15:00:13 <gundalow> #startmeeting Ansible PR review day
15:00:13 <zodbot> Meeting started Tue Dec  1 15:00:13 2020 UTC.
15:00:13 <zodbot> This meeting is logged and archived in a public location.
15:00:13 <zodbot> The chair is gundalow. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:13 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:13 <zodbot> The meeting name has been set to 'ansible_pr_review_day'
15:00:20 <gundalow> #topic Intros
15:00:27 <dmsimard> \o
15:00:31 <akasurde> gundalow, \o
15:01:25 <gundalow> I know the docs meeting is just finishing up, so will give them a few minutes.
15:01:25 <gundalow> In the meantime I'll past a few links and give some context
15:01:25 <gundalow> If you are joining please say so
15:02:03 <akasurde> Cool
15:02:23 <gundalow> #info What's this all about https://github.com/ansible/community/issues/407
15:02:36 <gundalow> Thanks for joining, this is very informal, so please do shout out at any point
15:02:46 <gundalow> #info If you are new to the Ansible Community I invite you to review the Code of Conduct https://docs.ansible.com/ansible/latest/community/code_of_conduct.html
15:04:39 <baptistemm> hello
15:05:08 <gundalow> #info this is the (long over due) 2nd PR review day since Collections have been in place. To start with we will be looking at `community.general` and `community.network`, `community.aws`
15:05:35 <gundalow> akasurde: not sure how long you are around for, though I see we have 21 PRs open in community.vmware, so we could take a look at that as well
15:05:51 <gundalow> Once jillr is here we will look at `community.aws`
15:06:21 <gundalow> basically what extra is here compared to PR discussions on github. hopefully i get some insights into the codebase - while reviewing PRs and also insight into process of contributing
15:08:12 <gundalow> #chair dmsimard akasurde baptistemm dericcrago
15:08:12 <zodbot> Current chairs: akasurde baptistemm dericcrago dmsimard gundalow
15:09:19 <dmsimard> #link https://github.com/ansible-collections/community.general/pulls
15:09:22 <maxamillion> PR REVIEW DAY!!!
15:09:24 <dmsimard> #link https://github.com/ansible-collections/community.network/pulls
15:09:25 <maxamillion> .hello2
15:09:26 <zodbot> maxamillion: maxamillion 'Adam Miller' <maxamillion@gmail.com>
15:09:29 <dmsimard> #link https://github.com/ansible-collections/community.aws/pulls
15:09:39 <dmsimard> #link https://github.com/ansible-collections/community.vmware/pulls
15:09:55 <gundalow> maxamillion: That's the type of excitement we want to see
15:10:00 <gundalow> #chair maxamillion
15:10:00 <zodbot> Current chairs: akasurde baptistemm dericcrago dmsimard gundalow maxamillion
15:10:06 <gundalow> Ok, lets get started
15:10:07 <maxamillion> :)
15:10:14 <gundalow> #topic community.general
15:10:23 <gundalow> #link https://github.com/ansible-collections/community.general/pulls?q=is%3Apr+is%3Aopen+sort%3Acreated-asc
15:10:31 <gundalow> We start with oldest first
15:10:40 <gundalow> https://github.com/ansible-collections/community.general/pull/53
15:10:41 <github-linkbot> https://github.com/ansible-collections/community.general/pull/53 | open, created 2020-03-26T09:52:00Z by matbu: Add pacemaker_resource module for managing cluster resources [affects_1,bug,ci_verified,module,needs_info,needs_revision,new_contributor,new_module,new_plugin,pr_day,stale_ci]
15:10:44 <felixfontein> gundalow: I'm mostly not here, but in case you need an opinion / comment /... on something just ping me
15:11:00 <gundalow> dmsimard: thanks for github-linkbot, it makes it much easier
15:11:04 <gundalow> felixfontein: ack, no chair for you
15:11:50 <gundalow> 53: I can see that the branch was updated 18 days ago, so this is still actuve
15:12:17 <felixfontein> gundalow: the branch hasn't been updated since May 26th
15:12:25 <felixfontein> gundalow: there was a commit which referenced this PR (for some reason)
15:12:27 <gundalow> oh, no, I'm miss reading the last comment on there
15:13:27 <gundalow> Looking at https://github.com/matbu we can see the author is still doing things on GitHub
15:13:29 <felixfontein> the comment didn't even reference this PR in particular, it referenced "ansible-collections#53"
15:15:16 <dmsimard> matbu is on freenode, can poke him
15:15:25 <gundalow> dmsimard: ah, thank you
15:15:42 <gundalow> #action dmsimard to ping matbu wrt community.general#53
15:15:58 <gundalow> https://github.com/ansible-collections/community.general/pull/85
15:15:59 <github-linkbot> https://github.com/ansible-collections/community.general/pull/85 | open, created 2020-03-31T13:45:37Z by Shaps: [WIP] Refactor github* modules,Added tests for github_deploy_key [WIP,affects_1,bug,ci_verified,deprecated,docs_fragments,integration,module,module_utils,needs_info,needs_rebase,new_plugin,pr_day,stale_ci,tests]
15:16:05 <baptistemm> general question: Should we keep PR for new modules / features if the reporter did not update for a while ? do we strive make this enter if the original poster is not responsive
15:16:44 <gundalow> baptistemm: That's a really good question. in gh/ansible/ansible we were not very good at closing old PRs
15:17:17 <gundalow> #action gundalow to ping shaps wrt community.general#85 (github modules)
15:17:22 <gundalow> I'll speak to the PR author
15:18:44 <felixfontein> maybe we should add needs_info if we see that a PR had no reaction for ~a month (except if there is an obvious reason, like PR needs to wait until 2.0.0 comes close enough or things like that :) ), so ansibullbot will close it automatically after some months
15:18:58 <gundalow> Sounds good
15:19:11 <dmsimard> baptistemm: there is precedent for bots eventually closing stale/outdated PRs in different github projects -- there's pros and cons
15:19:25 <maxamillion> I would vote that we close them if the initial PR submitter goes unresponsive, but have some sort of standard language for "we're closing this but if there's still interest in submitting and maintaining this module, please reopen and we'll gladly continue the review"
15:20:01 <gundalow> #action We need to define some standard language (and add to templates) along the lines of "we're closing this but if there's still interest in submitting and maintaining this module, please reopen and we'll gladly continue the review"
15:20:03 <gundalow> maxamillion: thanks
15:20:14 <maxamillion> maybe like ... 3 months stale? ... I'm trying to think of a good threshold that allows for long holidays and stuff like that because you know ... people have lives :)
15:20:35 <gundalow> I *think* the bot does 3 months for needs_info
15:20:35 <dmsimard> there should be a heads up before closing
15:20:51 <maxamillion> dmsimard: +1
15:21:04 <dmsimard> this is both on the maintainers and the contributors though
15:21:11 <maxamillion> dmsimard: something like a comment of "this will be closed if there's no response in 2 weeks" after a period of 2 or 3 months stale?
15:21:19 <felixfontein> needs_info takes ~3 months, and reminds regularly (once per month maybe?). maybe having a similar command (with text better suited for PRs) would be perfect
15:21:25 <dmsimard> I mean, the PR might be perfect and still be inactive due to lack of review
15:21:31 <maxamillion> felixfontein: +1
15:21:41 <maxamillion> dmsimard: ahhh, fair point
15:21:49 <jtanner> what would the bot say to a contributor's PR? "Sorry, we still don't have time to look at this?"
15:22:01 <gundalow> jtanner: Hi :)
15:22:03 <briantist> +1 it's ideal if a bot handles it, with reminders and predictable timelines
15:22:13 <gundalow> jtanner: this is for when PR author isn't responding
15:22:29 <felixfontein> https://github.com/ansible/ansible/issues/71860 here's an example of needs_info closing an issue
15:22:47 <dmsimard> gundalow: how do you tell if it's the PR author or the maintainers that aren't active though ?
15:22:56 <maxamillion> gundalow: can we decipher between the two scenarios with the bot?
15:23:05 <jtanner> if there's a change requested via comment or review, then the bot has logic to set needs_revision and similar logic for timeout+closure. It just probably isn't enabled because of the historical disagreement on who's fault and responsibility it is to get the PR over the finish line.
