Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2020-04-02
00:48:55 gmann brinzhang_: yeah, tests covering all those cases will be good
00:57:46 openstackgerrit Ghanshyam Mann proposed openstack/nova master: Introduce scope_types in server diagnostics https://review.opendev.org/716803
01:05:31 openstackgerrit Ghanshyam Mann proposed openstack/nova master: Add new default roles in security group policies https://review.opendev.org/716793
01:08:45 openstackgerrit Ghanshyam Mann proposed openstack/nova master: Add new default roles in server diagnostics policies https://review.opendev.org/716805
01:30:51 brinzhang_ gmann: context.can(va_policies.POLICY_ROOT % 'swap') this policy check, does we need to add target={} in it?
01:31:33 openstackgerrit Ghanshyam Mann proposed openstack/nova master: Pass the actual target in server diagnostics policy https://review.opendev.org/716810
01:36:45 openstackgerrit melanie witt proposed openstack/nova master: Move nova-manage placement heal_allocations to nova-audit https://review.opendev.org/708788
01:51:10 melwitt gmann: hey, just to clarify, if I understand correctly we're going to address the warning log issue in a later patch before FF with the plan being to remove the warning log and use Upgrade reno instead? and we'll have a nova doc page explaining the detail of the change in defaults with instructions for users on how to review the new defaults and use them? (enforce_scope = True and how to generate new defaults for review)?
03:45:05 openstackgerrit Brin Zhang proposed openstack/nova master: Separate update and swap volume policies https://review.opendev.org/711194
03:46:37 openstackgerrit Guo Jingyu proposed openstack/nova master: WIP: nova supports password encrypted VNC https://review.opendev.org/622336
03:48:35 brinzhang_ gibi, gmann: I addressed your comments in https://review.opendev.org/#/c/711194/11, but when I want to add the "policy check fails for swap + update due to update policy failure" case, that cannot get what I want, see http://paste.openstack.org/show/791501/
03:48:56 brinzhang_ gibi, gmann: Anything I missed?
07:08:08 gibi good morning nova
07:08:16 gibi brinzhang_: will get to it soon
07:08:47 brinzhang_ good morning gibi
07:08:48 gibi _erlon_: it is up as usual. I update the agenda as soon
07:08:58 brinzhang_ gibi: thanks, wait for your check
07:11:24 brinzhang_ gibi: the noVnc feature, my colleague pushed the PoC code, maybe we can done of that feature, I will give him help as soon as possiable https://review.opendev.org/#/c/622336/
07:12:06 brinzhang_ gibi: you dont too review now, when it's ready, I will respin
07:14:14 gibi brinzhang_: ack, I saw the patch getting up
07:49:34 gibi brinzhang_: regarding http://paste.openstack.org/show/791501/ the test fails as the calls https://review.opendev.org/#/c/711194/11/nova/api/openstack/compute/volumes.py@488 raises instead of returns False
07:51:05 gibi so somehow you setting @ for swap policy does not pass
07:51:17 brinzhang_ but I overwrite to self.policy_root % "swap": "@",
07:51:21 gibi yeah
07:51:30 brinzhang_ why?
07:51:40 gibi I have no clue yet
07:53:53 gibi there are 3 test classes all inheriting the unit test 2 of the fails 1 passes
07:54:01 gibi so something is different in those classes
07:55:14 gibi this passes VolumeAttachPolicyTest
07:55:30 gibi VolumeAttachNoLegacyPolicyTest and .VolumeAttachScopeTypePolicyTest fails
07:55:37 brinzhang_ yes, it's ok in VolumeAttachPolicyTest class, and not ok in VolumeAttachScopeTypePolicyTest and VolumeAttachNoLegacyPolicyTest class
07:56:32 brinzhang_ I tried move it to VolumeAttachScopeTypePolicyTest class, but I think I cannot do that, I should keep them in the base class, I think
07:56:49 brinzhang_ of course, in VolumeAttachScopeTypePolicyTest it's also failed
08:13:20 gibi I gave up, we need gmann to look at these tests
08:17:33 brinzhang_ gibi: yeah, thanks also ^^
08:32:09 gibi stephenfin: just to double check is this bp done now? https://blueprints.launchpad.net/nova/+spec/remove-nova-network-ussuri
08:32:21 gibi I don't see any open patches
08:32:48 gibi stephenfin: nvm, I see that the bp is in Implemented state
08:52:55 stephenfin Yeah, I updated that last week
08:53:34 gibi stephenfin: thanks
08:54:14 gibi I'm going through the BPs targeted to ussuri just by eye skipped a line an thought nova-net bp is not closed
08:54:21 gibi s/by/my
08:54:46 gibi stephenfin: how familiar you are with the oslo policy stuff?
