21:00:10 <notmyname> #startmeeting swift
21:00:11 <openstack> Meeting started Wed May  2 21:00:10 2018 UTC and is due to finish in 60 minutes.  The chair is notmyname. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:00:12 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:00:14 <openstack> The meeting name has been set to 'swift'
21:00:16 <notmyname> who's here for the swift team meeting?
21:00:18 <torgomatic> .
21:00:31 <tdasilva> eu
21:00:40 <acoles> hi
21:00:41 <kota_> hello
21:00:53 <patchbot> I'm here!
21:01:54 <notmyname> is mattoliverau around?
21:02:41 <notmyname> he may join later. let's get started
21:03:02 <mattoliverau> o/
21:03:02 <notmyname> #topic containers, containers, containers. deeeeeeeeeeep containers
21:03:11 <notmyname> #link https://wiki.openstack.org/wiki/Meetings/Swift
21:03:18 <notmyname> it's feature/deep-review time!
21:03:56 <notmyname> so -infra is currently upgrading gerrit, so gerrit links won't work in here during this meeting :-(
21:04:15 <notmyname> today acoles has proposed a patch chain to feature/deep-review
21:04:19 <kota_> :/
21:04:29 <notmyname> https://review.openstack.org/#/q/status:open+project:openstack/swift+branch:feature/deep-review
21:04:55 <notmyname> #link https://review.openstack.org/#/q/status:open+project:openstack/swift+branch:feature/deep-review
21:05:10 <notmyname> and earlier this week, tdasilva hosted a video chat where we talked about the architecture of the feature
21:05:15 <notmyname> a recording can be found at
21:05:21 <notmyname> #link https://bluejeans.com/s/3TyWf
21:05:33 <notmyname> and there's an etherpad with an outline of the proposed patches at
21:05:40 <notmyname> #link https://etherpad.openstack.org/p/deep-review
21:05:55 <notmyname> mattoliverau: kota_: did you get a chance to watch the video recording yet?
21:06:09 <kota_> not yet, sorry
21:06:39 <mattoliverau> Watched it, nice work acoles, and tdasilva for setting the meeting up
21:06:40 <notmyname> no worries. start with the outline, then watch the video. acoles did a great job. it's about 1h20m long
21:07:43 <notmyname> as of now, please consider master to have a soft freeze
21:07:48 <notmyname> please do not land new patches there
21:08:03 <mattoliverau> kk
21:08:24 <notmyname> soft freeze in this case means two things: (1) only critical patches should land and (2) all patches approved to land from master should be coordinated with acoles
21:08:33 <tdasilva> notmyname: mm...might want to cork pete's s3api patch then...
21:09:08 <notmyname> as a reminder, we do the freeze for two reasons. first is to avoid any merge conflict issues. second is to focus on the review of the featur ebranch
21:09:27 <tdasilva> https://review.openstack.org/#/c/565383
21:09:28 <patchbot> patch 565383 - swift - Simplify the code in s3api
21:10:00 <notmyname> do we know if it conflicts with feature/deep?
21:10:06 <tdasilva> don't think
21:10:08 <tdasilva> so
21:10:25 <notmyname> yeah, doesn't look like it could. only touches one s3api file
21:10:47 <notmyname> acoles: if you're ok with it, then let's let it land
21:11:14 <zaitcev> It was trivial and yet Sam already found a way to simplify it further.
21:11:14 <acoles> yep it only touches the middleware
21:12:04 <notmyname> container sharding is a huge feature, both in mental scope and in the amount of code added/changed
21:12:34 <notmyname> so it's going to take a while to review it. it's not like we'll have it approved by the end of the week
21:13:15 <notmyname> my hope (in a "wouldn't it be nice" sense) is that we'll be able to land container sharding on master by may 17
21:13:31 <notmyname> that's two weeks from tomorrow (in my time zone)
21:13:40 <clayg> do eet
21:13:59 <notmyname> one reason the May 17 date would be really neat is because may 17 is swift's birthday!
21:14:23 <kota_> sounds great :)
21:14:27 <notmyname> swift was first put into production and accepted user requests on may 17, 2010
21:14:44 <mattoliverau> They would be cool
21:15:21 <acoles> there are a few todo's on deep-review, for example where there are still unit tests to add. sorry I didn't get them all done, hopefully they are 'trivial' and I'll add them as we roll new patchsets.
21:15:22 <notmyname> so that being said, here's how I imagine the next two weeks going (spoiler, it's pretty much the same as the last time we did a big feature branch merge)
21:15:40 <notmyname> acoles is managing the feature/deep-review patch chain
21:15:54 <notmyname> I have added a -2 to the first one. this is only to "cork" it so that nothing lands
21:16:09 <notmyname> please review and add your reviews and comments on every patch
21:16:23 <acoles> and of course feel free to post a diff or follow on patch for me to squash in if you'd like to add tests, but please don't push over the existing patches.
