19:00:12 <notmyname> #startmeeting swift
19:00:13 <openstack> Meeting started Wed Apr  9 19:00:12 2014 UTC and is due to finish in 60 minutes.  The chair is notmyname. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:00:14 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
19:00:16 <openstack> The meeting name has been set to 'swift'
19:00:23 <notmyname> hello, everyone
19:00:25 <notmyname> who's here?
19:00:26 <peluse> hey hey hey
19:00:29 <creiht> howdy
19:00:36 <cschwede> Hello!
19:00:42 <portante> ola
19:01:07 <acoles> hi
19:01:18 <peluse> spoken like someone from back east :)  (its hola)
19:01:19 <notmyname> just a few things to go over this week https://wiki.openstack.org/wiki/Meetings/Swift
19:01:53 <notmyname> #topic openstack icehouse integrated release
19:02:08 <notmyname> We cut 1.13.1. it's the icehouse RC
19:02:12 <portante> peluse: cute
19:02:23 <notmyname> next thursday (the 17th) is the openstack release date
19:02:52 <notmyname> so if you have anything that needs to be backported, speak now or forever hold your peace
19:03:06 <creiht> we aren't aware of anything yet
19:03:11 <notmyname> (ie propose the patch to gerrit and shout about it in IRC)
19:03:11 <creiht> but we should start testing soon
19:03:13 <notmyname> creiht: cool
19:03:36 <notmyname> any questions on icehouse?
19:04:00 <notmyname> ok, next up
19:04:07 <zaitcev> what's the leading new feature now that SP miss it
19:04:10 <notmyname> #topic summit in atlanta
19:04:13 <clayg> i forget we do these every week now
19:04:36 <notmyname> zaitcev: I'll be updating https://wiki.openstack.org/wiki/ReleaseNotes/Icehouse#OpenStack_Object_Storage_.28Swift.29 this week
19:04:56 <peluse> clayg:  not this week you didn't I see...
19:05:00 <notmyname> zaitcev: but the good news is that there are 6 months of releases (ie not just 1.13.1)
19:05:30 <notmyname> for the summit, if you haven't registered and are planning on going, do so now. also buy flights and hotels, etc
19:05:41 <clayg> swing and a miss!
19:05:48 <notmyname> we currently have 18 session proposals at http://summit.openstack.org
19:06:12 <notmyname> if you have other things, please propose it soon. I'll need to start culling through them soon
19:06:30 <notmyname> any questions about the summit?
19:06:46 <creiht> notmyname: any change on the number of slots we get?
19:07:05 <clayg> creiht: how many do we get?
19:07:15 <clayg> creiht: and - UNCONFERENCE!  whooooo!
19:07:32 <notmyname> zaitcev: FYI here was a scratchpad I used for tracking swift-in-icehouse stuff https://gist.github.com/anonymous/606068be20d6fa314b5a
19:07:34 <creiht> now with more UN!
19:07:42 <notmyname> creiht: clayg: AFAIK still just with the 8
19:07:56 <creiht> :(
19:08:00 <ttx> there actually isn't any UNconference this time around
19:08:04 <notmyname> but at this summit there are "pods" or tables or something that will be more dedicated to use
19:08:11 <clayg> ttx: you can't stop us :P
19:08:11 <notmyname> ttx: hello! you have the answer here
19:08:14 <ttx> there is a project pod for continued discussions
19:08:22 <creiht> ttx: that's why I was saying now with more UN! :)
19:08:24 <ttx> (for all days)
19:08:29 <notmyname> that thing. ttx can you give more info on that?
19:08:33 <ttx> more UN!
19:08:51 <ttx> shall be a roundtable and a paperboard
19:09:08 <peluse> sounds so formal :)
19:09:12 <zaitcev> I settled in Quality Hotel about 1100m from the site entrance. By far the cheapest in that whole blasted town.
19:09:13 <notmyname> ttx: in a separate room or everyone's "pod" is in the same room?
19:09:14 <ttx> I don't have a lot of precisions on how much space there will be around it
19:09:29 <notmyname> zaitcev: and you know it's good because "quality" is in the name!
