18:00:08 #startmeeting Ansible Community Meeting 18:00:08 Meeting started Wed Sep 15 18:00:08 2021 UTC. 18:00:08 This meeting is logged and archived in a public location. 18:00:08 The chair is felixfontein. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:08 Useful Commands: #action #agreed #halp #info #idea #link #topic. 18:00:08 The meeting name has been set to 'ansible_community_meeting' 18:00:08 #topic Agenda https://github.com/ansible/community/issues/539 18:00:08 abadger1999 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! 18:00:11 o/ 18:00:12 #info Agenda: https://github.com/ansible/community/issues/539 / Topics: https://github.com/ansible-community/community-topics 18:00:14 o/ 18:00:15 o/ 18:00:17 #chair dmsimard jillr andersson007_ 18:00:17 Current chairs: andersson007_ dmsimard felixfontein jillr 18:00:25 o/ 18:00:27 Bom dia 18:00:28 o/ 18:00:38 #chair tadeboro abadger1999 gundalow 18:00:38 Current chairs: abadger1999 andersson007_ dmsimard felixfontein gundalow jillr tadeboro 18:00:40 o/ 18:00:44 * dericcrago waves 18:00:51 #chair cybette dericcrago 18:00:51 Current chairs: abadger1999 andersson007_ cybette dericcrago dmsimard felixfontein gundalow jillr tadeboro 18:00:52 #topic Updates 18:00:52 #topic ansible-core 2.11.5, ansible-base 2.10.14 and ansible 2.9.26 have been released 18:00:55 #info updates: EMS has plumbed #social:ansible.im (#ansible-social) so please feel free to start saying good morning, etc. News posts are welcome there too. There is an ongoing technical issue which is causing some display issues, but it's fine to use. 18:00:59 #undo 18:00:59 Removing item from minutes: INFO by felixfontein at 18:00:55 : updates: EMS has plumbed #social:ansible.im (#ansible-social) so please feel free to start saying good morning, etc. News posts are welcome there too. There is an ongoing technical issue which is causing some display issues, but it's fine to use. 18:01:02 #undo 18:01:02 Removing item from minutes: 18:01:07 #info ansible-core 2.11.5, ansible-base 2.10.14 and ansible 2.9.26 have been released 18:01:10 #info updates: EMS has plumbed #social:ansible.im (#ansible-social) so please feel free to start saying good morning, etc. News posts are welcome there too. There is an ongoing technical issue which is causing some display issues, but it's fine to use. 18:01:11 o/ 18:01:15 sorry... 18:01:17 #chair samccann 18:01:17 Current chairs: abadger1999 andersson007_ cybette dericcrago dmsimard felixfontein gundalow jillr samccann tadeboro 18:02:13 #info The concerns around `netbox.netbox` collection are been worked on. New maintainers have been setup. CI is been fixed. So we can keep the collection in the ansible package 18:02:33 that's great 18:03:25 #info cidrblock sends his apologies and he won't e able to make today's meeting. He authorized us to record him as a +0 on any votes that come up. 18:03:57 oh, I have something 18:05:08 #info The version of ansibullbot running on community.general has been updated with support for AZP amongst other things 18:05:23 good news:) 18:05:33 great news on netbox.netbox indeed! 18:05:46 and the new bot is great as well :) it finally supports AZP! 18:06:07 yeah:) 18:06:09 i.e. no more stale_ci / ci_missing labels, and it copies sanity check error messages into the PR 18:06:37 this will make life much easier 18:06:57 ok, so what we do want to discuss today? 18:07:13 as jillr noted earlier, we don't really have topics that must be discussed now 18:07:57 I'm sure we have plenty of things to spend the time on though :) 18:07:59 we could talk about the attributes feature, but I guess that feature already has enough discussions going on, so having one more won't improve the situation :) 18:08:11 (if someone is interested in this, please go to #ansible-docs) 18:08:38 Maybe we should just say a word or two about https://github.com/ansible-community/community-topics/issues/42 and close it? 18:08:40 how about resurfacing the "New content for community.general and community.network" discussions? 