19:59:49 #startmeeting barbican 19:59:50 Meeting started Mon Jun 22 19:59:49 2015 UTC and is due to finish in 60 minutes. The chair is redrobot. Information about MeetBot at http://wiki.debian.org/MeetBot. 19:59:52 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 19:59:54 The meeting name has been set to 'barbican' 20:00:04 #topic Roll Call 20:00:09 o/ 20:00:10 o/ 20:00:12 o/ 20:00:18 heyo/ 20:00:20 o/ 20:00:22 o/ 20:00:24 o/ 20:00:27 Greetings 20:00:28 o/ 20:00:39 Nice! Lots of barbicaneers here today! 20:01:02 As usual, the agenda for the meeting can be found here: 20:01:04 #link https://wiki.openstack.org/wiki/Meetings/Barbican 20:01:49 Everyone is encouraged to add agenda items ahead of time. Otherwise we end up with meetings like today, where I make up topics as I go along :-P 20:02:06 hehe, was gonna say that agenda needs some love ;) 20:02:26 elmiko hehe indeed 20:02:27 i have some minor castellan and barbican/fedora updates 20:02:34 elmiko noted. 20:02:35 o/ 20:02:38 as long as we don't end up talking about the weather, it's alright :P 20:02:42 lol 20:03:05 #topic Action items from previous meeting 20:03:10 #link http://eavesdrop.openstack.org/meetings/barbican/2015/barbican.2015-06-15-19.59.html 20:03:37 First action item was for me to look into the container-acl spec for a possible Google hangout this week 20:03:55 But I didn't actually start reviewing until today when kfox1111 poked me about it 20:04:08 still half-way through, I think it would be good for others to chime in on it. 20:04:23 #link https://review.openstack.org/#/c/190404/3 20:04:36 I don't think we need a hangout just yet though 20:04:44 next item 20:04:50 was also for me to look into Kite status 20:05:07 I sent an email to Jamie Lennox about it, but I haven't heard back from him yet 20:05:15 Jamie being the main driver/contributor for Kite 20:05:26 redrobot: I don't think there's much work going on regarding kite :/ 20:05:32 but I'll update you guys when I hear back from him. 20:05:49 jaosorior indeed. Last I heard he was re-tasked on something else 20:06:13 so that does it for the action items from last week 20:06:40 moving on 20:06:43 #topic liberty-1 20:06:49 #link https://wiki.openstack.org/wiki/Liberty_Release_Schedule 20:06:58 Liberty Milestone 1 is due this week. Preferably tomorrow. 20:07:29 tons of bugfixes are going into it 20:07:31 #link https://launchpad.net/barbican/+milestone/liberty-1 20:07:38 but we're a little light on blueprints 20:08:37 although I think one bp from jaosorior is close to landing 20:08:54 in any case it's just a milestone release, so it will be fine if it goes out with only the one feature 20:09:04 well... in theory two bps, just need reviewers ;) 20:09:15 jaosorior links? 20:09:44 jaosorior it would be good to drop them here in case some of our awesome reviewers want to tune out of the meeting ;) 20:10:04 This one has been there for a while https://review.openstack.org/#/c/178601/ ; already had 3 +2s, but decided to fix a minor issue on the last minute 20:10:19 and now it's back to....not being reviewed 20:10:28 =( 20:10:49 yeah, that's been pending review for 10 weeks :( 20:11:03 and these two https://review.openstack.org/#/c/194157/2 https://review.openstack.org/#/c/194283/ 20:11:12 Also, please please please review the spec CRs 20:11:36 at the last Summit we made it a goal to get all the specs reviewed by l-1 20:11:44 and we didn't quite make it 20:12:04 liberty-2 is going to be a hard deadline for specs, but it would be awesome if we can get stuff landed before then. 20:13:12 redrobot: some of the tls blueprints are pending Ade returning from vacation to finalize them...hopefully landing next week 20:13:22 woodster_ thanks for the reminder 20:13:35 Ade asked me to shepherd his blueprints while he's on vacation 20:13:52 so if you have a negative vote on any of his specs, please ping me for an update 20:14:16 * redrobot makes a note to go check on Ade's blueprints for negative votes 20:14:29 any questions/comments about liberty-1? 