Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2017-06-27
06:58:53 gcb mriedem: Could you please revisit https://review.openstack.org/#/c/466787 ?
07:10:44 openstackgerrit Chason Chan proposed openstack/nova master: DOC: Import glossary from openstack-manuals https://review.openstack.org/477788
07:30:55 openstackgerrit Chason Chan proposed openstack/nova master: DOC: Import glossary from openstack-manuals https://review.openstack.org/477788
07:38:22 openstackgerrit Chason Chan proposed openstack/nova master: import the admin guide content from openstack-manuals https://review.openstack.org/477497
07:46:48 openstackgerrit Artom Lifshitz proposed openstack/nova master: Delete nic metadata when detaching interface https://review.openstack.org/435387
07:46:56 openstackgerrit Artom Lifshitz proposed openstack/nova master: Delete disk metadata when detaching volume https://review.openstack.org/435082
07:46:57 openstackgerrit Artom Lifshitz proposed openstack/nova master: API support for tagged device attachment https://review.openstack.org/376021
07:46:57 openstackgerrit Chason Chan proposed openstack/nova master: switch to openstackdocstheme https://review.openstack.org/477751
08:05:13 openstackgerrit Rodolfo Alonso Hernandez proposed openstack/os-vif master: Read datapath_type from VIF object https://review.openstack.org/474914
08:05:37 openstackgerrit Rikimaru Honjo proposed openstack/nova master: compute: Do not create bdms before successfully reserving a volume https://review.openstack.org/290793
08:07:54 openstackgerrit Merged openstack/nova master: Plumbing for tagged nic attachment https://review.openstack.org/415622
08:09:36 openstackgerrit Rodolfo Alonso Hernandez proposed openstack/os-vif master: Rehome unit tests to ``tests\unit`` folder https://review.openstack.org/476623
08:13:17 openstackgerrit Gao Yuan proposed openstack/nova master: Remove mox from nova.tests.unit.api.openstack.compute.test_limits https://review.openstack.org/477812
08:16:36 openstackgerrit Gao Yuan proposed openstack/nova master: Remove mox from nova.tests.unit.compute.test_keypairs.py https://review.openstack.org/477812
08:29:14 openstackgerrit Stephen Finucane proposed openstack/nova master: Monkey patch the blockdiag extension https://review.openstack.org/476159
08:29:14 openstackgerrit Stephen Finucane proposed openstack/nova master: docs: Document the scheduler workflow https://review.openstack.org/475810
08:49:57 chandra_sekhar I have multiple compute nodes with LVM storage
08:50:08 chandra_sekhar will live migration work with this kind of setup ?
09:23:40 openstackgerrit Chason Chan proposed openstack/nova master: switch to openstackdocstheme https://review.openstack.org/477751
09:25:54 openstackgerrit Chason Chan proposed openstack/nova master: switch to openstackdocstheme https://review.openstack.org/477751
09:47:33 sfinucan artom: All of that series reviewed with just a few comments left
09:50:13 artom sfinucan, muchas gracias senor
09:51:12 artom sfinucan, I re-ordered the patches to put the delete nic one at the bottom so that it doesn't hold up the delete bdm one
09:51:30 artom If Matt's around and you both feel like re-+2ing it we can get it off our plate
09:52:02 artom I had to adjust the unit test since MagicMock didn't like the new 'tag' in vif construct
09:52:11 sfinucan artom: Yup, that's what I've done
09:53:57 sfinucan By "that", I mean +W'd. I don't think the unit test change is too controversial
09:59:18 artom sfinucan, in 10 years we'll find an obscure bug that would have been noticed right away had Matt reviewed the change and pointed out a tricky way in which the new unit test doesn't cover every case
10:00:03 sfinucan artom: Aye, but by then I'll have retired to southern Italy and you won't be able to find me
10:00:12 ralonsoh jaypipes: hi
10:03:23 openstackgerrit Merged openstack/nova master: Remove mox in test_availability_zone.py https://review.openstack.org/477439
10:20:04 openstackgerrit Rikimaru Honjo proposed openstack/nova master: compute: Do not create bdms before successfully reserving a volume https://review.openstack.org/290793
11:05:07 openstackgerrit iswarya vakati proposed openstack/nova master: Add python 3.5 in classifier https://review.openstack.org/477887
11:20:23 openstackgerrit Pavel Gluschak (scsnow) proposed openstack/nova master: VStorage: changed default log path https://review.openstack.org/458557
11:25:32 openstackgerrit Rodolfo Alonso Hernandez proposed openstack/os-vif master: Add abstract OVSDB API https://review.openstack.org/476612
11:26:43 artom The hyperv tests are... weird
11:36:23 artom As far as I can tell everything, including parameters passed to methods, and said methods return values, are mocked out
11:36:38 artom I'm not sure it's testing anything substantial
11:41:33 openstackgerrit Rodolfo Alonso Hernandez proposed openstack/os-vif master: Use ``assert_has_calls`` to check function calls https://review.openstack.org/477469
11:49:21 openstackgerrit OpenStack Proposal Bot proposed openstack/nova master: Imported Translations from Zanata https://review.openstack.org/477091
11:56:03 openstackgerrit Gao Yuan proposed openstack/nova master: Remove mox from nova.tests.unit.virt.test_block_device https://review.openstack.org/477904
12:05:37 bhagyashri_s johnthetubaguy: Hi, Waiting for long time, Could you please review patch https://review.openstack.org/#/c/409644 ? address all review comment. Thank you.
