| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2017-05-03 | |||
| 22:00:39 | EmilienM | mriedem: sounds like re-enabling WSGI didn't work well: http://logs.openstack.org/81/462281/1/check/gate-puppet-openstack-integration-4-scenario002-tempest-centos-7/7a37824/logs/nova/nova-compute.txt.gz#_2017-05-03_20_58_25_083 | |
| 22:02:06 | mriedem | EmilienM: that doesn't look like anything that would be related to running with wsgi | |
| 22:02:54 | EmilienM | I agree | |
| 22:26:18 | openstackgerrit | melanie witt proposed openstack/nova master: Use six.text_type() when logging Instance object https://review.openstack.org/461912 | |
| 22:29:33 | melwitt | mriedem, dansmith: there's the hopefully final version. ^ I'm not sure if the unit test is good enough, if I should instead do an exact comparison of the repr | |
| 22:36:02 | melwitt | clarkb: the repr is already converting to ascii with no errors. the UnicodeDecodeError is raised here https://github.com/python/cpython/blob/2e576f5/Lib/logging/__init__.py#L338 which implies it's trying to do an implicit conversion to unicode. I don't understand that | |
| 22:40:31 | mriedem | melwitt: nit inline | |
| 22:40:37 | mriedem | otherwise lgtm | |
| 22:40:53 | melwitt | mriedem: k, looking | |
| 22:41:29 | melwitt | ah, thanks | |
| 22:45:58 | openstackgerrit | melanie witt proposed openstack/nova master: Use six.text_type() when logging Instance object https://review.openstack.org/461912 | |
| 22:50:19 | mriedem | melwitt: should you be using assertIsInstance instead of assertIs? | |
| 22:50:42 | mriedem | i'm not entirely sure what the difference is, but i thought assertIs was like, == but without checking __eq__ | |
| 22:50:45 | mriedem | like same object in memory | |
| 22:51:04 | mriedem | https://docs.python.org/2/library/unittest.html#unittest.TestCase.assertIs | |
| 22:51:20 | melwitt | mriedem: I tried that first and it failed and said to check exact type, use assertIs https://docs.python.org/2/library/unittest.html#unittest.TestCase.assertIsInstance | |
| 22:51:59 | mriedem | oh huh, weird | |
| 22:52:00 | mriedem | ok | |
| 22:52:01 | mriedem | +2 | |
| 22:52:31 | melwitt | thanks | |
| 22:59:18 | mriedem | mtreinish: a few things in https://review.openstack.org/#/c/459470/ yet | |
| 23:01:47 | dansmith | mriedem: is and == are different | |
| 23:03:00 | mtreinish | mriedem: ok, I'll respin that now | |
| 23:03:00 | dansmith | >>> 1 == True | |
| 23:03:01 | dansmith | True | |
| 23:03:02 | dansmith | >>> 1 is True | |
| 23:03:03 | dansmith | False | |
| 23:04:38 | dansmith | mriedem: ah you said it above and I missed.. "same object in memory" is right | |
| 23:05:00 | dansmith | mriedem: but assertIsInstance is a subclass check | |
| 23:16:17 | openstackgerrit | Matthew Treinish proposed openstack/nova master: Remove cloudpipe APIs https://review.openstack.org/459470 | |
| 23:16:20 | mtreinish | mriedem: ^^^ that should address your comments | |
| #openstack-nova - 2017-05-04 | |||
| 00:00:25 | openstackgerrit | Matthew Treinish proposed openstack/nova master: Remove cloudpipe APIs https://review.openstack.org/459470 | |
| 00:26:25 | mriedem | dansmith: yeah gawd | |
| 00:43:00 | mriedem | mtreinish: we shall see | |
| 00:53:36 | mriedem | mtreinish: hells yes | |
| 00:53:37 | mriedem | +2 | |
| 00:53:49 | mriedem | alex_xu_: oomichi: https://review.openstack.org/#/c/459470/6 | |
| 01:09:38 | mtreinish | mriedem: awesome, thanks | |
| 01:11:33 | mtreinish | mriedem: I think once that merges it'll be my first nova bp implemented since the coverage extension back in grizzly | |
| 01:11:45 | mriedem | mtreinish: oh it's not done yet, | |
| 01:11:49 | mriedem | novaclient changes buddy | |
| 01:12:37 | mriedem | i don't think it mentions it | |
| 01:12:47 | mtreinish | it doesn't | |
| 01:13:07 | mriedem | certs was already deprecated in novaclient https://github.com/openstack/python-novaclient/commit/b6aea669630dc88296bdded963051a35cc9f995e | |
| 01:13:21 | mriedem | but cloudpipe wasn't https://github.com/openstack/python-novaclient/blob/master/novaclient/v2/cloudpipe.py | |
| 01:13:38 | mriedem | i can probably do that if you want, i've been doing a lot of that in novaclient lately | |
| 01:13:51 | mtreinish | mriedem: if you want, otherwise I'd push those up tomorrow | |
| 01:14:32 | mriedem | mtreinish: you should probably do it since it'll just enable me to procrastinate on summit prep | |
| 01:14:52 | mtreinish | mriedem: heh, what do you think it is for me? | |
| 01:14:54 | mriedem | don't use the certs one as an example, use the one that deprecated os-hosts in novaclient | |
