14:00:12 <licquia> #startmeeting LSB Bug Triage 2014 Apr 11
14:00:12 <lsbbot> Meeting started Fri Apr 11 14:00:12 2014 UTC.  The chair is licquia. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:12 <lsbbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
14:00:33 <licquia> hello everyone
14:01:53 <mwichmann> hiya
14:02:38 <mwichmann> couldn't reach buildbot page this morning, browser icon just spinning
14:02:49 <licquia> yup, just restarted
14:02:52 <mwichmann> k
14:03:14 <mwichmann> was my theory correct? you've used up most of the magic in your wand, and things are fading?
14:03:48 * licquia wonders
14:04:02 <licquia> some weird things about "not rotating the web site" or some such
14:04:16 <licquia> is that a symptom of fading magic powers? :-)
14:04:35 <licquia> #info lsb: 3 reopened, 2 pleasetest, 5 new
14:05:04 <mwichmann> need a tesla-style recharge station for indy area
14:05:32 <mwichmann> reopens first?
14:05:51 <licquia> makes sense
14:05:57 * licquia gets fhs stats, too
14:06:25 <licquia> #info fhs: 27 new
14:06:34 <mwichmann> for some value of new :)
14:06:40 <licquia> heh
14:06:46 <licquia> !lsbbug 3195
14:06:49 <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.
14:07:07 <mwichmann> irc discussion, seems we didn't get out qmtest-harness after its fix
14:07:32 <mwichmann> and a yocto dev is dealing with a strict rpm, and blowing up
14:08:05 <licquia> saw that; this is something that's fixed subsequent to the latest stable release?
14:08:13 <mwichmann> (I did note, we usually "resolve" when fix in in bzr and tested)
14:08:23 <mwichmann> yes
14:09:07 <licquia> hmm, that seems odd; the bug is old
14:09:17 <licquia> wonder why it didn't get out?
14:09:28 <mwichmann> -11, which is 49: Stew Benedict 2011-02-24 own the directories we populate (bug 3195)
14:09:31 <mwichmann> maybe not on our list
14:11:00 <licquia> yeah, that could be it
14:11:14 <licquia> so we just need to release our current version, basically
14:11:35 <licquia> shouldn't be a problem, since qmtest-harness is version-indep
14:12:21 <mwichmann> right
14:12:25 <licquia> so, do a release of qmtest-harness to get the thing released?
14:12:42 <mwichmann> I'd say so
14:13:21 <licquia> #action licquia do a release of qmtest-harness to push fix for bug 3195
14:13:44 <licquia> !lsbbug 3256
14:13:46 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3256 normal, P2, ---, stewb, REOPENED , CUPS/convenience tests fail or fail to report against cups-1.5-0.2.b2 (rawhide)
14:14:04 <mwichmann> more irc chatter here
14:14:37 * licquia is not happy with the state of the printing tests in general
14:14:51 <mwichmann> agree, but what can we do?
14:15:13 <licquia> same deal, unreleased fix?
14:15:27 <mwichmann> I don't know
14:15:55 <mwichmann> I actualyl preferred a new bug for this, but since he spotted similar results, maybe reopen was right until we know if it's this case
14:17:39 <mwichmann> to my completely educated eye it seems like there's not communication with the server, which could be the port thing stewb talked about
14:17:52 * mwichmann does not volunteer to look into this one
14:17:57 <licquia> could be
14:18:50 <licquia> for an old bug like this, it makes it hard to figure out what was done in the past
14:19:03 <licquia> so i think i'm inclined to break it off as a new bug
14:20:07 <licquia> ok with that?
14:20:14 <mwichmann> your call, I'm fine
14:21:45 <licquia> #agreed clone 3256 (to 3964), close original
14:25:46 <mwichmann> triage on 3964?
14:26:08 * licquia is fighting with bugzilla
14:26:14 <licquia> clone is stupid
14:27:59 <mwichmann> it is, I agree
14:28:06 * licquia notes for the future: *never* clone bugs, it doesn't do what you want
14:29:01 <licquia> and that first comment is really, really long; i'm inclined to go into the database and purge it
14:29:08 <licquia> ok
14:29:33 <mwichmann> wow!
