15:00:07 <thaumos> #startmeeting Ansible Core Meeting
15:00:07 <zodbot> Meeting started Thu May 18 15:00:07 2017 UTC.  The chair is thaumos. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:07 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:07 <zodbot> The meeting name has been set to 'ansible_core_meeting'
15:00:17 <thaumos> #chair thaumos bcoca gundalow
15:00:17 <zodbot> Current chairs: bcoca gundalow thaumos
15:00:24 <sivel> hello
15:00:29 <thaumos> #chair sivel
15:00:29 <zodbot> Current chairs: bcoca gundalow sivel thaumos
15:00:31 <thaumos> herro
15:00:34 <ryansb> hey team
15:00:44 <thaumos> #chair ryansb
15:00:44 <zodbot> Current chairs: bcoca gundalow ryansb sivel thaumos
15:00:48 <thaumos> mornin!
15:02:09 * mattclay waves
15:02:14 <thaumos> #chair mattclay
15:02:14 <zodbot> Current chairs: bcoca gundalow mattclay ryansb sivel thaumos
15:02:39 <thaumos> @mattclay go to sleep!
15:02:57 <gundalow> thaumos: It's not tomorrow yet :P
15:03:07 <thaumos> heh,
15:03:14 <mattclay> It's only 11 PM.
15:03:33 <thaumos> I am out cold by 11pm
15:03:46 <thaumos> I am savoring all the sleep I can before baby arrives in aug
15:03:57 * mattclay understands that
15:04:10 <thaumos> it's going to be all new for me 😞
15:04:33 <thaumos> #topic Quick discussion around backlog scrub
15:05:10 <thaumos> So, jimi|ansible brought up something yesterday that I wanted to throw by all of you.  He suggested that we could use some time to do some backlog scrub.
15:05:11 <sivel> haha, I am out by like 9:30pm. I also do wake up at 6am though
15:05:19 <thaumos> ^^ that's me!
15:06:24 <thaumos> So, bcoca, jimi|ansible, funzo and myself did some initial backlog scrubbing yesterday and got through 10 or so PR's labeled bugfix_pull_request.
15:07:10 <thaumos> So going off what jimi|ansible suggested, do you guys have any ideas on what would work best?  Should I pick out a couple ahead of time in the next meeting for us to scrub, or we just pick on a couple collectively, thoughts?
15:08:15 <thaumos> @bcoca I thought it worked rather well yesterday, but I think it's fair to think we could tackle more in a meeting once a week in here too.  Do you have any ideas how we could do it here?
15:08:39 <bcoca> as part of 'tirage' we have been doing 'individual backlog' scrubbing, i tink yesterday was good as many times the individual is not able to answer questions about the issue/pr
15:09:01 <bcoca> i think both things are useful, we might want the 'individual triage backlog' to add to the list for the weekly meeting
15:09:24 <sivel> Ideally shouldn't we start at the back? and work our way toward the newer PRs/issues. Just a random thought
15:09:34 <thaumos> we did start at the back yesterday
15:09:39 <thaumos> sorry for not clarifying that.
15:09:41 <sivel> cool
15:10:24 <bcoca> sivel: we cleared out 2015!
15:10:28 <gundalow> FYI the network team have a weekly meeting where we are going through all label:network items
15:10:36 <thaumos> So yeah, about group vs individual.  we'll have a group here too, right @bcoca.  so we can close some quick here as well.
15:10:50 <akasurde> hi all,
15:10:56 <gundalow> akasurde: Hi :)
15:10:57 <thaumos> #chair akasurde
15:10:57 <zodbot> Current chairs: akasurde bcoca gundalow mattclay ryansb sivel thaumos
15:10:58 <sivel> I don't spend much time going through the github.  I tend to pick up the things I see going through my email. Not ideal, but for me, I don't necessarily have the full ability with my $dayjob
15:10:58 * bcoca waves
15:11:10 <sivel> the github queue*
15:11:13 <thaumos> screw $dayjobs
15:11:23 <bcoca> sivel: unacceptable, we want to mandate how you do your fee labour to us!
