18:00:15 <felixfontein> #startmeeting Ansible Community Meeting
18:00:15 <zodbot> Meeting started Wed Nov  3 18:00:15 2021 UTC.
18:00:15 <zodbot> This meeting is logged and archived in a public location.
18:00:15 <zodbot> The chair is felixfontein. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
18:00:15 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
18:00:15 <zodbot> The meeting name has been set to 'ansible_community_meeting'
18:00:15 <felixfontein> #topic Agenda https://github.com/ansible/community/issues/539
18:00:15 <felixfontein> acozine andersson007_ baptistemm bcoca briantist cyberpear cybette dericcrago dmsimard felixfontein geerlingguy gundalow gwmngilfen ikhan_ jillr jtanner lmodemal misc nitzmahone resmo samccann tadeboro cidrblock thaumos zbr: ping!
18:00:19 <felixfontein> #info Agenda: https://github.com/ansible/community/issues/539 / Topics: https://github.com/ansible-community/community-topics
18:00:22 <felixfontein> #topic Updates
18:00:31 <tadeboro> o/
18:00:33 * dericcrago waves
18:00:33 <dmsimard> o/
18:00:37 <felixfontein> #chair tadeboro dericcrago dmsimard
18:00:37 <zodbot> Current chairs: dericcrago dmsimard felixfontein tadeboro
18:00:40 <cidrblock[m]> alo all!
18:01:06 <samccann> o/
18:01:12 <briantist> o/
18:01:13 <felixfontein> #chair cidrblock[m] samccann
18:01:13 <zodbot> Current chairs: cidrblock[m] dericcrago dmsimard felixfontein samccann tadeboro
18:01:15 <felixfontein> #chair briantist
18:01:15 <zodbot> Current chairs: briantist cidrblock[m] dericcrago dmsimard felixfontein samccann tadeboro
18:01:18 <cidrblock[m]> o/
18:01:27 <dmsimard> #info ansible 2.9 will end upstream support on May 23, 2022, for more information: https://groups.google.com/g/ansible-announce/c/kegIH5_okmg/
18:02:33 <cidrblock[m]> sivel: TY for putting that together & making the rounds with it
18:02:57 <sivel> np, I got a lot of help with the editing and final version :)
18:04:07 <felixfontein> I'm in the middle of cooking and didn't have time to collect some topics for today, so if anyone has a suggestion on which topics we could discuss... :)
18:04:08 <dmsimard> #info You will now find 'newsbot' in #ansible-social (or #social:ansible.com on matrix) and we are testing it as a potential replacement to the bullhorn issue (https://github.com/ansible/community/issues/546). You can mention the bot and it will record your news.
18:04:42 <dmsimard> I have another update
18:05:59 <dmsimard> #info The Ansible devtools team is seeking feedback towards establishing a regular meeting to discuss topics, please share your thoughts in https://github.com/ansible/community/issues/633
18:06:12 * acozine waves
18:06:13 <acozine> sorry I'm late
18:06:39 <dmsimard> #info Ansible 4.8.0 has been released: https://groups.google.com/g/ansible-announce/c/at_Cjdu-gh4
18:06:55 <dmsimard> #info Ansible 5.0.0a3 is available for testing: https://groups.google.com/g/ansible-devel/c/gm4oSevkabo
18:07:03 <dmsimard> no more updates from me :p
18:07:29 <felixfontein> #chair acozine
18:07:29 <zodbot> Current chairs: acozine briantist cidrblock[m] dericcrago dmsimard felixfontein samccann tadeboro
18:10:24 <felixfontein> so... any suggestions for a topic?
18:11:01 <dmsimard> felixfontein: does the 2.9 and 2.10 EOL announcement help clarify "Community collections: when to drop support for Ansible 2.9 and ansible-base 2.10?" ?
18:11:13 <cidrblock[m]> while it's top of mind, do we have a calendar to post vacation time? I know a few of us are planning for the holidays
18:11:14 <felixfontein> (there are quite few people around today)
18:11:31 <felixfontein> dmsimard: I think it does
18:11:36 <felixfontein> at least to me :)
18:11:42 <felixfontein> should we talk about that?
