16:00:25 #startmeeting oslo 16:00:25 courtesy ping for GheRivero, amotoki, amrith, bknudson, bnemec, dansmith, dhellmann, dougwig, e0ne, flaper87, garyk, harlowja, haypo, 16:00:25 courtesy ping for ihrachyshka, jd__, jecarey, johnsom, jungleboyj, kgiusti, kragniz, lifeless, lintan, ozamiatin, redrobot, rpodolyaka, spamaps 16:00:26 courtesy ping for sergmelikyan, sreshetnyak, sileht, sreshetnyak, stevemar, therve, thinrichs, toabctl, viktors, zhiyan, zzzeek, gcb 16:00:26 Meeting started Mon Nov 16 16:00:25 2015 UTC and is due to finish in 60 minutes. The chair is dims. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:28 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:30 The meeting name has been set to 'oslo' 16:00:30 o/ 16:00:31 o/ 16:00:32 yo yo 16:00:33 hi 16:00:36 hi 16:00:37 o/ 16:00:38 o/ 16:00:41 \o 16:00:41 ./ 16:00:41 o/ 16:00:45 hi 16:00:54 long time listener, first time caller 16:01:01 hello everyone 16:01:13 jimbobhickville welcome 16:01:20 ha 16:02:07 let's get started. 16:02:10 #topic Red flags for/from liaisons 16:02:22 Nothing to report in the lbaas/octavia side of the shop. 16:02:40 thanks johnsom 16:02:50 nothing from trove, we're trying to integrate oslo.db and consume the last oslo.incubator merge 16:02:53 one red flag is removing support for py26 since as far as I know it wasn't deprecated 16:03:02 but if we're forced to do it we'll do it 16:03:06 amrith cool 16:03:36 bknudson_ y, we have been talking about it forever, infra is yanking jobs in a week or two 16:03:53 let's come back to it, i have a topic for it 16:04:02 #topic Releases for Mitaka 16:04:02 #link https://review.openstack.org/#/c/245465/ 16:04:36 that's the lot i want to get out today, please check the output for http://logs.openstack.org/65/245465/1/check/gate-releases-tox-list-changes/426cf54/ 16:04:40 to see what's getting in 16:05:15 last week we managed to break a couple of projects with oslo.messaging release 16:05:52 :( the locking stuffs right? 16:05:53 harlowja_at_home does 1.24.0 have our decider fix? 16:05:59 my travis setup is green - https://travis-ci.org/dims/ but we'll have to act fast again if if we do 16:06:03 harlowja_at_home yes 16:06:47 johnsom, https://review.openstack.org/#/c/244922/ still reviewing so negative, i'll bug people about it to get it in, and after that i guess we can release 16:07:15 Ok, thanks. I will give the latest patch a test today too and give you a green light. 16:07:20 thx 16:07:37 ok. switching 16:07:40 #topic Steps to drop python 2.6 support 16:07:41 Steps : fix classifiers, bump major release number 16:08:17 big concern is the major version number bump (for Gordon) 16:08:30 any other concerns? 16:08:50 we have to release first and then python-*client will have to release their versions 16:09:14 we'll have to update g-r and u-c after our releases 16:09:23 in support of py2.6 16:09:27 We dropped 2.6 a long time ago, so no issue from us. 16:09:31 that's the tentative plan 16:09:33 bye bye 2.6 16:09:34 johnsom ack 16:10:07 anyone have other concerns? 16:10:10 just remove some workaround for 2.6 :-) 16:10:30 gcb : hold off on that, we need to do that after removing classifiers 16:10:49 or at least along with removing classifiers 16:10:59 are they used by pip? 16:11:27 * harlowja_at_home don't think so, mostly used by pypi for searching (cataloging?) and those pesky humans 16:11:34 ah, ok 16:11:34 gcb : we need to merge the project-config changes to stop running py26 jobs, then merge tox.ini changes and then classifiers and then code 16:11:38 and then release 16:11:58 ok, will hold on them 16:12:19 gcb : you can help file classifier changes if you want to 16:12:30 gcb base them on the tox.ini changs 16:12:38 so we know the sequence 16:12:50 and tack the code changes at the end of that chain 16:12:58 gcb sounds like a plan? 16:13:01 dims, yes, I will 16:13:13 yay, looking forward to your stats bump :) 16:13:19 lol 16:13:30 #topic Steps to cleanup oslo-incubator 16:13:56 so, for oslo-incubator the word is out, i've cleaned up launchpad, marking stuff as wont-fix, 16:14:04 step1, the first rule of incubator is u don't talk about incubator 16:14:04 filed last sync from oslo-incubator reviews 16:14:30 who wants to hunt down projects that have openstack-common.conf and nuke them? 16:14:42 * harlowja_at_home can do that 16:15:16 cool harlowja_at_home - if anyone wants to help harlowja_at_home please chime in 16:15:33 harlowja_at_home does not need a pesky stats bump :) 16:15:39 lol 16:15:42 that is true 16:15:51 :) 16:15:54 #topic First doc sprint for Mitaka 16:16:10 is next thu/fri good for everyone? 16:16:17 should be for me 16:16:34 harlowja_at_home will you be able to send reminders and setup etherpad etc based on last sprint? 16:16:50 dims, this thur or next thur/fri? 16:16:59 just thinking, isn't that thanksgiving in US 16:16:59 next week 16:17:12 oops right 16:17:26 let's do it this week then 16:17:30 works for me 16:17:48 ok let's give it a shot. please drop an email to the -dev list? 16:17:51 sure 16:18:01 thanks harlowja_at_home 16:18:07 gcb, you are up next 16:18:10 #topic oslo.config enforce_type=True in set_override 16:18:10 #link https://etherpad.openstack.org/p/enforce_type_true_by_default 16:18:15 yes 16:18:59 you can get details in that link 16:19:00 gcb : have you grepped sources to see which projects will have negative effects? 