15:11:25 <thaumos> oh wait
15:11:34 <sivel> lol
15:11:56 <bcoca> sivel: mostly discussing for those of us hired by RH how to interact better with community and reduce backlog
15:12:05 <thaumos> ^^
15:12:07 <thaumos> exactly
15:12:34 <thaumos> so just suggesting if I bring 3 -5 tickets for us as a group to scrub together, or we as a group pick 3-5 to scrub
15:12:51 <bcoca> so we ask YOU, what do you think about our approach and if anyone has better idea, we are open to it (no .. RH hiring 200 people for backlog is not feasable)
15:13:00 <thaumos> because then it isn't the individual problem still.  it's a group
15:13:03 <thaumos> yep
15:14:21 <thaumos> How about this.
15:14:30 <thaumos> Think it over, and let's revisit Tuesday?
15:15:08 <gundalow> End of each meeting to fill up the hour, or at least 2 PRs, whatevers greater?
15:15:13 <sivel> Yeah, I guess my suggestion, is bring the bottom 5, or so
15:15:41 <bcoca> https://github.com/ansible/ansible/pulls?q=is%3Apr+is%3Aopen+sort%3Acreated-asc
15:16:13 <thaumos> yeah, I was thinking start small at first.  so 3-5 I'd bring.  if we're successful here, then we can plow through more.
15:16:25 <bcoca> https://github.com/ansible/ansible/pulls?q=is%3Apr+is%3Aopen+sort%3Acreated-asc+label%3Abugfix_pull_request <= bugfixes as priority
15:16:40 <thaumos> thanks bcoca!
15:16:43 <akasurde> bcoca++
15:16:57 <thaumos> #link https://github.com/ansible/ansible/pulls?q=is%3Apr+is%3Aopen+sort%3Acreated-asc+label%3Abugfix_pull_request
15:17:01 <bcoca> list can be long, we just need to focus on 5 at time
15:17:09 <thaumos> 👍
15:17:16 <bcoca> also, we discussed and are 'working on' the first 5 of that one already
15:17:53 <thaumos> okay, cool!  So I will have five picked out for Tuesday's meeting
15:17:55 <bcoca> 1 &2 have been refered to cloud team, 3 i have alternate fix, 4 and 5 james is looking into (well, not right now, he is flying)
15:18:17 <thaumos> #action Thaumos to bring five PR's to Tuesday's meeting for scrubbing.
15:18:24 <thaumos> #action thaumos to bring five PR's to Tuesday's meeting for scrubbing.
15:18:47 <thaumos> #topic ansible/ansible#24751
15:18:50 <jtanner> hi
15:18:56 <thaumos> #chair jtanner
15:18:56 <zodbot> Current chairs: akasurde bcoca gundalow jtanner mattclay ryansb sivel thaumos
15:18:58 * jtanner forgot about meeting
15:19:07 <thaumos> welcome to west coast times
15:19:11 <shertel> me too
15:19:16 <thaumos> #chair shertel
15:19:16 <zodbot> Current chairs: akasurde bcoca gundalow jtanner mattclay ryansb shertel sivel thaumos
15:19:25 <thaumos> @shertel I was going to give you grief if you didn't show!
15:19:34 <jtanner> shame
15:19:35 <shertel> XD
15:19:41 <thaumos> #link https://github.com/ansible/ansible/pull/24751
15:20:00 <thaumos> So re: topic.  You all cool with the merge?  I always am going to ask for permission 😛
15:20:02 <bcoca> shertel: not sure if anyone touched base with you, but from discussion above we assigned 2 'old' tickets to you
15:20:18 <thaumos> ^^ thanks for bringing that up @bcoca
15:20:22 <bcoca> https://github.com/ansible/ansible/pull/11374 and https://github.com/ansible/ansible/pull/11658
15:20:23 <shertel> Okay, cool.
15:20:30 <shertel> thanks bcoca!
15:20:43 * bcoca expected cursing instead
15:20:44 <thaumos> we can give you more info offline if you need it @shertel 😄
15:20:56 <sivel> thaumos: the only question I have, is how are these decisions being made?
