21:00:07 <notmyname> #startmeeting swift 21:00:08 <openstack> Meeting started Wed Jan 4 21:00:07 2017 UTC and is due to finish in 60 minutes. The chair is notmyname. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:00:09 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:00:11 <openstack> The meeting name has been set to 'swift' 21:00:12 <notmyname> who's here for the swift team meeting? 21:00:19 <kota_> hi 21:00:24 <pdardeau> hi 21:00:25 <bkeller`> o/ 21:00:25 <sgundur_> hi 21:00:26 <jrichli> hello 21:00:28 <mmotiani1> Hi 21:01:15 <notmyname> hmm... kinda light today :-) 21:01:31 <notmyname> but to the faithful, welcome ;-) 21:01:36 <nadeem> o/ 21:02:17 <notmyname> this week I was hoping to catch up on what's going on, now that we're all back from holidays 21:02:39 <acoles> hi, sorry for being late 21:03:11 <notmyname> but first, I wanted to call out donagh in the meeting and thank him for working on swift. he's moving on to other projects at HPE and will no longer be working on swift https://review.openstack.org/#/c/416577/ 21:03:36 <notmyname> so if/when you see him around, tell him thanks and wish him well :-) 21:04:22 <notmyname> ok, so let's talk about what's going on 21:04:56 <notmyname> I'll admit I haven't completely caught up on patches since being off, so if I'm missing something that comes up, please remind me :-) 21:04:57 <cutforth> o/ 21:05:01 <tdasilva> hi 21:06:02 <notmyname> in our last meeting, I put forth a list of stuff that's being worked on that are big (either in impact or in effort) 21:06:26 <notmyname> golang object server, global ec, composite rings, policy migration, policy auto tiering, sync to elasticsearch, container sharding, symlinks, increase part power, container sharding 21:06:44 <notmyname> and I know there are other things going on too 21:08:15 <notmyname> here's what I'm working on personally, related to upstream work. I'm getting ready for a conference in two weeks (talk prep) and I'm working on an improved all-in-one environment that shoudl be really easy to consume 21:08:36 <notmyname> i hope to be done with that relatively soon, and it will be a public tool 21:08:56 <notmyname> so what are you working on? who's first? 21:09:05 <notmyname> ...or second, rather 21:09:45 <notmyname> kota_: what's going on with you and in the NTT world? 21:10:07 <kota_> notmyname: ok, so you know, global ec is my first priority 21:10:17 <notmyname> cool. I'm excited about that :-) 21:10:19 <kota_> notmyname: and still waiting reviews 21:10:23 <notmyname> how's progress? 21:10:28 <notmyname> ok, blocked by reviews 21:10:32 <kota_> not so much recently 21:10:35 <kota_> yes 21:10:45 <notmyname> ok 21:10:46 <kota_> and the good news in the end of the last year 21:10:48 <kota_> is 21:11:02 <kota_> storlets has been openstack official 21:11:11 <kota_> related project to swift 21:11:17 <jrichli> nice 21:11:23 <kota_> and i has been working on some stuff on that 21:11:30 <notmyname> cool 21:11:33 <acoles> congrats to storlets team 21:11:42 <kota_> the most impressive work has been writing jupyter/ipython extension 21:11:56 <kota_> https://review.openstack.org/#/c/414883/ 21:12:02 <kota_> thanks all 21:12:29 <kota_> now i'm thinking of sort of that could be used for swift too. 21:12:45 <notmyname> fun 21:13:16 <notmyname> pdardeau: you're next. what's up? 21:13:17 <kota_> e.g. %%swift-upload container object -> upload ipython cell to swift 21:13:36 <pdardeau> mmotiani1: and i are working on swift benchmarking 21:13:55 <pdardeau> ramping up to 20 storage nodes with 8 HDD each 21:14:19 <mmotiani1> +1 21:14:27 <notmyname> nice 21:14:36 <notmyname> anything upstream you're needing for that? 21:14:46 <pdardeau> not at the moment 21:14:51 <notmyname> ok 21:15:04 <notmyname> bkeller`: what's up with you? frozen in chicago? 21:15:10 <acoles> pdardeau: did you ever meet Mark Seger (in Bristol maybe?) - he has some nice tools for benchmarking/analysis 21:15:15 * notmyname always imagines chicago as very very cold 21:15:19 <bkeller`> it's 10F here today 21:15:42 <bkeller`> nothing new for notifications yet though 21:15:45 <pdardeau> acoles: yes, i did! thanks for reminding me. i'll look around to see what he has 21:16:08 <acoles> pdardeau: email me if you want to connect with him, he doesn't hang out in irc 21:16:23 <pdardeau> acoles: thx 21:16:38 <notmyname> bkeller`: ok. what do you need from upstream? 