16:19:35 gcb other than nova i mean 16:20:15 I just POC in Nova and Ironic, they don't work after I make force_type=True 16:20:53 I mean unit test failed 16:21:14 gcb : can you please grep other projects as well, we need to fix them first before we make a chance in oslo.config 16:22:02 * harlowja_at_home seems like a right change to me to do (from reading background) 16:22:23 yes, I have a plan to do that later before we change oslo.config 16:22:45 should we also send out mail about that to openstack-dev (just thinking it might help others become aware of the background + solution)? 16:23:12 harlowja_at_home ++ 16:23:13 I will send out the mail :-), 16:23:20 oslo_config: https://review.openstack.org/#/c/243001/ this blocks consuming projects clean up 16:23:20 thanks gcb 16:23:51 gcb ack will review after the meeting 16:23:59 dims thanks 16:24:30 #topic everyone based in europe safe? 16:24:41 jd__ : flaper87 ping 16:25:16 jd just reviewed my code :-) 16:25:44 gcb yes, have seen his tweets as well. my thoughts and prayers to everyone there 16:26:02 def +3 16:26:29 #topic Open discussion 16:26:31 Any stuck reviews? or specs? 16:26:48 :) 16:27:09 soooo https://review.openstack.org/#/c/198914/ is somewhat stuck, i made an etherpad listing all the changes and dates they were deprecated but i can't find the etherpad right now so maybe i can just explain :-P 16:27:11 ozamiatin toabctl kgiusti nkrinner jimbobhickville : anything to add? :) 16:27:30 nope :) 16:27:33 dims: nothing from my side) 16:27:41 nkrinner: no, thanks 16:27:42 * kgiusti seriously needs to be cloned - getting back to bugs 16:27:48 nope 16:28:03 I'd like to go through the taskflow blueprints and close down any of the "wontfix" ones. There's a bit of clutter there. Anyone want to join me in that effort? 16:28:18 jimbobhickville ++ 16:28:40 so the jist of that is, https://review.openstack.org/#/c/198914/ removes a bunch of deprecated stuff that was slotted for removal in 2.0, and lifeless brought up the point that nothing should be removed until its been deprecated for 3 cycles (1.5 years?) sooooo, i listed the lenght of time thoses current things existed in that etherpad and some is less than 1.6 years 16:29:12 buuuuut then it was like its a 2.0 release, sooooo, its not like imho 3 cycles should be required (semantic versioning and all allows for this stuff to happen)) 16:29:22 anyways, thoughts welcome :) 16:29:38 jimbobhickville, since i may have made the clutter i guess i can't volunteer :-P 16:30:12 lol, I'll probably just bug you before I close anything, just to make sure 16:30:18 harlowja_at_home if you can get some ack(s) from people using taskflow, that would be great 16:30:24 sure 16:30:42 harlowja_at_home that will help shorten the time line 16:30:45 dims, 1.x can also remain like bug-fix or whatever 16:30:48 we're good with it, fwiw. we pin to exact versions and don't upgrade until we test the new version 16:30:52 harlowja_at_home right 16:31:12 jimbobhickville good practice anyway :) 16:31:43 dims, but i'll ping some folks, i know lifeless had some concerns about 3 cycle policy and all that (1.5 years seems like alot, lol) 16:31:59 jimbobhickville mostly it's other openstack projects that follow global requirements we end up breaking as we dont pin for unit tests 16:32:21 harlowja_at_home that's something we can take to the cross project meeting agenda 16:32:24 k 16:32:42 other reviewww that is interesting and will likely need supporting help (beyond just code) 16:32:43 https://review.openstack.org/#/c/245362/ 16:32:44 jd__ glad to hear from you 16:32:48 WIP : Add Consul Driver 16:33:01 as others know that one will be more than just code and SpamapS rightly mentioned that there 16:33:08 ' Also this is going to need support in devstack (to spin up consul instead of zookeeper) so we can get a devstack run to pass with consul present and configured instead of zookeeper. ' .... 16:33:10 I'm working on the etcd driver 16:33:12 dims: yeah thanks :) 16:33:50 sooo if others have thoughts about what might be needed for that review to fully function (And/or code reviews) i'm sure vilobh would appreciate it 16:34:00 harlowja_at_home which one...etcd or consul would be better able to implement all the tooz functionality? 16:34:05 jd__ ^^ 16:34:19 theoretically 16:34:37 I think it's more or less the same, no? 16:34:47 they have the same limitation I'd say though I don't know Consul extensively 16:34:50 dims, unsure, both are capable, but i don't know the specifics of each to know exactly what will be missing when putting under the tooz apis 16:35:11 both don't have the zookeeper concept of sequenced nodes, so that makes fair locking and stuff hard 16:35:21 which makes things like fair read-writer/shared locks hard... 16:35:23 harlowja_at_home jd__ ok 16:35:39 and projects like ironic (i think want that kind of locking) 16:35:48 so still experiments... harlowja_at_home looks like you will need some libraries in g-r? 16:35:55 dims, correct 16:36:10 eventually 16:36:15 once experiemnts more flushed out 16:36:17 ok 16:36:29 jd__ etcd need any g-r updates? 16:37:14 looks like its probably just all ontop of requests 16:37:23 ok 16:37:43 let's revisit tooz stuff next week. 16:37:56 any other topics, questions... 16:38:16 going twice 16:38:39 thanks everyone, hope to see you all in the doc sprint or next meeting 16:38:51 #endmeeting