Tuesday, 2015-03-03

« Monday, 2015-03-02 Index Wednesday, 2015-03-04 »
*** stevemar has joined #openstack-oslo00:18
*** pmathews has quit IRC00:18
*** dims has quit IRC00:27
*** mtanino_ has joined #openstack-oslo00:28
*** mtanino has quit IRC00:29
openstackgerritMichael Bayer proposed openstack/oslo.db: Implement new oslo.db.sqlalchemy.enginefacade module  https://review.openstack.org/13821500:30
*** e0ne is now known as e0ne_00:33
*** e0ne_ is now known as e0ne00:33
*** e0ne has quit IRC00:34
*** bknudson has joined #openstack-oslo00:37
*** stpierre has quit IRC00:56
*** ujjain has quit IRC00:57
*** dimsum__ has joined #openstack-oslo01:08
*** noelbk has quit IRC01:12
openstackgerritJoshua Harlow proposed openstack/taskflow: Allow using ordered(set/dict) to retain ordering  https://review.openstack.org/16055801:13
openstackgerritJoshua Harlow proposed openstack/taskflow: Allow using ordered(set/dict) to retain ordering  https://review.openstack.org/16055801:20
*** achanda has quit IRC01:38
openstackgerritJoshua Harlow proposed openstack/taskflow: Allow using ordered(set/dict) to retain ordering  https://review.openstack.org/16055801:44
openstackgerritJoshua Harlow proposed openstack/taskflow: Allow using ordered(set/dict) to retain ordering  https://review.openstack.org/16055801:53
openstackgerritJoshua Harlow proposed openstack/taskflow: Add a fully functional orderedset  https://review.openstack.org/16021901:59
openstackgerritJoshua Harlow proposed openstack/taskflow: Allow using ordered(set/dict) to retain ordering  https://review.openstack.org/16055801:59
*** sigmavirus24 is now known as sigmavirus24_awa02:08
*** zzzeek has quit IRC02:10
*** takedakn has joined #openstack-oslo02:20
openstackgerritJoshua Harlow proposed openstack/taskflow: Add specific scoping documentation  https://review.openstack.org/15895702:22
*** mtanino_ has quit IRC02:24
*** tsekiyama has quit IRC02:27
*** amotoki has joined #openstack-oslo02:32
*** crc32 has joined #openstack-oslo02:49
openstackgerritJoshua Harlow proposed openstack/taskflow: Tweak functor used to find flatteners/storage routines  https://review.openstack.org/16061603:11
*** jungleboyj has joined #openstack-oslo03:20
*** yamahata has quit IRC03:20
*** takedakn has quit IRC03:40
*** sputnik13 has quit IRC03:41
*** dimsum__ has quit IRC03:45
*** amotoki has quit IRC04:02
*** takedakn has joined #openstack-oslo04:10
*** harlowja_ is now known as harlowja_away04:13
*** jamespage has quit IRC04:14
*** jamespage has joined #openstack-oslo04:14
*** crc32 has quit IRC04:22
*** harlowja_at_home has joined #openstack-oslo04:37
*** dimsum__ has joined #openstack-oslo04:46
*** dimsum__ has quit IRC04:51
openstackgerritJoshua Harlow proposed openstack/taskflow: Add a fully functional orderedset  https://review.openstack.org/16021905:00
openstackgerritJoshua Harlow proposed openstack/taskflow: Allow using ordered(set/dict) to retain ordering  https://review.openstack.org/16055805:01
*** takedakn has quit IRC05:03
*** david-lyle_afk has joined #openstack-oslo05:06
openstackgerritJoshua Harlow proposed openstack/taskflow: Add a frozen checking decorator  https://review.openstack.org/16053605:12
*** jecarey has quit IRC05:39
*** harlowja_at_home has quit IRC05:40
*** david-lyle_afk has quit IRC05:44
*** saikrishna has joined #openstack-oslo05:47
openstackgerritMerged openstack/taskflow: Fix lookup scoping multi-match ordering  https://review.openstack.org/15895605:54
*** takedakn has joined #openstack-oslo06:03
openstackgerritOpenStack Proposal Bot proposed openstack/oslo.utils: Imported Translations from Transifex  https://review.openstack.org/16064906:08
openstackgerritMerged openstack/taskflow: Provide more contextual information about invalid periodics  https://review.openstack.org/15902006:13
openstackgerritMerged openstack/taskflow: add get_flow_details and get_atom_details to all backends  https://review.openstack.org/15722506:13
*** achanda has joined #openstack-oslo06:14
*** redrobot has quit IRC06:21
*** redrobot has joined #openstack-oslo06:25
*** redrobot is now known as Guest3254406:25
*** jaosorior has joined #openstack-oslo06:30
*** russellb has quit IRC06:34
*** david-lyle_afk has joined #openstack-oslo06:44
*** jroll has quit IRC06:49
*** saikrishna has quit IRC06:58
*** jroll has joined #openstack-oslo06:59
*** inc0 has joined #openstack-oslo07:03
*** dulek has joined #openstack-oslo07:04
openstackgerritJoshua Harlow proposed openstack/taskflow: Add a frozen checking decorator  https://review.openstack.org/16053607:06
*** yamahata has joined #openstack-oslo07:21
*** openstackgerrit has quit IRC07:22
*** openstackgerrit has joined #openstack-oslo07:22
*** AnxiousGarlic has joined #openstack-oslo07:28
*** AnxiousGarlic has left #openstack-oslo07:29
*** takedakn has quit IRC07:35
*** ajo has quit IRC07:48
openstackgerritMerged openstack/taskflow: Add specific scoping documentation  https://review.openstack.org/15895707:48
openstackgerritMerged openstack/taskflow: Tweak functor used to find flatteners/storage routines  https://review.openstack.org/16061607:53
*** ajo has joined #openstack-oslo07:57
*** stevemar has quit IRC08:15
openstackgerritAkihiro Motoki proposed openstack-dev/pbr: Make setup.py --help-commands work without testrepository  https://review.openstack.org/12465208:15