19:09:34 <notmyname> ttx: ok, thanks
19:09:40 <ttx> multiple roomsn, but several tables per room
19:09:43 <q3k> bgorski: o/
19:10:18 <notmyname> any other questions about the summit?
19:10:22 <ttx> notmyname: if you want to make a lot of use for it, we could try to place you in a large corner
19:11:00 <zaitcev> Not specifically Swift, but what's going on on Friday? Schedule does not cover it, but other places say Design Summit is on Friday.
19:11:03 <ttx> maybe you cna use the wall and a picoprojector to do a poor man's design summit room :)
19:11:24 <clayg> ttx: I think what we *really* want is for swift to have more slots than 8, but because we have so many people and so much going on we'll find *somewhere* if we can't get the time we need on the schedule ;)
19:11:24 <notmyname> ttx: I certainly won't turn down a corner if it's offered :-)
19:11:25 <ttx> zaitcev: design summit statrs Tuesday and ends Friday
19:11:25 <creiht> ttx: or just give us some more slots? :)
19:12:03 <ttx> creiht, clayg: well.. swift already has the buggest ratio of slots per commit in a cycle, by far
19:12:23 <creiht> ttx: oh if it is based on commits, I can fix that real quick ;)
19:12:24 <clayg> oh, that's an interesting way to look at it...
19:12:25 <notmyname> new plan! each line of code as a new commit
19:12:28 <ttx> and we wanred that cross-project stuff would result in less sessions, overall
19:12:33 <torgomatic> oh, are we doing things per commit? I HAVE THE ANSWER! :)
19:12:47 <zaitcev> Ah-ha! So my effort to split up PBE into a dozed of miniature commits is going to help us politically, too!
19:12:48 <creiht> ttx: and yeah that sounds like a really bad metric to use
19:12:57 <ttx> it's not how we assigned stuff. Otherwise swift would have got 3 slots.
19:13:04 <clayg> heh
19:13:05 <ttx> or 2.
19:13:06 <peluse> ouch
19:13:16 <gholt> Hehe (I giggled at that at least)
19:13:18 <zaitcev> everyone who wants more slots go approve https://review.openstack.org/85909
19:13:28 <notmyname> yikes. (I had no idea that's how things were allocated)
19:13:33 <creiht> ttx: if it was measeard by quality commits, then swift would have the majority ;)
19:13:49 <ttx> notmyname: it's not. just a metric I looked at to see how far we were
19:13:50 <clayg> notmyname: he said it *wasn't* how it was allocated
19:14:12 <notmyname> creiht: that's not fair to a lot of people ;-)
19:14:29 <creiht> ttx: so how is the number of slots for each project computed?
19:14:30 <notmyname> ttx: clayg: right. I mean I didn't knwo that was an input to the function
19:14:33 <ttx> the allocation is more... tradition, with some feedback sprinkled on top. I know swift usually has a lot of sessions
19:14:45 <clayg> ttx: well you can't blame us for trying - i'm sure every project wants more slots, but it's going to keep getting contentious
19:14:50 <ttx> so I tried to preserve as many as I could
19:15:14 <gholt> Before long we'll have the OpenStack Storage Summit, etc.
19:15:15 * notmyname suspects this is the last time ttx will drop by a swift team meeting
19:15:21 <ttx> that said, if neighboring projects in the schedule use less, I already told notmyname he could have those extra slots
19:15:22 <creiht> haha
19:15:34 <clayg> gholt: genious!  we are doing that hackathon in June...
19:15:40 <ttx> that might make +1 or +2
19:15:43 <peluse> I was gonna say!
19:15:44 <zaitcev> gholt: We already have Swift Hackathon and I was going to miss Atlanta
19:15:54 <ttx> notmyname: especially in my evening.
19:16:16 <gholt> zaitcev: Damn, I took the opposite option. Silly me.
19:16:33 <peluse> gholt:  free beer in Colorado
19:16:39 <clayg> peluse: yum
19:16:52 <notmyname> ttx: do you know when the schedule will be finalized?