15:23:34 <jtanner> in some circumstances it was put on the core team to finish the PR and get it merged
15:23:42 <jtanner> which obviously didn't scale
15:23:56 <gundalow> Good background, thanks
15:25:38 <gundalow> #action Full review of needs_info inc if the ball is with author or reviewer.
15:26:03 <gundalow> ok, next
15:26:04 <gundalow> https://github.com/ansible-collections/community.general/pull/155
15:26:04 <github-linkbot> https://github.com/ansible-collections/community.general/pull/155 | open, created 2020-04-12T08:27:25Z by xiangge: New module to support bugzilla [affects_1,feature,module,needs_revision,new_contributor,new_module,new_plugin,stale_ci,tests,web_infrastructure]
15:26:06 <jtanner> / needs_revision
15:26:35 <felixfontein> now @full will get a mention :)
15:26:47 <gundalow> #undo
15:26:47 <zodbot> Removing item from minutes: <MeetBot.items.Link object at 0x7f379c085210>
15:27:06 <gundalow> #action Full review of needs_info & needs_revision inc if the ball is with author or reviewer.
15:27:16 <gundalow> hum, not sure what that undid
15:27:23 <gundalow> https://github.com/ansible-collections/community.general/pull/155
15:27:24 <github-linkbot> https://github.com/ansible-collections/community.general/pull/155 | open, created 2020-04-12T08:27:25Z by xiangge: New module to support bugzilla [affects_1,feature,module,needs_revision,new_contributor,new_module,new_plugin,stale_ci,tests,web_infrastructure]
15:27:28 <gundalow> felixfontein: hum?
15:28:00 <gundalow> 155: Authoris at Red Hat, I'll email them
15:28:13 <felixfontein> gundalow: your #action pings @full
15:28:37 <felixfontein> gundalow: I think for some reason what github-linkbot writes ends up in the meeting notes, so #undo could have un-done the last thing it wrote
15:29:00 <gundalow> felixfontein: I think meeting bot will only assign to someone that exists in the channel
15:29:16 <gundalow> 155: Net new module, though does have a unit test
15:29:19 <dmsimard> meetbot picks up standalone links
15:29:28 <felixfontein> gundalow: ah, that could be
15:29:37 <jtanner> "this moudle is not the idempotent" ... seems like the author could work around that and validate / sync the status.
15:29:37 <felixfontein> dmsimard: ok!
15:30:44 <baptistemm> did you see there is a strange label affects_1
15:31:38 <dmsimard> felixfontein: ah I see what you mean -- the undo undid the link, not the action
15:31:39 <jtanner> the numbers for that label are derived from the ansible version.
15:32:12 <felixfontein> dmsimard: yes, or at least I think so - that would explain the strange 'undo' message :)
15:32:29 <gundalow> 155: I've email author. Added needs_info, so bot will auto close
15:32:44 <gundalow> oh, I'm also adding `pr_day` GitHub label so we can track what we've looked at
15:32:50 <dmsimard> +1
15:32:56 <gundalow> https://github.com/ansible-collections/community.general/pull/175
15:32:57 <github-linkbot> https://github.com/ansible-collections/community.general/pull/175 | open, created 2020-04-13T19:49:58Z by achauvinhameau: add EfficientIP Device Manager dynamic inventory [affects_1,inventory,needs_revision,new_contributor,new_plugin,pr_day,stale_ci]
15:34:29 <gundalow> 175: No updates to the code since 7th Jul, even though various people have reviewed
15:34:44 <gundalow> I'll add a comment and set needs_info so it will be closed soon
15:34:46 <dmsimard> doesn't seem like there are huge blockers for that one, unit tests would be nice to have
15:35:39 <gundalow> https://github.com/ansible-collections/community.general/pull/219
15:35:40 <github-linkbot> https://github.com/ansible-collections/community.general/pull/219 | open, created 2020-04-22T12:29:38Z by traittinen: New Infoblox lookup plugin: nios_next_vlanid [affects_1,community_review,lookup,new_contributor,new_plugin,stale_ci]
15:35:47 <gundalow> 219 in wrong repo, I'm going to close
15:36:40 <gundalow> #info started with 76 open PRs in community.general
15:36:43 <dmsimard> gundalow: you're picking up PRs sorted  by oldest ?
15:36:55 <gundalow> dmsimard: https://github.com/ansible-collections/community.general/pulls?q=is%3Apr+is%3Aopen+sort%3Acreated-asc
15:36:58 <gundalow> yes
15:37:05 <gundalow> https://github.com/ansible-collections/community.general/pull/263
15:37:05 <github-linkbot> https://github.com/ansible-collections/community.general/pull/263 | open, created 2020-05-02T17:17:10Z by imjoseangel: New merge lists plugin from a given key [affects_1,community_review,feature,integration,lookup,new_plugin,pr_day,stale_ci,tests]
15:37:27 <baptistemm> ah there is a infoblox collection now
15:38:12 <felixfontein> baptistemm: there has been for some time, and eventally it will get included in ansible 2.10, but...
15:38:44 <dmsimard> looks like https://github.com/infobloxopen/infoblox-ansible if not mistaken
15:39:15 <felixfontein> yes, that one
15:39:29 <gundalow> 263: There is a lookup. Or there is a filter version in https://github.com/ansible-collections/community.general/pull/604
15:39:29 <github-linkbot> https://github.com/ansible-collections/community.general/pull/604 | open, created 2020-06-30T19:33:33Z by vbotka: Added filter plugin lists_mergeby. Fix #249 [affects_2.10,community_review,feature,integration,needs_triage,new_contributor,new_plugin,tests]
15:39:40 <acozine> o/
15:39:47 <gundalow> #chair acozine
15:39:47 <zodbot> Current chairs: acozine akasurde baptistemm dericcrago dmsimard gundalow maxamillion
15:39:49 <gundalow> Welcome :)
15:40:14 <lmodemal> Hi everyone!
15:40:22 <akasurde> lmodemal, hi
15:40:29 <akasurde> acozine, hi
15:40:37 <felixfontein> about 263 and 604, I personally prefer having a filter instead of a lookup
15:40:46 <samccann> o/
15:41:12 <samccann> 263 - I don't fully understand it, but is it in a similar set of utilities that are being gathered in https://github.com/ansible-collections/ansible.utils ?
15:41:17 <acozine> hi akasurde!
15:41:52 <gundalow> samccann: oh, good point, I'll ping Brad
15:42:07 <baptistemm> 604 does not have documentation + example for end-users
15:42:40 <briantist> dmsimard: re: c.g #1035 , any PRs and issues left related to hashi_vault I think are best re-submitted by the authors if they are still interested. I worked with Felix to review outstanding PRs before the collection split. I can go through again and leave comments for authors if you want. For some, I wanted to maybe post some guidelines in the new collection before pointing people, but I guess it's ok now
15:43:25 <dmsimard> briantist: works for me
15:43:44 <gundalow> felixfontein: Any reason you think filter rather than lookup?
15:43:52 <gundalow> What do others thing, filter or lookup?
15:44:02 <jtanner> what does it do?
15:44:07 <dmsimard> gundalow: it /feels/ more like a filter than a lookup to me
15:44:20 <jtanner> if it fetches data from the controller system, it's a lookup. if it munges input data, then it's a filter
15:44:24 <dmsimard> lookup is ... poke this thing and get data back
15:44:33 <dmsimard> what jtanner said
15:44:39 <dericcrago> yep, agree with jtanner
15:44:56 <jtanner> if it returns true/false for an input, it's a "test" filter
15:45:46 <felixfontein> gundalow: it does a simple transform on data without having to do fancy stuff (like talking to a server), so a filter seems like a good choice
15:46:19 <felixfontein> yeah, what jtanner said ;)
15:47:09 <felixfontein> a lookup can also be chosen if you need ansible variables as defaults for options, but that's also not the case here
15:47:25 <gundalow> OK, so do we close 263 (lookup)?
15:48:13 <acozine> agreed that a filter seems like lower overhead and a better match here
15:48:56 <gundalow> Cool, does someone want to do the honors?
15:49:21 <jtanner> couldn't sumitter just refactor PR?