*** ajo has quit IRC08:17
*** mtanino has joined #openstack-oslo08:21
*** dimsum__ has joined #openstack-oslo08:24
*** mtanino has quit IRC08:26
*** sigmavirus24_awa has quit IRC08:26
*** e0ne has joined #openstack-oslo08:26
*** dimsum__ has quit IRC08:29
*** dtantsur|afk is now known as dtantsur08:30
*** e0ne has quit IRC08:34
*** dulek has quit IRC08:38
*** e0ne has joined #openstack-oslo08:39
*** dulek has joined #openstack-oslo08:39
*** BrianShang has joined #openstack-oslo08:44
*** inc0 has quit IRC08:46
*** ajo has joined #openstack-oslo08:49
*** dulek_ has joined #openstack-oslo08:50
*** takedakn has joined #openstack-oslo08:51
*** inc0 has joined #openstack-oslo08:52
*** dulek has quit IRC08:53
*** achanda has quit IRC08:55
*** takedakn has quit IRC08:57
*** inc0_ has joined #openstack-oslo08:57
*** e0ne is now known as e0ne_08:59
*** dulek_ has quit IRC09:00
*** inc0 has quit IRC09:00
*** e0ne_ is now known as e0ne09:01
*** dulek has joined #openstack-oslo09:03
*** dulek has quit IRC09:04
*** lifeless has quit IRC09:05
*** lifeless has joined #openstack-oslo09:06
*** andreykurilin_ has joined #openstack-oslo09:06
*** e0ne is now known as e0ne_09:06
*** JoshNang has quit IRC09:07
*** sputnik13 has joined #openstack-oslo09:07
*** e0ne_ is now known as e0ne09:09
*** andreykurilin_ has quit IRC09:12
*** andreykurilin_ has joined #openstack-oslo09:12
*** yamahata has quit IRC09:13
*** e0ne has quit IRC09:15
*** andreykurilin_ has quit IRC09:26
*** andreykurilin_ has joined #openstack-oslo09:27
*** ajo has quit IRC10:19
*** ajo has joined #openstack-oslo10:19
*** ihrachyshka has joined #openstack-oslo10:20
*** ujjain has joined #openstack-oslo10:25
*** ujjain has quit IRC10:25
*** ujjain has joined #openstack-oslo10:25
*** dtantsur is now known as dtantsur|bbl10:28
openstackgerritMerged openstack/oslo-incubator: Add liberty release name to versionutils  https://review.openstack.org/16047910:29
*** inc0_ has quit IRC10:29
*** ajo has quit IRC10:30
*** ajo has joined #openstack-oslo10:31
*** inc0 has joined #openstack-oslo10:31
*** amotoki has joined #openstack-oslo10:34
*** e0ne has joined #openstack-oslo10:43
*** andreykurilin_ has quit IRC10:46
*** fnordahl has joined #openstack-oslo10:47
*** inc0 has quit IRC10:48
*** exploreshaifali has joined #openstack-oslo10:52
openstackgerritOpenStack Proposal Bot proposed openstack/debtcollector: Updated from global requirements  https://review.openstack.org/16073410:54
openstackgerritMerged openstack/tooz: Allow coordinator non-string options and use them  https://review.openstack.org/15560210:57
*** inc0 has joined #openstack-oslo10:58
*** e0ne is now known as e0ne_11:09
*** inc0 has quit IRC11:13
*** dimsum__ has joined #openstack-oslo11:16
*** e0ne_ has quit IRC11:20
*** e0ne has joined #openstack-oslo11:21
*** inc0 has joined #openstack-oslo11:44
*** inc0 has quit IRC11:45
*** cdent has joined #openstack-oslo11:49
*** ihrachyshka has quit IRC11:53
*** e0ne is now known as e0ne_11:53
*** e0ne_ is now known as e0ne12:01
*** ihrachyshka has joined #openstack-oslo12:05
*** russellb has joined #openstack-oslo12:08
openstackgerritSean Dague proposed openstack/oslo.policy: provide more descriptive exception  https://review.openstack.org/16076112:10
openstackgerritAbhijeet Malawade proposed openstack/tooz: fix mysql driver url parsing  https://review.openstack.org/15983512:12
*** ihrachyshka has quit IRC12:21
*** dtantsur|bbl is now known as dtantsur12:22
*** ihrachyshka has joined #openstack-oslo12:22
*** dimsum__ is now known as dims12:46
*** yassine has joined #openstack-oslo12:47
*** kgiusti has joined #openstack-oslo12:56
*** pblaho has quit IRC13:00
openstackgerritJulien Danjou proposed openstack-dev/hacking: Add a check for staticmethod  https://review.openstack.org/15195213:02
*** jroll has quit IRC13:07
*** jroll has joined #openstack-oslo13:07
openstackgerritIhar Hrachyshka proposed openstack/oslo.policy: Expose register and Check as part of public API  https://review.openstack.org/15952513:14
*** dims has quit IRC13:14
openstackgerritIhar Hrachyshka proposed openstack/oslo.policy: Expose register and Check as part of public API  https://review.openstack.org/15952513:15
*** dims has joined #openstack-oslo13:15
*** gordc has joined #openstack-oslo13:37
*** sreshetnyak has joined #openstack-oslo14:00
*** exploreshaifali has quit IRC14:08
*** exploreshaifali has joined #openstack-oslo14:10
*** exploreshaifali has quit IRC14:18
*** mriedem_away is now known as mriedem14:28
openstackgerritMerged openstack/oslo-incubator: Optimization of waiting subprocesses in ProcessLauncher  https://review.openstack.org/15634514:30
*** jungleboyj has quit IRC14:35
*** zz_jgrimm is now known as jgrimm14:43
*** salv-orlando has joined #openstack-oslo14:43
haypoihrachyshka, dhellmann, jd__ : hey, i opened http://bugs.python.org/issue23574 to generalize "support" of leap seconds in datetime (my patch simply ignores them in the datetime constructor)14:44
jd__thanks haypo14:45
ihrachyshkahaypo++14:45
haypoihrachyshka: the fix for oslo.utils is quite obvious, just replace second=60 with second=5914:46
ihrachyshkaright. that's a valid hack around14:46
haypoihrachyshka: i don't think that it's a hard, it's the right solution to handle leap seconds with datetime.datetime14:47
jd__not sure ignoring leap seconds is the good solution though14:47