21:17:33 <bkeller`> nothing yet that i know of 21:17:36 <notmyname> ok 21:17:40 <notmyname> jrichli: you're next 21:17:45 <jrichli> tdasilva, timburke, and I have done some work on symlinks recently 21:17:51 <jrichli> still more to do for sure 21:18:07 <jrichli> timburke left a lot of good feedback 21:18:13 <notmyname> cool 21:18:27 <jrichli> i uploaded a patch for api-ref docs - could use some feedback there 21:18:35 <notmyname> ok. link? 21:18:55 <jrichli> https://review.openstack.org/#/c/415943 21:19:17 <jrichli> clayg had mentioned it would be good for us to focus on hammering the api out before we go through perfecting the impl 21:19:45 <notmyname> yeah, I can see wisdom in that 21:20:00 <jrichli> and you all have prob seen in channel we are planning to have a conference call about keymaster v2 soon 21:20:14 <acoles> jrichli: do we have a time yet? 21:20:16 <notmyname> great 21:20:36 <jrichli> i guess tomorrow at 14:00 UTC - although, i still have a lot to go over for myself. but i dont want to hold it up. 21:21:59 <notmyname> I'll look forward to hearing what you talk about 21:21:59 <jrichli> i haven't created an invite yet. will today. let me know if you are interested - i already know about tdasilva, mahatic, acoles 21:22:35 <jrichli> well, i call it that, but its about mathiasb current patch - oh, and he will be on the list of course :-) 21:22:39 <notmyname> #info review https://review.openstack.org/#/c/415943 for symlink API 21:22:48 <acoles> specifically I want to suggest that future keymasters *only* keep config in a separate file (not in proxy-server.conf), and have others tell me I am wrong 21:23:00 <jrichli> https://review.openstack.org/364878/ 21:23:08 <notmyname> #info if interested, talk to jrichli about an invite to the conference call about keymaster v2 happening this week 21:24:12 <notmyname> jrichli: anything else? 21:24:27 <jrichli> that's all i have been up to lately 21:24:57 <notmyname> sgundur_: you're up. what are you working on this week? 21:25:54 <sgundur_> notmyname : Happy new year 21:26:04 <sgundur_> was on vacation , got back this week 21:26:24 <sgundur_> been learning about EC from Nadeem 21:26:25 <notmyname> hint: "mark all as read" is a *great* way to deal with email ;-) 21:26:36 <sgundur_> lol 21:26:54 <notmyname> cool on the EC learning 21:27:07 <sgundur_> I am inclined to do some meaningful reviews on EC and hopefully more in furture in EC work 21:27:38 <notmyname> sgundur_: great. it would be helpful for you to look at kota_'s global EC patches. both to learn and for getting that landed 21:28:02 <notmyname> sgundur_: nadeem: anything else you need from the rest of the community? 21:28:05 <sgundur_> yes, that's what I was thinking looking at Kota's update earlier 21:28:20 <notmyname> wonderful. thanks 21:28:27 <kota_> sgundur_: thanks ;-) 21:28:39 <joeljwright> Sorry I'm late! :S 21:29:06 <notmyname> joeljwright: no worries. welcome 21:29:38 <notmyname> acoles: your turn. what's going on? 21:30:17 <acoles> notmyname: before the break I was working with Pavel on this patch https://review.openstack.org/402043 which needs to land 21:30:32 <acoles> fixes regression in replication suffix hashing 21:31:04 <notmyname> ah, yes. thanks for the reminder. that's a big deal 21:31:15 <acoles> yes. 21:31:33 <notmyname> #info please review https://review.openstack.org/#/c/402043/ to close a high priority bug 21:32:07 <acoles> ^^ please go to a quiet place and think hard about whether we now have it right 21:32:18 <acoles> notmyname: now getting up to speed on the keymaster/barbican patch again. 21:32:23 <notmyname> cool 21:32:31 <notmyname> acoles: anything else you need from upstream? 21:32:46 <acoles> and found an ssync bug in the process :/ which I'll probably try to fix 21:32:54 <notmyname> yay 21:33:10 <notmyname> (sarcastically for the bug. seriously for the fix) 21:33:21 <acoles> notmyname: not feeling any need at the moment 21:33:31 <notmyname> ok 21:33:40 * acoles is grieving for donagh 21:33:44 <notmyname> +1 21:33:54 <jrichli> +1 21:33:59 <joeljwright> :( 21:34:13 <kota_> :( 21:34:40 <notmyname> cutforth: anything from you? are you working on anything related to upstream swift this week? 21:34:43 <acoles> I will still pester him! 21:35:37 <notmyname> tdasilva: your turn. what's new with you in the new year? 21:36:12 <tdasilva> hello! just before the break i was working on the libec/pyeclib merger 21:36:19 <notmyname> oh right! 