19:16:52 <ttx> notmyname: anyway, will try to save you a nice spot for the project pod
19:17:09 <ttx> notmyname: I expect PTLs to work on schedlue the week after release
19:17:27 <notmyname> ttx: thanks.
19:17:28 <ttx> I hope they will all publish by April 25
19:17:33 <notmyname> ok
19:18:31 * ttx exits the room and rubs the tomato stains from his face
19:18:36 <creiht> awww
19:18:36 <notmyname> lol :-)
19:18:46 <notmyname> ttx: thanks for coming by.
19:18:54 <creiht> ttx: yes thanks for answering the questions
19:18:56 <creiht> :)
19:19:10 <ttx> np :)
19:19:21 <notmyname> ttx: your input on what's happening at the summit is great
19:19:37 <notmyname> #topic hackathon in june
19:19:42 <notmyname> speaking of this...
19:20:02 <notmyname> https://www.eventbrite.com/e/openstack-swift-june-hackathon-tickets-8309569145
19:20:49 <notmyname> just a few spaces left
19:20:53 <notmyname> this is more a reminder
19:21:00 <notmyname> peluse can answer any questions you have
19:21:15 <peluse> fire away anytime..
19:22:34 <notmyname> #topic storage policy updates
19:22:45 <notmyname> now for the good stuff :-)
19:23:01 <notmyname> clayg: peluse: have been working on this
19:23:15 <notmyname> and we're very close to proposing stuff to master
19:23:25 <peluse> looking good on my end.  clayg, https://review.openstack.org/#/c/79731/ is as much yours as it is mine at this point but I think its ready for next steps
19:23:31 <clayg> I'm still adding polish to the reconciler and reviewing a few remaining outstanding items that peluse and his tem have been working on
19:23:41 <notmyname> clayg and I talked in the office this morning about it. clayg can you share some about that?
19:23:48 <notmyname> ie the pre-SP patches to master
19:23:58 <peluse> typing like mad on docs today - hope to have content before EOD tomorrow for review
19:24:09 <notmyname> peluse: great!
19:24:13 <clayg> peluse: I'm going to revisit that one shortly, I'm about done with the reconciler enqueue stuff enough to do the object-updater bits.
19:24:32 <peluse> clayg:  fantastic, let me know when it makes sense to start digesting the reconciler
19:24:47 <clayg> We need to start widdling away at the delta between master and feature/ec
19:25:06 <notmyname> whittling
19:25:09 <clayg> ^ that
19:25:15 <portante> FWIW, the in-process functional tests would be useful to help us get functional test coverage of the SP stuff
19:25:22 <notmyname> agreed
19:25:25 <clayg> portante: +1
19:25:54 <peluse> portante:  you've got something for that right?
19:26:03 <clayg> anyway, I think there's lots of "drive-by" changes (mostly to tests or just good ol' method extraction refactoring) that would look good on master by themselves
19:26:07 <portante> yes, all up and ready to be committed
19:26:09 <portante> peluse
19:26:21 <peluse> cool, will have a look.  thanks
19:26:34 <clayg> so I'm going to try and start putting those reviews up under a common topic so we can have a list of "pre-sotrage-policy" changes for folks that want to help with storage-policy reviews
19:26:35 <portante> clayg: are you breaking those out?
19:26:44 <portante> great
19:26:58 <clayg> portante: I have one already, and yuan has been doing some stuff too I think
19:27:13 <peluse> clayg:  you're sticking with the patch series that torgomatic started though right? (just updating)
19:27:17 <notmyname> and I'll be listing them on https://wiki.openstack.org/wiki/Swift/PriorityReviews (hopefully with a link to show them all in gerrit too, if the topic thing works)
19:27:19 <clayg> I need to go back to the earlier storage-policy commits and see if there's anything that could come out
19:28:07 <clayg> peluse: well torgomatic's sp-base branch is pretty out of date with feature/ec's merges
19:28:31 <peluse> yeah, so I was asking if you meant you're updating that?  Or, are you creating a new series or something else?
19:28:59 <clayg> peluse: but yes I still think a patch series for the storage policy change in compostable pieices will be the best strategy for reviewers
19:29:20 <clayg> peluse: I'll revisit it, but not until the container reconciler is finished
19:29:43 <cschwede> clayg: so you’ll split the existing patches?