21:16:42 * acoles know everyone loves writing tests :P
21:16:45 <notmyname> when we have a +A on every patch, we'll uncork the first one and everything will land. then we'll do a single merge commit to master and be done with it
21:16:58 <notmyname> yes, please include a suggested diff when leaving a review asking for an improvement
21:17:13 <notmyname> acoles: you expect to roll new patch sets every couple of days?
21:17:27 <mattoliverau> How many +2s? So it's in meeting logs
21:17:54 <acoles> notmyname: yes, depending on how much review traffic there is
21:18:00 <notmyname> acoles: ok, sounds good
21:18:03 <notmyname> mattoliverau: great question
21:18:38 <notmyname> at this point in the history of our community, we're smaller than we've been in the past. I don *not* think we need a +2 from everyone before adding a +A
21:18:51 <notmyname> however, I think we need 3-4 +2s for each patch
21:19:13 <zaitcev> ouch
21:20:33 <notmyname> with clayg, tdasilva, torgomatic, zaitcev, mattoliverau, and kota_ available to review (and maybe some of cschwede?) we should be able to make progress and get good reviews
21:20:44 <clayg> on i!
21:21:07 <notmyname> tim has had his baby, so he's not available for reviews. also, good news everyone! tim has replicated!
21:21:32 <kota_> wow! congrats Tim!
21:21:32 <mattoliverau> \o/ congrats
21:22:37 <notmyname> also, with reviews, note that the last patch in the chain is docs. please do not leave negative reviews on that patch. instead, please push over the patch with your changes and updates
21:23:09 <acoles> the docs need updating - I'll do that asap
21:23:35 <notmyname> in next week's meeting, we'll have a good check of where we are. and then if all the programming gods are smiling on us, then the following meeting we'll be celebrating the impending merge
21:24:03 <notmyname> are there any questions about the next two weeks of reviews on feature/deep-review? acoles, do you have anything to add or advice/requests for all of us?
21:24:36 <acoles> I'll try to answer questions
21:25:13 <acoles> reminder there is an overview of the patch chain here https://etherpad.openstack.org/p/deep-review
21:25:41 <acoles> feel free to expand that etherpad as you learn things
21:26:33 <notmyname> doesn't seem like there are any questions right now
21:26:35 <notmyname> :-)
21:26:45 <kota_> 1 thing
21:26:59 <notmyname> kota_: ok
21:27:10 <kota_> if you know, can i get the order of the patch chain?
21:27:27 <kota_> https://review.openstack.org/#/q/status:open+project:openstack/swift+branch:feature/deep-review seems not to be sorted...maybe?
21:27:48 <kota_> and /me is looking https://etherpad.openstack.org/p/deep-review yet...
21:28:18 <acoles> hmm I'm getting an error from gerrit right now
21:28:33 <mattoliverau> I have a pretty good grasp on the topic, so can try help you with doc acoles. Even if it's a bunch of content that you can further correct if I go wrong in places.
21:28:34 <notmyname> yeah, and I'm having issues with etherpad loading
21:28:36 <acoles> but the patches are in numerical sequence
21:29:34 <acoles> this is the first in the chain https://review.openstack.org/#/c/565742/
21:29:35 <patchbot> patch 565742 - swift (feature/deep-review) - Refactoring, test infrastructure changes and cleanup
21:29:48 <acoles> I'll annotate the etherpad with patch links
21:29:55 <zaitcev> thanks
21:30:03 <notmyname> acoles: thanks
21:30:16 <kota_> ok. and just an idea i had is fetching the last one (i.e. docs) then, check the order in the git history.
21:30:59 <kota_> thanks acoles
21:31:04 <acoles> so once again, and with apologies, the docs may be out of date in places - but in principle still correct - just some details and naming has changed
21:31:33 <kota_> ok
21:31:44 <acoles> also there is a glossary of terminology here https://etherpad.openstack.org/p/deep-containers and some notes to run probe tests
21:32:21 <notmyname> acoles: you've done a fantastic job preparing all of this for review, not to mention all the hard work of actually getting the thing written. thank you
21:32:53 <notmyname> any other questions from anyone?
21:33:56 <notmyname> all right then
21:34:11 <notmyname> thank you for coming today, and thank you for your work on swift
21:34:21 <notmyname> I'm excited about seeing container sharding land in swift
21:34:26 <notmyname> mattoliverau: it's finally happening!!
21:34:30 <notmyname> #endmeeting