18:09:02 Since that issue seems to be "done". 18:09:07 need to afk momentarily, be back in a bit 18:09:22 tadeboro and i reviewed community.ciscosmb collection - things look good there, does anyone else want to review it or we can vote on its inclusion. 18:09:23 tadeboro: let's assign it to Greg and let him close it, and perhaps write something before :) 18:10:07 #action GregSutcliffe write an update to https://github.com/ansible-community/community-topics/issues/42 and close it 18:10:15 felixfontein: OK, works for me. 18:10:40 andersson007_: I though we will vote on inclusions just before the release (forgot the exact date). 18:10:41 collection review sounds good to me 18:11:03 I guess we can already vote on candidates before the deadline, so we have more time for problematic ones on that date :) 18:11:09 +1 18:11:21 (I think last time we had a pretty long meeting because of that) 18:11:31 the maintainer of community.ciscosmb is very responsive 18:11:32 felixfontein sure :) 18:11:40 That is true ;) 18:11:54 #topic Inclusion of community.ciscosmb in Ansible 5 18:12:04 So maybe we can vote on two collections that have two positive reviews? 18:12:09 so i personally don't see any obstacles not to include the collection 18:12:12 tadeboro: yeah 18:12:16 thought of the same recently 18:12:17 #info Discussion: https://github.com/ansible-collections/ansible-inclusion/discussions/26 18:12:39 tadeboro: sure, if we have at least two :) 18:12:52 tadeboro: i mean the lower bound of review number 18:12:59 imo should be at least two 18:13:12 but two is enough 18:13:13 imo 18:13:15 #info Collection repo: https://github.com/ansible-collections/community.ciscosmb 18:13:24 andersson007_, felixfontein do we need another sample of a trivial community.general PR for hackfest? for example, c.g/plugins/modules/system/aix_inittab.py could be changed with module.get_bin_path(required=true) 18:13:29 I'm looking at that repo a bit 18:13:48 * jillr also looking at repo 18:14:03 mator sounds like a good idea, let me find the document 18:14:29 interesting, the readme for ciscosmb at the bottom points to what looked to be common contributor etc files. but if I click any of those links in Galaxy, they come up missing. 18:15:01 @cybette:ansible.im cyb-clock chimes every 15 minutes during the community meeting 18:15:14 samccann: they're not full URLs in the README.md file, they're relative paths to the codebase 18:15:28 mator: https://hackmd.io/CqUQNTEqRBCHBTC4WF-9Aw?both feel free to add the issue to the doc 18:15:36 `See [RELEASE_POLICY.md](./RELEASE_POLICY.md)` 18:15:47 jillr - makes the readme a bit useless in that section then. 18:15:51 fixing that would be good 18:16:13 ok I'll create an issue if one doesn't already exists for it 18:16:58 thanks! 18:17:28 Ugh, yeah. Maybe we should have a check for absolute links in README somewhere otherwise things will point into the void on galaxy. 18:18:03 i stumbled upon same thing 18:18:28 in the collections i maintaned 18:18:36 thanks felixfontein for spotting that 18:19:08 collection looks good to me, besides that 18:19:27 I didn't spot that, I think? 18:19:47 (or I'm confused what you mean) 18:19:55 i mean in collections that i maintain 18:19:59 ah 18:20:02 so some time ago :) 18:20:09 yeah 18:20:28 i think in this case the issue would be enough 18:20:37 and it's not an obstacle 18:20:43 agreed 18:20:51 as we've started to discuss its inclusion 18:20:57 opened the issue - https://github.com/ansible-collections/community.ciscosmb/issues/47 18:21:06 should we vote at :25? 