21:00:28 <notmyname> #startmeeting swift
21:00:29 <openstack> Meeting started Wed Oct  5 21:00:28 2016 UTC and is due to finish in 60 minutes.  The chair is notmyname. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:00:30 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:00:32 <openstack> The meeting name has been set to 'swift'
21:00:35 <notmyname> who's here for the swift team meeting?
21:00:40 <torgomatic>21:00:42 <dfg_> hey
21:00:43 <mattoliverau> o/
21:00:44 <jrichli> hello
21:00:47 <kota_> hi
21:00:48 <bkeller`> o/
21:00:48 <nadeem> hello
21:00:57 <mathiasb> o/
21:00:57 <hosanai> o/
21:01:34 <notmyname> welcome
21:01:42 <notmyname> a few less people this week than normal
21:01:48 <tdasilva> hi
21:01:53 <sgundur> hi
21:01:55 <dmorita> hi
21:02:04 <ntata> hello
21:02:13 <notmyname> oh, heh. here comes everyone else :-)
21:02:20 <m_kazuhiro> hi
21:02:24 <pdardeau> o/
21:02:24 <notmyname> agenda for this week is at
21:02:25 <notmyname> #link https://wiki.openstack.org/wiki/Meetings/Swift
21:02:48 <notmyname> #topic TC election
21:02:53 <notmyname> #link https://wiki.openstack.org/wiki/Meetings/Swift
21:02:56 <acoles> here
21:03:00 <notmyname> the tc election is going on now.
21:03:24 <notmyname> please read over the candidate statements and where candidates have been involved in the past
21:03:41 <notmyname> please vote. you should have a ballot in your email
21:03:54 <notmyname> lots of candidates for 6 spots
21:04:03 <notmyname> any questions about the election before we move on?
21:04:09 <mattoliverau> yeah, alot of reading, but worth it me thinks
21:04:24 <kota_> the link should be  https://governance.openstack.org/election/ ?
21:04:33 <notmyname> oh yeah, soory
21:04:35 <notmyname> #undo
21:04:35 <openstack> Removing item from minutes: <ircmeeting.items.Link object at 0x7f66f50bd8d0>
21:04:42 <timburke> if you didn't receive a ballot but believe you should have, contact the election officials per http://lists.openstack.org/pipermail/openstack-dev/2016-October/105002.html
21:04:43 <notmyname> #link https://governance.openstack.org/election/
21:05:28 <notmyname> ok, if no other questions, let's move on
21:05:34 <notmyname> #topic newton release
21:05:44 <notmyname> tomorrow is the big day for the openstack newton release
21:05:50 <notmyname> swift 2.10.0 will be in it
21:06:19 <notmyname> I didn't see anything come up since we tagged that which would necessitate another tag, so at this point, anything we find will be backports
21:06:34 <notmyname> thank you, everyone, for your work on swift during the openstack newton cycle
21:06:50 <notmyname> one last thing about this release... release notes and reno
21:07:13 <notmyname> in order to get a link on the right page, we need another sphinx doc tree in our repo. I added it in https://review.openstack.org/#/c/381382/
21:07:42 <notmyname> when that lands, it can be backported and then the URL will be created and I can update the release repo to include a link
21:08:03 <notmyname> any questions on the newton (or swift 2.10.0) release before we move on?
21:08:34 <notmyname> ok
21:08:38 <notmyname> #topic summit topics
21:09:00 <notmyname> barcelona is soon; please fill out the agenda etherpad
21:09:02 <notmyname> #link https://etherpad.openstack.org/p/ocata_swift_summit_topics
21:09:35 <notmyname> there's quite a few good things listed, but I know there's a lot more going on (and we have plenty of time). please add stuff ASAP -- this week if possible
21:09:55 <notmyname> I will start setting the schedule next week, and it will be based on that etherpad
21:10:04 <notmyname> so if you want to talk about it, write it down! :-)
21:11:04 <notmyname> in-progress patches, current designs for long-term stuff, ideas for new things, social/community improvements
21:11:08 <notmyname> all good stuff to include
21:11:48 <notmyname> again, any questions here?
