18:00:40 #startmeeting Infrastructure (2017-10-19) 18:00:40 Meeting started Thu Oct 19 18:00:40 2017 UTC. The chair is smooge. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:40 Useful Commands: #action #agreed #halp #info #idea #link #topic. 18:00:40 The meeting name has been set to 'infrastructure_(2017-10-19)' 18:00:40 #meetingname infrastructure 18:00:40 The meeting name has been set to 'infrastructure' 18:00:40 #topic aloha 18:00:40 #chair smooge relrod nirik dgilmore threebean pingou puiterwijk pbrobinson maxamillion 18:00:40 Current chairs: dgilmore maxamillion nirik pbrobinson pingou puiterwijk relrod smooge threebean 18:00:45 hi 18:00:48 * relrod waves 18:01:11 .hello2 18:01:12 nb: nb 'Nick Bebout' 18:01:33 .hello2 18:01:33 bowlofeggs: bowlofeggs 'Randy Barlow' 18:01:44 hello. There is currently a go/no-go meeting in #fedora-meeting-1 and a releng meeting going on in #fedora-meeting-2 18:01:49 morning 18:01:55 so I will keep this short 18:02:08 #topic New folks introductions 18:02:08 #info This is a place where people who are interested in Fedora Infrastructure can introduce themselves 18:02:18 Hi are there any new people looking to become apprentices? 18:03:21 ok well then off to the races 18:03:23 #topic announcements and information 18:03:23 #info PHX2 Colo Trip, Dec 4th - 9th 18:03:24 #info Infrastructure will be down during that trip. 18:03:24 #info Final freeze has started. PLEASE TEST ANY RCs. 18:03:24 #info EU daylight savings Oct 29/ US daylight savings Nov 5 18:03:25 #info bodhi-3.0.0 is going to be released "soon", and an FBR will be requested. This converts Bodhi to use pungi for "punging" (instead of "mashing") and enables modular updates 18:04:01 I wanted to bring up the DST as it always surprises everyone. 18:04:10 smooge: yeah, thanks fro that 18:04:19 In fact I can't remember if this is a "we stay at the same UTC or move to a different UTC" 18:04:36 I believe we stay.. 18:04:52 October is almost over. 18:05:15 Any other announcements? 18:06:17 * nirik hates DST 18:06:51 #topic Ticket cleanup 18:06:52 #info https://pagure.io/fedora-infrastructure/issue/5478 18:06:52 #info https://pagure.io/fedora-infrastructure/issue/4196 18:06:52 #info https://pagure.io/fedora-infrastructure/issue/5362 18:07:10 OK I randomly picked (ok I went to page 2 and rolled a dice) 18:07:36 some tickets. The 5478 looked like it was needing abompard to fix something so I assigned it 18:08:05 and I can't get to pagure at the moment :/ 18:08:16 same here 18:08:53 yeah, let me poke it 18:08:57 4196 is about autotweeting in ask. I think this is a "closed wont fix" item 18:09:10 but figured it should be brought up 18:09:40 it's still a bug/issue. 18:09:43 5362 is darkserver not working. Which needs to be fixed, and someone needs to take over maintenance now that the primary person is busy with other challenges probably 18:09:50 we could close it saying we won't fix it indeed. 18:10:04 I asked in the darkserver RFR about that, but no answer yet 18:10:04 Yeah, agreed on tweeting stuff 18:11:03 The darkserver I wasn't sure if it was still something that was being worked on so I figured it should be brought up 18:11:35 * bowlofeggs still plans to make twitterfs one day 18:12:00 before or after twitter goes away 18:12:07 I'd say give him a bit of time to answer, then if not, look at retiring the service and waiting to re-implement it. 18:12:22 ok cool on that 18:12:31 were there any other old tickets anyone wanted to look at? 18:12:37 I see we are down to 50 open ones 18:12:48 sadly no. 18:13:00 108 18:14:01 I was kinda starting from the oldest ones... but any are good to discuss. 18:14:24 107 18:15:14 https://pagure.io/fedora-infrastructure/issue/3654 is another ask one we will never fix and can probibly be closed -> upstream... 