haypoihrachyshka: as i wrote in my issue, datetime.datetime.fromtimestamp() exactly does that14:47
ihrachyshkajd__, datetime does not support them anyway, so we would not do anything worse with that14:48
haypojd__: which kind of issue do you see with ignoring leap seconds?14:48
ihrachyshkahaypo, that's an interesting observation - re fromtimestamp14:48
*** viktors|afk is now known as viktors14:54
openstackgerritTerry Wilson proposed openstack/oslo.rootwrap: Do not clear enviornment when env=None  https://review.openstack.org/16013014:54
jd__haypo: wrong computing of things like timedelta for example?14:55
jd__haypo: collision of timestamp that are identical where they are not?14:55
*** jecarey has joined #openstack-oslo15:02
*** yassine has quit IRC15:03
*** pmathews has joined #openstack-oslo15:04
*** pmathews1 has joined #openstack-oslo15:10
*** pmathews has quit IRC15:13
*** prad has joined #openstack-oslo15:17
viktorsdhellmann: hi! Can you please remove your -2 from this patch -    https://review.openstack.org/#/c/133962/ ?15:22
*** alexpilotti has joined #openstack-oslo15:25
*** exploreshaifali has joined #openstack-oslo15:28
*** jungleboyj has joined #openstack-oslo15:29
dhellmannviktors: did we come to an agreement on the approach?15:29
* dhellmann looks for his notes15:29
*** yassine has joined #openstack-oslo15:30
viktorsdhellmann: I went to vacation, when discussion started...15:30
viktorsbt I found no exact answer in mailing list15:31
viktorssame as objections15:31
dhellmannyeah, I'm not finding a resolution either15:31
dhellmannthis change just uses it for python 3, right?15:31
viktorsyes15:31
dhellmannI don't think it's safe to change python 2, but this seems ok for 315:32
viktorssounds reasonable15:32
*** sigmavirus24 has joined #openstack-oslo15:35
dhellmannviktors: is there a requirements change to add it to the list of global requirements, too? or did we already merge that?15:36
dhellmannviktors: found it: https://review.openstack.org/#/c/123737/15:36
*** daniel3_ has joined #openstack-oslo15:36
viktorsdhellmann: :)15:36
ihrachyshkadhellmann, hey, re oslo.log15:38
ihrachyshkadhellmann, I read migration notes15:38
ihrachyshkadhellmann, and now I'm confused15:38
* dhellmann declares victory and goes home15:38
ihrachyshkadhellmann, it says we need to pass unicode objects only15:38
ihrachyshkadhellmann, so it means including LOG.debug() calls15:38
dhellmannihrachyshka: yes15:38
ihrachyshkaand we generally pass str() there15:38
dhellmannright, and those will break if the thing being inserted into them can't be represented as ascii15:39
ihrachyshkaso the dumb way would be applying six.text_type to each occurance15:39
ihrachyshkabut there are lots of those15:39
dhellmannjust add the u in front of the literal strings15:39
ihrachyshkaso I wonder whether you see any smart way to mock LOG.debug to make it apply six.text_type() automatically15:39
dhellmannor use "from __future__ import unicode_literals"15:39
ihrachyshkadhellmann, u probably won't work in python3?15:40
dhellmannihrachyshka: it does with 3.3 and 3.4.15:40
ihrachyshkaHenryG, ^^15:40
ihrachyshkadhellmann, that's still a lot of LOC touched15:40
dhellmannihrachyshka: use "from __future__ import unicode_literals"15:41
ihrachyshkadhellmann, plus we'll need a hacking rule I guess15:41
ihrachyshkadhellmann, hm, that's probably a better option, though quite scary since it's global :)15:41
dhellmannonly to the module, but yeah15:41
ihrachyshkacan I put it in neutron/__init__.py and be done with it for the whole tree?15:41
dhellmannno, it's per-module15:41
ihrachyshka#sadpanda15:42
dhellmannit controls how the parser reads the rest of the file15:42
ihrachyshkagot it15:42
ihrachyshkaI suspect it means we should introduce a hacking check to make sure it's always on15:42
* dhellmann hates hacking rules15:42
dhellmannwe spend way way too much time fiddling with those instead of just fixing production code15:43
ihrachyshkadhellmann, well, how would you guarantee that nothing is broken otherwise?15:43
dhellmannwrite tests using unicode data15:43
*** sputnik13 has quit IRC15:43
ihrachyshkadhellmann, I appreciate your optimism when thinking about how neutron devs can enforce new rules to code reviewed15:44
ihrachyshka*effectively15:44
dansmithdhellmann: are you ready to unblock this? https://review.openstack.org/#/c/159380/15:45
dhellmannihrachyshka: I don't have a good answer to that, but I'm not convinced hacking rules are it15:45
dhellmanndansmith: looking15:45
ihrachyshkaHenryG, comments on ^^? :)15:45
dhellmanndansmith: do we need a new versionedobjects release first?15:45
* dhellmann is behind cutting releases this week15:45
dansmithdhellmann: no, why?15:45
dhellmanndansmith: oh, there are a few unreleased things that look like fixes, so I just wanted to doublecheck15:46
ihrachyshkadhellmann, are there any other projects that started on oslo.log path?15:46
dhellmanndansmith: "Allow passing serializer and indirection API objects to Fixture" looks like something someone needed in a test?15:46
dansmithdhellmann: just a few things I found in my nova PoC, but nothing that will affect heat's usage15:46
dhellmannihrachyshka: there's a patch for nova, and I believe keystone was where we found the unicode issue15:46
dhellmanndansmith: ok, sounds good15:46