20:15:20 Just agitating on arunkant's behalf the quota bp (just workflow needed) 20:15:23 #link https://review.openstack.org/#/c/159938/ 20:15:58 ok, moving on 20:16:08 #topic Castellan 20:16:15 elmiko tag. you're it! 20:16:32 ok 20:16:45 i have drafted this bp 20:16:46 https://blueprints.launchpad.net/castellan/+spec/improved-configuration-options 20:16:50 oops 20:16:51 #link https://blueprints.launchpad.net/castellan/+spec/improved-configuration-options 20:17:09 and i have a poc working locally for it, i need to generate some docs and some tests and i think it will be ready for review 20:17:24 although i might make it 2-3 reviews to make it easier on the eyes. 20:17:29 /buffer 25 20:17:33 elmiko awesome! 20:18:01 there are no specs yet for castellan, so i tried to be verbose in the bp. which might be a little unsusal, but i'd appreciate any feedback folks have =) 20:18:09 elmiko: +1 that is nice! 20:18:15 Thanks elmiko! 20:18:22 I'll have to go add a liberty series to the launchpad page, so we can track it. 20:18:27 =) 20:18:29 #action redrobot to add liberty series to castellan 20:18:45 that was all i had for castellan 20:18:51 thanks elmiko 20:19:09 #topic Barbican in Fedora 20:19:13 hehe 20:19:14 sooo 20:19:29 i've been looking at the rpm patch that xaeth has put together 20:19:40 and i'm working on some startup/shutdown scripts to use with systemd 20:19:52 sadly, i'm running into issues with the way barbican.sh does this 20:19:56 basically the start is fine 20:20:08 but the stop does a `killall uwsgi`, essentially 20:20:16 which is not gonna fly for a fedora package 20:20:25 woodster_: I haven't +2ed arunkant's audit bp because of the lack of testing specification, that's the only reason 20:20:36 so, i'm looking into something with recording the pid and using that to kill the uwsgi process. but, 20:20:45 i'd be curious to hear if anyone else has ideas about this? 20:20:59 also curious if barbican upstream would want the scripts when i finish? 20:21:36 elmiko: Would probably be useful, even if we keep them in the "contrib" folder or something of the sort 20:21:39 elmiko: I'd be all for a revamping of the barbican.sh script...as would a legion of others :) 20:21:54 ok, cool 20:21:58 elmiko: I don't use barbican.sh and have my systemd unit start uwsgi directly. 20:22:09 i am also working on something more openstacky for the barbican.cli scripts but that's a ways out 20:22:25 uwsgi has systemd integration, so systemd can handle it directly. 20:22:26 jkf: cool, i would love to talk more about that if you have time 20:22:36 I would be on board for a barbican.sh revamp, as well as dropping uwsgi from the rpm deps 20:22:37 awesome, i hadn't gotten to that point yet 20:22:43 elmiko: I thought we had removed uwsgi from teh barbican rpm? 20:22:50 elmiko: sure thing. 20:23:02 elmiko: at least the barbican-db-manage should be quite straight forward 20:23:13 woodster_: i think so, but for a standalone installation it will still need some sort of wsgi server, uwsgi/gunicorn/etc... 20:23:14 to... openstack....ize 20:23:42 elmiko I think that the Debian folks really want httpd support 20:23:50 so, if i just have a fedora system, and i want to `yum install openstack-barbican`, i need a script to start it 20:23:55 Hey, i have a change about moving the scripts (except barbican.sh) in gerrit into the module 20:24:01 redrobot: so do we, for the RDO/OSP products 20:24:15 Daviey: awesome, link? 20:24:16 https://review.openstack.org/#/c/193208/ 20:24:34 It isn't perfect, but a step in the right direction. 20:24:38 I would definitely +2 a change that s/uwsgi/httpd/ 20:24:51 very cool, i'll take a look Daviey 20:24:54 ta 20:25:12 Daviey: reviewing 20:25:20 Daviey: I think we've just moved to empty __init__ files now 20:25:38 woodster_: I thought all files had to have the header? 