08:55:04 stephenfin I've been learning it slowly
08:55:05 gibi stephenfin: brinzhang_ has a test that fails on policy problem and I didn't find the solution
08:56:33 gibi stephenfin: this is the problematic test (in the comment in the paste) https://review.opendev.org/#/c/711194/11/nova/tests/unit/policies/test_volumes.py@176 if you have time. If not then no worries we will ping gmann later
08:59:34 brinzhang_ gibi, thanks mention these, if stephenfin can give some help, I will be appreciate
09:03:07 stephenfin will do
09:03:35 luyao stephenfin: Hi, I hope you could continue looking at vpmem live migration support https://review.opendev.org/#/c/687856/ if you have time. :)
09:17:03 brinzhang_ gibi, dansmith: I will modify the SPEC in 2020/4/4 or 2020/4/5, to show the new PUT api changes, I think it's not later
09:17:38 gibi brinzhang_: sure. doc changes can go in after FF. and the spec repo is not even released so that is OK to update even later
09:20:55 brinzhang_ gibi: thank, if so, I will update that while I have more free time ^^
09:21:10 gibi brinzhang_: thanks
09:21:17 brinzhang_ gibi: np
09:39:11 openstackgerrit Stephen Finucane proposed openstack/nova master: hacking: force explicit import of python's mock https://review.opendev.org/708768
09:39:11 openstackgerrit Stephen Finucane proposed openstack/nova master: Use unittest.mock instead of third party mock https://review.opendev.org/714676
09:39:11 openstackgerrit Stephen Finucane proposed openstack/nova master: Remove future imports https://review.opendev.org/714675
10:28:51 stephenfin gibi, brinzhang_: RE: https://review.opendev.org/#/c/711194/11/nova/tests/unit/policies/test_volumes.py@176 - you're using the wrong HTTPRequest object
10:29:56 stephenfin you're modifying 'nova.context' on 'self.req' and then calling things with a local 'req' that doesn't have 'nova.context' configured
10:30:41 brinzhang_ stephenfin: I should add v2.85 support, what I do to change req = fakes.HTTPRequest.blank('', version='2.85')?
10:30:58 gibi stephenfin: good eyes! thanks!
10:31:23 stephenfin brinzhang_: I'm guessing but I think think self.req.version = '2.85' should do the trick??
10:31:37 stephenfin If not, you need to do req = fakes.HTTPRequest.blank('', version='2.85')
10:31:46 brinzhang_ stephenfin: let me try
10:31:57 stephenfin and the do e.g. req.environ['nova.context'] = self.system_foo_context
10:32:03 stephenfin ^ that's what I did locally
10:33:09 stephenfin brinzhang_: Actually, in the setUp, is there any reason not to configure 'self.req' with a version argument?
10:33:32 stephenfin You're testing policy, not microversion support, so you'll always want to use a known working version, right?
10:33:52 brinzhang_ yes
10:35:45 brinzhang_ stephenfin: you mean I should change self.req to req in setup()?
10:35:52 stephenfin no
10:36:24 stephenfin instead of self.req = fakes.HTTPRequest.blank('')
10:36:33 stephenfin do self.req = fakes.HTTPRequest.blank('', version='2.85')
10:36:49 stephenfin here https://review.opendev.org/#/c/711194/12/nova/tests/unit/policies/test_volumes.py@78
10:37:19 brinzhang_ no, it dosenot only for v2.85
10:37:35 brinzhang_ so I just used req in v2.85 policy tests
10:37:36 stephenfin ah, I see now, this is only for the 'update_volume' case
10:37:41 brinzhang_ yes
10:37:49 stephenfin gotcha
10:37:57 brinzhang_ req.environ['nova.context'] = self.system_foo_context
10:38:11 stephenfin yup, that'll do so ^
10:38:13 brinzhang_ I added this, it run ok now, will update
10:38:18 brinzhang_ thansk stephenfin, gibi
10:47:28 openstackgerrit Brin Zhang proposed openstack/nova master: Separate update and swap volume policies https://review.opendev.org/711194
10:47:33 brinzhang_ gibi, stephenfin: done thanks
10:48:43 brinzhang_ stephenfin: can you +A for this patch https://review.opendev.org/#/c/693828/25? it's ready for everything, that comment inline, I will fixed by FUP
10:50:39 stephenfin brinzhang_: I haven't looked at that in a while so I'll need a while
10:50:52 openstackgerrit Stephen Finucane proposed openstack/nova master: api: Add framework for extra spec validation https://review.opendev.org/704643
10:50:53 openstackgerrit Stephen Finucane proposed openstack/nova master: api: Add microversion 2.85, extra spec validation https://review.opendev.org/708436
10:50:53 openstackgerrit Stephen Finucane proposed openstack/nova master: api: Add support for new cyborg extra specs https://review.opendev.org/716222
10:50:54 openstackgerrit Stephen Finucane proposed openstack/nova master: docs: Add documentation for flavor extra specs https://review.opendev.org/710037
10:51:35 stephenfin gibi: addressed the comments from Sundar and removed those errant prints from the last doc patch ^
10:51:38 brinzhang_ stephenfin, gibi, gmann: thanks, hope you have a greate day ^^
10:52:47 openstackgerrit Sylvain Bauza proposed openstack/nova master: Provide the parent pGPU when creating a new vGPU https://review.opendev.org/715489
10:52:48 openstackgerrit Sylvain Bauza proposed openstack/nova master: WIP: Support different vGPU types per pGPU https://review.opendev.org/715490
10:52:53 bauzas sean-k-mooney: ^
11:03:45 openstackgerrit Merged openstack/nova master: Add new default roles in remote console policies https://review.opendev.org/716484
11:38:31 gibi stephenfin: ack, will look shortly
11:41:27 gibi stephenfin: I'm all the way +2 on that series. Do you have a novaclient change as well?
12:25:26 stephenfin bauzas: You think you could review https://review.opendev.org/#/c/710037/8/doc/ext/extra_specs.py ? It's not the least bit complicated, if that helps :)

Earlier   Later