19:04:31 #startmeeting Ansible Community Meeting 19:04:31 Meeting started Wed Feb 22 19:04:31 2023 UTC. 19:04:31 This meeting is logged and archived in a public location. 19:04:31 The chair is andersson007___. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions. 19:04:31 Useful Commands: #action #agreed #halp #info #idea #link #topic. 19:04:31 The meeting name has been set to 'ansible_community_meeting' 19:04:40 I'm not up to leading today 19:05:01 gotmax23: no problem 19:05:13 #topic Agenda https://github.com/ansible/community/issues/679 19:05:16 .hi 19:05:19 gotmax23: gotmax23 'Maxwell G' 19:05:21 hi 19:05:33 acozine, andersson007_, anwesha, baptistemm, bcoca, briantist, cidrblock, cyberpear, cybette, dericcrago, dmsimard, felixfontein, geerlingguy, gotmax, gundalow, gwmngilfen, ikhan_, jillr, jtanner, lmodemal, mariolenz[m], markuman, maxamillion, misc, nitzmahone, oranod, resmo, russoz, samccann, thaumos, zbr: The Ansible community meeting is starting now! 19:05:48 * gotmax23 has a new email btw 19:06:03 #info Agenda: https://github.com/ansible/community/issues/679 / Topics: https://github.com/ansible-community/community-topics 19:06:10 .hello2 19:06:11 maxamillion: maxamillion 'Adam Miller' 19:06:13 * acozine is half-here 19:06:17 hello hello 19:06:25 I"ll be afk for a few minutes, but I'll jump on soon 19:06:25 gotmax23: good to know, if you share it with me for SC reasons, it'd be great 19:06:47 It's the email that zodbot gave 19:06:47 #topic Updates 19:06:56 o/ 19:06:59 * zodbot gave the new one 19:07:21 ah, ok 19:07:50 any updates? 19:08:33 * gotmax23 has been busy with work and other projects 19:08:45 I had some thoughts about the collection removal process 19:08:59 #chair gotmax23 acozine samccann maxamillion oranod 19:08:59 Current chairs: acozine andersson007___ gotmax23 maxamillion oranod samccann 19:09:13 Also, I'd like to restart the discussion about removing unnecessary files from the ansible distribution once Felix is back 19:10:05 it doesn't look like anyone has any updates 19:10:15 so gotmax23 feel free to set the next topic 19:10:44 mariolenz[m]: are you around? 19:10:44 #topic Collection Removal Process 19:11:28 Yeah, I was actually going to bring up a discussion I had with him about this :) 19:12:09 What I wrote last week about the google.cloud vote: 19:12:22 > I guess I'll vote +1. I deliberately abstained from the cyberark.pas one. I don't like that the only way to get some collections to fix issues is to threaten them with removal. 19:12:31 > It's a lot of paperwork (filing a community-topics issue, filing a collection issue, starting a vote, filing an ansible-build-data issue, adding a deprecation changelog entry, etc.) if we're just going to revert it when the collection decides to get their act together. 19:13:00 > I guess I don't get to complain, because it's usually you doing those things, but I'm a bit conflicted about the whole thing :) 19:13:01 (Ultimately, I abstained) 19:13:02 everything sounds fair 19:13:51 yep, it's not good that threatening is the only thing that work but better than nothing? 19:14:00 also folks it can be related https://github.com/ansible-community/community-topics/issues/198 19:14:13 I'm happy that we have a tool to cleanup the package, but perhaps we can make it more efficient 19:14:14 it's another topic but a kinda related 19:14:52 Would you mind a bit of a tangent about how we do this in Fedora? It might inspire us :)? 19:15:10 @cybette:ansible.im cyb-clock chimes every 15 minutes during the community meeting 19:15:15 yep, sure 19:16:05 tangents can be fun (and educational) 19:16:13 * samccann puts on her virtual listening ears 19:16:23 anyway please don't forget to file a topic:) 19:17:51 * cyberpear arrives late 19:18:12 #chair cyberpear 19:18:12 Current chairs: acozine andersson007___ cyberpear gotmax23 maxamillion oranod samccann 19:19:19 meh, now that I think about it more it's completely irrelevant, but: packages that fail to build after two releases are retired. one the failure to build is identified a bug is filled against the package. six weeks before the planned retirement, there starts weekly announcements about the imminent retirement. then, the packages are mass retired. anyone can unretire the package within eight weeks. after that, a re-review is required. 