14:01:29 <jcapitao> #startmeeting RDO meeting - 2020-11-25 14:01:30 <openstack> Meeting started Wed Nov 25 14:01:29 2020 UTC and is due to finish in 60 minutes. The chair is jcapitao. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:01:31 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:01:33 <openstack> The meeting name has been set to 'rdo_meeting___2020_11_25' 14:01:41 <jcapitao> #topic Roll Call 14:02:05 <ykarel> o/ 14:02:32 <jcapitao> #chair ykarel 14:02:34 <openstack> Current chairs: jcapitao ykarel 14:05:59 <amoralej> o/ 14:06:23 <jcapitao> #chair amoralej 14:06:24 <openstack> Current chairs: amoralej jcapitao ykarel 14:08:00 <jcapitao> let's start 14:08:16 <jcapitao> #topic Deps update to match upper-constraints 14:08:25 <jcapitao> spotz: you're around ? 14:08:35 <jcapitao> #link https://review.rdoproject.org/etherpad/p/deps-match-uc 14:09:07 <jcapitao> so there is still work on-going on that topic 14:09:34 <jcapitao> dunno if we need to sync with the fedora cleanup topic 14:09:39 <jcapitao> maybe ? 14:09:49 <amoralej> jcapitao, wrt deps 14:09:58 <jcapitao> https://review.rdoproject.org/etherpad/p/fedora_cleanup 14:10:04 <amoralej> any issue you are hitting that you want to update? 14:10:16 <amoralej> rebuilds from fedora are mostly working? 14:10:56 <jcapitao> I scratch build updates on both Fedora and CBS koji 14:11:28 <jcapitao> generally it's allright 14:11:50 <jcapitao> but in some cases we hit issues because of python macros 14:12:06 <amoralej> what about automatic builddeps? 14:12:27 <amoralej> you are finding many packages with that enabled? 14:12:37 <jcapitao> not as much as i was expecting 14:12:48 <jcapitao> for now 14:12:58 <rdogerrit> Sorin Sbarnea (zbr) proposed rdo-infra/ci-config master: POC: Avoid running molecule from inside pytest https://review.rdoproject.org/r/30803 14:13:06 <jcapitao> but fedora packagers tend to take advantage of it 14:13:13 <amoralej> ok 14:13:30 <jcapitao> so we might hit the issue much more in the near future 14:13:38 <amoralej> let's see if they give us some time 14:13:39 <amoralej> :) 14:15:22 <jcapitao> yes, what would be the solution on our side ? 14:15:40 <amoralej> we'll need to fork in that case 14:16:02 <amoralej> i think backporting that feature is more that we can do from our side 14:16:11 <amoralej> as it needs certain support from rpm, iirc 14:17:35 <jcapitao> indeed.. 14:17:48 <jcapitao> we'll have to fork, sadly :( 14:19:26 <amoralej> yep 14:20:51 <jcapitao> fyi I noticed some zuul jobs on PR :) 14:20:53 <amoralej> wrt fedora cleanup 14:20:55 <jcapitao> https://src.fedoraproject.org/rpms/python-passlib/pull-request/6 14:20:59 <amoralej> oh, good 14:23:31 <amoralej> wrt fedora cleanup 14:23:54 <amoralej> #info some packages have been removed or orphaned from fedora 14:23:58 <amoralej> #link https://review.rdoproject.org/etherpad/p/fedora_cleanup 14:24:24 <amoralej> some are still pending on number80 to take care 14:24:35 <amoralej> as only the main admin can orphan packages 14:25:06 <amoralej> it seems karbor is out from rpm-packaging now 14:25:22 <amoralej> jpena, can we orphan python-abclient from fedora? 14:26:13 <jpena> amoralej: yes, let's move on with that one 14:26:50 <amoralej> chandankumar, https://src.fedoraproject.org/rpms/python-abclient one more package that can be orphaned 14:26:55 <amoralej> it has build in epel7 14:27:01 <amoralej> maybe someone is using it? 14:28:00 <chandankumar> amoralej: need to check if not then I will orphen it 14:28:13 <amoralej> it seems it was added to epel7 since initial build 14:28:18 <amoralej> so probably not 14:28:25 <amoralej> no explicit request, at least 14:29:14 <amoralej> other than that, cleanup is done 14:30:11 <ykarel> some less deps to maintain now, nice work 14:32:26 <jcapitao> nice, do you think we could improve package retirement ? in CI job I mean 14:33:08 <amoralej> you mean to add the README? 