15:49:31 <dmsimard> jtanner: there is a similar implementation as a filter plugin proposed already
15:49:37 <jtanner> ah, nvm me then
15:49:39 <dmsimard> #604
15:49:43 <dmsimard> gundalow: I'll take it
15:49:48 <gundalow> dmsimard: thanks
15:49:55 <gundalow> https://github.com/ansible-collections/community.general/pull/267
15:49:56 <github-linkbot> https://github.com/ansible-collections/community.general/pull/267 | open, created 2020-05-03T19:48:29Z by jdufresne: Fixed flatpak module to accept a list of names to install/uninstall [affects_1,bug,integration,module,needs_repo,needs_revision,stale_ci,tests]
15:50:58 <gundalow> 267: Bug fix to existing module, so would be good to get in
15:50:58 <dmsimard> gundalow: you already went out and asked, I've subscribed to the issues and will give them a chance to get back to us
15:51:25 <baptistemm> he deleted its ansible fork
15:51:41 <gundalow> dmsimard: Could you give the reasons for why we think filter is better
15:51:49 <dmsimard> yeah sure
15:51:51 <jtanner> i believe the bot should add "need_repo" in that case?
15:52:13 <felixfontein> 267: would be nice to know whether the changes / tests actually work
15:52:37 <baptistemm> felixfontein: the only way of doing is testing locally ?
15:52:54 <jtanner> for infoblox still?
15:52:54 <felixfontein> baptistemm: yes, since the tests don't run in CI
15:53:23 <felixfontein> baptistemm: at least not yet; I remember there was something, let me search
15:53:26 <jtanner> oh, flatpak ... nvm
15:53:52 <felixfontein> https://github.com/ansible-collections/community.general/pull/1246#discussion_r519419258
15:53:52 <github-linkbot> https://github.com/ansible-collections/community.general/pull/1246#discussion_r519419258 | closed, created 2020-11-08T11:36:40Z by TGuimbert: Add --non-interactive argurment to Flatpak [affects_2.10,backport-1,bug,community_review,module,new_contributor,os,packaging,plugins]
15:54:19 <felixfontein> one of the orignial authors already had a PR with tests that run in CI in ansible/ansible, but it never got merged and he gave up
15:54:29 <briantist> this one can be closed -> https://github.com/ansible-collections/community.general/pull/664
15:54:29 <github-linkbot> https://github.com/ansible-collections/community.general/pull/664 | open, created 2020-07-17T08:48:03Z by todorinskiz: Add support for listing secrets to hashi_vault [affects_2.10,feature,lookup,needs_revision,needs_triage,new_contributor,plugins,stale_ci]
15:54:33 <felixfontein> if someone wants to earn karma points, feel free to convert that PR to a c.g PR :)
15:55:08 <felixfontein> briantist: you can add close_me, then the bot should close it for your
15:55:20 <briantist> 👌
15:55:33 <briantist> is an edit ok or does it need a new comment?
15:55:40 <felixfontein> I think edit is fine
15:56:12 <felixfontein> briantist: we'll see in a few minutes :)
15:56:25 <cybette> hello o/
15:56:28 <felixfontein> hehe, or not
15:56:30 <felixfontein> hi cybette!
15:56:36 <gundalow> 664: closes
15:56:42 <gundalow> closed*
15:56:51 <gundalow> oh, didn't se you wanted to test the bot
15:56:55 <gundalow> #chair cybette
15:56:55 <zodbot> Current chairs: acozine akasurde baptistemm cybette dericcrago dmsimard gundalow maxamillion
15:56:57 <gundalow> Welcome :)
15:57:30 <cybette> hi felixfontein gundalow ! thanks :)
15:58:06 <gundalow> https://github.com/ansible-collections/community.general/pull/324
15:58:07 <github-linkbot> https://github.com/ansible-collections/community.general/pull/324 | open, created 2020-05-12T09:35:51Z by riemers: Yum versionlock support [affects_1,module,needs_revision,new_contributor,new_module,new_plugin,stale_ci]
15:58:28 <gundalow> 324, `It is not that i don't want to maintain it, i can handle that. But i cannot code/test on it anymore. I try to do the best i can.`
15:59:03 <maxamillion> that's an odd statement at face value
15:59:49 <gundalow> I'm guessing move job to somewhere that doesn't use yum
15:59:55 <felixfontein> ah, I rthink that can be closed, aminvakil created a new PR for that
16:00:02 <felixfontein> https://github.com/ansible-collections/community.general/pull/1405
16:00:03 <github-linkbot> https://github.com/ansible-collections/community.general/pull/1405 | open, created 2020-11-26T21:50:43Z by aminvakil: Newmodule yum_versionlock [affects_2.10,community_review,integration,module,needs_triage,new_module,new_plugin,os,packaging,tests]
16:00:50 <dmsimard> ok let's do that then
16:00:56 <dmsimard> taking it
16:01:12 <gundalow> dmsimard: thanks
16:01:22 <dmsimard> side note, these bot pings are insane: https://github.com/ansible-collections/community.general/pull/1405#issuecomment-734901174
16:01:23 <github-linkbot> https://github.com/ansible-collections/community.general/pull/1405#issuecomment-734901174 | open, created 2020-11-26T21:50:43Z by aminvakil: Newmodule yum_versionlock [affects_2.10,community_review,integration,module,needs_triage,new_module,new_plugin,os,packaging,tests]
16:02:02 <gundalow> #action bot pings are insane: https://github.com/ansible-collections/community.general/pull/1405#issuecomment-734901174
16:02:03 <github-linkbot> https://github.com/ansible-collections/community.general/pull/1405#issuecomment-734901174 | open, created 2020-11-26T21:50:43Z by aminvakil: Newmodule yum_versionlock [affects_2.10,community_review,integration,module,needs_triage,new_module,new_plugin,os,packaging,tests]
16:02:27 <felixfontein> dmsimard: indeed! it pings every maintainer / related person for every module in plugins/modules/packaging/os/ or even plugins/modules/packaging/ I think
16:02:53 <gundalow> dmsimard: Feel free to `#action` anything like that so we make a note of it
16:02:55 <gundalow> https://github.com/ansible-collections/community.general/pull/328
16:02:55 <github-linkbot> https://github.com/ansible-collections/community.general/pull/328 | open, created 2020-05-12T19:40:35Z by gthiemonge: Add gandi_livedns module [affects_1,ci_verified,community_review,module,module_utils,new_contributor,new_module,new_plugin,stale_ci]
16:03:27 <dmsimard> felixfontein: maybe we should limit to last N active contributors or something
16:03:44 <felixfontein> dmsimard: sounds good, feel free to implement that ;)
16:03:44 <gundalow> Do we know anyone at Gandi, feels like it should be a collection
16:03:54 <jtanner> "active" is hard to measure in the bot ... i think had code in there to try to figure that out
16:05:59 <dmsimard> gundalow: fwiw I don't see any other gandi modules/plugins
16:06:00 <baptistemm> misc did a comment on https://github.com/ansible-collections/community.general/pull/328
16:06:00 <github-linkbot> https://github.com/ansible-collections/community.general/pull/328 | open, created 2020-05-12T19:40:35Z by gthiemonge: Add gandi_livedns module [affects_1,ci_verified,community_review,module,module_utils,new_contributor,new_module,new_plugin,stale_ci]
16:06:20 <baptistemm> dmsimard: neither I did
16:06:25 <gundalow> hum, looks like https://github.com/Gandi/ansible/commit/b7322ac88bcedec4b88b214f9b2c4ea0257cd483#diff-86a3d82b0fa8ff013f9d883d9767d9ba0a16ca2c26d16c3f2db05da057af87c6 has some modules in
16:06:50 <dmsimard> brb going to hack github-linkbot to not expand the same link more than once so it's less spammy
16:07:05 <gundalow> #action gundalow to contact Gandi and see if they'd be interested in having their own collection, and include https://github.com/ansible-collections/community.general/pull/328
16:07:05 <github-linkbot> https://github.com/ansible-collections/community.general/pull/328 | open, created 2020-05-12T19:40:35Z by gthiemonge: Add gandi_livedns module [affects_1,ci_verified,community_review,module,module_utils,new_contributor,new_module,new_plugin,stale_ci]
16:07:10 <gundalow> dmsimard: +1
16:07:34 <baptistemm> dmsimard: expand it only each n minutes ?