15:20:57 <shertel> haha
15:21:00 <thaumos> @shertel isn't there with you yet @bcoca
15:21:07 <bcoca> thaumos: we need action items in those meetings to follow up with people we give work to
15:21:31 <sivel> not that I am concerned about any of the changes, just not sure who is making those decisions
15:21:36 <bcoca> sivel: mostly the people in meeting discussing, several core memeber there, think that is why thaumos brougth up here for 'oipening it up'
15:21:41 <thaumos> @bcoca, agreed.  I am happy to take charge on divvying those out.
15:21:57 <thaumos> @sivel, what bcoca said.
15:22:08 <bcoca> sivel: started out as internal effort to 'do more to reduce backlog'
15:22:22 * mikedlr waves
15:22:26 * bcoca waves back
15:22:28 <thaumos> #chair mikedlr
15:22:28 <zodbot> Current chairs: akasurde bcoca gundalow jtanner mattclay mikedlr ryansb shertel sivel thaumos
15:22:28 <sivel> cool, all I needed
15:22:47 <bcoca> sivel: no smoke filled room deciding, if that is what you were asking
15:23:00 <thaumos> bleah
15:24:29 <thaumos> so we good to merge?
15:24:55 <thaumos> or is everyone cool with me merging without permission on these sorts of updates?
15:25:02 <thaumos> as long as CI passes that is
15:25:25 <sivel> I'm good with it
15:25:26 <gundalow> thaumos: Ensure that "stale_ci" label isn't there
15:26:45 <thaumos> 👍
15:27:10 <thaumos> Anyone opposed?
15:27:52 <thaumos> k, cool
15:28:15 <thaumos> #action thaumos to merge pr, cherry-pick to stable-2.3 and merge.
15:28:23 <thaumos> #topic Open Floor
15:29:38 <sivel> https://github.com/ansible/ansible/pull/24732
15:29:43 <akasurde> I have one suggestion related to issues. Can we have easy_fix or low_hanging_fruit like label ?
15:30:00 <thaumos> #topic ansible/ansible#24732
15:30:01 <sivel> Wanted to get another set of eyes on that PR, and see if everyone agrees that it should be backported to 2.3.1
15:30:14 <thaumos> #link https://github.com/ansible/ansible/pull/24732
15:32:15 <sivel> Bueller...
15:32:19 <sivel> bcoca...
15:32:23 <jtanner> heh
15:32:33 <jtanner> i wonder why the unarchive maintainers didn't chime in
15:32:37 <thaumos> I think bcoca has passed out.
15:32:57 <jtanner> someone is wrong in the internet ... he out trying to fix that
15:33:16 <sivel> no idea.  I am comfortable with the patch, and I do believe it should be cherry picked into stable-2.3
15:33:22 <bcoca> sorry, can only handle 4 parallel discussions same time
15:33:37 <bcoca> also that is dag question
15:33:43 <bcoca> iirc unarchive is his?
15:33:54 <jtanner> and pileoffrogs
15:34:16 <jtanner> oh, this is 22 hours old
15:34:24 <jtanner> prob why they haven't seen it
15:34:26 <sivel> bcoca: well, partly I pinged you, because you introduced the bug iirc
15:34:48 <sivel> in https://github.com/ansible/ansible/commit/7b197d823ee775b0138495bcd3954dff28851dd5
15:35:04 <sivel> I just made a more robust fix
15:36:09 <sivel> https://github.com/ansible/ansible/commit/7b197d823ee775b0138495bcd3954dff28851dd5#diff-5254fe5279129755e6073ef35bd91025
15:36:15 <sivel> by stripping `remote_src` out
15:36:40 <sivel> anyway, we can move on I suppose.  I just wanted to get eyes on it, because I wanted to see about getting it in and cherry picked before a 2.3.1rc2 was cut
15:36:44 <bcoca> when i added encryption? that should not ahve affected remote_src
15:37:03 <sivel> bcoca: yeah, you stripped out `remote_src`
15:37:05 <sivel> for key in ('remote_src', 'decrypt'):
15:37:18 <sivel> not sure why `remote_src` was added in there
15:38:09 <bcoca> hhm, might have been me cpoy/pasting from copy
15:38:28 <sivel> but because `copy` is deprecated, and `remote_src` is the recommended option, I effectively reversed the logic, and stripped out `copy` instead
15:38:48 <sivel> and updated the module to reflect that change
15:40:18 <bcoca> no, meant from copy action plugin
15:40:24 <bcoca> not copy options
15:40:54 <sivel> I understood what you meant
15:40:59 <sivel> I was just talking about my implementation
15:41:04 <bcoca> so you might just fix by removing remote_src from that tuple
15:42:22 <sivel> Yes, that is possible as well.  Feel free to leave feedback on the issue.  As I mentioned, my change was just a bit more robust.