14:34:16 <jcapitao> I mean to write a tool which can give a status of dependencies 14:34:57 <amoralej> ah, you mean script that to create report 14:34:58 <jcapitao> wheter a dep is still required somewhere or not 14:35:10 <jcapitao> s/wheter/whether 14:35:33 <jcapitao> amoralej: yes 14:35:56 <amoralej> yes 14:36:03 <amoralej> well 14:36:08 <amoralej> there are two parts for that 14:36:22 <amoralej> what we did here 14:36:32 <amoralej> deps not in rdoinfo and not in fedora 14:36:44 <amoralej> this is the "easy" one 14:36:58 <amoralej> the harder is to find deps in rdoinfo which are not longer required 14:37:38 <chandankumar> amoralej: do we package karbor in rdo? 14:37:42 <amoralej> no 14:37:50 <amoralej> it was packaged in rpm-packaging 14:37:53 <amoralej> but it has been retired 14:38:07 <chandankumar> amoralej: then do we need to maintain it? 14:38:19 <amoralej> i don't think so chandankumar 14:38:26 <amoralej> that's why it's in the list :) 14:38:27 <chandankumar> amoralej: ok then retiring it 14:38:34 <amoralej> you can orphan it 14:39:08 <spotz> jcapitao: here now, I'm PTO for Thanksgiving:) 14:39:28 <chandankumar> amoralej: done, thanks! 14:39:42 <jcapitao> amoralej: ok, we may create a trello card with that topic 14:39:46 <amoralej> ok 14:40:10 <jcapitao> just wanted to know if it's possible technically 14:40:36 <jcapitao> spotz: oh happy Thanksgiving ! 14:41:04 <spotz> Thanks it's tomorrow need to go pick up the Turkey today and I had days to take soo...:) 14:41:15 <ykarel> amoralej, can't repoquery be used to detect unused deps in fedora? 14:41:23 <amoralej> yes 14:42:24 <ykarel> sorry didn't got <amoralej> the harder is to find deps in rdoinfo which are not longer required point 14:42:41 <amoralej> yes 14:42:58 <ykarel> you refering that for not fedora rpms, right? 14:43:04 <amoralej> so, it used to be more difficult to find when a dependency is not longer required in RDO 14:43:18 <amoralej> fedora is not a problem 14:43:22 <ykarel> yeap ok then, yes as deps are used outside of packages 14:43:30 <amoralej> exactly 14:43:32 <ykarel> in tools 14:43:32 <amoralej> and brs 14:43:33 <amoralej> and so on 14:43:36 <ykarel> deployment etcx 14:43:57 <amoralej> chained deps 14:44:35 <ykarel> yes okk 14:44:39 <amoralej> spotz, you want to discuss the topic about moving website and events to fedora or better on next meeting? 14:44:46 <amoralej> i meant to gerrit :) 14:44:51 <amoralej> not fedora 14:45:12 <spotz> amoralej: I'm game for today as I'm here. Thought I'd be out riding:) 14:45:42 <jcapitao> #action jcapitao to create trello card "Build tool to find deps in rdoinfo which are not longer required" 14:46:35 <spotz> I saw earlier in my backscroll talk about moving to a new server, I found it down on Sunday so thanks for whoever got us back up and running 14:46:35 <jcapitao> let's move then 14:46:40 <jcapitao> #topic Move website and events to Gerrit - initial discussion on 11/18 14:47:22 <amoralej> spotz, it was misc i think 14:48:41 <spotz> So we often talk about documentation being a great entry point for contributing but we currently have a totally different tooling and process for updating the website which houses our documentation. We should probably aslo look at moving events and adding them to the list of contributors as well 14:49:57 <amoralej> i'm fine with it 14:50:10 <amoralej> we'd just need to update the contributing doc 14:50:20 <amoralej> and remove (or update) the edit in github 14:50:26 <amoralej> button 14:51:11 <amoralej> also, we should create