16:00:19 #startmeeting oslo 16:00:20 Meeting started Mon Feb 22 16:00:19 2016 UTC and is due to finish in 60 minutes. The chair is harlowja_at_home. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:21 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:22 o/ 16:00:23 The meeting name has been set to 'oslo' 16:00:26 howday folks 16:00:26 o/ 16:00:33 ahoy 16:00:33 \o 16:00:33 hi 16:00:37 o/ 16:00:44 o/ 16:00:45 dims i think out (on a plane?) so i'm taking the battleship over for now, lol 16:00:57 *and no sinking my battleship* lol 16:00:59 aye aye captain 16:01:07 courtesy ping for GheRivero, amotoki, amrith, bknudson, bnemec, dansmith, dhellmann, dims, dougwig, e0ne, flaper87, garyk, haypo, 16:01:07 courtesy ping for ihrachyshka, jd__, jecarey, johnsom, jungleboyj, kgiusti, kragniz, lifeless, lintan, ozamiatin, redrobot, rpodolyaka, spamaps 16:01:07 courtesy ping for sergmelikyan, sreshetnyak, sileht, sreshetnyak, stevemar, therve, thinrichs, toabctl, viktors, zhiyan, zzzeek, gcb, Nakato 16:01:25 ok, let's see who else shows up :-P 16:01:36 o/ 16:01:45 o/ 16:02:01 \o 16:02:04 o/ 16:02:04 o/ 16:02:19 ./ 16:02:19 \o o/ 16:02:27 \0/ 16:02:30 #topic Red flags for/from liaisons 16:02:35 Oh, there are two Ron's? 16:02:39 ha 16:02:48 the man and the legend ;) 16:02:49 We are seeing an exception in cliff 1.16.0+ in python-neutronclient 16:02:53 https://bugs.launchpad.net/python-cliff/+bug/1547707 16:02:53 Launchpad bug 1547707 in cliff "cliff 1.16-17.0 breaks neutron lbaas-loadbalancer-list when no load balancers are present" [Undecided,New] 16:03:01 uh ohes 16:03:04 no so-called "red flags" for keystone that I know of 16:03:10 harlowja_at_home: Nothing from Cinder. I need to try to finish pulling out the incubator code. 16:03:13 no red flags from Trove at this time. 16:03:22 #link https://bugs.launchpad.net/python-cliff/+bug/1547707 16:03:27 Should have done link... 16:03:30 That is all I have 16:03:41 just a question about this oslo.log writeable logger change that is percolating its way through the system. Is that required for Mitaka 16:03:46 anyone looking into the cliff stuff? 16:04:01 I think neutron is good, nothing to raise that I am aware of 16:04:55 amrith, hmmmm, i'll have to followup with dims on that one, i don't think its required 16:05:16 thanks, I'll look for your response on #openstack-oslo 16:06:02 johnsom, as for your's i'll poke dims on that one to, unsure who the current cliff person is (i think dhellmann might know) 16:06:35 Ok, thanks 16:06:47 cool, seems like a not-so-hard-bug-to-address 16:06:58 *with extra dashes, ha 16:07:04 Yeah, we just started getting reports from users 16:07:14 kk 16:08:04 #topic Releases for Mitaka 16:08:12 so let's see if dims put up a release for the week review 16:08:23 yup, seems so, ha 16:08:24 https://review.openstack.org/#/c/282730/ 16:08:27 #link https://review.openstack.org/#/c/282730/ 16:08:56 so that looks like a subset, if anyone wants to add more please either comment on the review or file a separate one 16:09:08 (or if u object to those being released please comment on the review to) 16:09:08 this is supposedly the last week for oslo lib releases for M 16:09:27 * rbradfor so that's how releasenotes are seen in releases (oslo.log.yaml) 16:09:55 bknudson_, good to know, so if u want a release in M, get it added :-P 16:11:38 bogdando, yt 16:12:29 #topic Ongoing-specs 16:12:34 #link https://review.openstack.org/#/q/status:open+project:openstack/oslo-specs,n,z 16:12:53 so the only one, if people are interested (me and jd__ i think are, ha) is https://review.openstack.org/#/c/256342/ 16:13:17 tldr; about having workers, some kind of work/job queue in tooz that taskflow, mistral (others?) can share 16:13:25 jimbobhickville ^ 16:13:29 u might also be interested 16:14:06 but maybe can find bogdando (i think that name is bogdan dobrelya from the review) for a chat next week 16:14:24 looking... 