15:00:16 <licquia> #startmeeting LSB Bug Triage 2013 Jan 10
15:00:17 <lsbbot> Meeting started Fri Jan 10 15:00:16 2014 UTC.  The chair is licquia. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:17 <lsbbot> Useful Commands: #action #agreed #help #info #idea #link #topic.
15:00:24 <licquia> good morning
15:00:43 * licquia had a workstation meltdown at an inconvenient time, still recovering...
15:01:11 <mwichmann> you seem to have a lot of that... the fates have not been with you on equipment recently
15:01:25 * mwichmann will stay 1000+ miles away in case it's infections
15:02:30 <licquia> part of it is relying on iffy hardware; that laptop, this cheap kvm
15:03:20 * licquia notes that the laptop that failed was a hp laptop; mwichmann might remember our past adventures with a hp laptop
15:04:54 <licquia> #info 2 new
15:04:55 <mwichmann> and another round when Intel decided to give me one of those for one cycle
15:05:02 <mwichmann> and another when my daughter bought one
15:05:15 <mwichmann> and another when her current roomate, against all our advice, bought one
15:05:24 * mwichmann is inclined to run away screaming...
15:05:27 <licquia> heh
15:05:41 <licquia> i will note, for the record, that my son has a hp laptop which has actually been good
15:06:39 <mwichmann> and if that's one of your old ones, it was good for quite a while
15:06:45 <mwichmann> anyway... we able to start?
15:06:53 <licquia> #info fhs: 27 new, 1 reopened
15:07:13 <licquia> let's go on the lsb new, then the beta
15:07:16 <licquia> works?
15:07:20 <mwichmann> +1
15:07:36 <licquia> !lsbbug 3898
15:07:38 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3898 normal, P2, ---, mats, NEW , Package naming requirements should be weakened
15:07:50 <mwichmann> both new are "move remaining piece to a new bug, not blokcing 5.0, so we can close old"
15:08:08 <mwichmann> well, perhaps not quite true
15:08:20 <mwichmann> 3898 might still be a 5.0 issue if we like
15:08:33 <licquia> most important thing: do we want to make the pkg name requirements change for 5.0?
15:08:42 <licquia> otherwise, we can move this to "future" and move on
15:08:52 <mwichmann> awk! sun just popped over the hill, this time of year that means I'm blinded!
15:09:21 <licquia> fun
15:09:36 * mwichmann leans uncomfortably to hide it behind a monitor
15:09:47 * licquia has an office whose windows face north
15:10:09 <licquia> only get blinded by snow, which we happen to have
15:10:15 <licquia> anyway...
15:10:36 <licquia> i'm not sure this has a code impact; it's just spec wording at this stage
15:10:40 <mwichmann> yes
15:11:01 <mwichmann> possibly also checker
15:11:23 <licquia> so i'm willing to be persuaded either way
15:11:37 <mwichmann> it's hard to have a bug which has no champion
15:11:43 * licquia is inclined against change at this late stage
15:13:07 <mwichmann> that's a fair position
15:13:19 <licquia> ok, wins the day
15:13:39 <mwichmann> let's attach it to 3897 with a note "if someone wants to argue for..."
15:15:34 <licquia> !lsbbug 3899
15:15:36 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3899 normal, P2, ---, dsilakov, NEW , libtodb2 may reassign core base types to new lib
15:15:58 <licquia> #agreed postpone 3989 to next lsb release
15:16:19 <licquia> #info typo: 3989 -> 3898
15:17:03 <licquia> this seems like a "future" bug as well
15:17:13 <mwichmann> definitely
15:17:28 <mwichmann> hard to really give it a blocker, either
15:17:55 <licquia> #agreed unblock 5.0 release w/ 3899
15:18:37 <mwichmann> It seems we have a "pre-bug", from Till's email
15:18:59 <mwichmann> !lsbbug 1755
15:19:01 <lsbbot> mwichmann: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=1755 normal, P2, ---, licquia, RESOLVED FIXED, Add foomatic-rip to LSB 3.2
15:19:23 <mwichmann> apparently we got this wrong by picking the /usr/bin one, were supposed to pick the one in /usr/lib/cups/filter
15:20:25 <licquia> sigh
15:20:46 <mwichmann> I'm not entirely sure I'm buying this
15:20:53 <licquia> not sure how we want to address this; we could add /usr/lib/cups/filter/foomatic-bin to 5.0
15:21:14 <licquia> sorry, /usr/lib/.../foomatic-rip
15:21:28 <licquia> the one in /usr/bin would have to be deprecated
15:23:00 <licquia> anyway, looks like it should be filed; i'll explore options in the bug
15:23:06 <mwichmann> ok
15:23:42 <licquia> any inclination on whether to accept the /usr/lib path in 5.0?