a ci job for it?, or copy from github 14:52:20 <spotz> Ha deleted my long typing:) With one way to contribute we can flesh out our contributor guide that can hiopefully include not just the RDO processes but the basics of upstream so if someone finds an issue in TripleO they would be able to make a patch there with maybe just a little more reading of their docs 14:52:27 <amoralej> other than that, i'm fine 14:52:55 <amoralej> sounds reasonable 14:53:01 <spotz> If its not much added work we should make it as close to contributing to any other repo 14:53:24 <amoralej> i have the impression that PRs are more widely known that gerrit workflow for non openstack users 14:53:27 <amoralej> but tbh 14:53:48 <amoralej> most of our community will probably come from openstack world 14:53:57 <jcapitao> +1 14:54:09 <spotz> Correct on both parts, and if they're not an OpenStack contributor we want them to become one:) 14:54:16 <amoralej> and in the other side, it may be good way to onboard users to gerrit processes 14:54:18 <amoralej> yes 14:54:20 <amoralej> makes sense 14:54:24 <amoralej> you have my +1 14:55:20 <jcapitao> +1 for me as well 14:55:20 <spotz> I was hoping we'd have the cool stuff store worked out. But we can have a contest to name our duck and then make a user with that name to grab screenshots on how to register for accounts, etc 14:55:39 <spotz> Kinda like I made Stack McStackface upstream 14:56:32 <spotz> And we can have a part that says if its an upstream change they've found send them to the OpenStack contributor pages to create accounts and logins there 14:56:51 <spotz> And how to file a bug, etc 14:57:15 <amoralej> LGTM 14:57:38 <rdogerrit> Sorin Sbarnea (zbr) proposed rdo-infra/ci-config master: Avoid running molecule from inside pytest https://review.rdoproject.org/r/30803 14:58:22 <jcapitao> so do we agree now ? or should we vote next week ? 14:58:33 <spotz> So that was easy to decide and we can just make a plan to do it maybe after the server change. That way if there's a server issue we know our changes weren't involved 14:58:57 <amoralej> yes, also migration should happen soon afaik 14:59:09 <amoralej> let's wait for it 14:59:15 <jcapitao> good 14:59:25 <spotz> jcapitao: Voting next week sounds good and if we need to poll folks outside this group we'll have time to let them know 14:59:57 <jcapitao> spotz: ok 15:00:02 <jcapitao> let's move on then 15:00:06 <jcapitao> 1 min left 15:00:20 <jcapitao> #topic Next Weeks Chair 15:01:13 <spotz> And think how much easier it'll be to update the Meetings page:) 15:01:31 <amoralej> :) 15:01:46 <jcapitao> any volunteers ? 15:01:48 <amoralej> spotz, next meeting will be with video call? 15:01:59 <jcapitao> oh right 15:02:08 <spotz> amoralej: video with us typing some notes at the least in channel for logging 15:02:14 <amoralej> yes 15:02:17 <amoralej> exactly 15:02:36 <jcapitao> ok 15:02:44 * zbr wonders when we will get the new gerrit on review.rdoproject.org 15:02:53 <spotz> Why don't I plan on leading and maybe we'll need to designate a typer if we find out after this meeting we're not typing enough 15:03:37 <amoralej> spotz, yes, let's see how it goes 15:03:44 <spotz> zbr I'd think we'd make sure all the kinks are ironed out upstream then look into it so we can ask questions? 15:04:39 <jcapitao> #action next meeting will be on video call 15:04:58 <spotz> https://meet.google.com/uzo-tfkt-top 15:05:19 <jcapitao> #link https://meet.google.com/uzo-tfkt-top 15:05:24 <jcapitao> thanks spotz 15:05:27 <zbr> upstream upgrade work took a lot of time and the current performance issue would probably not apply to rdo deployment 15:05:31 <jcapitao> no topic open floor today 15:05:38 <jcapitao> #endmeeting