15:10:43 #startmeeting RDO meeting - 2016-12-21 15:10:43 Meeting started Wed Dec 21 15:10:43 2016 UTC. The chair is amoralej. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:10:43 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:10:43 The meeting name has been set to 'rdo_meeting_-_2016-12-21' 15:10:44 Meeting started Wed Dec 21 15:10:43 2016 UTC and is due to finish in 60 minutes. The chair is amoralej. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:10:45 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:10:48 The meeting name has been set to 'rdo_meeting___2016_12_21' 15:10:53 sorry for the delay 15:11:00 \o 15:11:03 #topic roll call 15:11:09 o/ 15:11:18 Oh, I thought we had canceled for today. Oops. 15:11:19 o/ 15:11:21 o/ 15:11:32 #chair dmsimard jpena rbowen number80 15:11:32 Current chairs: amoralej dmsimard jpena number80 rbowen 15:11:33 Current chairs: amoralej dmsimard jpena number80 rbowen 15:11:51 let's start with first topic 15:12:00 #topic New package review process clarifications 15:12:09 who added that? 15:12:13 I did it 15:12:26 go for it jpena 15:12:32 #info https://www.rdoproject.org/documentation/rdo-packaging/#how-to-add-a-new-package-to-rdo-trunk 15:12:45 so in https://www.rdoproject.org/documentation/rdo-packaging/#how-to-add-a-new-package-to-rdo-trunk we have a process, where we start by opening a bz, and then open the rdoinfo review and the rest 15:12:47 #info outline identifying missing docs https://review.rdoproject.org/etherpad/p/RDO-Documentation 15:12:56 o/ 15:13:03 #info vitrage packaging https://bugzilla.redhat.com/show_bug.cgi?id=1342987 what need to be done to move it forward (eyalb) 15:13:27 if I understood it correctly during some discussions with apevec, we agreed that the rdoinfo / initial spec review could be done in gerrit without the need to approve the bz 15:14:03 however, number80 disagrees, so I'd love to get some clarification (and update the docs with the result) 15:14:13 i usually don't create the repos until bz is approved 15:14:21 or very close to approved 15:15:05 so, at the time of merging first rdoinfo review, package review bz is approved, that was my understanding 15:15:14 I'm not aware of the legal ramifications, but being able to have the spec tested by DLRN and reviewed in Gerrit are really appealing 15:15:32 they simplify the review process a lot 15:16:03 so we could approve the bz once the spec review is done, and always before setting the tags in rdoinfo 15:16:05 yes, but if the approval process is in bz, i think it make sense that the discussion is in bz also, no in gerrit 15:16:38 in that case we would have to reference the gerrit review from bugzilla 15:16:39 \o/ 15:16:47 #chair trown chandankumar 15:16:48 Current chairs: amoralej chandankumar dmsimard jpena number80 rbowen trown 15:17:03 0/ 15:17:15 #chair jschlueter 15:17:15 Current chairs: amoralej chandankumar dmsimard jpena jschlueter number80 rbowen trown 15:17:16 amoralej, the image is this one http://artifacts.ci.centos.org/artifacts/rdo/images/master/delorean/current-tripleo/stable/overcloud-full.tar which I see was produced yesterday 15:17:28 The missing docs doc is really helpful. Should we go ahead and start creating stub docs for any of those? 15:17:30 amoralej: we always have a final round with the approved spec and srpm in the bz, so we can run fedora-review on it 15:17:40 * apevec is on different meeting... 15:18:04 jpena, where is that disagreement and about which part? 15:18:34 apevec: it's whether we should wait for rdo-approve+ in bugzilla before creating the project in rdoinfo 15:18:37 If we could create stub docs, and clearly tag them in a "to be written" kind of way, it would make it easy to script some process of asking folks to help with the writing. 