21:36:45 <tdasilva> I thought of two options, one was to copy files from pyeclib into libec repo and send gerrit patch 21:37:00 <tdasilva> the other was to work with infra to merge the repos, so that way we could keep history of changes 21:37:14 <notmyname> which way are you leaning so far? 21:37:30 <tdasilva> I took a stab at creating a throw away repo but still need to get feedback from infra team 21:37:52 <notmyname> ok. I'd prefer to keep the history, if possible. thanks for working on that 21:38:15 <tdasilva> yeah, i'd like that too, will continue trying to work with the infra team, let's see... 21:38:21 <notmyname> cool 21:38:31 <notmyname> anything you need from the rest of the community yet? 21:38:40 <tdasilva> over the break timburke left comments on symlink patch, so i started going through those 21:38:47 <notmyname> ok 21:38:51 <tdasilva> and jrichli send the api-ref patch which I still need to look over 21:39:15 <tdasilva> hoping to send an update to symlink patch soon 21:39:46 <notmyname> thanks 21:39:54 <cutforth> nothing from me, thank you for asking 21:40:07 <tdasilva> i think that's it. jrichli if you need video conf. call for tomorrow's meeting i can help 21:40:25 <jrichli> tdasilva: ok, cool. thx 21:40:39 <notmyname> timburke: your turn. what's up? 21:40:52 <timburke> mostly looking at swift3 stuff 21:41:07 <notmyname> ok. anything blocking that you need help with? 21:41:35 <timburke> can we clone kota_? 21:41:42 <timburke> :P 21:41:45 <notmyname> heh 21:42:01 <kota_> so my slow review blocks them, i know :/ 21:42:26 <timburke> well, i *could* just merge them, but then i'd feel dirty 21:43:35 <notmyname> joeljwright: I think you're the last nick I saw come in. your turn. what's going on with you, related to swift team work? 21:43:55 <joeljwright> I got some good feedback on the SLO preamble/postamble stuff 21:44:04 <joeljwright> need to talk to timburke about his comments 21:44:05 <notmyname> great 21:44:30 <joeljwright> also, the removing slo segments patch for the client 21:44:36 <notmyname> ah, nice 21:44:53 <notmyname> is that up yet and I'm forgetting about it, or is it goign to be proposed? 21:45:10 <joeljwright> which one? 21:45:16 <notmyname> removing slo segments 21:45:21 <joeljwright> the SLO segment removal patch had issues 21:45:31 <timburke> ugh, that reminds me that i need to think more about https://review.openstack.org/#/c/310075/ ... 21:45:40 <joeljwright> I'd like it fixed too, so that will be a chat with timburke 21:45:50 <notmyname> ok 21:45:52 <timburke> oh! "removal" -> "expiration" 21:46:15 <notmyname> anything else from anyone? did I skip anyone or did anyone remember something else? anything else to bring up? 21:46:16 <joeljwright> sorry - it's been a fun christmas 21:46:31 <joeljwright> I also have a machine set up to test container sharding, but no progress yet 21:46:57 <notmyname> joeljwright: nice 21:46:57 <joeljwright> will probably bug mattoliverau as soon as I have something 21:46:58 <nadeem> Was on vacation for last few weeks. I will working on trimming feature/hummingbird branch to isolate golang replicator 21:47:02 <pdardeau> any updates on anything golang related? 21:47:08 <timburke> nice! i've been wanting to look at that but haven't gotten around to it 21:47:39 <notmyname> nadeem: cool. please continue to talk with clayg about that too :-) 21:47:53 <nadeem> sure will do :) 21:47:59 <notmyname> thanks 21:48:22 <joeljwright> timburke: you got a few mins after the meeting? 21:48:34 <timburke> yeah, definitely 21:48:46 <notmyname> so in summary, https://review.openstack.org/#/c/402043/ needs some serious review to close a bug, https://review.openstack.org/#/c/415943 is a proposal for symlink API, and talk to jrichli if you're interested in the new keymaster call tomorrow 21:49:06 <notmyname> if there's nothing else.... 21:49:39 <notmyname> thanks, again, for all your work on swift 21:49:43 <notmyname> #endmeeting