19:30:00 <peluse> clayg:  OK, well for the next 24 or so I'll be doing docs and other stuff and obviously you've got a pile of work.  Let me know when we're ready to attack that patch set and what I can do to help.  Was planning on re-doijg the commit messages to make them flow better
19:30:10 <clayg> peluse: in the meantime I'm hoping we can cherry pick out of the feature/ec branch little bits that were just cleanup and pre-req's
19:30:27 <peluse> clayg:  Oh, and propose to master?
19:31:14 <clayg> peluse: yes that's a bigger priority for me that merging feature/ec to torgomatic's sp-base and rebasing on current master - but both are part of the endgame strategy
19:31:29 <notmyname> clayg: when do you expect to start proposing the little bits to master?
19:31:33 <peluse> gotcha
19:31:35 <notmyname> later this week or next week?
19:32:00 <clayg> well I'm not sure, i have https://review.openstack.org/#/c/84621/ now
19:33:11 <clayg> notmyname: but in particular as I'm tieing all the merge-storage-policy-index and container-reconciler changes together it seems like there's room to simplify by pulling out changes to other commits, if something falls out I'll try submitting to master with the pre-storage-policy topic
19:33:21 <notmyname> ok
19:33:38 <peluse> you da man
19:33:49 <notmyname> clayg: let me know when you propose the first of those
19:33:59 <clayg> notmyname: if you or peluse had bandwidth to give the unified diff of feature/ec and master a once over looking for changes that could collapse out I think every line we can pull out of the storage policy patch train will help
19:34:23 <clayg> notmyname: the first of the ones with the topic?  i.e. keep status_changed_at up-to-date doesn't count?
19:34:42 <notmyname> clayg: ya, with the topic (that other one totally counts though)
19:34:50 <clayg> notmyname: will do
19:34:53 <peluse> clayg:  I'll try to spot opportunities as I authro doc content, I'm doing a feature/ec - master diff as I write
19:35:28 <clayg> peluse: oh nice, super helpful!
19:35:35 <notmyname> what questions are there from not peluse or clayg? this is getting closer to mainline, so more people will be involved
19:36:01 <notmyname> I want to make sure we're on the same page and expectations are set properly
19:36:15 <cschwede> i hope i can help with smaller patches for sp
19:36:24 <cschwede> eh, i mean reviewing smaller patches
19:36:38 <cschwede> so i think this will be very helpful
19:36:46 <clayg> cschwede: hopefully we'll hve a topic for you by next meeting!
19:37:29 <notmyname> creiht: gholt: portante: zaitcev: are you ok with the proposed process?
19:37:33 <notmyname> or is it still unclear?
19:38:06 <creiht> notmyname:  I guess we will find out :)
19:38:12 <portante> notmyname: yes
19:38:18 <creiht> seems reasonable, but I guess we will see how it goes
19:38:28 <peluse> FYI I'll post the doc content submission to IRC when ready and hopefully it will paint a decent picture of what's been done and why to help tie all of this together
19:38:39 <portante> I think the assumption in this process is that we'll flag issues that come up as we go, and adapt
19:38:48 <notmyname> creiht: oh, I think we're all learning the right process as we go here :-)
19:38:57 <notmyname> portante: ya, exactly
19:39:05 <peluse> inventing might be a better word than learning :)
19:39:10 <notmyname> peluse: thanks. I think that's helpful
19:39:16 <notmyname> peluse: yes :-)
19:39:37 <notmyname> anything else on SP for right now?
19:40:09 <notmyname> #topic open discussion
19:40:19 <notmyname> anything else that needs to be brought up here today?
19:41:28 <portante> I don't have anything on this end
19:41:36 <notmyname> I'm thinking silence is a good thing ;-)
19:41:39 <notmyname> ok, then
19:41:45 <clayg> thanks all!
19:41:49 <notmyname> thanks for coming. talk to you next week!
19:41:49 <peluse> later, thanks
19:41:58 <notmyname> #endmeeting