16:14:32 rakhmerov, yt 16:15:30 anyways, seeing as nobody is around for that one to talk about it let's skip that (maybe another time) 16:16:13 there's a few other specs that would be good to have looked at (if people get time) 16:16:39 but probably to late for mitaka for accomplishing them 16:17:17 #topic Stuck reviews 16:17:30 soooo anyone have any reviews that i can help prioritize or get eyes on 16:17:38 i know jimbobhickville and me want some taskflow ones :-P 16:18:01 yeah, pretty big backlog there 16:18:02 but besides those (which i can post a bunch on if people want) anyone have any others? 16:18:49 If neutron guys are here I would ask to take a look at https://review.openstack.org/#/c/276842/ 16:19:09 it is not oslo review but related to it 16:19:22 lxsli has a olso.log review that I'm sure we should get into M, but it's got a cross dependency on oslo.config (in current release) https://review.openstack.org/#/c/254821/ 16:19:23 #link bit.ly/1oDpFQ5 16:19:26 ^ review dashboard 16:20:44 rbradfor, k, 254821 looks like its getting worked on, guess we will see how that goes (especially if this week is the last oslo-release-in-M) 16:21:25 dukhlov_, have u poked some of the neutron folks for that one? 16:21:30 harlowja_at_home, agreed, it looks tough as it requires a release and its the last one 16:21:46 (unsure when that neutron weekly meeting is) 16:22:08 harlowja_at_home: just added a lot as reviewers, haven't poke in chat 16:22:14 harlowja_at_home, https://review.openstack.org/#/c/281790/ is also pending a +2 and is holding up two other approved changes. also lxsli work for adding release notes to oslo.config (not critical I would assume) 16:22:16 dukhlov_, or an email to the ML with [neutron] might help also 16:22:34 harlowja_at_home: ok, will try, thank you 16:23:02 dukhlov_: it's in my review queue, sorry for not getting to it prev week, I was sick 16:23:13 rbradfor, cool i'll get that one looked over, seems like a releasenotes stuffs 16:23:29 i gotta learn more about reno 16:23:45 ihrachys: ah, np. get better 16:23:55 harlowja_at_home, thanks. I'm learning about reno myself, had it successfully added to oslo.log last week. 16:24:00 :) 16:24:32 rbradfor, does it just make a prettier http://docs-draft.openstack.org/90/281790/3/check/gate-oslo.config-docs/1dcabc1//doc/build/html/history.html ? 16:25:03 harlowja_at_home, you can see it add a highlight to release yaml (which I didn't know) nice to help https://review.openstack.org/#/c/282730/2/deliverables/mitaka/oslo.log.yaml 16:25:07 or maybe that docs in that review is differnt 16:25:25 harlowja_at_home, that does not seem to include release notes. 16:25:28 k 16:25:36 wonder how those 2 get inter-connected 16:25:42 since that's like i guess the old-style release notes 16:25:56 (which most oslo projects have) 16:26:00 harlowja_at_home, see http://docs-draft.openstack.org/85/263785/7/gate/gate-oslo.log-releasenotes/f5724bb//releasenotes/build/html/ 16:26:21 rbradfor, cool 16:26:41 so i guess some work is for folks to remove the current release notes that show up in docs, with a link to that one ? 16:26:48 harlowja_at_home, I assume as part of releases (dhellmann would confirm) they move unreleased.rst to mitaka.rst 16:27:10 for example the oslo.log one still has http://docs.openstack.org/developer/oslo.log/history.html 16:27:13 harlowja_at_home, that's a good point, I'm not sure how you see it in developer docs. 16:27:18 which maybe should just point to the other one now ? 