16:08:07 <felixfontein> baptistemm: dmsimard: only once every n minutes would be great
16:08:18 <felixfontein> where n is not too big :)
16:08:31 <gundalow> https://github.com/ansible-collections/community.general/pull/335
16:08:32 <github-linkbot> https://github.com/ansible-collections/community.general/pull/335 | open, created 2020-05-13T17:40:12Z by C0rn3j: [icinga2_host.py] Actually return codes instead of data [affects_1,bug,community_review,module,new_contributor,stale_ci]
16:08:43 <baptistemm> and if it was not expanded in the last N lines also
16:08:56 <gundalow> 335 bug fix, so would be nice to get in
16:08:56 <felixfontein> alternatively, add some insults if the link has already been pasted in the last 5 minutes ;)
16:09:59 <baptistemm> felixfontein: +1
16:10:30 <felixfontein> gundalow: maybe it would be better to output both code and data in 335
16:10:37 <gundalow> 335: maintainer (t794104) doesn't seem to been active on GitHub since Jan
16:10:51 <baptistemm> gundalow: doing to ping one of the commiter of the Gandi fork
16:10:59 <baptistemm> s/doing/going/
16:11:22 <gundalow> baptistemm: Thanks :)
16:11:51 <gundalow> 155 (bugzilla from earlier): PR author has replied to my email and said they'll take a look
16:12:06 <gundalow> baptistemm: please could you `@gundalow` in there as well
16:14:39 <gundalow> Do we think https://github.com/ansible-collections/community.general/pull/335/files is OK to merge?
16:14:44 <dericcrago> re 335 maintainer, looks like module add was 10/2017 and then a follow up on 12/2019
16:15:25 <samccann> re: 335 - someone needs to apply the suggested edit to the changelog fragment
16:15:50 <dmsimard> reloaded github-linkbot with https://github.com/dmsimard/linkbot/commit/017355f79fd0e5f237d80aa3bf62c88c24ddab7d should be good enough for now
16:15:51 <gundalow> 335: actually, someone might be able to test that PR by just connecting to any random http server (or ever something that isn't running) and comparing error message before/after patch
16:16:19 <gundalow> samccann: good point, I've pushed that change
16:16:26 <samccann> thanks!
16:17:35 <gundalow> dericcrago: Do you fancy testing 335, no need to setup icinga
16:17:53 <dericcrago> sure, gimme a few
16:18:15 <gundalow> #action dericcrago to test 335
16:18:17 <gundalow> Thanks
16:18:34 <gundalow> https://github.com/ansible-collections/community.general/pull/356
16:18:34 <github-linkbot> https://github.com/ansible-collections/community.general/pull/356 | open, created 2020-05-17T18:21:56Z by AshishVarshneyy: New feature for logging runtime [affects_1,callback,community_review,feature,has_issue,new_contributor,new_plugin,stale_ci]
16:18:38 <gundalow> https://github.com/ansible-collections/community.general/pull/356
16:18:43 <gundalow> dmsimard: thanks for the bot fix :)
16:18:44 <dmsimard> hey, it works \o/
16:18:56 <felixfontein> thanks! :)
16:19:07 <felixfontein> for how long does it cache links?
16:20:26 <dmsimard> felixfontein: simple implementation in the link slightly above, it just keeps a list of the last 5 expanded links and doesn't expand them again
16:21:32 <felixfontein> dmsimard: ah, that also works :)
16:21:57 <dmsimard> based on duration and stuff would've been nice but might've taken more than 5 minutes
16:22:03 <gundalow> maxamillion: What you think (since you reviewed previously) given felixfontein comments in https://github.com/ansible-collections/community.general/pull/356#issuecomment-706548491
16:22:04 <github-linkbot> https://github.com/ansible-collections/community.general/pull/356#issuecomment-706548491 | open, created 2020-05-17T18:21:56Z by AshishVarshneyy: New feature for logging runtime [affects_1,callback,community_review,feature,has_issue,new_contributor,new_plugin,stale_ci]
16:22:19 <maxamillion> gundalow: looking
16:22:25 <maxamillion> (sorry, I'm multitasking badly)
16:22:33 <dmsimard> ah, it picked up the link again because of the #issuecomment T_T
16:22:45 * dmsimard fixes
16:23:46 <gundalow> maxamillion: nps, didn't expect you to be following every PR :)
16:24:21 <maxamillion> I agree with felixfontein's review, I honestly don't remember reviewing this but I give it a +1 for merge so I assume I at least looked it over at some point in the past
16:25:54 <samccann> 356 - needs changelog fragment?
16:27:00 <samccann> Also - we're getting away from using 'whitelist' as a term. Can't remember the replacement? is it allow/deny?
16:27:49 <briantist> allow list? accept list?
16:28:35 <baptistemm> (felixfontein: using gh command line to pull PR is good)
16:30:40 <felixfontein> baptistemm: sorry I lost context :)
16:31:09 <abadger1999> Morning.
16:31:16 <dmsimard> abadger1999 o/
16:31:20 <dmsimard> #chair abadger1999
16:31:20 <zodbot> Current chairs: abadger1999 acozine akasurde baptistemm cybette dericcrago dmsimard gundalow maxamillion
16:31:32 <acozine> re: replacing `whitelist` as a term: https://github.com/ansible/ansible/pull/70028/files
16:31:39 <felixfontein> morning abadger1999!
16:32:12 <abadger1999> release day today.  I'm going to cut another antsibull release before that to pick up felixfontein's addition of version to the docs.
16:32:24 <felixfontein> yay!
16:32:30 <felixfontein> thanks :)
16:32:54 <samccann> thanks acozine!
16:33:04 <samccann> I put in a suggested edit to remove whitelist
16:33:23 <samccann> (re 356 ^^)
16:34:47 <gundalow> samccann: ah, good spot, didn't notice that
16:34:51 <felixfontein> samccann: thanks!
16:37:10 <dmsimard> gundalow: I'll pick up #356 and ask author to get back to us regarding last comments
16:37:45 <gundalow> dmsimard: Thank you
16:37:48 <abadger1999> Oops, I hsould have squashed that last PR but rebase-merged it instead.
16:38:30 <gundalow> https://github.com/ansible-collections/community.general/pull/357
16:38:30 <github-linkbot> https://github.com/ansible-collections/community.general/pull/357 | open, created 2020-05-18T13:28:16Z by Akasurde: rpm_ostree_pkg: new module [affects_1,bug,community_review,module,new_module,new_plugin,stale_ci]
16:38:35 <gundalow> one from akasurde :)
16:39:43 <felixfontein> I guess the main missing part are tests
16:40:58 <akasurde> yes I need to work on test
16:41:07 <akasurde> I will take a look at this week
16:41:22 <dmsimard> akasurde++
16:41:22 <zodbot> dmsimard: Karma for akasurde changed to 1 (for the current release cycle):  https://badges.fedoraproject.org/tags/cookie/any
16:42:00 <baptistemm> wot ?
16:42:42 <dmsimard> baptistemm: have a question ?
16:43:11 <felixfontein> briantist: before you close more issues, these can simply be moved to the new repo (though not by you)
16:43:22 <felixfontein> briantist: also your closing message says "PR" even though it is an issue :)
16:43:23 <akasurde> anyone want to help on 357, let me know
16:43:30 <baptistemm> dmsimard: no :)
16:43:46 <dmsimard> baptistemm: ¯\_(ツ)_/¯
16:44:04 <felixfontein> briantist: I don't have sufficient rights to transfer them though. I guess someone from the community team has to do it
16:45:17 <dmsimard> felixfontein, briantist: yeah we can transfer issues, just point them out
16:45:22 <acozine> akasurde: if you accept felixfontein's suggestion, it should pass CI
16:45:40 <acozine> the only failure is the version-mismatch one
16:45:45 <abadger1999> Ohhhh...
16:45:59 <abadger1999> felixfontein, gundalow: Should community.postgresql be in the ansible tarball?
16:46:09 <felixfontein> abadger1999: yes, it should be
16:46:14 <dmsimard> abadger1999: it's in ansible.in
16:46:15 <abadger1999> hmmm....
16:46:22 <abadger1999> I need to check and see if it is.