15:42:36 <sivel> we can move on, or if nothing else for open floor
15:43:31 <sivel> we actually just got another issue for that same bug
15:43:49 <bcoca> sivel: thinkign we can backport thei simple change (minimal) and push yourse into devel (might make a 2.3.2
15:44:22 <sivel> bcoca: I'll put together a 2nd pr for the 2.3 fix
15:44:49 <bcoca> so random idea after talking to networking folk yesterday (duck)
15:44:57 <thaumos> one sec
15:45:17 <bcoca> they want a 'discovery' to figure out Os of net appliance, does it make sense to expand to windows/linux/other flavors?
15:45:21 <thaumos> @sivel, so if I follow, you're going to create a new PR for 2.3?
15:45:26 * gundalow sees network mentioned and looks up
15:45:39 <sivel> thaumos: yes, I'll create a 2nd PR for a fix for stable-2.3
15:45:44 <thaumos> kk
15:45:48 <thaumos> and the current is for devel?
15:46:00 <thaumos> or you'll abandon it?
15:46:15 <bcoca> thaumos: both appy to devel, current is more robust but also requires more review
15:46:28 <thaumos> cool, thx for clarifying!
15:46:48 <thaumos> #action sivel to create new pr for fix to stable-2.3
15:47:12 <thaumos> #action more review to occur on ansible/ansible#24732
15:47:37 <thaumos> #topic quick Network Os discussion
15:47:39 <nitzmahone> discovery of things with that don't use SSH would be ... tricky. And slow.
15:47:47 <bcoca> sivel: ping me tomorrow about the curernt PR, should be able to use some cycles on it
15:47:59 <sivel> bcoca: will do
15:48:05 <bcoca> nitzmahone: good indicator of looking at winrm ports
15:48:25 <bcoca> nitzmahone: closed port not that bad to look for, its 'bad ssh port' that is harder
15:48:43 <bcoca> allowed, listened to, but not actually ssh (yes i run my webservers on port 22)
15:48:55 <thaumos> you would
15:49:04 <sivel> can you define what "Os of net appliance" means?
15:49:17 <bcoca> ios vs junos vs cumulos vs etc
15:49:23 <sivel> oh, lol
15:49:38 <sivel> OS, I was thinking it was like plural of the letter O
15:49:40 <sivel> and was confused
15:49:44 <gundalow> lol
15:49:50 <bcoca> just thinking, discovery plugin might be good for 'new setups' or peole with high turnover
15:49:54 <akasurde> hehehe
15:50:02 <gundalow> high turnover?
15:50:11 <bcoca> gundalow: cloud/virtual
15:50:13 <sivel> We have been talking about this pluggable "setup" phase for a while
15:50:28 <bcoca> sivel: this would be 'before that'
15:50:46 <nitzmahone> The hard part is deciding how to connect once you think it's WinRM (which authtype? ignore self-signed certs?). By the time you've covered all that, seems like you'd do better to just stuff the rest of it in there too. I understand that there are scenarios where that might make sense, but it seems like something that'd be better prototyped in dynamic
15:50:46 <nitzmahone> inventory (as it's really an inventory problem) than something we want to set in stone in core right away.