dansmithdhellmann: we can cut a release if you'd prefer15:46
dhellmanndansmith: it'll wait +215:47
dansmithdhellmann: cool, thanks!15:47
HenryGihrachyshka: no comment yet, still processing the info15:48
ihrachyshkadhellmann, failed to find those patches. have a link?15:48
haypojd__, ihrachyshka : ok, i played more with time.mktime & time.localtime. they don't handle leap seconds as i expected :-p http://bugs.python.org/issue2357415:48
jd__haypo: sigh15:50
*** stevemar has joined #openstack-oslo15:53
*** zzzeek has joined #openstack-oslo15:54
dhellmannihrachyshka: the nova patch has merged already16:00
dhellmannihrachyshka: https://review.openstack.org/15426016:00
HenryGdhellmann: and reverted?16:01
dhellmannstevemar, bknudson: I did a soft release of oslo.policy 0.2.0 with those bug fixes and new methods. How are we looking for a more public release?16:01
dhellmannHenryG: ?16:02
HenryGdhellmann: Oh the revert was abandoned16:02
ihrachyshkadhellmann, there seems to be no unicode related updates though16:02
dhellmannihrachyshka: yeah, you don't have to go rewrite your whole app to adopt the library16:02
ihrachyshkadhellmann, https://review.openstack.org/159525 ?16:03
ihrachyshkawon't it be included in the release?16:03
dhellmannihrachyshka: not this week, unless it merges and we do another one tomorrow16:03
*** ekarlso has quit IRC16:03
ihrachyshkaack16:03
*** SpamapS has quit IRC16:03
ihrachyshkadhellmann, so in the end, we are fine without unicode markers? I am even more confused now. :)16:03
dhellmannihrachyshka: we'll merge it, I'm just clearing our release backlog again16:04
stevemardhellmann, if theres a 0.2.0 then keystone code use that, ihrachyshka could use 0.3.0 when his bug is fixed16:04
*** SpamapS has joined #openstack-oslo16:04
ihrachyshkastevemar, not really we can use it actually (unless we access private methods)16:04
stevemardhellmann, theres also the issue with the policy_dirs that sdague pointed out16:05
dhellmannihrachyshka: you may have subtle bugs, so I recommend fixing them, but you don't have to block adoption on that.16:05
dhellmannstevemar: ah, right, forgot that one16:05
dhellmannihrachyshka: do not use the private names16:06
ihrachyshkadhellmann, I won't, that's why I told that we won't consume it in K :)16:07
*** ekarlso has joined #openstack-oslo16:08
*** ChuckC has joined #openstack-oslo16:08
*** JoshNang has joined #openstack-oslo16:14
*** amotoki has quit IRC16:16
stevemardhellmann, one more too: https://review.openstack.org/#/c/160761/16:17
*** pmathews1 has quit IRC16:18
*** ChuckC has quit IRC16:20
ihrachyshkadhellmann, 'subtle bugs' hm, sounds thrilling16:25
stevemardhellmann, thoughts on the policy_dir patch?16:25
dhellmannstevemar: bnemec has a patch to oslo.config to let us deprecate an option that we're removing: https://review.openstack.org/#/c/160475/16:26
dhellmannstevemar: that would let us update sdague's patch to register the option as deprecated with a warning, for now16:27
*** enikanorov has quit IRC16:28
stevemardhellmann, oh okay, so we could do it for any release then?16:28
dhellmannstevemar: we would need to follow the usual option deprecation period16:28
*** enikanorov has joined #openstack-oslo16:28
dhellmannright now we only have warnings on renames, this gives us a warning for something that is going to be removed16:29
stevemardhellmann, yeah, what i meant was that we don't have to necessarily fix the policy_dir stuff before announcing policy for public consumption16:29
dhellmannstevemar: right, we would do it well after a public release16:30
stevemaryeah, o/w we are waiting for another oslo.config release, and a g-r sync16:30
dhellmannright16:30
*** alexpilotti has quit IRC16:31
*** yamahata has joined #openstack-oslo16:33
dansmithdims: okay, I'm missing something16:36
dansmithdims: I've added o.vo to the oslo setup bit in devstack, but when I run it, I don't get any oslo.* libraries installed in /opt/stack16:36
dansmithdims: and I'm having a hard time wrapping my head around when this library would be installed from git in devstack, except when testing changes against itself16:37
*** david-lyle_afk has quit IRC16:37
dimsone sec16:37
dansmithI see where install_oslo is being called, which seems to be "always" even though I don't see them installed16:37
dansmithsure16:37
dhellmanndansmith: there's a variable to control whether it's installed from source or from pypi, let me find that16:38
dimsthe way the code is structured, if you want o.vo in /opt/stack you need LIBS_FROM_GIT=oslo.versionedobjects16:39
dimsin your stackrc16:39
dansmithah16:39
dhellmannright, that's it16:39
dansmithdims: so I looked and didn't see that set to anything in an oslo.messaging job16:39
dansmithhttp://logs.openstack.org/47/146047/26/check/gate-oslo.messaging-dsvm-functional-rabbit/1f5f5a2/logs/localrc.txt.gz16:39
dansmithwhich had me confused16:39
dimsthat's running against the oslo.messaging project, so it will pull down oslo.messaging from trunk and not pypi. in the same one if you wanted oslo.log from git then you would add the variable above16:40
dansmithbut that's an oslo.messaging job, shouldn't it have oslo.messaging set?16:40
dimsdansmith: no16:41
*** tsekiyama has joined #openstack-oslo16:41
dhellmanndims: so it installs from source automatically because it's the oslo.messaging project?16:41