18:11:45 <cidrblock[m]> (for meetings that may be cancelled, or meetings we can't attend due to pto, etc)
18:12:08 <felixfontein> cidrblock[m]: I don't think we have anything like that
18:12:09 <dmsimard> cidrblock[m]: we don't have a public facing PTO calendar that I am aware of
18:13:16 <dmsimard> felixfontein: re: topic, I guess we can now set a date so that answers the "when" -- do we need to discuss the "how" ?
18:15:01 <remindbot[m]> @cybette:ansible.im cyb-clock chimes every 15 minutes during the community meeting
18:15:50 <dmsimard> as in: is it just about no longer testing against ansible 2.9 ? or are there other things to think about
18:15:58 <felixfontein> #topic Community collections: when to drop support for Ansible 2.9 and ansible-base 2.10?
18:16:04 <felixfontein> #info Discussion: https://github.com/ansible-community/community-topics/issues/50
18:16:24 <felixfontein> dmsimard: for c.g, there are all the symlinks that are needed only for Ansible 2.9 that we could remove
18:16:57 <felixfontein> basically I would suggest to actively drop support by also increasing the `requires_ansible` version to `>=2.11.0`
18:17:29 <felixfontein> IMO we shouldn't claim we still support something if we don't test it
18:17:51 <dmsimard> yeah that's fair enough
18:18:39 <felixfontein> I guess every collection can decide by itself when to drop support, some won't have major releases in that timeframe anyway so they can't drop support
18:19:50 <felixfontein> I'm not sure what's the current community.network release cycle, but dropping all the symlinks in there is probably also a good thing, especially since symlinks end up as duplicates thanks to setuptools/pip
18:20:15 <dmsimard> I suppose we will also need to update/refresh bits in the collection inclusion requirements doc
18:21:17 <tadeboro> What does `requires_ansible` actually affect? Collection installation? Runtime execution? ...
18:21:31 <sivel> right now just a runtime warning
18:21:42 <felixfontein> dmsimard: what does needs to be updated? I'm not aware of anything
18:21:47 <dmsimard> tadeboro: the one thing I know is that if it's not defined the galaxy importer fails
18:21:49 <sivel> eventually we hope the galaxy api and ansible-galaxy cli can use it to filter the correct versions
18:23:31 <dmsimard> felixfontein: I ctrl+f'd 2.9 in https://github.com/ansible-collections/overview/blob/main/collection_requirements.rst and you're right, the mentions of it are not particularly problematic
18:24:25 <felixfontein> dmsimard: basically collections must test against what they support, and they only have to support the latest stable-2.x branch (i.e. the one used in the Ansible version including it) if they want to get included in ansible
18:25:00 <tadeboro> OK, if warning is all users get right now, I do not mind bumping the requirement just to make it clear to the users they are runnign an unsupported combination of components and that they are on their own when it comes to support. Preventing execution would make things a bit  more complex.
18:25:29 <felixfontein> in fact c.g will stop working for Ansible 2.9 once we remove the symlinks. at least if you try a module or action plugin :)
18:25:50 <felixfontein> for most other collections, it should make no difference (assuming they don't add some other intentional breakage)
18:25:56 <dmsimard> felixfontein: right, my line of thinking was around "what burden (in terms of requirements) can we lift as a result of 2.9 going EOL
18:26:34 <dmsimard> it's two less versions of ansible to test against, maybe versions of python, etc
18:27:16 <felixfontein> no version of Python :)
18:27:34 <felixfontein> but quite some CI matrix entries in many collections
18:27:42 <tadeboro> felixfontein: I do not mind if cexecution dies because there is an actuall reason for it. But preventing users to run things because there might be an issue is not something I would support.
18:28:19 <dmsimard> sivel: announcement for ansible 2.9/2.10 was great, now do python2 next :p
18:28:35 <briantist> 2.9 is also much slower in general, by a lot, so it takes a lot of time in my CI
18:28:55 <briantist> and I can't really control which jobs start in which order, and 2.9 tends to start last which really drags it out
18:29:32 * tadeboro needs to run because scheduling kids is NP-hard ;)
18:29:38 <briantist> the docker test images alone for 2.9 take ~2 full minutes to pull
18:29:45 <dmsimard> tadeboro: thanks for your time o/
18:30:01 <remindbot[m]> @cybette:ansible.im cyb-clock chimes every 15 minutes during the community meeting
18:30:06 <felixfontein> tadeboro: thanks and bye!