21:12:02 <mattoliverau> we seem to be a chatty bunch today
21:12:09 <acoles> looks like we have sessions all day thurs and friday https://docs.google.com/spreadsheets/d/1TQ-RSlbiBBEclkonIbfUP7R1ExZSJylF1uiEKV2G_Cw/pubhtml?gid=1107826458&single=true
21:12:16 * notmyname thinks this whole meeting is "<notmyname>: blahh. and 'any questions?'"
21:12:26 <mattoliverau> lol
21:12:36 <notmyname> acoles: correct
21:12:44 <acoles> and a late finish friday
21:12:52 <notmyname> we have 2 full days
21:13:03 <notmyname> tuesday and wednesday are cross-project and ops sessions
21:13:38 <notmyname> we start with 2 fishbowls (eg our previous ops-feedbask sessions) and have the rest as working sessions
21:14:18 <notmyname> like the last couple of summits, I'll group topics into bunches and schedule them for blocks of time. eg 11-12:30 and 1:50-4:10 so we can get a lot of things covered
21:14:47 <mattoliverau> and that worked awesomely!
21:15:00 <notmyname> I liked it. seems like a lot of other people did too :-)
21:15:21 <notmyname> also, if you have a conflict (eg a talk you're giving), please mention it at the bottom
21:16:05 <acoles> someone suggested last time that people have a chance to give a  v short elvator pitch for a topic that we may not all be familiar with, to attract interest
21:16:18 <acoles> hmm, or was that at hackathon?
21:16:31 <notmyname> either way, it's a good idea :-)
21:16:34 <acoles> anyway, seems it could be useful
21:16:36 <mattoliverau> +1
21:16:53 <notmyname> we could do that at the end of the last fishbowl, perhaps, as a teaser for the rest of the sessions
21:17:02 <notmyname> depends on what other fishbowl we have scheduled
21:17:14 <acoles> like, we probably don't need it for sharding?, but if someone is bringing a newish topic
21:17:55 <mattoliverau> Barcelona is quite far, so dont know how many people are making the trip, maybe those who can't make it but want to give an elevator can still put a topic down and provide some info (link) and someone will chapion it ;)
21:18:08 <mattoliverau> acoles: you just dont want to hear me talk again :P
21:18:18 <notmyname> it's not bad to have a quick intro to each of them, because there will be new people there too
21:18:33 <notmyname> but still needs to be 2 sentences!
21:18:40 <acoles> mattoliverau: no, I want to hear you talk lots more than an elevator ride would allow ;)
21:19:14 <mattoliverau> acoles: I'll even provide some background elevator music if you'd like.. for the _whole_ session :P
21:19:30 <notmyname> dfg_: will anyone from rax be there?
21:19:47 <dfg_> barcelona? not from over here. mattoliverau?
21:20:11 <notmyname> dfg_: ok. yeah, was thinking of gophers and hummingbird
21:20:14 <mattoliverau> I just made the cut.. but many of my colleges here in Oz didin't
21:20:24 <dfg_> notmyname: no i don't think so
21:20:27 <notmyname> kk
21:20:38 <notmyname> I mean, that's not ok, but ack to your answer ;-)
21:20:45 <dfg_> :)
21:21:16 <notmyname> I'd like to have some discussion on the golang work in barca, though. we'll need to figure out how to do that
21:21:56 <dfg_> ya- just let us know.
21:22:01 <notmyname> ok
21:22:31 <nadeem> maybe video conferencing like we did in previous conference
21:22:49 <notmyname> not sure the facilities we'll have.
21:22:55 <nadeem> true
21:23:15 <notmyname> but let's figure out what all is proposed and what can work and who/how to have the conversation
21:23:35 <nadeem> cool
21:23:55 <notmyname> anything else about the summit or summit planning someone would like to bring up in this meeting today?