dimsdhellmann: yes16:41
dansmithhmm16:41
dimsthen it looks for its dependencies and pulls either from pypi or git16:41
dimsdhellmann: see devstack-gate-setup-workspace-new.txt.gz which runs before devstack16:43
dimshttp://logs.openstack.org/47/146047/26/check/gate-oslo.messaging-dsvm-functional-rabbit/1f5f5a2/logs/devstack-gate-setup-workspace-new.txt.gz#_2015-03-02_12_17_52_99716:43
dimshttp://logs.openstack.org/47/146047/26/check/gate-oslo.messaging-dsvm-functional-rabbit/1f5f5a2/logs/devstacklog.txt.gz#_2015-03-02_12_27_37_36516:43
dhellmannit looks like LIBS_FROM_GIT defaults to DEVSTACK_PROJECT_FROM_GIT and the job definition sets that16:43
dhellmannsometimes tracking this stuff feels like a verse of http://www.timmyabell.com/music/lyrics/ol/oldlady.htm16:45
dimsack. so it's not set in the job itself16:45
dims:)16:45
openstackgerritMerged openstack/taskflow: Use debtcollector library to replace internal utility  https://review.openstack.org/15374616:45
dimsLOL16:45
dansmithheh16:46
*** cdent has quit IRC16:46
dimsdhellmann: queued up the muppet show rendition :) https://www.youtube.com/watch?v=qC_xO2aN_IA16:48
*** mtanino has joined #openstack-oslo16:49
dhellmannheh16:49
stevemardhellmann, just +3'd the patch to publicize register and Checks16:49
*** Guest32544 is now known as redrobot16:51
openstackgerritJoshua Harlow proposed openstack/taskflow: Add a fully functional orderedset  https://review.openstack.org/16021916:51
dhellmannstevemar: cool16:51
dhellmannstevemar: we should get everyone to agree on a plan for dealing with sdague's patch to remove the config option before making a public release16:52
dansmithdims: so after this devstack change lands, I need to create a job to run devstack+o.vo against o.vo patches, yes?16:54
openstackgerritJoshua Harlow proposed openstack/taskflow: Allow using ordered(set/dict) to retain ordering  https://review.openstack.org/16055816:54
dimsdansmith: looking what exists/runs now16:55
dhellmanndansmith: if you followed the instructions in the wiki you should already have that job16:55
stevemardhellmann, i figure the plan would be to deprecate the option as soon as a new oslo.config is available, then wait the usual amount of time16:56
dansmithdhellmann: hmm, I don't see anything on there about that16:56
stevemarunless i'm missing something in formalizing the process :)16:56
dansmithdhellmann: if I already had the job I wouldn't be able to merge code, right?16:56
dhellmannstevemar: sdague left a comment that he was hoping to skip that, so I figured we should at least talk about it so we're all on the same page16:56
dimsdansmith: looking at oslo.messaging, there are 2 tempest-dsvm jobs at check and one at gate, we could do the same for o.vo16:56
dansmithdims: okay.. where does that live? I thought it was in the config repo16:57
dimsproject-config repo16:57
dansmithI never maintain enough state in my head to remember all the linkages16:58
dhellmanndansmith: ah, you're missing "lib-forward-testing"16:58
dhellmanndims: ^^16:58
dimsdhellmann: yep16:58
dansmithmissing from where?16:58
openstackgerritDan Krause proposed openstack/taskflow: Persistence backend refactor  https://review.openstack.org/16088916:58
dhellmanndansmith: openstack-infra/project-config/zuul/layout.yaml16:58
dimsdansmith: see how i added dsvm large ops to oslo.log - https://review.openstack.org/#/c/160113/16:59
dhellmanndansmith: add "- name: lib-forward-testing" to the end of the template list around line 226616:59
dimsdansmith: y, that one would add the neutron dsvm job16:59
dimsif you need the dsvm large ops then see review above17:00
dhellmanndims: do we need large ops as a special case?17:00
dansmithdhellmann: dims I have lib-forward-testing already17:00
dimsdhellmann: y17:00
dhellmannbecause of how logging affects those jobs?17:01
dhellmanndansmith: do you? I don't see it in my copy...17:01
dansmithhmm, not in master, but in the stale version I had locally I did17:01
dhellmannah17:01
dansmithdims: did you add that early and remove it at some point maybe?17:01
*** dtantsur is now known as dtantsur|afk17:02
dimsdansmith: probably17:03
*** andreykurilin_ has joined #openstack-oslo17:03
dansmithokay17:03
dansmithhttps://review.openstack.org/16089317:03
dansmithdims: ^ yes?17:03
dhellmanndansmith: I believe that's what you want, yes17:04
dansmithcool17:04
dimsneed to add a line in project-config/jenkins/jobs/projects.yaml17:05
dimsdansmith: - 'lib-forward-testing-{name}'17:05
*** noelbk has joined #openstack-oslo17:06
dansmithdims: better?17:06
dimsdansmith: y17:07
dimslooks good17:07
dimsnow we wait for the jobs to verify the configuration17:07
dansmithdims: thanks for all the help :)17:07
dansmithyeah17:07
dimsmy pleasure!17:08
*** alexpilotti has joined #openstack-oslo17:11
*** andreykurilin_ has quit IRC17:12
*** andreykurilin_ has joined #openstack-oslo17:13
*** viktors is now known as viktors|afk17:13
openstackgerritMerged openstack/oslo.policy: provide more descriptive exception  https://review.openstack.org/16076117:16
*** ChuckC has joined #openstack-oslo17:23
*** ihrachyshka has quit IRC17:25
*** sputnik13 has joined #openstack-oslo17:26
openstackgerritDoug Hellmann proposed openstack-dev/oslo-cookiecutter: Drop py33 from default env list  https://review.openstack.org/16091217:27