16:46:33 <baptistemm> https://github.com/ansible-collections/community.general/pull/267 works fine. I pulled the PR and did a sudo /home/baptistemm/Code/venv/ansible-stable/bin/ansible-test integration flatpak --allow-unsupported --allow-destructive --allow-root --docker --docker-privileged
16:46:33 <github-linkbot> https://github.com/ansible-collections/community.general/pull/267 | open, created 2020-05-03T19:48:29Z by jdufresne: Fixed flatpak module to accept a list of names to install/uninstall [affects_1,bug,integration,module,needs_rebase,needs_repo,needs_revision,pr_day,stale_ci,tests]
16:46:35 <felixfontein> abadger1999: https://github.com/ansible-community/ansible-build-data/pull/39#issuecomment-736376036
16:46:35 <github-linkbot> https://github.com/ansible-community/ansible-build-data/pull/39#issuecomment-736376036 | closed, created 2020-12-01T09:38:10Z by abadger: Add community.okd, a collection which was broken out of the community.general collection, for 2.10.4
16:46:45 <abadger1999> When I added community.okd last night, I thought it needed to be added to ansible-2.10.build as well.
16:46:59 <abadger1999> I'll go through and check for other things that might be left out.
16:47:04 <felixfontein> we only added the new collections to ansible.in
16:47:08 <briantist> oof, copy paste :)  but I was just about to post. I'm done closing issues for the time being; the ones I closed I preferred not moved. I think I'm down to the last 4 remaining (issues, there should be no remaining PRs), and those I think are worth moving
16:47:22 <dmsimard> baptistemm: thanks for doing that, can you add that info in the PR? it'd be helpful
16:47:58 <briantist> can be moved to `community.hashi_vault`: https://github.com/ansible-collections/community.general/issues/1207
16:48:14 <dmsimard> i'll do it, just provide the links :)
16:48:27 <abadger1999> <nod>
16:48:28 <felixfontein> abadger1999: looks like the four collections we added last week need an entry in ansible-2.10.build then
16:48:46 <briantist> can be moved to `community.hashi_vault`: https://github.com/ansible-collections/community.general/issues/1261
16:48:54 <felixfontein> abadger1999: from a quick line count, there seem to be four collections missing in ansible-2.10.built
16:49:20 <abadger1999> felixfontein: Cool, I'll confirm whether or not that was needed and make sure that those are added.
16:49:37 <felixfontein> abadger1999: thanks!
16:50:26 <baptistemm> need to leave 20 min
16:50:58 <briantist> This one, I think I'd like to look into using the Projects feature in the repo. Is it better to move it as an issue first? Or should I leave it where it is and create something new (it was my "issue"): https://github.com/ansible-collections/community.general/issues/1271
16:51:12 <abadger1999> Yep, looks like they were needed.  /me opening a PR to add them.
16:51:43 <briantist> can be moved to `community.hashi_vault`: https://github.com/ansible-collections/community.general/issues/1227
16:52:00 <abadger1999> I also opened a ticket on antsibull last night to remember to add functionality that either checks ansible.in or will take care of adding to both ansible.in and ansible-X.Y.build in the future
16:52:19 <felixfontein> :+1:
16:53:07 <abadger1999> Thinking about this a bit more, I should probably create the 2.11/ansible.in file soon, too..  Maybe once we've approved guidelines for adding new collections to 3.0 (next week, IIRC)
16:53:52 <dmsimard> briantist: for 1271 I suggest we can move it first and you can do whatever afterwards
16:54:02 <briantist> sgtm
16:54:25 <dmsimard> ok all of them are moved, I've notified issue authors that weren't you :)
16:54:42 <dericcrago> tested 335 locally and added comment
16:54:49 <briantist> thank you dmsimard
16:54:58 <jillr> o/
16:55:15 <dmsimard> jillr: o/
16:55:27 <felixfontein> hi jillr!
16:55:55 <felixfontein> dericcrago: maybe it would make sense to output both code and data, in case data contains a useful message (which f.ex. can happen if it talks to an actual instance)
16:56:07 <gundalow> #chair jillr
16:56:07 <zodbot> Current chairs: abadger1999 acozine akasurde baptistemm cybette dericcrago dmsimard gundalow jillr maxamillion
16:56:17 <dmsimard> jillr: we're still working through community.general, aws is on the todo-list
16:57:01 <jillr> dmsimard: sounds good
16:58:20 <gundalow> abadger1999: dericcrago: dmsimard felixfontein andersson007_ Forgot to mention earlier, though can we add `pr_day` label to any PRs we look at today, means we can filter them out next time
16:58:42 <dmsimard> gundalow: we got a bit sidetracked with other topics but I think we're good with #357
16:59:40 <gundalow> cool
16:59:58 <gundalow> https://github.com/ansible-collections/community.general/pull/358
16:59:58 <github-linkbot> https://github.com/ansible-collections/community.general/pull/358 | open, created 2020-05-18T13:43:36Z by notok: Bugfix: Fix parsing array values from osx_defaults [affects_1,bug,community_review,module,new_contributor,small_patch,stale_ci]
17:00:11 <dmsimard> gundalow: until when are we running this btw ?
17:00:21 <dmsimard> or rather for how much longer
17:01:04 <gundalow> I think we have 3 more hours, though I'll be dropping off for food in a bit
17:01:45 <gundalow> Though I don't expect we will actually do another three hours
17:02:05 <dmsimard> I would +1 a break for food at some point too but it was also in respect to time management whether we wanted to spend all the time we had today on community.general or split between the other collections we wanted to look at
17:02:16 <gundalow> yup
17:03:31 <gundalow> jillr: Anyone else that we should see if they are around?
17:03:54 <briantist> felixfontein: I'm going to merge the FAQ PR in a few mins unless you'd like to take another look first: https://github.com/ansible-collections/community.hashi_vault/pull/5
17:03:55 <github-linkbot> https://github.com/ansible-collections/community.hashi_vault/pull/5 | open, created 2020-11-30T22:55:37Z by briantist: Add FAQ to README
17:04:19 <briantist> dmsimard: I'll merge the CI update too unless you want to have a look first: https://github.com/ansible-collections/community.hashi_vault/pull/6
17:04:19 <github-linkbot> https://github.com/ansible-collections/community.hashi_vault/pull/6 | open, created 2020-12-01T00:10:41Z by briantist: Use conditional interpreter for package module in CI
17:07:53 <dmsimard> briantist: reviewed
17:08:08 <felixfontein> briantist: LGTM
17:08:55 <dmsimard> gundalow: I'll take 358, doesn't look like it's missing much to land but author hasn't been responsive -- I'll ping
17:09:03 <briantist> thank you both
17:09:36 <felixfontein> abadger1999: I think the idea that adding them to ansible.in is enough comes from the time when we rebuilt ansible-2.10.build all the time before freeze :)
17:09:41 <abadger1999> Okay, ansible-build-data change has been made.  We're ready for release.  (I'll be working with dmsimard and dericcrago later today to do that... making sure there's someone trained to take over for me ;-)
17:09:49 <gundalow> dmsimard: perfect, thank you
17:10:21 <abadger1999> felixfontein: yeah :-)  For a non-stable release, adding it just to ansible.in is the right way to do it.
17:10:23 <felixfontein> abadger1999: hehe, so you can finally go on vacation when it's release time ;)
17:10:43 <gundalow> jillr: tremble dericcrago As people that know AWS. How about we swap to `community.aws` in ~50 minutes
17:10:47 <abadger1999> That's right :-)
17:10:58 <jillr> gundalow: works for me
17:12:27 <gundalow> Cool
17:12:34 <gundalow> #chair
17:12:34 <zodbot> Current chairs: abadger1999 acozine akasurde baptistemm cybette dericcrago dmsimard gundalow jillr maxamillion
17:12:45 <gundalow> We will swap to community.aws in ~50 minutes
17:12:51 <gundalow> Shall we break till then?
17:12:55 <dmsimard> +1
17:13:00 <acozine> +1 to both
17:13:26 <gundalow> Cool
17:13:33 <gundalow> Thank you everybody
17:13:38 <gundalow> #topic break
17:14:22 <gundalow> If anyone wants to continue looking through https://github.com/ansible-collections/community.general/pulls?q=is%3Apr+is%3Aopen+sort%3Acreated-asc+-label%3Apr_day that's cool. mention links here
17:14:40 <gundalow> Otherwise will see you back in 45 minutes when we will go through community.aws
17:14:45 <gundalow> Thanks again :)
17:25:22 <dericcrago> looks like something happened to the tests on 335 after that last commit
17:25:52 * dericcrago steps away for a few
17:43:55 * acozine goes to grab lunch, will be back but might be a bit late for the start of the AWS stuff
17:44:42 <aminvakil> hello everyone!