18:15:40 ok I can do so later after pagure is not pounded 18:16:17 #topic Apprentice Open office hours 18:17:08 Any questions from people today? I expect we need more easy fixes versus other ones 18:18:01 #topic Open Floor 18:18:11 ok anyone have items for the floor? 18:18:17 I'll note (because I forgot to earlier): 18:18:28 right, I wanted to show some demo of copr+src.fp.o CI 18:18:31 so... 18:18:36 I moved the staging fedmsg irc stream to it's own new channel... #fedora-fedmsg-stg 18:18:45 and out of #fedora-fedmsg 18:18:51 clime: oh, cool. Go for it. 18:18:56 thx 18:18:56 #topic clime on copr 18:19:01 haha 18:19:09 I will open a PR here: https://src.stg.fedoraproject.org/rpms/python-copr 18:19:37 it should create a project and launch a build of the PR here: http://copr-fe-dev.cloud.fedoraproject.org/coprs/ 18:19:59 done! 18:20:14 (maybe you can't access the second url because it's http) 18:20:22 (so chromium will work) 18:20:53 Builds are running in the clime/rpms-python-copr 18:21:15 and ppc64le are failing...but ppc64le is not quite working today 18:21:36 This might be of interest: http://copr-be-dev.cloud.fedoraproject.org/results/clime/rpms-python-copr/srpm-builds/00276866/builder-live.log 18:21:44 it's a log about the SRPM generation process 18:22:10 nice. 18:22:20 and the script responsible for catching the femsgs and launching the build is here: https://pagure.io/copr/copr/blob/d84a92d6d813105e22d7c463aa48e984eedc3780/f/frontend/coprs_frontend/run/src-fp-stg-ci.py 18:22:49 It's not yet reporting back the results or anything. I will need to figure that part out if this will show some potential. 18:22:53 So that's it. 18:23:43 yeah, hopefully it wouldn't be too hard to plug into pagure 18:24:18 where do these get built? 18:24:34 on copr builders. 18:25:08 build is launched from copr-frontend locally by invoking some copr python logic and the rest is handled 18:25:16 ...as usual 18:25:19 ah ok 18:25:24 thanks 18:25:34 any questions for clime? 18:26:21 I guess those projects stay around forever after that? 18:26:31 yup, if not manually deleted. 18:26:58 ooof 18:27:15 might be nice to purge X amount of time after PR is merged... 18:27:24 or closed. 18:27:38 is closed the same as rejected? 18:27:44 right. 18:27:57 Actually, this is asking for a dedicated COPR instance :) 18:28:29 because there might be also conficts in project names in what user creates manually and what is auto-created by PR at src.fp.o 18:28:31 perhaps yeah, that acts somewhat differently. 18:28:36 yup 18:29:03 but we can look at that after the new cloud future. 18:29:24 okay 18:29:57 This is also similar to pingou's one for koji, but copr does provide some more flexability. 18:30:22 Yup, the new SCM method should be pretty powerful. 18:30:31 my main worry is how much disk space this will need and where will it go 18:30:44 yep. Always the worry. 18:31:13 Well, with a dedicated instance, any purging might be in place. 18:31:35 so, it should be possible to adjust to the actual needs. 18:32:01 sure, worth pondering 18:32:33 the dedicated instance needs storage also. do you know what kind of hardware is needed 18:32:44 but that is outside this meeting I think. 18:33:44 yes, I think thats down the road some more and outside. 18:34:26 By the way, we also plan to integrate copr with Taskotron. 18:34:50 So automatic tests can be launched and be accessible from the build. 18:35:45 I think that's all I've got. 18:36:33 ok thanks 18:36:39 #topic Open Floor 18:36:49 ok anything for the open floor? 18:38:04 Oh, one reminder... 18:38:35 I am going to actually be gone next wed -> sunday. Might appear for meetings thursday morning, but otherwise I will be not around. ;) bug smooge and relrod and puiterwijk instead 18:39:04 #info nirik is going on his annual halloween wizard ride 18:39:11 ok that seems to be it 18:39:15 #endmeeting