Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2018-04-17
15:56:56 openstackgerrit Matt Riedemann proposed openstack/nova master: Cleanup ugly stub in TestLocalDeleteAllocations https://review.openstack.org/561997
15:58:11 jaypipes arvindn05: sure
15:59:22 arvindn05 jaypipes: ty...its hard to find bauzas in IRC because of timezone...if you know of suggestions for when i can find him in IRC that would help :)
15:59:38 bauzas arvindn05: jaypipes: I'm here but it's 6pm here
15:59:46 bauzas so I'll litterally disappear in like the next mins
16:00:00 arvindn05 ahh...its 9 AM here...
16:00:17 arvindn05 will try to catch you earlier next time then
16:00:26 bauzas arvindn05: you're US based ?
16:00:31 bauzas West coast I guess
16:00:39 arvindn05 yup...CA
16:01:11 openstackgerrit Merged openstack/nova stable/ocata: Add regression test for BFV+IsolatedHostsFilter failure https://review.openstack.org/543634
16:02:15 arvindn05 bauzas: please do respond to the comment offline if you get a chance....my second patch for the BP depends on this change so it is starting to block progress for me
16:02:42 openstackgerrit Ed Leafe proposed openstack/nova master: Add columns for generation to consumers https://review.openstack.org/557958
16:02:43 openstackgerrit Ed Leafe proposed openstack/nova master: Add the Consumer object https://review.openstack.org/561406
16:02:44 openstackgerrit Ed Leafe proposed openstack/nova master: Add Consumer objects to Allocations https://review.openstack.org/561407
16:02:45 openstackgerrit Ed Leafe proposed openstack/nova master: Add generation checking logic to Consumer https://review.openstack.org/561408
16:02:46 openstackgerrit Ed Leafe proposed openstack/nova master: Add a microversion for consumer generation https://review.openstack.org/562000
16:03:03 edleafe jaypipes: efried: cdent: ^^ now with microversion!
16:03:28 efried ack
16:03:43 edleafe jaypipes: efried: cdent: I used microversion 1.99 to avoid having to rebase as things merge. It will be changed to a real version once this gets close
16:04:04 bauzas arvindn05: okay, will try
16:04:06 efried edleafe: You'll still have to rebase, of course.
16:04:11 bauzas it was a busy day for me today
16:04:12 arvindn05 bauzas: ty
16:04:22 bauzas so hoping tomorrow will be different
16:04:35 bauzas for folks wondering, if you see me silent here, that means something else
16:05:35 openstackgerrit Merged openstack/nova stable/ocata: Handle volume-backed instances in IsolatedHostsFilter https://review.openstack.org/543635
16:05:42 edleafe efried: yeah, but just once more
16:05:43 openstackgerrit Merged openstack/nova stable/ocata: Fix docs for IsolatedHostsFilter https://review.openstack.org/543636
16:07:16 efried kashyap: You still around?
16:07:27 kashyap efried: Hi, yeah. But will step out in 10 minutes
16:07:34 kashyap Need to feed self
16:07:36 efried kashyap: Problem in your paste http://paste.openstack.org/show/719390/ is at L3
16:07:50 efried @mock.patch takes one positional arg.
16:08:16 kashyap efried: Yeah, I forgot to note that, I tried by removing the "return_value" from the @mock.patch
16:08:30 efried kashyap: If you change it to
16:08:30 efried @mock.patch('nova.virt.libvirt.migration.get_updated_guest_xml',
16:08:30 efried return_value='')
16:08:30 efried you get by that error (and on to the next :)
16:08:46 kashyap Right
16:08:49 efried if you wanted to mock.patch.object instead, then the first arg needs to be an object, the second a string naming a symbol.
16:09:07 kashyap I see
16:09:24 kashyap efried: I'm squashing a couple of commits that belong to the same logical change.
16:09:41 efried kashyap: Does this get you unwedged, or did I misunderstand what you were stuck on?
16:09:42 kashyap And then I'll duke it out tomorrow with the rest. I'll let you know
16:10:28 kashyap efried: A bit ago, I was tinkering by removing the return_value'' in the @mock.patch, and that got me somewhere
16:10:54 efried kashyap: The only thing you were probably accomplishing there was making the return value the name of the method you thought you were mocking.