12:05:59 artom alex_xu, hi, questions about your feedback in https://review.openstack.org/#/c/376021/71/nova/tests/unit/api/openstack/compute/test_attach_interfaces.py@251
12:07:06 artom Having those tests in InterfaceAttachTestsV21 allowed me to use all the scaffolding that's already there. Should the new class inherit from InterfaceAttachTestsV21? That would replicate a lot of tests needlessly
12:08:21 artom If it doesn't inherit, we have to copy-paste said scaffolding
12:08:33 artom Well, maybe not the *entire* scaffolding
12:08:40 artom Let me see what I can get away with
12:08:48 artom Err, thanks for being my rubber duck :/
12:09:18 artom And if I was talking to myself the whole time and no one's here, even better, it means I haven't wasted anyone's time :)
12:15:17 alex_xu artom: we can create a new class, and you needn't copy too much code, since your test is only about schema validation, so the api method will fail at very early stage, so you needn't mock or stub_out anything
12:16:33 alex_xu artom: the new class just inherit from test.NoDBTestCase
12:16:51 openstackgerrit OpenStack Proposal Bot proposed openstack/nova master: Updated from global requirements https://review.openstack.org/477982
12:17:26 artom alex_xu, yep, that totally works, thanks :)
12:17:36 alex_xu artom: np :)
12:19:45 openstackgerrit OpenStack Proposal Bot proposed openstack/os-vif master: Updated from global requirements https://review.openstack.org/467173
12:21:51 jaypipes ralonsoh: heya
12:22:10 openstackgerrit OpenStack Proposal Bot proposed openstack/python-novaclient master: Updated from global requirements https://review.openstack.org/478083
12:24:39 jaypipes ralonsoh: did you need me for something?
12:27:09 openstackgerrit Artom Lifshitz proposed openstack/nova master: Delete disk metadata when detaching volume https://review.openstack.org/435082
12:27:10 openstackgerrit Artom Lifshitz proposed openstack/nova master: API support for tagged device attachment https://review.openstack.org/376021
12:34:42 ralonsoh jaypipes: hi, I left a comment in https://review.openstack.org/#/c/476612/
12:34:54 ralonsoh jaypipes: about the abstraction of the ovsdb access
12:38:44 jaypipes ralonsoh: ah, gotcha
12:47:44 sfinucan sdague, alex_xu
12:47:49 sfinucan Can I kill this docs? https://docs.openstack.org/developer/nova/api_plugins.html
12:47:51 sfinucan *doc
12:48:45 alex_xu sfinucan: yea, but probably need new doc to teach people how to add new API to the nova
12:49:00 sfinucan Well, we don't do it as plugins anymore, right?
12:49:28 sfinucan alex_xu: Is this what you're looking for? https://docs.openstack.org/developer/nova/api_microversion_dev.html
12:55:19 alex_xu sfinucan: yea, although it is specific to the microversion
12:56:31 sfinucan alex_xu: Is there any other way to extend the API than with a microversion now?
12:56:41 sfinucan I thought we didn't let people extend the API anymore?
12:57:11 sdague sfinucan: we don't
12:57:21 alex_xu sfinucan: yes, that doc is for the nova developer add API to the upstream
12:57:32 sdague however the api_plugins page is the only one that talks about the how you add new bits of api code
12:57:38 sfinucan Ohhh, is it now
12:57:50 sdague the microversion page is just about changing responses in existing ones
12:57:56 sfinucan OK, I won't delete it then
12:58:05 sdague it definitely needs a rewrite
12:58:07 sfinucan But I think it will need to be reworked
12:58:08 sfinucan yup
12:58:11 sfinucan I'll include a TODO
12:58:25 sdague ++
12:58:31 alex_xu yea, probably we need to mention this new route list https://github.com/openstack/nova/blob/master/nova/api/openstack/compute/routes.py
12:58:48 sfinucan indeed
13:11:06 jaypipes bauzas, edleafe, Guest63234: will be doing my best to push revisions on the allocation requests series while on a 5-hour long sales call today. :(
13:12:42 openstackgerrit Alex Szarka proposed openstack/nova master: Refactor init_instance:resume_guests_state https://review.openstack.org/349582
13:16:00 edleafe jaypipes: ugh
13:16:07 jaypipes edleafe: yeah.
13:16:13 edleafe jaypipes: will reply to your reply on the patch
13:16:17 jaypipes edleafe: regarding that...
13:16:48 jaypipes edleafe: so like I mentioned on my comment, I actually tried that approach first and ended up having to change just as many lines eventually in the unit tests.
13:17:12 jaypipes edleafe: It really comes down to whether we want the filter scheduler driver to be the thing that claims resources or not.
13:17:39 jaypipes edleafe: or the manager class.
13:18:16 edleafe jaypipes: can you explain why we can't have chance scheduler do claims? And I also thought that we weren't going to sidetrack this for chance scheduler
13:18:21 jaypipes edleafe: my objective was to make as few changes to the filter scheduler driver class as possible.
13:18:40 edleafe jaypipes: because if there is no claim made when a compute node gets a build request, it will claim
13:18:44 jaypipes edleafe: chance scheduler doesn't know uuids.
13:19:27 jaypipes edleafe: it's more than the chance scheduler. it's a separation of concerns thing.
13:19:47 jaypipes edleafe: I'm trying to leave the filter scheduler driver doing filtering and weighing.

Earlier   Later