15:23:49 <licquia> should be a simple change
15:23:52 <mwichmann> this also has some test implications, I think... there was a separate bug to add a test for this
15:23:54 <licquia> no new docs needed
15:24:51 <mwichmann> once there's a bug I'll look at the text
15:25:09 <mwichmann> I did eventually grep, and we never do mention /usr/lib/cups
15:25:14 <mwichmann> wonder if that path is in fhs?
15:26:08 <mwichmann> ... it isn't, that I can find
15:26:25 <licquia> it's very cups-specific; basically, that's where cups looks for filters by default
15:26:27 <mwichmann> !lsbbug 3164
15:26:29 <lsbbot> mwichmann: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3164 blocker, P1, 5.0, licquia, ASSIGNED , LSB 5.0 Complete Release
15:26:44 * orc_fedo looks in after wrapping a support tc
15:27:17 <mwichmann> hi
15:27:40 <orc_fedo> hi
15:28:50 <licquia> !lsbbug 3900
15:28:52 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3900 normal, P2, ---, mats, NEW , foomatic-rip is specified in the wrong place
15:30:06 <mwichmann> back in a sec
15:31:20 <orc_fedo> I have to think that NOT adding per package additional compatability paths is the 'correct; course --- the trend is away from path profiveration, and there is a finite limit of the PATH variable's capacity that will be hit, as it is non-extensivle under some scripting languages
15:31:20 <mwichmann> back
15:31:37 <licquia> i'm inclined to keep 3900 as new, and reconsider next week after giving people a chance at feedback
15:31:59 <licquia> orc_fedo: /usr/lib/cups/filter doesn't need to be part of the PATH
15:32:10 <licquia> cups will look for filters at that specific path
15:32:50 <mwichmann> we have some cmds which have an absolute path because they need to appear in certain scripts, but try to avoid it
15:33:03 <orc_fedo> reading the email it does not directly say that, but hokay
15:33:40 <mwichmann> so 5.0 bugs?
15:33:46 <licquia> think so
15:34:29 <mwichmann> dependencies of 3164
15:34:51 <licquia> #info lsb 5.0 has 114 open bugs
15:35:47 <licquia> mwichmann: any particular ones that need group attention, or should we just move down the line?
15:36:04 <mwichmann> lemme look
15:36:30 <mwichmann> 2143 wanted some look
15:36:39 <licquia> !lsbbug 2143
15:36:41 <lsbbot> licquia: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=2143 enhancement, P2, 5.0, mats, ASSIGNED , Add *ifaddrs to LSB
15:37:30 <mwichmann> manpage need; plus the last thing was a question about doing it differently than upstream?
15:37:59 <mwichmann> do we need to do anything here?
15:38:19 <licquia> how odd
15:38:26 * licquia wonders why one doesn't include the other
15:38:40 <licquia> maybe some issue w/ include loops or something
15:39:00 <mwichmann> gah, tryign to juggle two keyboards, keep typing on wrong one :(
15:39:25 <mwichmann> linux manpage says nothing on this, just says #include <ifaddrs.h>
15:39:30 <mwichmann> are we okay just staying with that?
15:39:43 <licquia> i'd say let's stick with that
15:39:59 <licquia> if it's a problem, it'll come up when someone tries to build thigns
15:40:00 <licquia> things
15:40:09 * licquia adds a note
15:40:22 <mwichmann> sold.  will gladly accept manpage volunteers :)
15:41:01 <mwichmann> !lsbbug 2902
15:41:04 <licquia> we have a few of those, as i recall
15:41:04 <lsbbot> mwichmann: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=2902 enhancement, P2, 4.1-updates, mats, ASSIGNED , RFE: Add symbols for libnspr4
15:41:10 <licquia> (manpage-missing bugs)
15:41:19 <mwichmann> yes, this is one of them
15:42:01 <licquia> i think that' s just "someone needs to slog through and get those added"
15:42:06 <mwichmann> I added a bunch of functions, but state of upstream docs is not helping us
15:42:14 <licquia> someone -> probably me, unless someone has time and volunteers some
15:42:22 <mwichmann> licquia is right, and it will probably be him
15:42:50 <licquia> of course, manpages added to bugs will be gratefully accepted :-)
15:42:59 <licquia> for those with the inclination
15:43:36 <licquia> ok, anything else that "floats to the top"?
15:43:56 <mwichmann> yeah, looking in my list
15:44:48 <mwichmann> !lsbbug 3754
15:44:50 <lsbbot> mwichmann: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3754 blocker, P2, 5.0, mats, ASSIGNED , missing interface definitions
15:45:15 <mwichmann> #info that was a typo, ignor
15:45:21 <mwichmann> !lsbbug 3745
15:45:23 <lsbbot> mwichmann: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3745 normal, P2, 5.0, mats, ASSIGNED , Type entries can't move
15:46:15 <mwichmann> propose unhooking this from 5.0
15:46:40 <mwichmann> we need to think of how to fix this, but not really now
15:46:48 <licquia> ok; is there something about type moves that has to be done for 5?
15:46:57 <mwichmann> I unwound the gfloat thing
15:47:07 <licquia> ok, then i agree
15:48:07 <licquia> catching up on meeting notes...