17:45:03 <aminvakil> i think this can be closed https://github.com/ansible-collections/community.general/pull/478#issuecomment-734460818
17:45:04 <github-linkbot> https://github.com/ansible-collections/community.general/pull/478#issuecomment-734460818 | open, created 2020-06-08T06:51:18Z by Akasurde: nmcli - Use 'connection.autoconnect' instead of 'autoconnect' [affects_1,bug,has_issue,module,needs_maintainer,needs_rebase,needs_revision,stale_ci,tests,unit]
17:46:08 <aminvakil> i guess this has been done by https://github.com/ansible-collections/community.general/pull/1113
17:46:09 <github-linkbot> https://github.com/ansible-collections/community.general/pull/1113 | closed, created 2020-10-17T02:19:17Z by jbronn: nmcli: module refactor [affects_2.10,backport-1,community_review,module,net_tools,plugins,tests,unit]
17:55:54 <tremble> jillr, How do you feel about https://github.com/mattclay/aws-terminator/pull/117 ?  Specific context being getting https://github.com/ansible-collections/community.aws/pull/295 to a state where we can at least run the tests when someone pushes a PR
17:55:54 <github-linkbot> https://github.com/mattclay/aws-terminator/pull/117 | open, created 2020-11-25T10:25:58Z by tremble: Increase age_limits on resources
17:56:14 * jillr looks
17:58:33 <jillr> tremble: personally I'd prefer to start a bit more conservative for the default, like 15min and see how we do
17:58:54 <jillr> I'd also like mattclay to chime in, since this will potentially affect CI billing
17:59:34 <dmsimard> aminvakil: closed, thanks
17:59:46 <tremble> Is mattclay about?  In theory it shouldn't affect billing too much since it's only the fall-back, but it does mean some tests will run longer
18:01:00 <mattclay> tremble: Approved. :)
18:01:16 <jillr> hehe, works for me  :)
18:01:40 <gundalow> OK
18:02:19 <jillr> tremble: your latest comment on 295 I'll have to dig into a bit
18:02:45 <gundalow> #topic community.aws
18:02:59 <gundalow> jillr: Shall we start with oldest PR first?
18:03:05 <jillr> gundalow: sure
18:03:30 <gundalow> markuman: tremble Please let me know when you are around
18:03:37 * tremble is around
18:03:44 * markuman also
18:03:52 <gundalow> Full house
18:04:06 <gundalow> dericcrago: You back from lunch?
18:04:23 <gundalow> https://github.com/ansible-collections/community.aws/pulls?q=is%3Apr+is%3Aopen+sort%3Acreated-asc+-label%3Apr_day (Oldest PR first
18:04:25 <gundalow> )
18:04:52 <gundalow> FYI we add the `pr_day` label when we look at PRs, so we can skip them next time
18:04:59 <gundalow> https://github.com/ansible-collections/community.aws/pull/27
18:05:00 <github-linkbot> https://github.com/ansible-collections/community.aws/pull/27 | open, created 2020-04-07T13:44:27Z by Tyler-2: fix: Don't mark kstreams changed when no encryption actions taken [affects_2.10,bug,module,needs_revision,plugins,stale_ci]
18:06:38 * dericcrago is back
18:07:24 <gundalow> > Looks like something's a little funky with the unit tests here. Given the results of my integration testing I suspect the unit test is buggy rather than your code.
18:07:32 <gundalow> Do we know what's up with the unit tests?
18:08:00 <markuman> 02:09 [gw1] [ 55%] FAILED tests/unit/modules/test_kinesis_stream.py::AnsibleKinesisStreamFunctions::test_dsbale_stream_encryption
18:08:01 <aminvakil> dmsimard: thank you!
18:08:15 <markuman> was it broken before the pull request?
18:08:25 <jillr> I'm running it locally from main now to check
18:08:31 <jillr> appears to pass
18:09:01 <gundalow> jillr: cool, does it pass if rebased on latest `main`?
18:09:10 * jillr checking
18:09:52 <tremble> IIRC I wasn't sure if it was checking the 'right' thing
18:09:53 <jillr> nope
18:10:27 * jillr comments pr
18:12:39 * samccann strolls in late
18:15:52 <gundalow> #chair samccann acozine markuman jillr tremble
18:15:52 <zodbot> Current chairs: abadger1999 acozine akasurde baptistemm cybette dericcrago dmsimard gundalow jillr markuman maxamillion samccann tremble
18:15:52 <tremble> https://github.com/ansible-collections/community.aws/pull/32 - This feels like the wrong approach to me and iam_user / iam_role both support attaching managed policies
18:15:53 <github-linkbot> https://github.com/ansible-collections/community.aws/pull/32 | open, created 2020-04-08T20:46:59Z by rbobrowicz: add iam_managed_policy_attachment module [affects_2.10,community_review,module,needs_triage,new_contributor,new_module,new_plugin,plugins,stale_ci]
18:17:33 <jillr> I'm inclined to defer to tremble's greater experience on anything IAM related
18:17:43 <markuman> tremble: I agree
18:18:24 * tremble comments on it to give the author time to point out something I missed
18:18:42 <markuman> ok
18:21:35 <jillr> https://github.com/ansible-collections/community.aws/pull/39
18:21:35 <github-linkbot> https://github.com/ansible-collections/community.aws/pull/39 | open, created 2020-04-16T16:09:36Z by TheOptimisticFactory: fix: Cloudfront distribution now uses provided origin_access_identity [affects_2.10,bug,integration,module,needs_revision,new_contributor,plugins,stale_ci,tests]
18:22:55 <tremble> I couldn't get the CI tests to run to a successful conclusion and I've not really poked at cloudfront much
18:23:19 <jillr> cloudfront is hard to CI because updates take so long
18:24:02 <jillr> oh and it looks like the tests themselves might be broken
18:24:18 <tremble> yeah
18:25:42 <tremble> I'd be tempted to accept the PR and debug the tests separately but I can't make any guarantees about following up to fix the tests
18:26:04 <jillr> same. there's already an open issue for it
18:26:38 <jillr> you've look at this more recently that me tremble, do you have confidence about this specific change?
18:26:45 <Goneri> Can we mock CloudFront? This may be done in another PR later.
18:27:23 <tremble> jillr, about 90% confidence
18:27:24 <Goneri> I know we've got a couple of painful cases with community.vmware that we should cover properly, and maybe we can address this one at the same time.
18:27:35 <jillr> Goneri: mocking AWS in general is challenging. we can use the placebo library to record API responses but it's still overhead and no guarantee to remain the same.
18:27:50 <jillr> not to say we shouldn't do it
18:27:54 <Goneri> (just a thought, I don't want to derail the triage with a long CI discussion)
18:27:57 <gundalow> #chair Goneri
18:27:57 <zodbot> Current chairs: Goneri abadger1999 acozine akasurde baptistemm cybette dericcrago dmsimard gundalow jillr markuman maxamillion samccann tremble
18:28:04 <gundalow> Goneri: welcome :)
18:28:14 <Goneri> Aloha :-)
18:28:34 <tremble> Goneri, The big problem with AWS is that it's got massively complex APIs.  it's almost like needing to mock VMWare for each of the AWS service endpoints.
18:29:01 <tremble> They're also somewhat fluid.
18:29:32 <Goneri> yes, I know. This is the reason why we moved away from govcsim.
18:29:36 <jillr> aiui there's nothing about this change that would break existing use cases, right?
18:29:37 <dmsimard> if only we could spin up our own versions of these to test ¯\_(ツ)_/¯
18:29:51 <tremble> jillr, I don't believe so.
18:30:19 <jillr> tremble: shall we ship it then?
18:30:30 <tremble> yeah
18:30:57 <jillr> dmsimard: we can, but our test suite for this is currently broken, which has been blocking this so far
18:31:15 * jillr comments the pr
18:32:38 <jillr> tremble: do you want to update your review?
18:33:00 <tremble> sure
18:35:13 <jillr> https://github.com/ansible-collections/community.aws/pull/44
18:35:13 <github-linkbot> https://github.com/ansible-collections/community.aws/pull/44 | open, created 2020-04-21T10:02:20Z by stefanhorning: New aws_api_gateway_domain module for adding custom domains [affects_2.10,ci_verified,integration,module,needs_revision,new_module,new_plugin,plugins,stale_ci,tests]
18:37:21 <tremble> Oh, it's been broken by us dropping the redundant 'aws' in module_utils.