14:30:15 <mwichmann> yes, that part of clone is wonky.  a single line "this is a clone of foo" and you past what you want would be fine
14:30:33 <licquia> sigh
14:30:40 <licquia> !lsbbug 3964
14:30:42 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3964 normal, P2, ---, licquia, NEW , printing-test failures on cups 1.6.1
14:32:07 <licquia> ok, i'm going to vote it goes on the "next stable release" rollup
14:32:30 <licquia> and probably deserves a "problemreport" tag
14:32:54 <mwichmann> sure
14:33:10 <mwichmann> I thought there was already problemreport on this, but tag will let us (you) check
14:33:41 <licquia> there might be for the old bug
14:35:04 <mwichmann> ah... pr needs both test and bug #?  then you would need new
14:36:09 <licquia> #agreed target 3964 at next stable update, write problem report
14:36:39 <licquia> !lsbbug 3599
14:36:41 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3599 normal, P2, ---, licquia, REOPENED , rework development Release naming to add a leading '0.' before the YYYYMMDD
14:37:27 <licquia> that's the release version thing we talked about yesterday
14:37:35 <mwichmann> yes
14:37:44 <licquia> so, if i can fast-forward a bit...
14:37:50 <mwichmann> one more minor tweak does it; pushed the runtime-test one as sample
14:37:58 <licquia> !lsbbug 3962
14:38:00 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3962 normal, P2, ---, licquia, NEW , versioning mechanism is not correct in lsb-test-core
14:38:13 <licquia> close as duplicate of 3599?
14:38:43 <mwichmann> yes, once I got pointed to that one, 3962 makes no sense as independent bug
14:38:49 <licquia> ok, cool
14:38:49 <mwichmann> (to me anyway)
14:39:09 <licquia> #agreed close 3962 as duplicate of 3599
14:39:46 <licquia> #action get buy-in on latest version change for 3599, roll out to other pkgs
14:40:18 <licquia> going to otherwise leave 3599 alone, so we're reminded of it next time
14:41:28 <mwichmann> orc_fedo_ +1'd in other channel a bit ago
14:41:44 <licquia> ok, so maybe it's just me that needs to be ok with it
14:42:10 <mwichmann> ppc64 never came back, it looks like
14:42:21 <licquia> no, i need to follow up with osuosl
14:42:36 <licquia> filed a ticket with them last week; assume they've been heartbleed-ed
14:42:37 <mwichmann> lsb-test-core-4.1.91-0.20140410.6.x86_64.rpm
14:42:52 <mwichmann> the entire change is we got back the trailing ".6"
14:42:58 <licquia> ok
14:43:32 <licquia> i can see if the diff in lsb-test-core can be automated in any way to apply to the others
14:43:34 <mwichmann> I went off in the weeds because the thing I looked at for comparison was lsb-build-*, which did not get converted (and it's a mess, so not easy), and got the wrong impression
14:43:53 <licquia> yup, makes sense
14:44:02 * orc_fedo_ gets off heartbleed support call
14:44:02 * licquia thought he had converted lsb-build, but apparently not
14:44:20 <mwichmann> tread very carefully there if you look,
14:44:49 <mwichmann> it's got strings for each pkg, repeated multiple times, then building on the base one
14:45:01 <licquia> yeah, might see about refactoring that at the same time
14:45:09 <licquia> of course, not until the gtk3 script is done
14:45:14 <mwichmann> low prio
14:45:29 <licquia> !lsbbug 3946
14:45:31 <mwichmann> development-version things really only affect the two of us these days, I think
14:45:32 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3946 normal, P2, 4.1-updates, licquia, PLEASETEST , Add scandir and alphasort interface with tests to olver-core
14:46:13 <licquia> i recall; started a test for this one and the next, but it ran for days and never finished
14:46:23 <mwichmann> loveley
14:46:33 <licquia> not sure if that's an artifact of some weirdness here or a real bug
14:47:04 <licquia> !lsbbug 3947
14:47:06 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3947 normal, P2, 4.1-updates, licquia, PLEASETEST , Extend mutex interface in OLVER to support prioceiling and protocol
14:47:12 <licquia> that's the second one
14:47:29 <licquia> so, just need to re-test those
14:47:51 <licquia> #action test 3946 and 3947 with a snapshot olver-core test run
14:47:52 <mwichmann> in your copious free time
14:48:02 <licquia> at least i can just kick it off and let it percolate
14:48:31 <licquia> !lsbbug 3959
14:48:33 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3959 enhancement, P2, ---, licquia, NEW , add limited scope, keyed ssh scripts to perform buildbot status queries, and restarts
14:48:49 <orc_fedo_> speak of that devil
14:49:22 <orc_fedo_> when I filed it and heartbleed had not surfaced, SOMEONE said I was a paranoid
14:49:35 <licquia> in actuality, i wonder if the easier thing here would be to just make sure orc_fedo_ and mwichmann can get in to the build master and restart it
14:49:59 <orc_fedo_> licquia: works for me if so, hand to me and make a 'handwavy future' bug
14:50:17 <licquia> ok
14:51:01 <orc_fedo_> licquia: your approach would just be a puppet recipe edit, now?
