Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2020-08-28
08:52:27 brinzhang0 stephenfin: I got your comment in the new patch, later we will consider to deal with that patch
08:53:13 brinzhang0 the cyborg evacuate support patch was updated, pls review if you are free today, thanks
08:53:22 stephenfin brinzhang0: Yes, I'm saying we need to check whether the new flavor contains dp_name also. Currently we are only checking the existing flavor
08:53:30 stephenfin brinzhang0: This is what I did for vTPM https://github.com/openstack/nova/blob/75b5535e3448335cb26972b5ca37afbd24dba01d/nova/compute/api.py#L3997-L4020
08:53:56 stephenfin But yes, this should be dealt with separately in a follow-up
08:56:43 brinzhang0 stephenfin: yeah, may we should add same_instance_type=False as the request parameter in the block_accelerators(), while we do resize, it shuold be set same_instance_type=True
09:09:45 brinzhang0 stephenfin: https://github.com/openstack/nova/blob/75b5535e3448335cb26972b5ca37afbd24dba01d/nova/compute/api.py#L3994-L3995
09:11:10 brinzhang0 if same_instance_type and flavor_id: raise(), if the same_flavor_type is True, the flavor_id must be true, do we need this judgement?
09:12:34 brinzhang0 sorry, if the this is a migrate operation, it will be fine \o
09:18:04 openstackgerrit Wenping Song proposed openstack/nova master: Process exception of delete resource provider https://review.opendev.org/748339
09:22:03 stephenfin brinzhang0: Yes. If you're cold migrating, then the flavor_id parameter should be unset. It should only be set if resizing
09:23:01 stephenfin and if it's set and is set to the ID of the same flavor as the existing instance flavor, that's an error
09:23:29 brinzhang0 stephenfin: yeah, short-circuited of my brain \o/
09:28:14 openstackgerrit Takashi Natsume proposed openstack/nova-specs master: Make 'Feature Liaison' optional in test https://review.opendev.org/748591
09:30:18 openstackgerrit Merged openstack/nova stable/train: libvirt: Handle VIR_ERR_DEVICE_MISSING when detaching devices https://review.opendev.org/742415
09:37:12 jkulik Did we ever think of bubbling down a rename of an Instance to the actual driver? At least the vmwareapi driver names the VMs in the vCenter "{display_name} ({uuid})", which doesn't match after a rename anymore.
09:37:56 openstackgerrit Takashi Natsume proposed openstack/nova-specs master: Make 'Feature Liaison' optional in test https://review.opendev.org/748591
09:40:13 openstackgerrit Merged openstack/nova stable/train: libvirt: Do not reference VIR_ERR_DEVICE_MISSING when libvirt is < v4.1.0 https://review.opendev.org/747358
09:40:20 openstackgerrit Merged openstack/nova stable/stein: libvirt: Provide VIR_MIGRATE_PARAM_PERSIST_XML during live migration https://review.opendev.org/747974
09:40:28 openstackgerrit Merged openstack/nova master: zuul: Start to migrate nova-live-migration to zuulv3 https://review.opendev.org/711604
09:40:45 gibi o/
09:44:24 openstackgerrit Wenping Song proposed openstack/nova master: Reject resize operation for accelerator https://review.opendev.org/748593
10:10:18 jsuchome stephenfin gibi lyarwood: could you please agree on a way https://review.opendev.org/#/c/574301/ should be treated? I already said that I think it's better to follow what was decided in blueprint, but if you all agree it should be different, let's do it differently
10:14:35 gibi so as far as I see artom and stephen on the side of a new config. I have no hard oppionon on either side so I will fall in line with the rest and support the new config then
11:04:18 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Provide default 'host_info' for '_get_connection' https://review.opendev.org/746942
11:04:19 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Add 'start_compute' helper https://review.opendev.org/746943
11:04:19 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Use tempdir for CONF.instances_path https://review.opendev.org/746944
11:04:20 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Clean up PCI tests https://review.opendev.org/746945
11:04:20 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Add resize tests for PCI servers https://review.opendev.org/746946
11:04:21 openstackgerrit Stephen Finucane proposed openstack/nova master: conductor: Don't use setattr https://review.opendev.org/746947
11:04:21 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Add test for SR-IOV neutron ports https://review.opendev.org/746948
11:04:22 openstackgerrit Stephen Finucane proposed openstack/nova master: fixtures: Handle binding of first port https://review.opendev.org/746949
11:04:22 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Add live migration tests for PCI, SR-IOV servers https://review.opendev.org/746950
11:12:24 openstackgerrit Merged openstack/nova master: Change default num_retries for glance to 3 https://review.opendev.org/740389
11:20:13 elod lyarwood gibi : Train release patch, if you'll have time: https://review.opendev.org/#/c/748383/
11:27:30 gibi elod: +1
11:27:58 elod thx \o/
11:33:23 swp20 Stephenin: hi, when i run `python3 -m unittest nova.tests.unit.compute.test_api._ComputeAPIUnitTestMixIn.test_resize__with_vtpm`, it raise `TypeError: test_resize__with_vtpm() missing 1 required positional argument: 'mock_get_flavor'`, what's wrong?
