desk, you'd understand why the wireless bit is important to me... mad tangle otherwise 14:32:26 <mwichmann> I brought up scribus, that's not one of the four "new" bugs 14:32:31 * licquia has done a pretty good job keeping the wires under control 14:33:08 <mwichmann> I'd be fine if there was space, but there isn't; have to reach for this keyboard, then that keyboard 14:33:11 <licquia> grabbing designer 14:35:02 <mwichmann> observation on the "final" packaging version thingy: 14:35:30 <mwichmann> "new day" overrides "new buildno" 14:35:51 <licquia> well, that's that; confirmed on f20 14:35:57 <mwichmann> fails? 14:36:02 <licquia> segfault 14:36:17 <licquia> - error='Cannot load library /usr/lib64/qt4/plugins/designer/libqt3supportwidgets.so: (/usr/lib64/qt4/plugins/designer/libqt3supportwidgets.so: undefined symbol: _ZNK22QDesignerPropertySheet24dynamicPropertiesAllowedEv)' 14:36:26 <mwichmann> sounds like need to mark 3970 to investiage then 14:36:37 <mwichmann> or some word like that 14:36:55 <mwichmann> can't blame that one on wireless, just my typing 14:37:26 * mwichmann considers getting a special bug ready to be 4000 14:38:07 <mwichmann> qt3support? 14:38:11 <mwichmann> why? 14:38:14 <licquia> and the missing symbol... QDesignerPropertySheet::dynamicPropertiesAllowed() const 14:38:36 * licquia wonders if that's a callback 14:39:37 <mwichmann> no match on QDesignerPropertySheet in our specdb 14:39:48 <licquia> yup 14:40:18 <mwichmann> so not /directly/ our fault it looks like... 14:40:23 <licquia> i'm actually wondering if we're pulling in something that's outside in our build, and it's not interacting well with newer versions 14:40:47 <licquia> so, target stable update? 14:41:06 <mwichmann> until we decide it's an impl/distro bug, yes 14:43:19 <mwichmann> back in <3mins 14:43:28 <licquia> #agreed confirmed 3970, target next stable update 14:44:43 <licquia> !lsbbug 3972 14:44:45 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3972 normal, P2, ---, licquia, NEW , Apache Tests(2.2.14-8) Fail on x86_64: Failed to prepare apache 14:45:28 <mwichmann> trying to sort... pkg signing causing them trouble? 14:45:57 <mwichmann> no, "scriptlet failed", but not much hint why 14:46:06 <licquia> at one point, they didn't have /lib/lsb/init-functions, so i'm thinking simple fails 14:46:11 <licquia> i.e. distro problems 14:46:24 <mwichmann> I see now, next comment 14:46:32 <licquia> i vote add a comment for a question, leave as new 14:46:39 <mwichmann> yes 14:46:41 <licquia> until we're sure this is a problem 14:47:16 <mwichmann> this is one of our efforts to make sure the initscript support is working - wanted "real-life example" 14:47:59 <licquia> #info asked questions for 3972, leaving as new for now 14:48:14 <licquia> !lsbbug 3973 14:48:16 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3973 normal, P2, ---, licquia, NEW , LSB4.1:xpdf on test1.pdf does not show anything 14:48:53 <licquia> orc_fedo is working with this one 14:49:05 * licquia votes we leave this one as new as well 14:49:15 <mwichmann> +1 14:49:44 <licquia> #info still working with bug filer on 3973; leaving new for now 14:50:09 <licquia> !lsbbug 3974 14:50:12 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3974 normal, P2, ---, licquia, NEW , LSB 4.1: Celestia Test : OpenGL GLX extension not supported by display ':0' 14:51:00 * licquia tries this on f20 14:51:34 <mwichmann> I think that's a distro thing: either the xserver is configured with such an extension, or it isn't 14:51:57 <mwichmann> I'm thinking we require the support 14:52:56 <mwichmann> we do 14:53:00 <licquia> confirming celestia works on f20 14:55:31 <licquia> voting to treat like the others: comment, ask for more info, leave as NEW 14:55:48 <mwichmann> sure 14:56:39 <licquia> #info on 3974, confirmed celestia works on f20, asking for more info, leaving as new 14:56:51 <licquia> ok, that's the whole list of lsb bugs 14:57:04 <mwichmann> submitter provided info asked for by orc_fedo 14:57:13 <mwichmann> do we need to ask for something different? 14:57:24 <licquia> yup, but not necessarily helpful here 14:57:33 * licquia asked for glxinfo output 14:58:10 <mwichmann> that should be useful 14:58:41 <mwichmann> as we run out of time again 14:58:43 <mwichmann> any others? 14:58:54 <licquia> that's it for the usual list 14:58:58 <mwichmann> did we need to start scheduling a "new LSB" session also? 14:59:12 <licquia> yeah, we talked about doing that on the call this week 14:59:46 <licquia> think robjo was going to post something to the list to that effect 15:00:26 <licquia> #endmeeting