13:03:27 #startmeeting rpm_packaging 13:03:28 Meeting started Thu Dec 8 13:03:27 2016 UTC and is due to finish in 60 minutes. The chair is number80. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:03:29 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:03:31 The meeting name has been set to 'rpm_packaging' 13:03:33 agenda's here 13:03:37 https://etherpad.openstack.org/p/openstack-rpm-packaging 13:06:27 nobody? 13:07:01 o/ 13:07:12 #chair dirk 13:07:13 Current chairs: dirk number80 13:07:44 let's wait few more minutes or I'll just review our current review queue 13:08:12 toabctl is also here but I keep him busy :) 13:09:28 np 13:10:36 any controversies you want us to look at? 13:12:55 ok, closing meeting in 2 min :) 13:13:12 the renderspec version thing 13:13:27 ok 13:13:28 https://review.openstack.org/#/c/407102/ 13:13:47 I wonder if you really want to throw away everything after the 3rd digit 13:13:48 number80: I added potential topcs to the meeting agenda 13:13:53 #action number80 focus on renderspec reviews 13:13:57 ok 13:13:59 number80, is that what fedora wants to have? 13:14:30 toabctl: it's close, yes 13:14:42 #topic PTG planning 13:15:24 We are likely to have a representative from RDO team there, anyone else? 13:15:31 so I have seen that we have a room 13:15:40 for monday-tuesday and I booked a flight 13:15:42 Yes 13:15:45 goood 13:15:45 I'm pretty sure that I'll be there 13:15:46 but I'll not stay a whole week 13:16:14 number80: so you're not going? 13:16:32 I haven't seen any response from travel support yet though :/ 13:16:32 I have been designated, but I have yet to get final decision 13:17:29 ok, so I guess then it would start to make sense to think about planning those two days 13:17:38 dirk, travel support will answer at 16th dec 13:17:42 Yes 13:17:42 would be cool to get some more progress, on e.g. upstream building or improved gating 13:17:54 ack for both 13:18:00 do we want to create an etherpad or the like for collecting topics? 13:18:08 dirk, yes 13:18:15 also don't forget to hand in t-shirt sizes :) 13:18:29 for our absolutely rocking packaging donkey shirt 13:18:31 oh. where? 13:18:45 ah. the registration? 13:18:45 dirk: for now, I'm adding suggestion on meeting etherpad 13:18:50 Yes 13:19:39 toabctl: I think its via the registration. I didn't process that part yet, I just read about it somewhere 13:19:47 * dirk still has troubles finding a hotel though 13:20:49 sorry folks, can I have some recommendation about https://review.openstack.org/#/c/401200/ and https://review.openstack.org/#/c/401260/? 13:21:11 ack 13:21:31 number80: whats the link to meeting etherpad? 13:21:43 kaslcrof, the topic is on the agenda 13:22:11 kaslcrof: its the "xstatic packaging" meeting point on the agenda 13:22:15 kaslcrof: agenda is here: 13:22:25 #link https://etherpad.openstack.org/p/openstack-rpm-packaging 13:22:58 number80: ah, I see you added it to the normal etherpad 13:23:01 yes 13:23:46 SUSE CI need wheels, but centos tests ignore this field in setup.py and passes..What can I do to make it pass? 13:25:11 hm. I wonder why upstream has wheel in the requirements 13:25:41 sorry, I say >> ok, closing meeting in 2 min :) and decided to ask earlier 13:25:50 *saw 13:26:09 ah right 13:26:13 next topic 13:26:32 #topic xstatic packaging 13:27:16 kaslcrof: I'm not sure this is related to wheels 13:27:25 kaslcrof: you'r enot failing on build but on a suse specific packaging check 13:27:51 kaslcrof: this is about the Requires: py2pack('XStatic') that I wanted to get removed in an earlier incarnation of that patch 13:28:00 but maybe my request to remove it isn't correct 13:28:22 I wanted to reseearch that before this meeting but was in other meetings all day so far and had no second of spare time :/ 13:28:48 currently I think if the Rrequires: would be added back then build would pass for SUSE CI 13:29:02 but I darkly remember that there were file conflicts (multiple packages owning the same file) afterwards and that caused other issues 13:29:07 but maybe I misremember that 13:30:00 dirk, i will check it now 13:30:27 kaslcrof: just add the Requires back, and we'll get an answer from suse ci before the end of this meeting :) 13:31:03 so in general, it seems the Requires on XStatic is in the MOS packages 13:31:16 number80: how is the situation for RDO? do the XStatic-foo packages runtime-require XStatic? 