16:00:36 #startmeeting fpc 16:00:36 Meeting started Thu Aug 11 16:00:36 2016 UTC. The chair is geppetto. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:36 Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:00:36 The meeting name has been set to 'fpc' 16:00:36 #meetingname fpc 16:00:36 The meeting name has been set to 'fpc' 16:00:36 #topic Roll Call 16:00:47 hello 16:00:49 #chair tibbs 16:00:49 Current chairs: geppetto tibbs 16:00:52 #chair limburgher 16:00:52 Current chairs: geppetto limburgher tibbs 16:04:16 #chair racor 16:04:16 Current chairs: geppetto limburgher racor tibbs 16:04:28 So... 16:04:46 La? 16:04:52 I talked to Xavier at Flock and his hours just haven't lined up. 16:04:59 * geppetto nods 16:05:11 boo 16:05:23 #chair orionp 16:05:23 Current chairs: geppetto limburgher orionp racor tibbs 16:05:34 orionp: sorry, didn't see you there :-o 16:06:19 * geppetto remembers to send out ping, but we have 5 so we'll start in a couple of minutes 16:07:36 Hi 16:07:55 #chair mboddu 16:07:55 Current chairs: geppetto limburgher mboddu orionp racor tibbs 16:08:03 #undo 16:08:03 Removing item from minutes: 16:08:12 #unmod mboddu 16:08:19 #chair mbooth 16:08:19 Current chairs: geppetto limburgher mboddu mbooth orionp racor tibbs 16:08:24 #unchair mboddu 16:08:24 Current chairs: geppetto limburgher mbooth orionp racor tibbs 16:08:28 ok 16:08:47 Not sure what my undo did ... probably removed the roll call topic, sigh 16:08:49 #redo 16:08:53 ha, no 16:09:10 #topic Schedule 16:09:14 https://lists.fedoraproject.org/archives/list/packaging@lists.fedoraproject.org/message/FZWRWHLWLIXX77SODQYQBEQIQRTK7EMF/ 16:09:30 #topic #641 Proposal to improve wiki page of cmake packaging 16:09:35 .fpc 641 16:09:36 geppetto: #641 (Proposal to improve wiki page of cmake packaging) – fpc - https://fedorahosted.org/fpc/ticket/641 16:10:25 I will admit to not really understanding cmake. 16:10:55 I'm the cmake packager, so I suppose I do a bit :) 16:10:55 I only know the very basics 16:11:06 * geppetto hands the mic to orionp 16:11:14 rdieter's opinion would also be useful, I guess. 16:12:07 I have no problems with cleaning up the page, but I guess there's disagreement about whether the stuff in that ticket is good. 16:12:08 yeah, the kde folks live in the cmake world 16:12:14 hrm, could've sworn I'd commented in the ticket before 16:12:40 but offhand, the rpath suggestion is bad 16:13:00 Generally we request a draft, which seems like the place to start 16:13:27 the out-of-src-tree suggestion... meh, some packages work well, others don't, so maybe not a good idea to make any official recommendation there either 16:14:17 Yeh, people love to do drafts via. giving us a hand patch :-o 16:14:19 so as is, given the suggestions aren't *great* and lack of any formal draft, there's probably little/nothing to do from fpc side 16:14:53 #action Need a real draft 16:15:14 #action Out-of-src-tree suggestion won't work for all packages. 16:15:22 I would think that we'd recommend out of tree builds when they work, but if they work less than 80% of the time then maybe not. 16:15:34 I guess for rpm building it doesn't really matter. 16:15:57 sure, even autoconf-based projects could theoretically to out-of-tree builds 16:16:03 s/to/do/ 16:16:15 yeh, it's roughly the same there 16:16:21 some work well, some fail 16:16:36 Then almost certainly not worth the efort. 16:16:42 some require VPATH builds ;) 16:16:44 I do a lot of out of tree builds, but then I package MPI stuff which does multiple builds 16:17:11 orionp: , for that use-case, it's almost essential 16:17:17 Right, in the RPM building contezt that's really the only time you have to do it. 16:17:25 Otherwise it shouldn't really matter. 16:17:29 * geppetto nods ... I've used it a lot for optimized vs. debug builds 16:17:50 It's useful, but it requires testing 16:18:03 Anyway, moving on unless anyone wants to add more actions? 16:18:11 or data. 16:18:20 #action - rpath sed is rejected 16:18:22 For the sake of keeping the guideline page clean, I'd suggest just ignoring it. Or maybe adding a short appendix about it. 16:19:19 I'm done 16:19:29 #topic #642 Clarify in the packaging guidelines that the "nobody" user is not OK to use for system daemons 16:19:32 .fpc 642 16:19:33 geppetto: #642 (Clarify in the packaging guidelines that the "nobody" user is not OK to use for system daemons) – fpc - https://fedorahosted.org/fpc/ticket/642 16:20:37 I was +1 in the ticket. 