21:25:52 <notmyname> ok, moving on
21:25:58 <notmyname> #topic top patches and bugs
21:26:06 <notmyname> ok, more time this week than last :-)
21:26:36 <notmyname> we've done several things as a group to help prioritize patches and bugs in the past
21:27:09 <notmyname> as I said last week, i want to start highlighting just a few things in the meeting so that they get some awareness (and hopefully some reviews/patches)
21:27:18 <notmyname> but, first, and importantly
21:27:49 <notmyname> if you are working on something different than something mentioned as a "priority" in here (or elsewhere), that's ok. doesn't take away from your work at all
21:28:08 <notmyname> we all have different forces pulling us in different ways, asking us to do different things upstream
21:28:40 <notmyname> the point here is not to shame anyone in to writing or reviewing code, but simply highlight some things going on
21:29:03 <notmyname> ok, so last week, right at the end of the meeting, I highlighted these
21:29:11 <notmyname> https://bugs.launchpad.net/swift/+bug/1328735 -- Object-updater gives up updating container with no success if all containers are placed at handoff
21:29:12 <openstack> Launchpad bug 1328735 in OpenStack Object Storage (swift) "Object-updater gives up updating container with no success if all containers are placed at handoff" [High,In progress] - Assigned to Takashi Kajinami (kajinamit)
21:29:19 <notmyname> which has a patch at https://review.openstack.org/#/c/99598/
21:29:33 <notmyname> and https://bugs.launchpad.net/swift/+bug/1624088 -- EC missing durable can prevent reconstruction
21:29:33 <openstack> Launchpad bug 1624088 in OpenStack Object Storage (swift) "EC missing durable can prevent reconstruction" [High,Confirmed]
21:29:43 <notmyname> which has a fix proposed at https://review.openstack.org/#/c/376630/
21:30:06 <notmyname> those are pretty big issues, currently each at "high" priority
21:30:31 <notmyname> I also pointed out https://review.openstack.org/#/c/210099/ -- process level concurrency for container sync
21:31:03 <notmyname> there's 2 more I wrote down this week:
21:31:11 <notmyname> https://review.openstack.org/#/c/355958/ -- eliminate .durable files
21:31:11 <notmyname> https://review.openstack.org/#/c/376630/ -- reconstruct with non-durable fragments
21:31:26 <notmyname> whew. 5 is enough :-)
21:31:47 <acoles> the last is a repeat of the fix for bug 1624088
21:31:49 <openstack> bug 1624088 in OpenStack Object Storage (swift) "EC missing durable can prevent reconstruction" [High,Confirmed] https://launchpad.net/bugs/1624088
21:32:40 <notmyname> ah, thanks
21:32:43 <kota_> wow, it looks like mine is getting low.
21:32:44 <notmyname> double important :-)
21:33:21 <notmyname> kota_: your global EC work is important. but I'd like to see the elimination of the .durables first, and the two patches conflicted
21:33:39 <kota_> notmyname: fair point
21:33:49 <mattoliverau> pre-release I was reviewing patch 355958 (.durables), I'll swing back round to that while it's still reletively close in my head :)
21:34:02 <kota_> let's land eliminate durable asap :P
21:34:16 <notmyname> so on the general topic of EC, fixing the reconstructor and eliminating the .durables is a big deal
21:34:29 <acoles> IIRC most of the change is in test_diskfile :/
21:34:32 <mattoliverau> kota_: its almost our timezone... just saying ;)
21:34:47 <tdasilva> lol
21:35:05 <notmyname> mattoliverau: thanks. and torgomatic has also already put a +1 on that patch too
21:35:08 <kota_> acoles: the conflicting area is just proxy/test_server.py :/
21:35:32 <acoles> kota_: yep, sorry! we should stop writing tests, they just cause problems ;)
21:35:44 <mattoliverau> \o/ no more tests!
21:36:14 <kota_> acoles: yeah, it might be good to make a talk at barcelona how we can cleanup test_server.py which is so long line test cases.