| 01:15:07 | mriedem | use this as the example https://github.com/openstack/python-novaclient/commit/ea3b9f7fef28a7f06b51a3b176c85a1fc33cecaa | |
| 01:15:10 | mtreinish | ok | |
| 01:15:54 | mriedem | mtreinish: there is a nova onboarding session at the summit you can attend too | |
| 01:15:57 | mriedem | since you're a new contributor again | |
| 01:17:05 | mtreinish | mriedem: heh, I'll consider it | |
| 01:17:26 | mtreinish | I just looked my last nova commit before the cert removal was in feb. 2016 | |
| 01:18:14 | mriedem | probably when you were doing the tenant isolation stuff | |
| 01:18:17 | mriedem | no, that was older | |
| 01:18:38 | mtreinish | mriedem: https://review.openstack.org/#/c/279799/ | |
| 01:19:01 | mriedem | ah i remember that | |
| 01:22:36 | mtreinish | mriedem: at some point this cycle I'm also going to work on adding docs on how to install nova (or any project) from a tarball, because there were no instructions I could find for that (something from my dirty clouds talk) | |
| 01:23:42 | mriedem | ok | |
| 01:23:55 | mriedem | i noticed your talk was accepted but i think i have a conflict... | |
| 01:23:57 | mriedem | which sucks | |
| 01:26:20 | openstackgerrit | Vladik Romanovsky proposed openstack/nova master: neutron: retrieve physical network name from a multi-provider network https://review.openstack.org/430026 | |
| 01:28:40 | mtreinish | mriedem: oh well, there will always be the video | |
| 01:28:51 | mtreinish | mriedem: I linked you the spoiler picture of what the cloud looked like right? | |
| 01:29:12 | vladikr | jaypipes, dansmith, thanks for the reviews. I've update the multi-segments patch (comments mostly), could you please take a look? https://review.openstack.org/430026 | |
| 01:29:25 | vladikr | s/update/updated | |
| 01:30:50 | mriedem | mtreinish: i think so yeah | |
| 01:30:53 | jaypipes | vladikr: done! | |
| 01:31:17 | vladikr | jaypipes, thanks! | |
| 01:31:21 | jaypipes | np :) | |
| 01:32:53 | openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Use plain routes list for server-password endpoint instead of stevedore https://review.openstack.org/462049 | |
| 01:35:15 | openstackgerrit | Zhenyu Zheng proposed openstack/nova master: Use plain routes list for server-password endpoint instead of stevedore https://review.openstack.org/462049 | |
| 01:41:34 | mriedem | Kevin_Zheng: do you want to rebase this? https://review.openstack.org/#/c/459593/ | |
| 01:42:02 | mriedem | i suppose i can in gerrit | |
| 01:42:20 | openstackgerrit | Matt Riedemann proposed openstack/nova master: Support tag instances when boot(3/4) https://review.openstack.org/459593 | |
| 01:42:31 | Kevin_Zheng | mriedem sure, but why? | |
| 01:42:57 | mriedem | just because it messes up the patch set order when going through the changes in gerrit if it's not rebased | |
| 01:43:10 | mriedem | nothing major | |
| 01:43:35 | Kevin_Zheng | AH, OK, I was going to ask you to review this one if you have time | |
| 01:44:33 | mriedem | i can run through it quick | |
| 01:46:56 | Kevin_Zheng | One question | |
| 01:48:15 | Kevin_Zheng | if we are using placement, which is mandatory in O, it seems that we are unable to use AggregateCoreFilter properly | |
| 01:57:08 | mriedem | Kevin_Zheng: i'm not familiar with that filter; comments in your tags 3/4 patch btw | |
| 01:57:22 | mriedem | edleafe: bauzas: jaypipes: ^ re: placement + AggregateCoreFilter | |
| 01:58:03 | jaypipes | Kevin_Zheng: that filter wasn't touched by the placement service. | |
| 01:58:21 | mriedem | i don't think that's the question | |
| 01:58:48 | jaypipes | mriedem: I don't see a question... | |
| 01:58:53 | mriedem | this is a host aggregate with cpu_allocation_ratio meta set on it? | |
| 01:59:00 | mriedem | is that the idea? | |
| 01:59:34 | mriedem | jaypipes: so with placement in the filter scheduler we basically ignore the corefilter because we get compute node RPs via the VCPU resource class filtering right? | |
| 01:59:42 | openstackgerrit | fengzhr proposed openstack/nova master: Modify the description of flat_injected in nova.conf https://review.openstack.org/458789 | |
| 01:59:42 | mriedem | we don't ignore it, but it's redundant | |
| 01:59:58 | mriedem | and i think we said you could/should disable the CoreFilter, RAMFilter and DiskFilter once you're using placement | |
| 02:00:20 | jaypipes | yeah... | |
| 02:00:34 | jaypipes | but aggregate metadata should still be processed, IIRC. | |
| 02:01:00 | mriedem | yeah i'm not sure why this would be a problem, | |
| 02:01:15 | mriedem | it's just extra filtering once we get the list of compute node resource providers filtered on VCPU to start | |