15:18:53 o/ 15:18:54 (Sorry, I'm being pulled several different ways right now.) 15:19:07 #chair weshay 15:19:07 Current chairs: amoralej chandankumar dmsimard jpena jschlueter number80 rbowen trown weshay 15:19:29 jpena, if we block on that then we cannot use gerrit for initial review, that was the point 15:19:31 rbowen: I think amoralej mixed two topics in one, we should cover missing docs as the next topic 15:19:42 we can block on + before removing under-review tag 15:19:45 * trown is confused 15:20:01 #undo 15:20:01 Removing item from minutes: INFO by amoralej at 15:13:03 : vitrage packaging https://bugzilla.redhat.com/show_bug.cgi?id=1342987 what need to be done to move it forward (eyalb) 15:20:11 having the review up and being able to use normal git review -d is really useful when reviewing new spec 15:20:12 #undo 15:20:12 Removing item from minutes: INFO by amoralej at 15:12:47 : outline identifying missing docs https://review.rdoproject.org/etherpad/p/RDO-Documentation 15:20:18 apevec: that's what I understood as well, but it's not documented and not all of us seem to be on the same page 15:20:21 jpena: Oh, ok. I was confused. 15:20:35 my fault rbowen, sorry 15:20:59 how do gerrit reviews for new package get created? 15:21:42 users create the reviews against rdoinfo project trown 15:22:08 trown: first, we create a review for rdoinfo, tagging the package as "under-review" so it doesn't get built by DLRN. Someone (us) creates the project in review.rdo, then approves the rdoinfo change. And then, we can open an initial spec review using the distgit project 15:22:56 so, iiuc, we can do the spec discussion in gerrit and once we have something that can be approved, we add it to the bz and do formal review in bz 15:22:58 is it? 15:23:23 if we all agree, I think that's the best approach 15:23:33 but we're missing number80's opinion 15:23:50 I think this is not at all clear... I got asked to review https://bugzilla.redhat.com/show_bug.cgi?id=1373821 by radez yesterday... seems like there is missing step of rdoinfo review? 15:23:55 no problem from my side, just adding a reference to the gerrit review in the bz 15:24:04 would be convenient 15:24:22 ya, I am +1 to doing reviews in gerrit... bugzilla is an awful review tool 15:24:24 trown: that's a Fedora package review 15:24:32 it follows the fedora process 15:24:37 k 15:25:03 jpena: but it has RDO-OCATA in "Blocks" 15:25:09 * jpena feels like a bureaucrat, all the time talking about "the process" 15:25:16 number80 ^ wdyt? 15:25:35 trown: it's a requirement for another package (congress, I think). So once it's part of Fedora we'll rebuild it in CBS 15:26:00 ah sorry 15:26:09 jpena: cool, it is deps not dlrn... sorry for the noise 15:26:33 actually, Dan needs to create a new ticket 15:26:44 reviewee is creator of the ticket, assignee is the reviewer 15:27:39 number80: k, I think I muddied the actual discussion here 15:28:05 amoralej, you still tjere? 15:28:05 that being whether for packages we will track with dlrn, if we can do package review in gerrit 15:28:14 main question is doing initial spec review on gerrit before having approved the bz number80 15:28:18 rasca: it is rdo meeting 15:28:19 we are in a meeting rasca 15:28:25 i'll back to you later 15:30:47 i'm not sure if i'm explaining it right, number80, sorry 15:30:48 amoralej, that's not a question, we really want/need that :) 15:31:00 reviewing in bz comments is just soo bad 15:31:04 ok, then let's put the aggrement 15:31:33 what if I propose a change to the packaging page, and we discuss the details there? 15:31:44 ack 15:31:44 #agreed to do initial spec review in review.r.o before having the package review bz approved 15:31:47 I mean https://www.rdoproject.org/documentation/rdo-packaging/#how-to-add-a-new-package-to-rdo-trunk 15:31:57 send docs PR and we review it there 15:32:12 #action jpena to propose a change in the process in the packaging page to discuss in the PR 15:32:17 ok, let's move on 15:32:32 #topic outline identifying missing docs https://review.rdoproject.org/etherpad/p/RDO-Documentation 15:32:45 rbowen, now it's for you 15:33:04 I believe jpena made that document. I'm just an avid supporter of it. 15:33:05 that outline was actually from jpena 15:33:25 I just pasted as support for the previous topic 15:33:36 Identifying docs that need to be written is a huge help in encouraging people to write them. 15:33:59 rbowen: if we could get stubs