Thursday, 2018-11-22

*** takashin has joined #openstack-placement00:54
*** tetsuro has joined #openstack-placement01:41
*** tetsuro has quit IRC04:05
openstackgerritTakashi NATSUME proposed openstack/placement master: Fix a bug tag for placement doc  https://review.openstack.org/61943805:40
*** tetsuro has joined #openstack-placement07:27
*** takashin has left #openstack-placement08:00
*** helenafm has joined #openstack-placement08:25
*** e0ne has joined #openstack-placement09:17
*** tssurya has joined #openstack-placement09:23
*** dtantsur|afk is now known as dtantsur09:54
*** cdent has joined #openstack-placement09:58
* cdent waves09:58
*** dims has quit IRC11:45
*** e0ne has quit IRC11:46
*** e0ne has joined #openstack-placement11:59
lyarwoodcdent: https://bugs.launchpad.net/oslo.config/+bug/1804420 - would you mind taking a look at this as it's blocking RDO builds of placement at the moment, I'm not sure if there's anything we can do to workaround what I believe is an oslo.config bug but any ideas would be appreciated!12:08
openstackLaunchpad bug 1804420 in oslo.config "placement unit test placement.tests.unit.cmd.test_manage.TestCommandParsers.test_commands_associated fails on CentOS 7" [Undecided,New]12:08
cdentlyarwood: yeah, thanks for the heads up12:09
cdentweird12:10
openstackgerritChris Dent proposed openstack/placement master: Add integrated-gate-py35 template to .zuul.yaml  https://review.openstack.org/61756512:12
openstackgerritChris Dent proposed openstack/placement master: Remove keystoneauth1 opts from placement config group  https://review.openstack.org/61929912:12
cdentlyarwood: I suspect there is some kind of clue in the way the help handling is different between python2 and 3, but I'm not really figuring it out yet12:23
*** tetsuro has quit IRC12:58
*** diga has joined #openstack-placement13:21
lyarwoodcdent: https://github.com/python/cpython/commit/af26c15110b76195e62a06d17e39176d42c0511c found it finally13:21
lyarwoodcdent: that makes sense given the reproducer13:21
* cdent nods13:22
cdentWhat does that mean for Centos?13:22
lyarwoodI can asked to get this backported but it's going to take an age to land13:23
lyarwoodor we can work around this in placement itself13:23
cdentdid you see my comment on the bug about installing argparse from pypi?13:23
cdentif that fixes it, is that sufficient?13:24
lyarwoodno unfortunatley, RDO doesn't use pypi and I don't think that would override the version of argparse in stdlib anyway right?13:25
cdentI thought it did? And I guess I was assuming that if there's an argparse on pypi, there would be an rpm for it. If that's not the case, then I guess working around it in placement would be fine.13:26
cdentI know that when I created that big of code I did slightly differently from what nova does because doing a direct copy of nova made the help output be misleading about needing to say 'placement-manage command'13:27
cdentand there is no command command13:27
lyarwoodyeah there's no package for argparse AFAIK, I'll double check quickly13:28
openstackgerritChris Dent proposed openstack/placement master: Start a contributor goals document  https://review.openstack.org/61881113:28
lyarwoodyeah there's not, okay so I'll take a look at how Nova is doing this and post something this afternoon13:29
lyarwoodthanks again13:29
cdentno, thank you, I did very little. You and yours found (and are fixing) a bug. You people get the gold star.13:29
openstackgerritChris Dent proposed openstack/placement master: Remove keystoneauth1 opts from placement config group  https://review.openstack.org/61929913:34
*** cdent_ has joined #openstack-placement13:39
*** cdent has quit IRC13:39
*** cdent_ is now known as cdent13:39
cdentlyarwood: despite the test failing, does the command itself mostly work?13:46
lyarwoodcdent: I've not really tested it yet, I assumed it wouldn't because of this failure tbh13:52
openstackgerritLee Yarwood proposed openstack/placement master: manage: Do not use set_defaults on parent parser  https://review.openstack.org/61957413:52
cdentlyarwood: that somewhat fixes it but is a change for python314:04
cdentso we might some version conditionals14:04
lyarwoodcdent: ah, I did run the py36 tests and it appeared happy14:05
* lyarwood respins14:06
cdentthe tests probably don't evaluate the output sufficiently tight14:06
cdentthat is: they don't check the output to confirm the expected subcommands, it should14:06
cdent<- sloppy14:06
cdentor perhaps14:06
cdent<- rushed14:07
cdentif you're happy to add that test, please do, otherwise I'll do it in a followup14:07
lyarwoodcdent: ah no issues, I can add these now14:09
cdentthanks14:10
*** dims has joined #openstack-placement14:14
* cdent waves at dims14:17
openstackgerritLee Yarwood proposed openstack/placement master: manage: Do not use set_defaults on parent parsers with py2  https://review.openstack.org/61957414:56
lyarwoodcdent: ^ sorry just getting dragged into something else, I'm going to leave the output checks to a follow up, tbh I'm not sure how useful they are anyway if I'm honest.14:57
cdentno problem. I'll look into it if I have a chance. now wrestling with grenade weirdness. it's always something14:58
*** diga has quit IRC15:46
*** e0ne has quit IRC16:28
openstackgerritChris Dent proposed openstack/placement master: Add assertions to verify placement-manage error output  https://review.openstack.org/61961916:38
*** helenafm has quit IRC16:59
*** e0ne has joined #openstack-placement17:25
*** dtantsur is now known as dtantsur|afk17:35
* cdent glares at grenade17:37
openstackgerritMerged openstack/placement master: Fix a bug tag for placement doc  https://review.openstack.org/61943817:48
*** e0ne has quit IRC17:49
*** e0ne has joined #openstack-placement18:07
*** e0ne has quit IRC18:15
*** tssurya has quit IRC18:39

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!