16:27:49 (pretty sure most oslo libraries have that same history.html file) 16:28:05 dhellmann, yt 16:28:12 harlowja_at_home, it's a different directory structure and tox directive, so not sure how it would get to docs.o.o 16:28:15 kk 16:28:29 maybe dhellmann , reno expert , can explain 16:28:31 harlowja_at_home, dhellmann is out most of today to my knowledge 16:28:35 oh 16:28:41 well guess he won't explain then 16:28:42 ha 16:29:05 harlowja_at_home, not here, we need to stumble across the reno uses and benefits! 16:29:19 #action harlowja_at_home find dhellmann and see what to do about the history.rst file now that reno exists 16:29:21 release notes aren't developer docs 16:29:49 bknudson_, k, so i guess both will still exist? 16:30:06 if so, seems like the developer doc history should at least point to the release notes? 16:30:24 (where release notes from reno would have notes/more details about each review) 16:30:26 If a developer wants to see the history they can git log 16:30:33 *about each release 16:30:41 bknudson_, right, that's basically what the history.rst is, lol 16:30:53 *with a little more prettiness added 16:31:25 history seems to be git commit messages, while reno is human generated. I suspect it's more for operators, highlighting changes, new features etc in releases. 16:31:48 rbradfor, agreed, seems like that 16:31:59 i'll figure out my own confusion around it :) 16:32:05 once i find dhellmann ha 16:32:12 for those interested, here is the etherpad of notes for reno rollout https://etherpad.openstack.org/p/reno-rollout 16:32:39 #link https://etherpad.openstack.org/p/reno-rollout 16:32:44 * dhellmann arrives late 16:32:53 dhellmann, well well, ha 16:32:59 u have been assigned all the work, ha 16:33:13 dhellmann, we just blamed you for something, but alas you called our bluff 16:33:25 harlowja_at_home : cliff is managed by the sdk team now. the fix for the bug has landed, it just needs to be released. 16:33:25 :) 16:33:30 johnsom, ^ 16:33:37 thx dhellmann 16:34:28 Cool, thanks dhellmann 16:34:48 harlowja_at_home , rbradfor : reno is meant to be used for release notes for deployers of openstack, the pbr-generator history file is useful for developers consuming the lib 16:35:05 dhellmann, ok, so i guess cross-linking them is up to each project 16:35:12 like a for-detailed-developer-docs click here 16:35:13 dhellmann, thanks, that is what we kinda got to 16:35:25 johnsom : the release request is https://review.openstack.org/#/c/282905/ it just needs dtroyer to sign off on it 16:35:52 rbradfor : ah, yeah, I'm caught up with scrollback and I see that now 16:36:39 #topic Open discussion 16:36:51 anything else that people want to talk about? 16:36:57 I'd be interested in comments from oslo on https://review.openstack.org/#/c/281579/ 16:37:04 patchbot?? 16:37:12 "Stop using oslotest.BaseTestCase" 16:37:14 in keystone 16:37:18 interesting 16:37:51 bknudson_ : yeah, that follows with http://specs.openstack.org/openstack/oslo-specs/specs/policy/test-tools.html and http://specs.openstack.org/openstack/oslo-specs/specs/liberty/oslotest-refactor-test-base-class.html 16:38:00 I'm not sure what the goal of oslotest's base test case is -- was it to bring some commonality to openstack tests? 16:38:17 bknudson_, afaik yes 16:38:27 I hadn't seen these specs 16:38:37 bknudson_ : it was, but we've found a lot of issues with having multiple inheritance in test classes, so we thought fixtures would make a better way to compose things into tests 16:38:40 although i didn't know BaseTestCase was going bye bye 16:38:57 harlowja_at_home : it probably won't actually be deleted, we just want people to move off of it 16:38:59 k 16:39:07 at least not for a while 16:39:23 cool 16:39:49 ya, fixtures exist for all the stuff its doing anyway (afaik) so probably nicer to just use the fixtures u want as needed 16:41:08 bknudson_, are u thinking in that review that there would be a oslotest mega-fixture? 