openstackgerritDoug Hellmann proposed openstack/oslotest: Import nova's wrapper for nice testr output  https://review.openstack.org/15175717:32
*** e0ne has quit IRC17:32
openstackgerritMerged openstack/oslo.policy: Expose register and Check as part of public API  https://review.openstack.org/15952517:42
*** JoshNang has left #openstack-oslo17:43
openstackgerritOpenStack Proposal Bot proposed openstack/oslo.versionedobjects: Updated from global requirements  https://review.openstack.org/16091817:47
*** yamahata has quit IRC17:49
*** exploreshaifali has quit IRC17:56
*** shardy has quit IRC18:01
*** stpierre has joined #openstack-oslo18:03
*** ihrachyshka has joined #openstack-oslo18:06
*** pmathews has joined #openstack-oslo18:09
*** harlowja_away is now known as harlowja_18:12
*** alexpilotti_ has joined #openstack-oslo18:16
*** alexpilotti has quit IRC18:16
*** alexpilotti_ is now known as alexpilotti18:16
*** yamahata has joined #openstack-oslo18:19
*** crc32 has joined #openstack-oslo18:20
*** yamahata has quit IRC18:21
*** yamahata has joined #openstack-oslo18:22
*** andreykurilin_ has quit IRC18:22
*** andreykurilin_ has joined #openstack-oslo18:23
*** achanda has joined #openstack-oslo18:23
*** pasquier-s has quit IRC18:34
openstackgerritDoug Hellmann proposed openstack/stevedore: Fix test for finding multiple drivers  https://review.openstack.org/16093118:36
dhellmanndims, bnemec, jd__ : stevedore is currently blocked because of a poorly written test that has succumbed to a change outside of our code: https://review.openstack.org/16093118:38
bnemecLooking18:39
*** pasquier-s has joined #openstack-oslo18:40
openstackgerritDarragh Bailey proposed openstack-dev/pbr: Allow overwriting sphinx builder from command line  https://review.openstack.org/11747818:44
jungleboyjThere was a new oslo.messaging release yesterday?18:53
openstackgerritBen Nemec proposed openstack/oslo.config: Add ability to deprecate opts for removal  https://review.openstack.org/16047518:55
jungleboyjWondering if anyone else has seen things failing like this:  http://paste.openstack.org/show/186691/18:55
ihrachyshkadhellmann, hey. one more issue seeing with oslo.log, not sure how to debug.18:56
jungleboyjcinder-manage blows up with ImportError: cannot import name ResourceError in Juno.18:56
ihrachyshkadhellmann, I'm getting unit tests timing out18:56
ihrachyshkahttp://logs.openstack.org/38/159638/5/check/gate-neutron-python27/e4c56fd/console.html18:56
dhellmannjungleboyj: oslo.messaging was last tagged on 25 feb18:57
*** achanda has quit IRC18:57
*** achanda has joined #openstack-oslo18:57
jungleboyjdhellmann: Ok.  Any idea what could be causing that error?18:58
dhellmannjungleboyj: i haven't seen that particular failure, but can you import the module where the configured cinder driver is?18:58
dhellmannoh, it's the messaging driver18:58
ihrachyshkadhellmann, looks like test workers lock one by one, until no workers are left active18:58
dhellmannwell, still, can you import that module by hand?18:58
*** yamahata has quit IRC18:59
*** yamahata has joined #openstack-oslo18:59
dhellmannihrachyshka: do the neutron tests set up fake loggers to avoid writing to files?18:59
dhellmannwe usually see this sort of locking in logs caused by eventlet19:00
ihrachyshkadhellmann, right, there is FakeLogger in BaseTestCase19:00
ihrachyshkahttp://git.openstack.org/cgit/openstack/neutron/tree/neutron/tests/base.py#n7919:00
jungleboyjdhellmann: I am not sure where that import is coming from.  Oh, wait ...19:00
jungleboyjSo, it looks like kombu tries to use that.19:01
jungleboyjkombu is imported by /oslo/messaging/_drivers/impl_rabbit.py19:01
jungleboyjBut we are trying to use qpid I believe.19:01
*** daniel3_ has quit IRC19:01
ihrachyshkadhellmann, though I don't follow how it can lock workers19:01
*** yassine has quit IRC19:01
*** mriedem has quit IRC19:02
dhellmannihrachyshka: using the fake logger should help prevent this. Can you reproduce the problem by running just one of the neutron tests?19:02
dhellmannmaybe you have several locking, but can you get just one to lock?19:03
*** sputnik13 has quit IRC19:03
ihrachyshkadhellmann, it's not clear which of them lock since tox reports successful only19:03
dhellmannjungleboyj: the traceback doesn't show which module it's trying to load (we should improve that exception)19:04
jungleboyjdhellmann: :-)  Agreed.19:04
dhellmannihrachyshka: what about this problem points to oslo.log?19:05
*** noelbk has quit IRC19:06
*** daniel3_ has joined #openstack-oslo19:06
ihrachyshkadhellmann, it shows up in oslo.log migration patch only19:06
*** inc0 has joined #openstack-oslo19:07
ihrachyshkaI will execute tests locally on 4 core machine, I wonder whether it will lock. seems like it locks quickly in gate.19:07
jungleboyjdhellmann: Interesting.  It is trying to use rabbit and I don't think it should.19:07
ihrachyshkaaround 10 mins in gate19:08
dhellmannihrachyshka: ok, I don't know what to tell you. The log library is returning loggers from the stdlib and the fixture should be capturing their output. Unless you're also setting up logging to a file in the tests, there's no reason why oslo.log should behave any differently than what was there before.19:08
jungleboyjdhellmann: Cinder.conf has qpid specified.19:08
*** noelbk has joined #openstack-oslo19:08
dhellmannjungleboyj: is it in the right group?19:08