20:25:53 Daviey: otherwise this is good stuff. 20:27:15 elmiko anything else on the Fedora front? 20:27:38 Daviey: we've been using blank __init__ of late...but maybe that has changed back to header based now? 20:29:35 * redrobot thinks elmiko got distracted with a review 20:29:39 ok, moving on 20:29:49 any other topics we should talk about? 20:30:00 #topic Open Discussion 20:30:16 sorry, had a bad net thingie 20:30:29 elmiko__ no worries 20:31:08 so now would be a good time to post links to pending reviews that need attention 20:31:13 was gonna say, Daviey that patch looks good to me. my only suggestion would be to make the package named barbican.cli instead of barbican.cmd to be more explicit. otherwise, i'll give it a try locally. 20:32:31 elmiko: ok, i can change that 20:32:43 https://review.openstack.org/#/c/178601/ well, again, I would REALLY like that people check this one out :/ 20:33:04 * redrobot pinky promises to reivew jaosorior 's CR 20:33:06 jaosorior: i added that to my review queue ;) 20:33:21 elmiko__: I think 'cmd' is how I've seen other projects handle this 20:33:57 woodster_: If you have time to check the above CR, you had a +2 there before 20:34:14 woodster_: lol, i've seen cli as well 20:34:21 aaaand this one only needs a workflow :D https://review.openstack.org/#/c/194157/2 20:34:56 woodster_: either way though, it's not a -1 for me. just a stylistic thing 20:35:03 looks like kfarr already reviewed that last one... maybe she can just +W right away? 20:35:14 woodster_, Daviey, elmiko__: Well... keystone has cmd... honestly I don't think that folder namingg is a big deal 20:35:18 My first workflow! 20:35:33 kfarr \o/ 20:35:42 yay! 20:35:45 kfarr: nice! 20:36:05 jaosorior: done 20:36:07 jaosorior: agreed, not a big deal. i was just adding my $0.02 20:36:09 =) 20:36:21 kfarr: http://funny-pics.co/wp-content/uploads/HighFiveLikeABoss.jpg 20:36:43 elmiko__: I still have memories from the incubation process...don't like to stray too far from the openstack fold :) 20:36:53 woodster_: Thanks dude! :D 20:37:16 woodster_: i hear ya, i don't think its a big deal, i just think cli is more clear 20:38:09 * redrobot agrees with elmiko__ 20:38:25 I've got a quick update on the symantec ca plugin. 20:38:38 #topic Symantec CA 20:38:42 jkf go ahead 20:39:31 I'm not sure how the conversation went outside of my pervue, but a scope document its way to the product manager of the trust services group at Symantec. 20:39:54 jkf progress is good! 20:40:16 and based on the work in the scope doc, we're looking to get a developer from that org to do the development of the plugin. 20:40:50 Based on those estimates, I don't have nearly the time to do the work myself on a community basis, but they want to look at doing it as part of the product development. 20:41:12 So they're still trying to identify who can do so, and I've offered my assistance to get them up to speed once they do. 20:41:43 Unfortunately, with my deployment being my top priority, its meant I haven't had a whole lot of time to put into the plugin yet. :( 20:42:25 So, that's where we stand with the ca plugin at the moment. 20:43:32 jkf awesome, thanks for the udpate 20:43:40 update even 20:43:54 anything else y'all? 20:43:57 I proposed a new spec today: https://review.openstack.org/#/c/194298/ Really interested in seeing what people think about it. 20:44:00 if not we can have our 15 min back. 20:44:28 ah yes, I think kfarr and rellerreller will be interested in that one silos1 20:44:55 redrobot: cool 20:44:57 silos1 cool, adding it to my queue 20:45:15 kfarr: :) thx 20:46:20 alrighty... seems like we're done early today. 20:46:24 thanks everyone for coming! 20:46:33 thanks redrobot 20:46:38 Don't forget to add topics to the wiki for next week! 20:46:50 #endmeeting