18:21:06 samccann: thanks! 18:21:12 in case someone wants a bit more time to browse 18:21:22 that works for me 18:21:23 if someone wants even more time, please announce it before we vote so we can wait longer 18:22:26 +1 for voting in this meeting (if there are no major issues spotted) 18:23:49 modules just named "facts" and "command" feels like a trap for users 18:24:33 why? name conflicts? 18:24:35 don't use `collections:` :) 18:24:44 IIRC, this is somewhat standard practice in the netowrking collections. 18:24:50 andersson007_: yeah, if you have more than one colleciton using that name and don't FQCN every task 18:24:56 tadeboro: fair enough then 18:25:01 other network collections also do that 18:25:24 I only cross-checked the cisco.ios collection which are all ios_$something 18:25:25 thankfully examples use FQCNs:) 18:25:44 heh just checked that same one jillr! 18:25:54 ok, it's :25, anyone wants to have more time? 18:25:59 (before we vote?) 18:26:07 with fqcn, it would be community.ciscosmb.facts and community.ciscosmb.command.... Is there something else that could be added to the module names that won't duplicate the namespace.colletcion information? 18:26:13 but that typically means all the network collections handled by the internal ansible team probably use _facts 18:26:13 the sysadmin in me dislikes the potential for "surprise network device management" but if that's the expectation for networking *shrug* 18:26:33 ios stuff came from ansible 2.9 and was not renamed would be my guess. 18:26:37 Alternately, should there be a short list of module names that are too central to some ansible concept that they shouldn't be used? 18:26:51 (community.routeros also uses command and facts) 18:27:16 facts module is actually used by the fact gathering mechanism on network collections. 18:27:32 eos uses eos_facts, iosxr uses iosxr_command etc etc 18:27:38 I don't think it's a blocking thing, it's just definitely not my personal preference 18:27:53 so in terms of certified collections, the trend is to NOT have it just facts or command as the module name. 18:27:57 tadeboro: Ah, so it's already a convention that's being made use of in code. 18:28:42 hmm, looks like community.routeros is the only collection currently included in Ansible which uses `command` and `facts` 18:29:29 yeah looking at the other collections in the Cisco space, the four I've looked at all include the platform in all module names 18:29:58 I personally do not mind the short names since we advocate using FQCNs and refix contains enough info for disambiguation. 18:30:01 @cybette:ansible.im cyb-clock chimes every 15 minutes during the community meeting 18:30:23 Im okay either accepting or having them changed. I just want us to encode the rule in the guidelines if we decide to make them change it. 18:30:25 namespaces exist to prevent name collisions and we encourage users to use FQCNs 18:30:28 man of those collections use runtime.yml to redirect 18:30:32 many* 18:30:48 I agree with tadeboro 18:30:53 https://github.com/ansible-collections/cisco.ios/blob/main/meta/runtime.yml#L27-L28 18:31:36 +1 to what abadger1999 is saying 18:31:50 fwiw - i've checked 10 of the cisco collections and only 2 so far do not include the platform name. 18:32:07 sivel: interesting, so ios_command is a module, and command simply points to the action plugin? (which might use the ios_command module?) 18:32:15 tho i realize we aren't in the business of enforcing consistancy across their own collections 18:32:31 it's done for a module too: https://github.com/ansible-collections/cisco.ios/blob/main/meta/runtime.yml#L167-L168 18:32:52 ah 18:32:53 I just linked the first redirect I found :) 18:33:31 The openstack went the other way IIRC: they removed the os_ prefix from all of their modules. 