15:48:36 <licquia> #agreed don't worry about undocumented header usage for 2143
15:48:49 <licquia> #agreed unhook 3745 from lsb 5.0
15:49:20 <licquia> ok, done; anything else?
15:49:25 <mwichmann> !lsbbug 3850
15:49:27 <lsbbot> mwichmann: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3850 normal, P1, ---, mats, ASSIGNED , mkheader does not look at arch-specifics for ATattribute
15:49:46 <mwichmann> this is twinned with 3849, where we actually have a specific bug
15:50:44 <mwichmann> two arches have wrong epoll struct, and it won't go away until the script gets some work
15:51:01 <licquia> right, so should be fixed
15:51:16 <licquia> probably a higher priority, since we can't fix the other bug w/o it
15:51:39 <mwichmann> db change is done, fwiw
15:52:00 * licquia notes that the priority is already dialed up to 11
15:52:24 <orc_fedo> one last thing mentioned in the call --- could either of you please apply this patch: https://gallery.herrold.com/herrold-2014-key.patch on lsb-puppet-upstream/modules/user/manifests
15:52:39 <orc_fedo> oops ... CR slipped in too early
15:52:51 <licquia> heh
15:53:14 <orc_fedo> licquia: is P11 sort of a 'crank up the volume' urgency?
15:53:52 <licquia> something like that; don't think we have a "P11", referencing the whole "dial up to 11" meme
15:54:36 <mwichmann> !lsbbug 3853
15:54:39 <lsbbot> mwichmann: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3853 normal, P2, ---, mats, ASSIGNED , Commands do not have a refspecurl
15:54:59 <mwichmann> this is a poke at licquia: it's ready to close once you can see it in the master navigator
15:55:12 <mwichmann> means db/nav sync with bzr
15:55:29 <licquia> ok
15:55:43 <mwichmann> !lsbbug 3856
15:55:45 <lsbbot> mwichmann: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=3856 minor, P2, 5.0, mats, ASSIGNED , RFC: should libpam be in "utility libraries" instead of "base libraries"?
15:55:52 * licquia would love an easy way to update the master db, but no such luck
15:56:14 <mwichmann> we put this on the 5.0 list to get an answer... given the time the answer may be next-release
15:56:36 <mwichmann> this only affects spec organization
15:56:45 <mwichmann> pam sits in a place where it doesn't seem to logically fit
15:57:05 <licquia> how much work to move?
15:57:25 <mwichmann> good question, I *think* it's easy
15:58:18 <mwichmann> to move to utility is just editing makefiles
15:58:34 <mwichmann> to move to security is a db change too
15:59:00 <licquia> only argument i have: libpam isn't really a "utility", since it hooks into the os in a way that makes it hard to statically link
15:59:15 <mwichmann> anyway, was going to say, this sort of thing sounds less compelling the closer you get to release panic time....
15:59:18 <licquia> i think the arg could be made that it's "base" in that way
15:59:40 <licquia> "security" could work, but security is all crypto stuff atm
15:59:52 <mwichmann> it is
16:00:04 <mwichmann> but pam does put its headers in /usr/include/security :)
16:00:05 <licquia> not saying i'm persuaded, but between that and the release panic you mentioned, makes sense to not do
16:00:20 <mwichmann> okay, let's pop it to next release
16:00:40 <licquia> ok
16:01:05 <mwichmann> I'm run us out of our hour and have many more questions, sigh....
16:01:44 <mwichmann> do you have any date in mind for wrapping things up?  is there a point where we need to do a short bug session every day, say?
16:02:03 <orc_fedo> perhaps just do ad hoc in main?
16:02:22 <licquia> i have a task list for beta 2, which mostly amounts to doing a feature-complete release (specs, all archs, etc.)
16:02:50 <licquia> imho, post-beta-2 seems like the time
16:03:10 <mwichmann> just as long as we've scanned through this list to make sure we really are feature-complete
16:03:22 <licquia> #agreed unhook 3856 from lsb 5.0 release
16:03:30 <licquia> good point
16:04:06 <licquia> that should be part of the beta 2 release process, certainly
16:04:19 <mwichmann> we can also unhook 2249 from 5.0
16:04:24 <mwichmann> !lsbbug 2249
16:04:26 <lsbbot> mwichmann: 04Bug http://lsbbugs.linuxfoundation.org/show_bug.cgi?id=2249 major, P2, 5.0, mats, ASSIGNED , Convert spec from SGML to XML
16:04:56 <mwichmann> I have more experience now, and can do this, BUT, due to the number of existing files and generator scripts, it's not happening for this release
16:05:23 <licquia> #agreed unhook 2249 from lsb 5.0 release
16:05:24 <mwichmann> I fiddled with one yesterday as an experiment ("could this happen really quickly") and broke stuff
16:05:59 <licquia> ok, done
16:06:08 <licquia> think we can call it for this time
16:06:12 <licquia> #endmeeting