*** andreykurilin_ has quit IRC19:08
ihrachyshkadhellmann, well another case is me doing smth wrong and not understanding it in my patch19:08
dhellmannihrachyshka: that's possible, where's the patch?19:09
ihrachyshkadhellmann, you can check the list of non-obvious things I did for the patch in the commit message: https://review.openstack.org/15963819:09
*** mriedem has joined #openstack-oslo19:09
jungleboyjdhellmann: What do you mean?  In the config file?  It is in [DEFAULT[19:09
*** pblaho has joined #openstack-oslo19:09
dhellmannjungleboyj: ok, some of the options for the drivers moved around but that one should be ok19:10
jungleboyjOk.  Strange.19:10
dhellmannihrachyshka: I don't see anything obviously wrong in that patch19:15
dimsdhellmann: looks like ben +A'ed it19:19
*** amrith is now known as _amrith_19:23
openstackgerritDoug Hellmann proposed openstack/stevedore: Fix test for finding multiple drivers  https://review.openstack.org/16093119:24
dhellmanndims, bnemec : had to fix a pep8 error ^^19:24
dhellmannjd__: ^^19:25
openstackgerritDoug Hellmann proposed openstack/stevedore: Updated from global requirements  https://review.openstack.org/15780119:26
*** inc0 has quit IRC19:36
openstackgerritDan Krause proposed openstack/taskflow: Persistence backend refactor  https://review.openstack.org/16088919:36
*** dtantsur|afk has quit IRC19:36
*** dtantsur has joined #openstack-oslo19:41
*** ChuckC has quit IRC20:05
openstackgerritDan Krause proposed openstack/taskflow: Persistence backend refactor  https://review.openstack.org/16088920:08
*** liusheng has quit IRC20:10
*** liusheng has joined #openstack-oslo20:11
*** e0ne has joined #openstack-oslo20:11
openstackgerritJoshua Harlow proposed openstack/taskflow: Allow node finding to not do a deep search  https://review.openstack.org/16096520:17
openstackgerritMerged openstack/oslo.config: Typo in StrOpt docstring: Integer to String  https://review.openstack.org/15969820:20
*** yassine has joined #openstack-oslo20:25
ihrachyshkadhellmann, ok, the tests lock locally too, and I think I have one detected that locks20:29
ihrachyshka(neutron.tests.unit.test_wsgi.TestWSGIServer.test_app)20:29
*** e0ne is now known as e0ne_20:29
*** e0ne_ has quit IRC20:30
ihrachyshkadhellmann, oh, once I execute it via testr, I immediately get an error, smth about my patch20:30
*** e0ne has joined #openstack-oslo20:31
*** e0ne has quit IRC20:33
bknudsonwe've got some reviews up to use choices= option in oslo.config...20:33
bknudsonis it expected that the choices are listed in the help text?20:34
lbragstadwhen using oslo config generator20:34
bknudsonSeems like the config generator should be able to build the help text including the allowed options.20:34
bknudsonI'd ask dhellmann but he's busy in meetings.20:35
ihrachyshkadhellmann, heh, found a problem. I wonder why that made tests lock, but that's probably something in our test suite20:36
openstackgerritJoshua Harlow proposed openstack/taskflow: Allow node finding to not do a deep search  https://review.openstack.org/16096520:37
*** achanda has quit IRC20:38
*** e0ne has joined #openstack-oslo20:39
*** yassine has quit IRC20:43
*** jecarey has quit IRC20:44
*** achanda has joined #openstack-oslo20:46
*** pblaho has quit IRC20:50
harlowja_dhellmann so qq, wanted thoughts on, taskflow via a patch i added has new support for tooz to find things (using tooz group concepts), but taskflow is 2.6+ and tooz is 2.7+ i think most projects should be on 2.7+ now, and the last taskflow release that people are using is 2.6+; but to merge that patch in taskflow would need to drop 2.6 (i think) to align itself with 2.7+ that would be required to use that tooz code, any ideas on20:54
harlowja_ how that should be managed, just announce on the ML that the next version of taskflow will be 2.7+ only (and after that no more 2.6...)20:54
*** e0ne has quit IRC20:55
harlowja_or wait until kilo is done, and then do this, or something else...20:55
openstackgerritJoshua Harlow proposed openstack/taskflow: Use tooz to find/advertise workers (WIP, need to drop 2.6)  https://review.openstack.org/15149521:00
harlowja_mainly ^ that review, ha21:00
openstackgerritJoshua Harlow proposed openstack/taskflow: Allow node finding to not do a deep search  https://review.openstack.org/16096521:02
*** sputnik13 has joined #openstack-oslo21:03
*** pblaho has joined #openstack-oslo21:05
openstackgerritMin Pae proposed openstack/taskflow: Added a map and a reduce task  https://review.openstack.org/15767721:09
*** BrianShang_ has joined #openstack-oslo21:13
*** BrianShang has quit IRC21:15
*** _amrith_ is now known as amrith21:15
*** pblaho has quit IRC21:16
*** jecarey has joined #openstack-oslo21:16
*** kgiusti has left #openstack-oslo21:19
*** ChuckC has joined #openstack-oslo21:26
*** ihrachyshka has quit IRC21:33
*** stevemar has quit IRC21:41
*** pasquier-s has quit IRC21:44
*** pasquier-s has joined #openstack-oslo21:47
*** mtanino has quit IRC21:53
*** jgrimm is now known as zz_jgrimm21:57
*** mtanino has joined #openstack-oslo21:59
*** achanda has quit IRC22:01
sdaguedhellmann: just so I have my tomorrow list sorted out in my head - https://review.openstack.org/#/c/160407/ is blocked on a new feature landing and releasing in oslo.config, right?22:01