16:11:17 kashyap Hmm
16:11:25 kashyap Let me get the full test, for a quick once over, when you have time
16:13:49 kashyap efried: That's the updated test: http://paste.openstack.org/show/719400/
16:15:19 efried kashyap: Change L3 to use a single string: @mock.patch('nova.virt.libvirt.migration.get_updated_guest_xml')
16:16:03 openstackgerrit Merged openstack/nova stable/queens: ironic: Get correct inventory for deployed node https://review.openstack.org/561923
16:17:03 openstackgerrit Merged openstack/nova stable/ocata: Functional test for regression bug #1713783 https://review.openstack.org/505160
16:17:03 openstack bug 1713783 in OpenStack Compute (nova) ocata "After failed evacuation the recovered source compute tries to delete the instance" [High,In progress] https://launchpad.net/bugs/1713783 - Assigned to Illes Elod (elod-illes)
16:17:11 openstackgerrit Merged openstack/nova master: Deduplicate notification samples Rocky - 1 https://review.openstack.org/560296
16:32:04 openstackgerrit Matt Riedemann proposed openstack/nova stable/ocata: Pass the correct image to build_request_spec in conductor.rebuild_instance https://review.openstack.org/516404
16:32:05 openstackgerrit Matt Riedemann proposed openstack/nova stable/ocata: Set error state after failed evacuation https://review.openstack.org/518733
16:42:48 openstackgerrit Jay Pipes proposed openstack/nova master: support multiple member_of qparams https://review.openstack.org/561315
16:46:35 openstackgerrit Chris Dent proposed openstack/nova master: Parse forbidden in extra_specs https://review.openstack.org/561677
16:52:34 mriedem melwitt: if we can get this novaclient change in https://review.openstack.org/#/c/561520/ we can do a novaclient release to align with rocky-1
16:52:55 mriedem which we should probably do either way; i think i waited too long in queens to release novaclient and it caused some havoc with OSC and horizon late in queens
16:53:21 melwitt mriedem: okay, yeah I did want to release novaclient for r-1. I'll take a look
16:58:09 openstackgerrit Balazs Gibizer proposed openstack/nova master: Deduplicate notification samples Rocky - 4 https://review.openstack.org/560731
16:58:10 openstackgerrit Balazs Gibizer proposed openstack/nova master: Deduplicate notification samples Rocky - 5 https://review.openstack.org/560734
16:58:11 openstackgerrit Balazs Gibizer proposed openstack/nova master: Deduplicate notification samples Rocky - 6 https://review.openstack.org/560737
16:58:12 openstackgerrit Balazs Gibizer proposed openstack/nova master: Add request_id to instance action notifications https://review.openstack.org/553288
17:09:32 melwitt mriedem: why do you think disabling the UCA could cause the ceph job problem again? problems occur when installs by pip get mixed with installs by apt-get. devstack will only pip install over a apt-get installed package if upper-constraints aren't being met
17:12:37 melwitt with the Pike UCA, the version of python-paste installed by apt-get did not meet upper-constraints so devstack pip installed a newer version over it. so the matching python-pastedeploy package installed by apt-get had to be manually pip force installed in order to have paste and pastedeploy both installed by pip so pastedeploy could refer to modules in paste
17:13:10 mriedem melwitt: so that's not a problem if you're just using stock ubuntu 16.04?
17:13:51 melwitt it depends on what versions of python-paste and python-pastedeploy are in the repo for 16.04 and if only *one* of them does not meet upper-constraints
17:16:08 melwitt problems occur when only one of them does not meet upper-constraints because then devstack will upgrade only one with pip and then they can't refer to each other anymore. really, the hack probably should have been to pip force install both of the packages to guarantee they will always be able to find each other
17:16:58 jaypipes mriedem, melwitt, dansmith: any of you seen "nova.exception.InvalidDiskInfo: Disk info file is invalid: qemu-img failed to execute on /fake/path : Unexpected error while running command." pop up in unit test failures recently?
17:17:33 melwitt jaypipes: I have not
17:17:50 mriedem unit tests?
17:17:55 mriedem i probably know the one that is causing it
17:18:23 jaypipes mriedem: ya. if I run tox -epy27, I get a single failure, always in nova.tests.unit.virt.test_images.QemuTestCase.test_qemu_info_with_errors
17:18:29 mriedem https://review.openstack.org/#/c/553067/
17:19:08 openstackgerrit Chris Dent proposed openstack/nova master: Parse forbidden in extra_specs https://review.openstack.org/561677
17:20:30 openstackgerrit Jay Pipes proposed openstack/nova master: support multiple member_of qparams https://review.openstack.org/561315
17:20:48 jaypipes mriedem, melwitt, dansmith: only seems to happen for me locally. so not really concerned about it...
17:21:11 mriedem probably a bad global mock on os.exists
17:22:09 melwitt do those tests require that you have qemu-img installed maybe?
17:29:18 jaypipes melwitt: not sure... I can check
17:35:02 dansmith sean-k-mooney: if you're still around and can re-apply your +1 to this I think that'd be good: https://review.openstack.org/#/c/484997/20
17:37:53 sean-k-mooney dansmith: done. it looks like v20 is just some pep8 fixs/whitespace clean up for v19 so ya it looks fine to me
17:38:25 dansmith sean-k-mooney: yes, just want your +1 on there :)
17:38:31 dansmith s/just/I just/
17:44:52 mriedem melwitt: unit tests should never require that you have native packages installed
17:45:12 melwitt I know, but I've run into issues with that in the past
17:52:54 melwitt Kevin_Zheng: your blueprint "Add request_id to instance action notifications [END DATE: 2018-05-01]" has been added to a review runway. please let us know if the next two weeks are not a good time for fast iteration on review https://etherpad.openstack.org/p/nova-runways-rocky
17:53:38 dansmith I assume we won't hear back on that until later this afternoon yeah?
17:54:07 melwitt maybe not. I didn't know what time zone he is
17:55:27 openstackgerrit Chris Dent proposed openstack/nova-specs master: Update forbidden spec to reflect chosen nova-side format https://review.openstack.org/562012
17:55:30 dansmith melwitt: china
17:55:41 melwitt a-ha, cool
17:55:59 melwitt the series is half approved already and currently actively reviewed/responded so I expect it's going to be okay
18:07:39 imacdonn mriedem dansmith efried: need to decide what to do with https://review.openstack.org/#/c/558089/ - I'm leaning back towards just removing the check in init_host(). If it needs something more than that, I probably need to let it go to someone else
18:07:58 openstackgerrit Merged openstack/python-novaclient master: Fix comments in novaclient/tests/unit/fakes.py https://review.openstack.org/561772
18:08:48 efried imacdonn: I thought we decided to basically leave the old check as is but switch to checking auth_type instead of region
18:09:07 imacdonn efried: I tried that, but it gets all complicated with unit testing again

Earlier   Later