13:31:36 no, since it's broken we don't 13:33:37 number80: do you remember what is broken? I also darkly remember there were problems but I don't recall them 13:34:17 * number80 trying to remember 13:35:19 https://bitbucket.org/thomaswaldmann/xstatic/issues/2/cannot-build-a-new-xstatic-package-with 13:35:39 I think it's related to that issue (still unsolved) 13:36:18 Actually, I was wrong we have python-XStatic as Requires but not BuildRequires 13:40:28 next topic? 13:40:44 #topic systemd macros 13:42:11 hm. jpena|off is not here 13:42:30 I think the idea was to just use the macros from upstream instead of the ones suse introduced 13:42:51 yep, that's what we already do 13:42:54 dirk, are the upstream macros available in our CI now? 13:43:42 toabctl: I think yes, didn't confirm yet 13:43:47 toabctl: lets assume yes if not I'll fix it 13:44:22 ok.then we can abandon https://review.openstack.org/#/c/407026/ 13:44:30 number80: ok, so we agree to have XStatic in requires? can we record that as agreed? 13:44:42 number80: then I can go revisit my -1's and we can finally get the swoop of xstatics done 13:45:06 toabctl: imho yes. I wanted to refresh the keystone or mistral review iwth the new macros 13:45:19 I can do that now 13:45:38 what? abandon or refresh the review? 13:46:07 refresh keystone review 13:46:18 dirk: yes 13:46:30 #agreed have XStatic as Requires in XStatic packages 13:48:20 let's also agree on using the upstream systemd macros 13:48:43 sorry, but why we should add Xstatic as Requires? 13:49:30 ok 13:49:42 kaslcrof: some directories are owned by base package 13:50:39 kaslcrof: all Xstatic-$foo packages should runtime-require python-XStatic 13:50:52 kaslcrof: to track ownership of directories. thats what the suse ci is complaining about 13:51:54 It's correct, only BR is breaking 13:52:00 R works fine 13:54:19 next topic? 13:54:25 dirk, keystone does not use any systemd macros so there is no need for a refresh 13:54:27 yes 13:55:30 yep 13:55:50 let's skip RDO gate as jpena is not here 13:55:58 #topic pending reviews 13:56:07 packages reviews (https://review.openstack.org/#/q/project:openstack/rpm-packaging+status:open ) 13:56:16 pymod2pkg reviews (https://review.openstack.org/#/q/project:openstack/pymod2pkg+status:open ) 13:56:24 renderspec reviews (https://review.openstack.org/#/q/project:openstack/renderspec+status:open ) 13:56:34 number80, https://review.openstack.org/#/c/406901/ could be merged imo 13:56:51 also https://review.openstack.org/#/c/407808/ 13:57:08 https://review.openstack.org/#/c/406464/1 <- merged 13:57:11 ack 13:57:12 IgorYozhikov wanted to discuss https://review.openstack.org/#/c/408371/ 13:57:18 Yep 13:57:36 but imo we should just accept it (with a normal review). nothing special there I thinkj 13:58:11 Yep, maybe I have some nitpicks about some comestic changes 13:58:16 there is no reason to not map package names for ubuntu if they need that 13:58:19 number80, yes 13:58:50 pymod2pkg, so such changes are acceptable by definition :) 13:59:14 https://review.openstack.org/#/c/406465/ could be merged 14:00:00 agreed 14:00:15 any other controversies before we close themeeting? 14:01:32 not from my side 14:03:22 okay, then thanks for attending and see you next week! 14:03:27 #endmeeting