15:50:50 <sivel> well, something related at least.  A pluggable way to run some form of script that could do some determination early
15:51:28 <nitzmahone> I've kicked it around before too, but I've always found other ways around it.
15:51:34 <thaumos> kind of like the old fashioned way of pinging a host and looking at it's response.
15:51:41 <sivel> I mean, it can be done currently with a playbook also, at least determining the port, and then some logic to determine what connection method
15:51:55 <bcoca> nitzmahone: not saying it should be default, but an option
15:52:01 <bcoca> thaumos: yep
15:52:04 <sivel> `gather_facts: false` and then run some determination steps
15:52:26 <thaumos> I think it's something to play around with.
15:52:32 <thaumos> definitely an option
15:52:37 <bcoca> gather_facts: false .. tasks : - fingerprint: connections=ssh,paramiko OS=bsd,linux.aix
15:52:39 <nitzmahone> I'm just worried that it'll turn into a boondoggle like the "fallback authtype" stuff in WinRM that's a huge pain in my ace
15:52:54 <thaumos> @bcoca, can you write up a proposal on it and we can discuss further?
15:53:08 <sivel> ooh, I kinda like the idea of fingerprint
15:53:20 * bcoca has 6 proposals still open, just wanted to test waters for this idea
15:53:25 <thaumos> I like it
15:53:25 <sivel> not so sold on OS determination from that
15:53:35 <sivel> but at least port/connection type
15:53:41 <thaumos> I want to hear more.
15:53:54 <bcoca> sivel: its easy to do one, the other or both
15:53:56 <bcoca> OS=None
15:54:01 <thaumos> Lol, but Sivel, I always loved suprirsing airheads in the NOC when I could pretty much tell them if it were a windows or unix box.
15:54:03 <bcoca> os: []
15:54:11 <bcoca> ^ both take lists, try 'all by default'
15:54:31 <sivel> yeah, I just think it will be not so easy to determine OS type, unless we nmap the box
15:54:35 <bcoca> again, just brain storm here
15:54:53 <bcoca> sivel: or once you have ssh/winrm port, try connection plpugin + uname
15:54:54 <sivel> port and connection type, are likely easiest
15:55:12 <thaumos> yeah, let's table for now.  write it up and let's go from there
15:55:13 <bcoca> agreed
15:55:33 <bcoca> thaumos: just wanted to gague intereset
15:55:41 <thaumos> I know :smile
15:55:48 <thaumos> 😄
15:55:50 <thaumos> I am interested
15:56:02 <thaumos> There's been customer requests of the same
15:56:13 <thaumos> so I think there's general interest.
15:56:29 <bcoca> probably more intereset in 'bigger shops'
15:56:38 <thaumos> yep, because more of those noc airheads exists.
15:56:47 <bcoca> guy that has 20 servers probably knows them by name and can recite the MAC addresse
15:56:56 <thaumos> You'd hope
15:57:00 <thaumos> okay, really quick
15:57:07 <thaumos> some housekeeping from me!
15:57:17 <thaumos> 1. Contributor summit!
15:57:43 <thaumos> #link https://public.etherpad-mozilla.org/p/ansible-summit-june-2017
15:57:46 <thaumos> Proposed topics
15:58:04 <sivel> I'll see about proposing some things, but I won't be able to participate
15:58:05 <thaumos> Time: Wednesday 21st June, from 9:30am to 4pm
15:58:11 <thaumos> Come on sivel!
15:58:14 <gundalow> #topic Contributor Summit
15:58:25 <thaumos> thx gundalow
15:58:28 <gundalow> #info Agenda, feel free to add items https://public.etherpad-mozilla.org/p/ansible-summit-june-2017
15:58:32 <thaumos> was lazy in essence of time
15:58:36 <gundalow> :)
15:58:39 <sivel> That is the first day of my sabbatical, and prep for 6 weeks out of the country
15:58:45 <thaumos> fun!
15:59:52 <thaumos> I forgot #2, so not important I guess.
16:00:08 <thaumos> #endmeeting