18:37:46 <tremble> Author says he'll take a look, I guess we move on
18:37:59 <samccann> needs some docs edits... I'll add a review
18:38:15 <jillr> samccann: and metadata, thanks!
18:38:28 <jillr> https://github.com/ansible-collections/community.aws/pull/51
18:38:29 <github-linkbot> https://github.com/ansible-collections/community.aws/pull/51 | open, created 2020-04-23T11:13:00Z by tomwwright: ec2_imagebuilder_component module and unit tests [affects_2.10,module,needs_revision,new_contributor,new_module,new_plugin,plugins,stale_ci,tests]
18:39:40 <jillr> looks like there's unresolved issues from your review, tremble
18:40:01 <tremble> jillr, yeah, docs are almost empty
18:40:45 <jillr> 57 is WIP..
18:40:48 <jillr> https://github.com/ansible-collections/community.aws/pull/65
18:40:48 <github-linkbot> https://github.com/ansible-collections/community.aws/pull/65 | open, created 2020-05-05T16:11:53Z by marknet15: dynamodb_table: boto3, AnsibleAwsModule, billing_mode, point_in_time_recovery and more [affects_2.10,bug,has_issue,integration,module,needs_revision,needs_triage,new_contributor,plugins,stale_ci,tests]
18:41:04 <jillr> ah, and 65 is a draft
18:41:37 * tremble updates the title for consistency
18:41:43 <jillr> thx
18:41:54 <jillr> https://github.com/ansible-collections/community.aws/pull/71
18:41:54 <github-linkbot> https://github.com/ansible-collections/community.aws/pull/71 | open, created 2020-05-14T10:11:51Z by stefanhorning: New module for AWS SES DKIM identity [affects_2.10,integration,module,needs_rebase,needs_revision,new_module,new_plugin,plugins,stale_ci,tests]
18:43:15 <gundalow> jillr: Would you be able to add `pr_day` label when you link the PRs here. That means next time we can filter them out. Also gwmngilfen can do stats to see they progress better than usual PRs
18:43:27 <jillr> gundalow: ack, will do
18:44:17 <gwmngilfen> ty!
18:44:52 <jillr> looks like we're waiting on the author in 71
18:46:22 <jillr> https://github.com/ansible-collections/community.aws/pull/91
18:46:23 <github-linkbot> https://github.com/ansible-collections/community.aws/pull/91 | open, created 2020-06-07T13:03:35Z by raags: Add validation to ensure ecs service is deployed successfully [affects_2.10,community_review,feature,module,needs_triage,new_contributor,plugins,stale_ci]
18:47:10 <markuman> hmm the idea is good, but the implementation is not sufficient imho.
18:47:34 <markuman> when the desired count equals running task, the container can still fail the target groups healthcheck and will be stopped.
18:47:58 <markuman> ...depends on the healtchecks settings in the related target group of the container
18:48:01 <tremble> markuman, Probably needs a 'waiter' but we don't have much documentation on them...
18:49:46 <jillr> we can point them to another module with waiters as an example
18:50:43 <markuman> and it must wait for target group healthy, not for desired count == running count.
18:50:56 <tremble> it took me a while to figure out how they work, they're not exactly new-contributor friendly.
18:51:12 <samccann> as an aside - how frequently do developers need to understand these `waiters` (aka do we need docs to cover this or is it an infrequent thing?)
18:51:27 <jillr> tremble: yeah I think we've relied on the boto docs for them, which are not amazing either
18:52:03 <jillr> samccann: we should cover them, they're based on the sdk but aws' docs for them are not great
18:52:20 <jillr> upwards of half of our modules might use them?  offhand?
18:52:47 <tremble> samccann, We *should* use them much more than we do, some of the test flakes I've been chasing are because they're *not* used
18:53:40 <tremble> jillr, Not sure it's that high, but there's a mix of our waiters and boto3's default waiters.
18:53:45 <markuman> tremble: then we need to do more support :) it's a nice to have feature. we use it internaly be running `elb_target_group_info` with retries and delays after ecs service update.
18:54:19 <jillr> tremble: I phrased that badly - we should use them more than we do but lack of awareness definitely means we don't have that mich now,
18:54:37 <jillr> but if we addressed them better we *might* see as many as half our modules benefit from them, maybe
18:54:37 <samccann> ah so waiter docs need to live in Ansible AWS docs 'somewhere'
18:54:56 <jillr> s/mich/much
18:55:44 <jillr> samccann: yep
18:56:21 <jillr> they're mentioned in https://www.ansible.com/blog/getting-started-with-aws-ansible-module-development but that hasn't been backfilled into any gaps in the dev guide
18:56:40 <jillr> markuman: did you want to leave a review on #91?
18:57:11 <markuman> jillr: Ok, I'll do
18:57:16 <jillr> thanks!
18:57:28 <jillr> https://github.com/ansible-collections/community.aws/pull/95
18:57:29 <github-linkbot> https://github.com/ansible-collections/community.aws/pull/95 | open, created 2020-06-10T22:40:59Z by TylerLubeck: Implement the list_global_accelerators module [affects_2.10,module,needs_revision,needs_triage,new_contributor,new_module,new_plugin,plugins,stale_ci]
18:58:37 <tremble> Should be an _info module?
18:59:03 <jillr> probably
18:59:28 <samccann> fyi - docs tracking issue to update aws dev guide based on jillr's blog post - https://github.com/ansible/ansible/issues/72777
18:59:39 <jillr> thanks samccann
19:01:04 <tremble> samccann++
19:02:48 <acozine> o/
19:04:32 <jillr> there's another pr, #97, with the create/update module for global accelerator
19:04:46 <jillr> neither has tests
19:12:53 <tremble> #95 looks to have done a lot of things right
19:13:12 <jillr> yeah that one seems straightforward
19:14:29 <tremble> Yuck list_accelerators is another of AWS's APIs which doesn't support filtering.
19:14:53 <jillr> also gotta love "Global Accelerators must currently be created and modified in us-west-2"
19:15:12 <jillr> sucks to be in gov cloud?
19:16:22 <jillr> I'm not sure about the approach in 97
19:17:47 <tremble> Gov Cloud not offer us-west-2 ?
19:19:27 <jillr> I thought it was east?  I could totally be wrong though
19:20:21 <tremble> Looks like they expanded
19:20:50 <tremble> AWS GovCloud (US-East) and (US-West) Regions are operated by employees who are U.S. citizens on U.S. soil ....
19:21:04 <jillr> ah, well then
19:22:16 <jillr> wrt 97, it feels like it's more complex than it needs to be but I'm not familiar with the global accelerator api so I'm just slowly reading my way through this pr
19:25:12 <sgw_> can someone tell me how to rename a role on galaxy? or "redirect" one name to a new one (if the role is moved to another organization on github)?
19:26:08 <sgw_> I might be in the wrong room .... right?
19:26:47 <dmsimard> sgw_: I don't have an answer for you but this isn't a bad place to ask
19:26:51 <felixfontein> sgw_: #ansible-galaxy might be better, though since you already asked here, you might be lucky and someone knows
19:27:01 <dmsimard> we just happen to be in the middle of reviewing PRs
19:27:43 <sgw_> didn't see #ansible-galaxy in my overview. will hop there, thanks. sorry for the noise, I am not used to this medium
19:28:53 <jillr> tremble: markuman: are y'all still looking at 97? have any thoughts on this module?
19:29:57 <markuman> jillr: sorry. no thoughts about it
19:30:17 <jillr> markuman: nw, just realized I was taking quite a while with it  :)
19:30:46 * jillr tries to wrap it up
19:32:35 <tremble> This accelerator API is not particularly friendly
19:32:46 <tremble> (AWS API not module API)
19:32:57 <jillr> https://github.com/ansible-collections/community.aws/pull/100
19:32:58 <github-linkbot> https://github.com/ansible-collections/community.aws/pull/100 | open, created 2020-06-16T22:56:47Z by sydoluciani: Adding a new module - route53_crrs module [affects_2.10,community_review,module,needs_triage,new_contributor,new_module,new_plugin,plugins,stale_ci]
19:33:36 <jillr> dmsimard: how long do we go for today?
19:34:02 <markuman> is it not possible to move `route53` to boto3?
19:34:09 <markuman> does it really need a new module?
