19:30:20 <notmyname> #startmeeting swift
19:30:21 <openstack> Meeting started Wed Oct  8 19:30:20 2014 UTC and is due to finish in 60 minutes.  The chair is notmyname. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:30:22 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
19:30:24 <openstack> The meeting name has been set to 'swift'
19:30:33 <notmyname> hello, who's here froe the swift team meeting?
19:30:35 <elambert> o/
19:30:36 <mattoliverau> o/
19:30:37 <cschwede> o/
19:30:49 <acoles> \o
19:31:17 <torgomatic> hi
19:31:19 <peluse> yo
19:31:32 <notmyname> thanks for coming :-)
19:31:38 <notmyname> #link https://wiki.openstack.org/wiki/Meetings/Swift
19:31:43 <cutforth> here
19:32:02 <notmyname> thanks for letting us start late. we had a company all-hands that overlapped
19:32:04 <lpabon> here
19:32:32 <notmyname> hmm..looks like I forgot to add the hackathon review to the agenda. we'll talk about that too
19:32:37 <notmyname> ok, first up..juno release
19:32:50 <notmyname> #topic swift 2.2.0 (juno integrated release)
19:33:09 <notmyname> thansk for helping get patches written, reviewed, and merged
19:33:25 <notmyname> the RC-1 for swift 2.2.0 was cut early this week
19:33:54 <notmyname> unless something else comes up, that will be the final release for 2.2.0 and what is included in the juno integrated release
19:34:10 <notmyname> juno is scheduled to be cut on the 16th (one week from tomorrow)
19:34:46 <notmyname> so....although I'm pretty confident about the state of it, please continue to look at ti, do QA, and (quickly!) report any showstoppers you find
19:35:29 <notmyname> #link https://github.com/openstack/swift/blob/master/CHANGELOG
19:35:31 <zaitcev> is there a summary of highlights for Juno? A PM asked me and I said "umm... Keystone v3"
19:35:38 <notmyname> zaitcev: there ^^
19:35:40 <notmyname> :-)
19:35:45 <ppai> here
19:35:49 <notmyname> juno includes swift 2.0 - 2.2
19:35:59 <notmyname> zaitcev: storage policies are kinda big ;-)
19:36:54 <notmyname> here were my notes for a phone call I had with the foundation about that very question (what's new/big in juno for swift): https://gist.github.com/notmyname/918dec4be5424275934f
19:36:59 <notmyname> zaitcev: ^^
19:37:27 <notmyname> any questions about either the juno release or specifically about swift 2.2.0?
19:37:31 <ppai> zaitcev, https://github.com/openstack/swift/blob/master/CHANGELOG#L1-L173 ;)
19:37:53 <notmyname> ppai: cool! I didn't know you could link ranges in github :-)
19:38:20 <ppai> :)
19:38:28 <notmyname> ok, moving on then to summary of the hackathon
19:38:34 <notmyname> #topic hackathon followup
19:38:51 <notmyname> first, again I'd like to thank tdasilva (and lpabon) for helping organize it
19:39:11 <peluse> heck yeah, thanks guys!
19:39:13 <notmyname> I thought it was a great week. very productive and, as always, great to see people in person
19:39:14 <lpabon> aah, it's all tdasilva :-)
19:39:20 <tdasilva> thanks to everyone for coming
19:39:21 <lpabon> ;-)
19:39:22 <tdasilva> it was great
19:39:28 <notmyname> also, I thought westford was very nice
19:39:33 <ppai> I did learn a lot
19:39:33 <zaitcev> I think we were more productive this time around, although perhaps it was merely me. In Colorado the SP were dominating and I didn't do much to advance and conclude that work, just poked here and there.
19:39:47 <acoles> yeah, thanks hosts - great job!
19:39:54 <notmyname> zaitcev: I agree. I thought this was the most productive one we've had so far
19:40:05 * tdasilva is looking forward to Hawaii
19:40:14 <peluse> oh geeze
19:40:30 <acoles> tdasilva: you just let the secret out :)
19:40:32 <notmyname> I haven't yet had a chance to write up a blog post about it, be here's my (very rough) notes https://gist.github.com/notmyname/f13d8aaeda3d065909a0
19:41:12 <notmyname> yeah, I'll be working on the next one sooner than later so that we can have it not as close to the vancouver summit
19:41:42 <notmyname> we've got a few people who have already offered to sponsor/host, which is pretty cool
19:41:58 <mattoliverau> that would be awesome, I should be able to swing that with work ;)
19:42:08 <notmyname> mattoliverau: nice!
19:42:33 <notmyname> anything else to ask or say about the hackathon?
19:42:54 <lpabon> how awesome I am at pool?
19:42:56 <cutforth> will final notes be posted somewhere?
19:43:13 <zaitcev> Someone was smoking weed in Regency a lot and stunk up the place so much that cops with AR-15 stormed it
19:43:25 <cutforth> notes = minutes of hackathon
19:43:29 <peluse> interesting...
19:43:31 <notmyname> zaitcev: clearly MA and not CA or CO ;-)
19:43:48 <peluse> BTW, I was at the Marriott, wasn't me!