16:41:30 harlowja_at_home: yes, if we want to keep projects consistent 16:41:30 where the mega-fixture would activate all those other fixtures? 16:41:58 right, mega-fixture would help the consistency part 16:42:14 the thing is that these fixtures are actually controlled by env vars that come from tox.ini for example 16:42:30 right, or testr.conf 16:42:34 (i think) 16:42:53 I don't think we want a megafixture. I think we want projects to choose which of these fixtures to use in the base class(es) they define in their own test suites. 16:43:15 ah, yes: http://git.openstack.org/cgit/openstack/keystone/tree/.testr.conf#n2 16:43:54 so I don't know if infra is setting OS_STDOUT_CAPTURE OS_STDERR_CAPTURE OS_LOG_CAPTURE ? 16:44:08 i don't think so 16:44:18 is infra depending on those being available? I don't know what affect it has if you actually change them. 16:44:19 (but i might be wrong) 16:44:30 the defaults can be set in .testr.conf but they're really mean to be used for debugging tests locally I think 16:44:31 the tests might not even pass if these are changed. 16:45:22 maybe it's better to simplify and not use these env vars or fixtures 16:45:30 bknudson_, understood, i myself have been stumped by that log/stdout capturing one 16:45:37 harlowja_at_home : did you cover the fact that this is our last week to release changes before the freeze? 16:45:45 dhellmann, yup 16:46:32 harlowja_at_home : https://review.openstack.org/282435 would be good to get in, if I can make it pass the test suite :-/ 16:47:05 dhellmann, cool, looks like a good one once it passes 16:47:11 recheck * 10, ha 16:47:13 lol 16:47:30 there's something weird going on in neutron, I think one of their plugins has an old incubated oslo.config 16:47:41 dhellmann, one of lxsli mutate changes to get released is dependent on a release first https://review.openstack.org/#/c/254821/ 16:47:59 rbradfor : ack, is there a release request for that? 16:48:18 I see https://review.openstack.org/282730 from dims 16:48:46 yeah, that should cover it. dims is traveling today, but we can get that release done this afternoon. 16:48:50 dhellmann, so https://review.openstack.org/#/c/280847/ the pre-requisite I believe is in release 16:49:07 right 16:51:17 dhellmann, amrith had a question about oslo.log and the writeable option, wonder if u know much about that 16:51:30 * dhellmann scans scrollback again 16:52:06 hmm, I'm not sure what change he means 16:52:12 let me get a link 16:52:32 k 16:52:53 https://bugs.launchpad.net/oslo.log/+bug/1440773 16:52:53 Launchpad bug 1440773 in Trove "Remove WritableLogger as eventlet has a real logger interface in 0.17.2" [Undecided,In progress] - Assigned to Chaozhe Chen (chaozhe-chen) 16:52:53 ah, WritableLogger is deprecated 16:53:02 I think this was working around -- right, an issue in eventlet 16:53:13 and the change in Trove, https://review.openstack.org/#/c/283021/ 16:53:21 the class is still present, and I don't expect us to remove it from the library this week 16:53:30 I was wondering whether this is something critical that we should merge for Mitaka. 16:53:33 it's likely to happen early next cycle, though, if we keep on top of it 16:55:28 thanks dhellmann harlowja_at_home 16:55:42 np 16:55:53 ok, guess we running out of time 16:55:58 amrith, I'll add it to my ongoing deprecation work I'm trying to do across oslo config 16:56:22 anything else from folks before time runs up? get it while its hot, lol 16:57:30 ok welllllll if more stuff people think about #openstack-oslo 16:57:44 dims should also be around later to 16:57:56 (or find me or dhellmann or others there) 16:58:03 #endmeeting