18:33:55 should we first vote on whether we accept the "short names"? 18:34:02 So I would personally stay from the naming business and let maintainers use whatever they feel is OK. 18:34:08 (or do you want to discuss this further before voting?) 18:34:13 tadeboro: +1 18:34:32 yeah leaning toward tadeboro's idea that maybe we aren't in the business of enforcing module names like this 18:34:36 assuming some standard convention are satisfied, like using the _info and _facts suffixes correctly :) 18:35:02 I'm assuming someone checked that facts is really facts in the code? ;-) 18:35:06 Re openstack: yeah, I don't think we should make "short names" illegal in general. If we make them illegal, I'd like to see that scoped to specific short names or when there's extra information that needs to be included in the name. 18:35:35 samccann: for this collection, I looked at it :) 18:35:40 it does get repetative (ffr.ffr.ffr... i'm lookin at you!) 18:35:44 we have a mechanism (encouraged to use) - FQCNs 18:36:02 awx.awx.awx? 18:36:07 heh yep 18:36:32 feYeah, let's vote on allowing command and facts as valid module names and if that passes we can move on to vote on inclusion of this collection. 18:36:52 +1 18:36:53 VOTE: are collections that want to get included in Ansible free to use short names (like `command` or `facts`) for modules/actions? 18:36:57 #chair 18:36:57 Current chairs: abadger1999 andersson007_ cybette dericcrago dmsimard felixfontein gundalow jillr samccann tadeboro 18:37:04 +1 18:37:05 +1 18:37:06 +1 18:37:15 +1 18:37:15 +1 18:37:16 +1 (I dont have to like it, but I'll support it :)) 18:37:27 jillr: heh :-) 18:37:35 y'all have convinced me :) 18:37:50 +1 18:37:58 +1 18:38:02 "`collections:` is evil!" ;) 18:38:03 (same as jillr lol) 18:38:13 jillr: now you can say as cyberpear "I told you..." 18:38:19 you canNot 18:38:31 later 18:38:38 lol, I can say whatever I want on this side of my keyboard anyway! 18:38:43 hehe :) 18:38:58 I think we need more steering committee member votes 18:39:05 * cyberpear late 18:39:10 #chair cyberpear 18:39:10 Current chairs: abadger1999 andersson007_ cyberpear cybette dericcrago dmsimard felixfontein gundalow jillr samccann tadeboro 18:39:19 cyberpear: just in time for a vote :) 18:39:30 There is one +0 (cidrblock) as well. 18:39:48 +1 allow short names IIUC 18:40:09 +1 since they are new modules & plugins and will always need `collections:` or FQCN 18:40:13 #agreed collections that want to get included in Ansible are free to use short names (like `command` or `facts`) for modules/actions 18:40:25 cool :) 18:40:36 so how about the inclusion? 18:40:42 ok, more discussion on community.ciscosmb, or should we vote on whether to include it? 18:41:03 cyberpear: it's a small network collection having basically two modules (command and facts), repo is https://github.com/ansible-collections/community.ciscosmb 18:41:25 tadeboro and andersson007_ reviewed it and are happy (https://github.com/ansible-collections/ansible-inclusion/discussions/26) 18:41:32 I am +1 on vote and +1 on inclusion 18:41:51 VOTE: should we include community.ciscosmb in Ansible 5? 18:41:52 #chair 18:41:52 Current chairs: abadger1999 andersson007_ cyberpear cybette dericcrago dmsimard felixfontein gundalow jillr samccann tadeboro 18:41:56 +1 18:41:57 +1 18:41:57 +1 18:42:14 +1 18:42:42 +1 18:42:52 (not sure if you want nonSteering people but +1) 18:43:06 +1 18:43:09 +1 18:43:16 * samccann considers calling it the rabble-vote +1 18:43:20 +1 18:44:04 samccann: sure, everyone can vote! 18:44:19 we just have to make sure we got enough steering committee votes for the vote to be valid in the end 18:44:31 ok cool. didn't want to complicate your math any 18:45:00 @cybette:ansible.im cyb-clock chimes every 15 minutes during the community meeting 18:45:07 i think non steering committee votes should be encouraged 18:45:19 definitely! 