dhellmannsdague: well, only if we want to follow the deprecation policy. Your argument for just going ahead seems like it might set a bad precedent, but I'm willing to discuss it.22:02
sdagueI'm ok either way22:02
dhellmannin that case, I would rather play it safe22:03
*** andreykurilin_ has joined #openstack-oslo22:03
sdagueif it's blocked, could I ask you to WIP the change, and unWIP it once the feature needed is released? once I see it unWIP again I'll dig into it.22:03
*** zz_jgrimm is now known as jgrimm22:03
sdagueor bnemec or anyone that's going to stay on top of the oslo.config change getting out there22:04
dhellmannsdague: sure22:04
sdaguejust makes it easier to look at my Outgoing Review list and know what I can make progress on22:05
dhellmannsdague: I just +2 bnemec's config change, so if we can get that merged tomorrow we can release it and you can update22:05
sdaguethanks22:05
dhellmannsure22:05
*** achanda has joined #openstack-oslo22:07
harlowja_haypo message me when u get back, think i have some idea of what the meta-spec should be for our asyncio/threading specs22:12
*** devlaps has joined #openstack-oslo22:13
*** devlaps has quit IRC22:13
harlowja_have u guys seen https://cloud.google.com/dataflow/what-is-google-cloud-dataflow22:14
harlowja_how interesting...22:15
harlowja_copying taskflow, lol22:15
harlowja_dhellmann when u get free, let me know what u think about the py2.7 taskflow question (if u missed it i can repaste)22:17
*** jungleboyj has quit IRC22:17
dhellmannharlowja_: is taskflow used in any projects that are still using 2.6? we're worried about backports there22:18
harlowja_let me double check the current list i know about22:18
harlowja_hmmm, seems like cue is; might be useful to send out a ML question (to pick up projects idk about)22:20
harlowja_glance - no more 2.622:20
harlowja_cinder - no more 2.622:20
harlowja_cue - 2.6 (hmm)22:20
harlowja_pumphouse - no more 2.622:20
harlowja_others (that i can't think of right now, lol)22:20
dhellmannharlowja_: how about in stable branches? does taskflow have stable branches for releasing bug fixes, yet?22:23
harlowja_it doesn't have those yet22:23
harlowja_*although can create them if needed22:23
dhellmannok, I think we'll want to set that up before dropping 2.6 support so we have a way to do backports22:24
harlowja_agreed22:24
dhellmannother than that, I don't foresee any issues -- we could probably do the same across oslo, now, if we wanted to22:25
*** e0ne has joined #openstack-oslo22:26
harlowja_something to talk about next monday i guess22:29
dhellmannharlowja_: we have enough going on -- I'd rather have people doing reviews than making that change right now :-)22:30
harlowja_seems like octavia (also using taskflow) not supporting 2.6 (put up a couple of patches that removed listing 2.6 in there setup.cfg files)22:30
harlowja_dhellmann agreed22:30
harlowja_lifeless do u want to start a mini-CSP section, maybe on an etherpad somewhere; that we can merge into this meta/mega-spec, lol22:31
* harlowja_ is unsure if i should then also write up a dataflow one also (for taskflow), lol22:31
harlowja_how many approaches should we list, lol22:31
dhellmannharlowja_: for now, focus on convincing folks to move off of eventlet at all22:32
harlowja_k22:32
lifelessright22:32
harlowja_fair enough22:32
openstackgerritAmaury Medeiros proposed openstack/oslo-incubator: Use fill method to simplify string wrapping  https://review.openstack.org/15895122:32
lifelessthere is an angle of 'there is something much better out there' in any move-off story.22:32
lifelessbut I think we have enough pain points to not need to do that22:33
harlowja_well there are multiple paradigms out there22:33
harlowja_which is why i am sorta of leaning to not just saying 1 paradigm is the best22:33
harlowja_because each is different22:33
harlowja_*and has its own benefits/drawbacks...22:33
harlowja_CSP, dataflow, threads, asyncio, blah blah22:34
openstackgerritMerged openstack/debtcollector: Add a 'removed_kwarg' function/method decorator  https://review.openstack.org/15550622:43
*** stevemar has joined #openstack-oslo22:44
*** mriedem has quit IRC22:51
*** openstackgerrit has quit IRC22:51
*** openstackgerrit has joined #openstack-oslo22:51
*** andreykurilin_ has quit IRC22:55
openstackgerritJoshua Harlow proposed openstack/taskflow: Mark all current deprecations to get removed before 1.0  https://review.openstack.org/15681623:08
*** alexpilotti has quit IRC23:19
openstackgerritMerged openstack/stevedore: Fix test for finding multiple drivers  https://review.openstack.org/16093123:20
openstackgerritMerged openstack/stevedore: Updated from global requirements  https://review.openstack.org/15780123:25
*** jungleboyj has joined #openstack-oslo23:25
*** stevemar has quit IRC23:29
*** e0ne has quit IRC23:30
*** e0ne has joined #openstack-oslo23:34
*** prad has quit IRC23:36
*** sigmavirus24 is now known as sigmavirus24_awa23:36
*** e0ne has quit IRC23:47
*** achanda has quit IRC23:49
*** jgrimm is now known as zz_jgrimm23:50
*** achanda has joined #openstack-oslo23:56
*** gordc has quit IRC23:57
openstackgerritDean Troyer proposed openstack/cliff: Pass user command text to the Command object  https://review.openstack.org/14738123:59
« Monday, 2015-03-02 Index Wednesday, 2015-03-04 »

Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!