14:51:14 <licquia> maybe not even that; it might be in place already
14:52:05 <mwichmann> my time is ticking, quickly hit the last few?
14:52:45 <licquia> sure
14:52:53 <licquia> !lsbbug 3960
14:52:55 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3960 normal, P2, ---, licquia, NEW , url_checker should also examine ISrefspecurl
14:53:22 <licquia> #agreed on 3959, see if simple ssh access works
14:53:40 <mwichmann> this ties in to earlier one, just noticed it doesn't check the subsidiary links
14:54:00 <licquia> ok; tie to any particular milestone?
14:54:03 <orc_fedo_> add as ablocker on the earlier?
14:54:14 <mwichmann> this is useful, but these links affect only navigator (yes, can tie to the earlier one)
14:54:42 <mwichmann> we could release 5.0 and continue to fiddle links afterwards, so I think not strictly 5.0 blocker
14:54:55 <mwichmann> individual-page links are not in spec
14:54:58 <licquia> ok; do we know the bug # for the previous bug?
14:55:26 <mwichmann> 3953?
14:55:27 <orc_fedo_> 3953
14:56:10 <orc_fedo_> added '60 to '53
14:56:53 <licquia> !lsbbug 3961
14:56:55 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3961 normal, P2, ---, licquia, NEW , network tests do not follow normal testing policy
14:57:10 <licquia> #agreed tie 3960 to 3953, mark assigned, fix after 5.0
14:57:21 <mwichmann> generalization of the other bugs I've wasted time chasing
14:57:41 <licquia> this is the "each test is not independent" bug, right?
14:57:54 <mwichmann> pretty much
14:58:02 <mwichmann> 3524 and 3957
14:58:04 <licquia> next stable update?
14:58:33 <mwichmann> they're not independent because they're careless on error, pretty much only do the right thing if everything succeeds
14:58:55 <mwichmann> reads to me like the developers did not fully understand the methodology/framework
14:59:14 <mwichmann> this /should/ be fixed, but who's going to do all of it?
14:59:23 <licquia> #agreed tie 3961 to next stable update
14:59:28 <mwichmann> I think we chatted earlier about doing this on-demand
14:59:29 <licquia> we'll worry about that later
14:59:39 <licquia> !lsbbug 3963
14:59:41 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3963 normal, P2, ---, mats, NEW , document method of  use. freshening of local 'look-aside' cache archives (bzr, git, ...)
14:59:42 <mwichmann> the four or five areas that have actual issues get attention, wait on others
15:00:15 <orc_fedo_> I filed taht as you two were jawing about undocumneted workflow ;)
15:00:19 <licquia> this is the bzr.freestandards.org bug
15:00:19 <orc_fedo_> no rush
15:01:33 <licquia> #agreed fix 3963 after 5.0, or as needed
15:01:38 <licquia> and that's it!
15:01:42 <licquia> just under the wire
15:01:43 <mwichmann> I think as a practical matter,
15:02:05 <mwichmann> if fixing package/Makefile for the dev-release thing, kicking the freestandards line at the same time makes sense
15:02:08 <mwichmann> I did for runtime-test
15:02:12 <mwichmann> don't have to tie to anything
15:02:14 <licquia> definitely
15:02:20 <mwichmann> just if you're in there already...
15:02:33 <licquia> #endmeeting