19:43:51 <notmyname> cutforth: no official minutes. I'll write up my notes (above) in prose and post a blog on it
19:43:54 <zaitcev> Well, it was when I left for dinner. So it's not like when I was in Portland for Summit where they woke me up 5th Element style
19:44:04 <cutforth> notmyname: thanks
19:44:57 <notmyname> so again, a great week. thanks to you all
19:45:07 <notmyname> #topic paris summit
19:45:20 <notmyname> ...and in a month we'll all* be together again
19:45:48 <notmyname> so for Paris, we have 6 scheduled time slots and then a half-day freeform
19:46:03 <notmyname> this is less assigned slots, but more aggregate time than we had in atlanta
19:46:29 <notmyname> here's how scheduling topics for those time slots will work
19:46:44 <notmyname> I've started an etherpad where we need to collect the topic ideas
19:46:45 <notmyname> #link https://etherpad.openstack.org/p/kilo-swift-summit-topics
19:46:54 <peluse> is anyone going that can lead a topic on the single process work to date/status/next steps?
19:47:02 <notmyname> add your topics there, with a description
19:47:18 <notmyname> and (peluse) you can add topics even if you aren't going to be in paris
19:47:39 <notmyname> if you do add something but you aren't going to be there, please make a note of that
19:47:48 <notmyname> and overall, when you add something, please add your name too
19:47:57 <peluse> well, I'll be there but am not wanting to lead that discussion but will participate if tdasilva or someone will be there
19:48:20 <tdasilva> yeah, i'll be there
19:48:31 <tdasilva> i believe Ignacio is also going
19:48:36 <tdasilva> i'll add a note
19:48:37 <notmyname> from the list of topics that is collected, we'll prioritize and schedule the 6 time slots. the rest will be discussed either during the half-day or free-form at a table
19:48:40 <peluse> great!
19:48:42 <lpabon> i'll most likely will be there also.
19:48:42 <notmyname> tdasilva: great, thanks!
19:48:50 <notmyname> I think that's a great topic
19:49:21 <notmyname> acoles: I added an entry for service tokens, but I don't know if donnagh will be there
19:50:00 <acoles> i think not but i will and can pull in guys form glance, cinder if appropriate
19:50:06 <notmyname> great
19:50:21 <acoles> or is that cross-project?
19:50:31 <notmyname> acoles: could be either
19:50:51 <notmyname> so consider the etherpad document a place to brainstorm. we can condense it down later
19:51:08 <notmyname> any other questions about paris?
19:51:27 <notmyname> oh, this is important: the half-day swift stuff is on friday, so don't fly out too soon
19:51:53 <notmyname> cross project stuff on wednesday, swift sessions on thursday, swift "meetup" on friday
19:52:12 <notmyname> the currenst schedule is available online
19:52:14 <notmyname> #link http://kilodesignsummit.sched.org
19:52:47 <notmyname> anything else?
19:53:02 <notmyname> ok, quickly moving on in the interest of time...
19:53:11 <notmyname> #topic EC status
19:53:14 <notmyname> briefly...
19:53:16 <peluse> Trello is up to date...
19:53:17 <peluse> Big topic at hackathon was around the reconstructor.  WIP to update the design spec, ETA 1-2 days.
19:53:17 <peluse> I think the main patch for review now is https://review.openstack.org/#/c/124903/ from torgomatic
19:53:17 <peluse> tsg will need it and some of the reconstructor work will need tsg's patches
19:53:17 <peluse> ...also https://review.openstack.org/#/c/112449/ can be reviewed anytime
19:53:22 <notmyname> peluse: lol
19:53:31 <notmyname> thanks :-)
19:53:36 <peluse> heh, typed that in advance :)
19:54:00 <notmyname> peluse: can you add those 2 patches to the priority reviews page under the EC section?
19:54:01 <torgomatic> peluse: yeah, that one could use some feedback, keeping in mind that it's really more about groundwork for EC than it is for that particular etag-missing case
19:54:07 <peluse> yup
19:54:15 <tsg> ack
19:54:16 <torgomatic> because dfg's right; that's way too much work for just that one little edge case
19:54:26 <peluse> torgomatic, I'm guilty of not reviewing it myself but will do so today
19:54:36 <torgomatic> sure, whenever :)
19:54:49 <tsg> torgomatic: is the bigger concern md5 or mime?
19:55:02 <torgomatic> tsg: both?
19:55:16 <peluse> tsg:  see the comments in the patch from dfg
19:55:37 <torgomatic> md5 is extra CPU, but the wacky MIME protocol is a big departure from how Swift's done things before, and it is definitely more complex
19:55:52 <tsg> peluse: I did look .. it seemed like dfg was concerned about md5
19:56:20 <tsg> torgomatic: we can't not do md5 in the cases where we mangle data at proxy
19:56:50 <tsg> would going back to our non-mime approach make it a little less painful?
19:57:15 <notmyname> tsg: rather than have the design discussion in here (for the next 4 minutes), might be better to bring it to -swift
19:57:27 <tsg> notmyname: sure
19:57:31 <notmyname> thanks :-)
19:57:32 <peluse> or in the patch comments
19:57:36 <notmyname> yeah..
19:57:44 <notmyname> #topic other stuff
19:57:51 <notmyname> anything else to bring up this week?
19:58:01 <notmyname> hi dfg
19:58:49 <notmyname> thank you all for coming!
19:58:53 <zaitcev> sure
19:58:56 <peluse> hasta
19:58:58 <notmyname> see you next week at the normal time
19:59:03 <notmyname> #endmeeting