18:45:25 #agreed community.ciscosmb will get included in Ansible 5 18:45:35 cool! one collection less to look at ;0 18:45:36 ;) 18:45:41 woohoo! 18:45:45 šŸŽ‰ 18:45:45 is there another one that's close to inclusion? 18:45:52 (though probably for next week and not today) 18:46:05 https://github.com/ansible-collections/ansible-inclusion/discussions/22 18:46:30 i thought we included that one:) 18:46:37 That one also has two +1 reviews. 18:46:40 that one still needs a 2.0.0 release ;) but that will come soon 18:47:04 I'm basically waiting for feedback from someone who's a bit short on time (since he became father recently) 18:47:39 2.0.0 is a nice to have as far as the inclusion goes. The version I reviewed was OK for inclusion. 18:47:51 yeah 18:48:04 But we can postpone the vote with no problem. 18:48:38 do you want to vote on it today? 18:48:50 i'd vote 18:48:57 Me too. 18:49:03 I'm more asking all the others, who probably haven't reviewed it yet ;) 18:49:15 * jillr looking but happy to vote in a couple minutes 18:49:25 #topic Inclusion of community.dns in Ansible 5 18:49:31 #info Discussion: https://github.com/ansible-collections/ansible-inclusion/discussions/22 18:49:42 #info Repository: https://github.com/ansible-collections/community.dns/ 18:50:08 #info Documentation: https://ansible.fontein.de/collections/community/dns/ 18:51:16 this collection is a bit special since it includes a MPL 2.0 licensed data file 18:51:48 (it fortunately does not invoke the ā€œIncompatible With Secondary Licensesā€ notice) 18:53:34 lgtm 18:54:03 +1 from me as well 18:54:19 let's vote, before time runs out 18:54:19 lgtm from the docs side 18:54:30 VOTE: should we include community.dns in Ansible 5? 18:54:31 :) 18:54:35 #chair 18:54:35 Current chairs: abadger1999 andersson007_ cyberpear cybette dericcrago dmsimard felixfontein gundalow jillr samccann tadeboro 18:54:35 +1 18:54:37 +1 18:54:39 +1 18:54:39 +1 18:54:51 I abstain since I'm involved a lot in that collection 18:55:09 :) 18:55:18 +1 18:56:22 +1 18:56:39 I assume that Toshio[m] == abadger1999? 18:56:56 :) 18:56:56 Oops, sorry. yes. I hve both matrix and irccloud open :-) 18:57:37 :) 18:58:07 hmm, we need one more steering committee vote I think 18:58:24 (next to cidrblock[m]'s +0) 18:58:30 (and my +0) 18:58:39 maybe the folks we're missing can look over the next couple days and vote in the GH discussion? 18:58:55 hum, or wait, does abstain mean +0? 18:59:00 acozine is I think on vacation until the end of next week 18:59:25 would be great to see more votes from non steering committee who are present in the meeting.. in general 19:00:00 maybe they just hesitate or something 19:00:12 ok, let's do a quick open floor 19:00:14 or don't know that they should 19:00:29 felixfontein: I think they're equivalent, yeah. 19:00:40 #info Vote for inclusion: 5 x +1, 2 x +0 (steering committe); 1 x +1 (non-steering committee) 19:00:49 #topic Open floor 19:01:21 ok, does anyone have something for the open floor? 19:01:23 congratulations felixfontein and markuman and maybe someone else! 19:01:33 on the inclusion 19:01:51 and thanks for your hard work there 19:01:59 thanks! 19:02:19 nothing from me besides that:) 19:02:22 (I guess I could have added an #agreed then...) 19:02:29 (vote counting is hard :D ) 19:02:46 :) 19:04:21 I've contributed just one single character of code. felixfontein did all the hard work! 19:04:44 markuman: I did some copy'n'pasting from your code, though not much survived :) 19:05:00 heh 19:05:15 ok, if there's no topic for the open floor, let's close the meeting! (just 5 mins over time ;) ) 19:05:18 #endmeeting