21:36:36 <notmyname> EC (fix reconstructor, eliminate .durable), fix object-updater if containers on handoff, finally land container sync concurrency
21:36:38 <notmyname> that's the summary
21:36:40 <kota_> acoles: and freaquently making conflicts.
21:37:12 <acoles> kota_: +1
21:37:26 <mattoliverau> kota_: add it to the etherpad :)
21:37:33 <notmyname> you know, landing those things would be worthy of doing a release to get them out the door
21:37:40 <kota_> acoles, mattoliveau: :)
21:38:33 <notmyname> if anyone is looking for something to review, those are great patches to start with
21:38:38 <acoles> notmyname: we should land this before another release https://review.openstack.org/379649
21:38:54 <acoles> fixes a func test failure
21:39:05 <notmyname> acoles: ah, good
21:39:11 <notmyname> I'll take a look at that
21:39:27 <acoles> not in gate but fails if you have >1 policy configured
21:39:42 <acoles> kota_ has already +2
21:39:48 <notmyname> sounds like we need to get another policy in the gate :-)
21:40:17 <acoles> I wrote a bug report with some thoughts for that ;) think it is linked off that patch
21:40:20 <notmyname> speaking of stuff in the gate, I want to get xfs on the CI workers so we can get torgomatic's "checksum metadata" patch landed
21:40:25 <notmyname> acoles: oh, good
21:40:33 <torgomatic> that'd be nice, wouldn't it?
21:41:00 <notmyname> torgomatic: yep. i'll do my best to get it done without needed a TC governance resolution
21:41:05 <acoles> https://bugs.launchpad.net/swift/+bug/1629234
21:41:06 <openstack> Launchpad bug 1629234 in OpenStack Object Storage (swift) "Add a second policy to in process functional tests" [Wishlist,New]
21:41:19 <torgomatic> notmyname: and that sentence right there is why I've been avoiding dealing with that patch :/
21:41:24 <mattoliverau> what me to poke my infra workmate to get some ideas (or how hard it could be)
21:41:29 <notmyname> torgomatic: i know :-)
21:41:53 <notmyname> mattoliverau: not yet. it's not a lack of knowledge on how to get it done...we can go over it after the meeting, if you want
21:42:18 <mattoliverau> right kk
21:42:36 <notmyname> but since it involves CI image changes, I've held off while the combined releases were being done
21:42:47 <notmyname> #topic open discussion
21:42:59 <notmyname> anything to bring up? on the patches/bugs or anything else?
21:44:24 <mattoliverau> update: I've almost got a readonly large database sharding POC running, hopefully get it running by the end of the week and see if I can start getting some numbers for barcelona
21:44:43 <notmyname> great
21:44:52 <acoles> mattoliverau: nice
21:44:55 <mattoliverau> coding is "done" there is more to tweak, but enough to start seeing how well it works
21:45:32 <mattoliverau> **cough** famous last words **cough**
21:46:00 <kota_> lol
21:46:35 <notmyname> ok, last call for something else to talk about before this meeting is over
21:46:39 <nadeem> Also we looked at https://etherpad.openstack.org/p/humming-replication-server and have decided on Solution B. I will be coding for that.
21:46:49 <notmyname> nadeem: "we"?
21:47:11 <nadeem> dfg_ ,me & I guess clay too like this option
21:47:12 <notmyname> nadeem: also, good. i was hoping for that one :-)
21:47:33 <nadeem> cool
21:48:08 <notmyname> anything else?
21:48:39 <notmyname> thanks everyone for coming this week. please continue to fill out https://etherpad.openstack.org/p/ocata_swift_summit_topics
21:48:42 <notmyname> also, add ideas to https://etherpad.openstack.org/p/ocata_swift_summit_topics
21:48:48 <notmyname> and comment on https://etherpad.openstack.org/p/ocata_swift_summit_topics with what you like
21:49:17 <notmyname> and, as always, thanks for working on swift
21:49:23 <notmyname> #endmeeting