19:34:20 <tremble> felixfontein, you about?
19:34:20 * markuman looks at the code
19:34:21 <jillr> +1 boto3
19:34:30 <jillr> I'm not aware of anything that would block doing that
19:34:37 <tremble> time :)
19:34:47 <jillr> ok yeah, that  :)
19:34:48 <tremble> Not sure it has test suites either
19:35:08 <jillr> route53 does
19:35:15 <jillr> they're even stable and supported!
19:35:20 <tremble> ok, so it's just time :)
19:35:21 <markuman> sydoluciani is a new contributor...all is invest will be lost than
19:35:23 <dmsimard> jillr: I think we have another hour, we can leave some for next time and move on to other collections
19:35:31 <dmsimard> gundalow ^
19:35:40 <felixfontein> so route53_crrs is doing essentially the same as route53, except that it uses boto3?
19:35:43 <jillr> dmsimard: oh there will be plenty for next time  :)
19:35:58 <markuman> felixfontein: yeah it looks like
19:36:32 <markuman> so ... plus geo location as the author claims
19:36:33 <tremble> markuman, How does the interface compare?
19:36:38 <felixfontein> looks like someone is a big fan of max 80 chars per line
19:36:56 <jillr> nothing wrong with that  :)
19:36:58 <dmsimard> jillr: I lost track but was there any aws PRs closed ? We wrote down the amount of PRs we had started with for c.g
19:37:10 <jillr> dmsimard: we've merged 1
19:37:42 <felixfontein> the PR definitely needs some work
19:37:59 <tremble> jillr, If you include the 2 I looked at earlier it's 3 ;)
19:38:06 <gundalow> dmsimard: jillr I'm happy to continue with community.aws. Next PR review day is Thursday December
19:38:14 <jillr> y'all read faster than I do, want to leave a review?
19:38:25 <gundalow> tremble: did both of those get `pr_day` label added?
19:38:25 <jillr> tremble: thanks!
19:38:33 <felixfontein> the interface seems to be different, I wonder whether that's really necessary
19:38:46 <tremble> gundalow, No, but I can add them :)
19:40:15 <gundalow> tremble: thank you :)
19:40:54 <nitzmahone> gundalow: was looking over the core/community docs proposal that Alicia pointed to us- at a glance, I think that solves the problems, but the one thing I might suggest is that the first major version bump of `ansible` be significantly higher than the core version to help avoid the initial confusion of "wait, 3.0"?
19:40:54 <jillr> as much as I would love for py27 to go away, I'm not sure there's a reason to require python>=3.5 here
19:41:13 <jillr> well, until that^ anyway  ;)
19:41:42 <nitzmahone> (eg, call it "10.0" or something)
19:42:17 <gundalow> nitzmahone: yup, that's something I wondered, what you think abadger1999 dericcrago dmsimard felixfontein andersson007_ should the next major `ansible`release be $LARGE_NUMBER
19:42:17 <jillr> but we shouldnt bump the supported python version for the aws collection just for a new module (wrt reviewing #100)
19:42:33 <felixfontein> gundalow: we already started using 3.0.0 in changelog entries
19:42:40 <dmsimard> nitzmahone: how would Ansible v10 be less confusing than v3 ?
19:42:48 <nitzmahone> If not for the desire for semver, a date-based thing would make more sense to me (and be less confusing), but yeah, if you really want strict semver, that's not an option
19:43:06 <felixfontein> hmm, though none that has been in a release yet
19:43:10 <gundalow> nitzmahone: We would be doing multiple major versions a year
19:43:14 <gundalow> felixfontein: `sed` :)
19:43:26 <nitzmahone> The general consensus seems to be that as the core and community distribution numbers drift wildly apart, people will be more inclined to figure it out
19:43:29 <felixfontein> gundalow: I was mainly worried that we already had that in a published changelog :)
19:43:47 <abadger1999> gundalow: changelogs are distributed, though.
19:43:52 <nitzmahone> (well, the general consensus around core and other folks we've mentioned it to ;) )
19:44:08 <abadger1999> <nod> Yeah... you could make ansible-core-1.0 since it's a new name, too.
19:44:16 <dericcrago> calver would make it 21.something
19:44:47 <dmsimard> let's put that on the agenda for tomorrow's meeting ?
19:44:55 <abadger1999> I'm not sure that moving to 10 is a great idea, though, since the idea of the ansible package is to keep continuity for end users.
19:45:09 <nitzmahone> abadger1999: Heh, if not for the repo and tags and history and Python package name, I'd be +1 to that, actually...
19:45:30 <abadger1999> Feels like the natural drift from making releases approximately twice as fast as ansible-core will differentiate the versions rather rapidly.
19:45:37 <abadger1999> <nod>
19:45:47 <gundalow> AFAIK `YY.` doesn't work if you ever want to make 2+ major releases in a year
19:45:52 <abadger1999> nitzmahone: yeah, and coordination with tower, etc.
19:46:13 <gundalow> dmsimard: could you please add to the agenda, thanks
19:46:33 <felixfontein> I think tomorrow's agenda is already packed with inclusion criteria
19:46:44 <felixfontein> we should make sure we spend only very little time on other things :)
19:47:00 <abadger1999> Yeah, it probably can't come up for several weeks.... by which time we've already gotten people used to 3.0.
19:47:48 <nitzmahone> Yeah, esp if you're talking about multiple major bumps per year, it kinda solves itself reasonably quickly- the whole thing is a minor nit, but just something that occurred to me recently that I thought I'd at least bring up for folks to poke holes in ;)
19:47:48 <dmsimard> gundalow: will do
19:48:50 <felixfontein> nitzmahone: shortly after ansible-core 2.11 is out, we'll already have ansible 4.0.0
19:49:41 <nitzmahone> Yeah, and we'll have something clearer in our CLI `--version` output by then as well (probably also backported to 2.10)
19:50:09 <jillr> dmsimard: gundalow: I think we're wrapped up with community.aws for today
19:50:13 <nitzmahone> Anyway, no biggy- just thought I'd see if anyone thought it was a fantastic idea ;)
19:50:28 <dericcrago> gundalow - shrug, ubuntu versioning comes to mind 20.04 & 20.10
19:50:50 <gundalow> dericcrago: sure, though that's not semver :)
19:51:20 <gundalow> nitzmahone: backporting `--version` to `ansible-base==2.10` would be nice
19:52:55 <abadger1999> backporting ++, that will be awesome
19:53:26 <nitzmahone> I'm angling for core to do semver as well whenever we decide to call something 3.0, but there are lots of implications to that, and our major release cadence will likely stay pretty slow
19:56:29 <felixfontein> nitzmahone: if you stick to only one major release per year, you can also combine semver with year-based versioning ;)
19:56:47 <felixfontein> (you'll just overtake ansible's version numbers then :) )
19:57:47 <nitzmahone> We've also kicked around the date-based versioning thing as well, and just basically keep the assumption that every major release has breaking changes, but 🤷
19:58:00 <acozine> I kinda like the sound of ansible-core-2021
19:58:40 <nitzmahone> IMO the value of semver is much lower in core (as is the likelihood of our properly using it ;) )
19:58:44 <felixfontein> acozine: you're assuming they already change in 2021? ;)
19:59:11 <acozine> felixfontein: heh
20:11:23 <actatux> I have a failing test in shippable that pass locally.
20:11:33 <actatux> Do you know if shippable rebase on top of master?
20:11:41 <felixfontein> acozine: yes, it does
20:12:08 <gundalow> actatux: yes it does
20:12:25 <actatux> Great. I understand better then. ;)
20:13:16 <gundalow> #chair
20:13:16 <zodbot> Current chairs: Goneri abadger1999 acozine akasurde baptistemm cybette dericcrago dmsimard gundalow jillr markuman maxamillion samccann tremble
20:13:22 <gundalow> Thank you all for your help today
20:13:35 <acozine> \o/
20:13:37 <gundalow> We reviewed many PRs and found some things to improve. That's what today is all about
20:13:56 <gundalow> anyone got any final `#action` to make a note of?
20:16:24 <GoKEV> I’m trying to delete a namespace in galaxy-ng… is that a “here” type of question or an #Ansible type of question?
20:16:46 <gundalow> GoKEV: in #ansible-galaxy
20:16:55 <GoKEV> thx!
20:17:24 <gundalow> #endmeeting