in the web, it would be great and we could start filling the gaps 15:34:25 Ok, I'll try to figure out where these would fit in the site structure, and start creating those docs. 15:35:12 some of that info is distributed among blogs, etc... having it in a more structured way will help 15:35:33 specially for people who don't work 100% of their time with RDO :) 15:36:00 #action rbowen to start creating doc stubs according to structure in https://review.rdoproject.org/etherpad/p/RDO-Documentation 15:36:07 +1 15:36:33 #topic vitrage packaging https://bugzilla.redhat.com/show_bug.cgi?id=1342987 what need to be done to move it forward (eyalb) 15:36:47 thats me 15:36:50 #info feedback from mrunge in https://bugzilla.redhat.com/show_bug.cgi?id=1342987#c17 also see c15 15:36:51 so that's kind of related to the first topic 15:37:07 eyalb, have you seen reply from mrunge in c17 ? 15:37:24 no 15:37:49 but discussed in the 1. topic we can iterate on spec changes in the gerrit review 15:37:57 I mean I didnt understand what need to be done 15:37:59 it will be more convenient then in bz comment 15:38:37 o/ 15:38:38 mrunge, ^ can you help explaining your c17 in vitrage bz ? 15:39:06 looks like master is already good? 15:39:36 also the vitrage-dashbord is ready 15:39:41 so I'd say let's proceed with project creation and do initial spec review in review.rdo 15:39:58 eyalb, which bz# is that? 15:40:37 https://bugzilla.redhat.com/show_bug.cgi?id=1390608 15:41:11 it's strange that https://review.rdoproject.org/r/#/c/1504/ is merged but repos are not created 15:41:50 amoralej, yeah, we don't have automation for project creation yet 15:42:21 jpena, ^ let's proceed with project creation 15:42:47 i can take care of creating the proyect for vitrage so eyalb can send an initial spec, ok? 15:42:49 eyalb, once that's done, we can create initial review in gerrit 15:42:54 amoralej, ack 15:43:12 and for vitrage-dashboard we'll need a new review to rdoinfo 15:43:26 there is a link to the spec in the bz 15:43:45 #action amoralej to create vitrage project 15:44:07 eyalb, ok, I'll provide steps how to create initial spec review in gerrit 15:44:16 ...and we'll add that to the docs 15:44:24 as discussed earlier 15:44:28 I will add a vitrage-dashbord to the rdoinfo 15:44:43 ok 15:45:05 #action eyalb to send a review to create vitrage-dashboard project 15:45:20 i think plan is more clear now for vitrage? 15:45:49 vitrageclient is already done iirc ? 15:45:54 eyyes 15:45:55 yes 15:46:51 #topic open floor 15:47:04 any other topic you'd like to bring? 15:47:34 Quick one: the DLRN API review has been updated, covering the latest comments. If you have some time, https://review.rdoproject.org/r/3838 15:48:34 apevec, would it be useful to do some session about all issues related to 7.3 release with CentOS team? 15:48:47 so we can improve on next release 15:51:03 amoralej, yeah, not sure which venue would work, maybe CBS meeting on Monday? 15:51:08 rbowen, number80 ^ wdyt 15:51:30 I guess next Monday is probably cancelled 15:51:36 Possibly, however, a LOT of people are out for the next few weeks. 15:51:50 yeah, so early Jan then 15:51:59 Jan 2 is also out 15:52:00 Would make Monday 9th the next likely date. 15:52:05 yep 15:52:21 amoralej, add it on agenda, not sure where that is 15:52:29 ok, i'll look for it 15:52:39 btw, when are we having the next meeting? 15:52:52 January 4 or 11? 15:52:57 good question 15:52:59 https://www.centos.org/community/calendar/#community-buildsystem-infrastructure-meeting doesn't link to an agenda ... 15:53:08 I'll be out on Jan 4, but I expect some people will be here. 15:53:16 i'm on pto on 4th 15:53:38 let's keep Jan 4, I'll take chairing 15:53:50 to keep the lights on :) 15:54:34 that solves the standing last topic 15:54:38 #info meeting on Dec 28 is canceled 15:54:55 #info apevec will chair next meeting on Jan 4th 15:55:06 #info happy holidays everyone! 15:55:36 Thanks! You too! 15:55:45 happy xmas to rdo-ers!!! 15:56:09 so i think we can end the meeting if there is no more topics 15:56:15 3 15:56:16 2 15:56:17 1 15:56:21 bye 15:56:21 #endmeeting