19:20:12 if I'm counting correctly, maintainers have ~a year to fix these bugs 19:21:07 what unretiring includes? 19:21:25 just to fix the bugs? 19:21:28 there's also an orphaning process. package maintainers can orphan a package that they no longer have time to maintain. there's ~weekly announcements. if nobody picks it up after ~6 weeks, it's also retired from the distribution. 19:21:40 there are interesting points in that tho... about fixed deadlines and fixed points at which the owner has to react or it's out. We could also include something that if a collection manages to fix things before getting the boot, they are also on 'probation' for the next couple of releases.. in case they swoop in, fix, and then ignore bugs for another year or so etc 19:21:55 andersson007___: they have to file a ticket with release engieneerig 19:22:16 *engineering 19:23:06 there's a couple other processes that ultimately result in retirement. depending on the offense, the package maintainers have different amounts of time to fix it before it gets retired 19:23:55 +1 to modeling after the Fedora process 19:24:21 (I'm admittedly biased though) 19:24:41 lets file the topic? i'm happy to consider any options 19:25:16 there's a lot more packages in Fedora, but in my opinion, the main takeaways that we could apply are that things are announced all at once and the process is more automated 19:26:09 yes this sounds like a good thing to discuss in a topic and come to consensus. Some great ideas here 19:26:39 +1 19:28:10 we could use passing sanity tests btw as one of criteria 19:28:21 gotmax23: would you like to create the topic? 19:29:43 * gotmax23 is currently writing the issue 19:29:59 https://github.com/ansible-community/community-topics/issues/200 19:30:00 I'm game to move on to andersson007___ 's topic if we want to continue this on the issue 19:30:01 @cybette:ansible.im cyb-clock chimes every 15 minutes during the community meeting 19:30:03 in all fairness, the Fedora process also ends up being a kludge to get people to fix issues, but at the end of the day, I'm happy that we're trying to identify which collections are actually dead and doing something about it. I just want to try to simplify the process. 19:30:34 gotmax23: SGTM, thanks! 19:30:55 #topic "Community-approved collections" instead of file inclusion in the Ansible package. 19:31:03 #link https://github.com/ansible-community/community-topics/issues/198 19:31:04 #info Dear community please take a look at the collection community-approval status instead of inclusion topic https://github.com/ansible-community/community-topics/issues/198. Any feedback will be much appreciated! 19:31:13 I guess this is the next topic to discuss :)? 19:31:22 I'm not sure if it's worth discussing it today 19:31:48 * gotmax23 realizes he missed last discussion 19:31:50 Open floor I guess? 19:31:58 we're still collecting feedback there 19:32:18 does anyone have any particular topic to discuss? 19:32:20 * gotmax23 nods 19:35:17 does anyone remember how to start openfloor?:)) 19:35:40 #topic Open Floor 19:35:41 that :)? 19:35:52 yep, thanks gotmax23 :) 19:36:02 wasnt' sure 19:37:04 feeling dumb now , usually not so as today evening 19:37:24 must be a long day (and night) for ya! 19:37:39 another reason not to start that collection approval topic:) 19:37:41 yeah 19:37:49 * gotmax23 is also jumbled and it's not even 14:00 here 19:38:16 hehe, tough 19:40:13 before new year started to take the first caffeine in 90-120 minutes after awakening. Heard it as an advice on how to avoid afternoon crash. 19:40:49 i think it doesn't solve the issue entirely but now it's better:) 19:40:52 in my case 19:42:08 ok, i'll close the meeting in a minute if there are no topics 19:42:25 👍️ 19:43:04 i meant to say in 90-120 minutes instead of immediate:) 19:43:14 it's important:)) 19:43:42 oh I don't know. caffeine first thing is so good though. 19:44:04 thanks andersson007___ and all 19:44:08 the wisdom is that adenosin should have time to attach to neurons 19:44:23 #endmeeting