18:30:14 <felixfontein> #unchair tadeboro
18:30:14 <zodbot> Current chairs: acozine briantist cidrblock[m] dericcrago dmsimard felixfontein samccann
18:30:33 <acozine> bye tadeboro!
18:30:47 <felixfontein> briantist: you can already drop 2.9 support earlier if you want, there's no need to wait until next May. but next May would be a good time due to 2.9 EOL :)
18:30:48 * acozine is only half here, debugging something mid-meeting
18:31:36 <sivel> dmsimard: Working on it. That'll take little more time
18:31:47 <felixfontein> hmm. any more comments on this topic? I think otherwise I'll switch to 'open floor' if nobody has another suggestion :)
18:31:49 <briantist> of course, I'm aware. As it happens, my company still uses 2.9 too.. (we're trying to move off of it), and I would like to maintain some parity with what the community  generally supports. Just adding in some examples of additional burden that 2.9 support adds.
18:32:28 <dmsimard> no burning topic from me though I do have something appropriate for open floor
18:33:03 <acozine> did we make a decision about the meeting time for next week?
18:33:28 <acozine> are we staying with the same UTC time, at least for the foreseeable future?
18:34:09 <felixfontein> acozine: we didn't make a decision, and there were no further votes
18:34:19 <felixfontein> we had quite a few 0's and one -1, and just a few +1's for moving
18:34:37 <dmsimard> last week > During today's meeting, we had 7 steering committee votes (3 x +1, 3 x 0, 1 x -1), which is inconclusive. So next week will again be at 18:00 UTC.
18:35:14 <acozine> okay, thanks
18:36:05 <felixfontein> I hope that this week's very low attendance is not because of the meeting time
18:37:12 <dmsimard> I know that jillr and andersson007_ are on PTO while gundalow couldn't make it
18:37:34 <dmsimard> can't speak for the others
18:38:20 <felixfontein> #topic open floor
18:40:05 <bungalow> felixfontein: `18:25 <felixfontein> in fact c.g will stop working for Ansible 2.9 once we remove the symlinks. at least if you try a module or action plugin :)` and c.network as that uses symlinks as well (fyi dericcrago)
18:40:13 <samccann> I couldn't open an issue against `community.kubernetes` to ask for a deprecation notice doc so it shows up on docs.ansible.com. So I created one on kubernetes.core in the hopes it's the same people on both? - https://github.com/ansible-collections/kubernetes.core/issues/275
18:41:42 <bungalow> samccann: interesting, what stopped you creating an issue?
18:41:44 <dmsimard> For open floor: I have nothing formal to present yet but I did want to mention we are working to fill a gap between contributor summits and ansiblefests in what will likely be a short (i.e, not full day! ~90 minutes) monthly event (at least at first) suitable for news/updates/presentations/demos. Kind of like a mini version of what we had on the
18:41:44 <dmsimard> friday contributor summit last time.
18:41:53 <bungalow> Woot
18:42:25 <felixfontein> dmsimard: sounds like a really cool idea!
18:42:40 * samccann uploaded an image: (34KiB) < https://libera.ems.host/_matrix/media/r0/download/matrix.org/ApkQDNSaEhmGuHOOLboQmzuA/image.png >
18:42:49 <samccann> bungalow: there was no issue button on it!
18:43:50 <dmsimard> I will share more info as we put structure around the idea -- stay tuned :)
18:43:57 <bungalow> Lol, not sure why that's been disabled, I'll take a look
18:45:01 <remindbot[m]> @cybette:ansible.im cyb-clock chimes every 15 minutes during the community meeting
18:45:47 <dmsimard> We can close the meeting early unless someone has anything else for open floor
18:47:40 <felixfontein> I guess if nobody suggests something in ~1 min I'll close
18:47:45 <felixfontein> then we had open floor open for 10 mins
18:47:46 <felixfontein> new record :D
18:47:50 <acozine> heh
18:48:40 <acozine> re: time change - I mistakenly put this meeting later on my calendar for this week, thinking we HAD changed it, so I ended up double-booked
18:48:52 <acozine> because the US hasn't changed time yet . . .
18:48:58 <felixfontein> acozine: no problem, there were too few people around for today anyway :)
18:49:22 <felixfontein> #endmeeting