16:20:41 Yeh 16:20:56 I'm happy with where you want to put it too 16:20:59 I'm +1. And I'm the distcc maintainer. 16:21:01 +1 and that seems like the right location. 16:21:03 * limburgher groans 16:21:11 limburgher: :) 16:21:51 I'd think it would be safe to change a package to do it between releases. Probably not within a release. 16:22:04 (where "do it" is use a proper user. 16:22:14 Yeah. 16:22:31 #chair Rathann 16:22:31 Current chairs: Rathann geppetto limburgher mbooth orionp racor tibbs 16:22:52 hi, terribly sorry for being late 16:22:58 No problem 16:23:25 Wea re on 642 atm. ... and at +4 with tibbs recommendation in the comment 16:23:27 * Rathann has an acute neck ache 16:23:32 :( 16:23:47 but painkillers started to work now 16:23:53 ok, let me check 16:24:01 +1 16:24:05 +1 from me, too 16:24:26 yep, +1 16:25:02 #action Clarify guidelines to not use "nobody" userid for daemons (+1:7, 0:0, -1:0) 16:25:11 #topic #643 Exceptions for bootstrapping Rust and Cargo 16:25:21 .fpc 643 16:25:22 geppetto: #643 (Exceptions for bootstrapping Rust and Cargo) – fpc - https://fedorahosted.org/fpc/ticket/643 16:25:53 Seems like a simple +1 here 16:26:17 +1 16:26:21 +1 16:26:42 +1 16:27:43 +1 16:28:47 +1 16:31:03 Rathann: That's +6 want to vote for the record? 16:31:14 yes, +1 16:31:19 #action Exceptions for bootstrapping Rust and Cargo (+1:7, 0:0, -1:0) 16:31:31 #topic #644 Allow %systemd_ordering to be used 16:31:34 .fpc 644 16:31:36 geppetto: #644 (Allow %systemd_ordering to be used) – fpc - https://fedorahosted.org/fpc/ticket/644 16:33:21 Midding paragraph of last comment seems fine, although it seems weird to have uid allocation of all of systemd grouped together. Sigh. 16:33:40 Looks reasonable. 16:36:12 Anyone else +1 or want something changed first? 16:37:41 looks fine to me, +1 16:39:14 Sure +1 16:39:39 I've lost the context; let me see if my concern about the language was addressed. 16:41:22 Looks like it. Although, there's mention of systemd UID and GID services, which we actually don't use (or at least we have no guidelines for them). 16:41:52 yeh 16:42:35 I''m +1, though I think while writing it up I'd probably remove the UID/GID bits.. 16:42:46 We should probably look into that at some point. 16:42:50 s/e.g. to install UID, GID, or tmpfiles/e.g. tmpfiles/ ? 16:43:00 Yeh, I'm shocked nobody else has come forward 16:43:07 By which I mean that someone who knows what it does should tell us about it. 16:43:10 +1, yeah, drop UID/GID for now 16:43:20 I thought it was a big thing on the container side to want to move away from scriptlets to allocate uid/guis 16:43:21 MmmHmm. 16:43:38 atomic too 16:44:04 I don't have a problem with moving away from scriptlets at all. It could potentially solve some other problems (like install-time allocation). Maybe. 16:44:19 * geppetto nods 16:44:28 But... that's a separate topic. 16:45:13 mbooth: I think that's everyone but you is +1 without the uid/gid bits 16:45:25 damnit ... not you 16:45:33 * limburgher does a shot, in accordance with the "is systemd fixing more than it breaks" drinking game 16:46:03 * limburgher loves systemd, to be clear and avoid flames. :) 16:46:37 geppetto: Thank you for having missed my vote ;) 16:46:42 racor: It's you! 16:46:47 I knew there was one missing 16:46:47 0, I am not convinced, but I apparently do not know enough about docker/containers to have a strong opinion. 16:46:53 I couldn't work out who it was 16:47:25 limburgher: Actually you didn't actually say +1, I think 16:47:34 +1 16:47:42 #action Allow %systemd_ordering to be used (+1:6, 0:1, -1:0) 16:47:51 ok 16:48:03 #topic Open Floor 16:48:15 I'm pretty sure none of the old tickets have moved/changed 16:48:26 But anyone want to talk about them, or anything else? 16:48:37 I obviously have a huge queue. 16:49:00 Hopefully I'll be able to make a dent in it over the next few days. Sadly I came home from Flock sick. 16:49:05 Nothing pressing here. 16:49:07 Booo. 16:49:19 I've heard that's a thing. 16:49:26 tibbs: You aren't the only one ... I also heard a bunch of people spent > 24hrs traveling :-o 16:50:00 If, someday, I get to attend one of these things, I'm pounding vitamin C and wearing a mask. 16:50:15 :) 16:51:06 need to attend as one of those drivable video robots, like sheldon ;) 16:51:16 best of both worlds 16:51:37 * limburgher celebrates passing Turing test 16:51:42 ha 16:53:01 Ok, if nobody has anything I'll close in a minute or so 16:53:29 nothing from me today, thanks 16:54:40 #endmeeting