11:34:09 swp20 stephenfin: hi, when i run `python3 -m unittest nova.tests.unit.compute.test_api._ComputeAPIUnitTestMixIn.test_resize__with_vtpm`, it raise `TypeError: test_resize__with_vtpm() missing 1 required positional argument: 'mock_get_flavor'`, what's wrong?
11:37:35 elod lyarwood: btw, do you want to wait with Stein release until these get merged: https://review.opendev.org/#/q/topic:bug/1887946+branch:stable/stein ?
11:40:12 openstackgerrit Balazs Gibizer proposed openstack/nova master: Make PCI claim NUMA aware during live migration https://review.opendev.org/748453
11:40:13 openstackgerrit Balazs Gibizer proposed openstack/nova master: [WIP] Support SRIOV interface attach and detach https://review.opendev.org/740995
11:40:13 openstackgerrit Balazs Gibizer proposed openstack/nova master: DNM: tracing the PCI allocation code path https://review.opendev.org/748454
11:45:56 openstackgerrit Elod Illes proposed openstack/nova stable/stein: libvirt: Handle VIR_ERR_DEVICE_MISSING when detaching devices https://review.opendev.org/742416
11:45:57 openstackgerrit Elod Illes proposed openstack/nova stable/stein: libvirt: Do not reference VIR_ERR_DEVICE_MISSING when libvirt is < v4.1.0 https://review.opendev.org/747360
12:29:44 openstackgerrit Brin Zhang proposed openstack/nova master: Cyborg evacuate support https://review.opendev.org/715326
12:48:47 openstackgerrit Luyao Zhong proposed openstack/nova master: Introduce a new driver interface 'cleanup_instance' https://review.opendev.org/714653
12:48:48 openstackgerrit Luyao Zhong proposed openstack/nova master: Track error migrations in resource tracker https://review.opendev.org/745281
12:48:48 openstackgerrit Luyao Zhong proposed openstack/nova master: Improve orphans tracking in resource tracker https://review.opendev.org/745282
12:48:49 openstackgerrit Luyao Zhong proposed openstack/nova master: Rollback when reset running live migration https://review.opendev.org/747899
13:16:39 openstackgerrit Alexandre Arents proposed openstack/nova master: Add a lock to prevent race during detach/attach of interface https://review.opendev.org/747957
13:18:06 aarents Hum commit message edit re-trigger all the stuff.. Now I now
13:18:17 aarents know
13:21:15 sean-k-mooney yes but i dont think it clears +2
13:21:23 sean-k-mooney i do think it clears +w
13:21:49 sean-k-mooney assuming you only edit the comit and not the code
13:22:24 sean-k-mooney its treated like a rebase in that way
13:40:25 aarents sean-k-mooney: it seems in my case to have drop +2, NVM
13:48:49 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Add 'start_compute' helper https://review.opendev.org/746943
13:48:50 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Use tempdir for CONF.instances_path https://review.opendev.org/746944
13:48:50 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Clean up PCI tests https://review.opendev.org/746945
13:48:51 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Add resize tests for PCI servers https://review.opendev.org/746946
13:48:51 openstackgerrit Stephen Finucane proposed openstack/nova master: conductor: Don't use setattr https://review.opendev.org/746947
13:48:52 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Add test for SR-IOV neutron ports https://review.opendev.org/746948
13:48:52 openstackgerrit Stephen Finucane proposed openstack/nova master: fixtures: Handle binding of first port https://review.opendev.org/746949
13:48:53 openstackgerrit Stephen Finucane proposed openstack/nova master: functional: Add live migration tests for PCI, SR-IOV servers https://review.opendev.org/746950
13:50:45 gmann stephenfin: done
14:08:53 stephenfin aarents: Could I get the comment here and I can fast approve, seeing as gibi was +2 before? https://review.opendev.org/#/c/747957/
14:14:43 openstackgerrit Merged openstack/nova master: vmware: Handle exception in destroy with attached volumes https://review.opendev.org/738416
14:15:33 stephenfin_ hurrah, my znc instance is alive again \o/
14:27:00 aarents stephenfin_: done thks
14:34:56 stephenfin_ aarents: Sorry, I meant can I get a comment in the code? :D
14:36:05 stephenfin_ aarents: Basically so some poor sucker isn't looking at this in 5 years time saying "why on earth did Alexandre add that call?" :P
14:36:34 aarents stephenfin_: sure.. friday..
14:40:11 stephenfin gibi: Could you weigh in on https://review.opendev.org/#/c/748411/ ? I think it's a fair request, but in fairness it's not what the BP stated so maybe I'm being unfair
14:41:07 gibi I did so on IRC but I can put that in the patch too
14:41:07 gibi 12:14 < gibi> so as far as I see artom and stephen on the side of a new config. I have no
14:41:10 gibi hard oppionon on either side so I will fall in line with the rest and
14:41:13 gibi support the new config then
14:41:49 stephenfin ah, sorry. Had ZNC issues so I missed any discussion this morning /o\
14:41:51 stephenfin Thanks
14:41:59 gibi doen
14:42:01 gibi done
14:42:09 gibi no worries
14:48:20 artom gibi, yeah, I don't really have strong opinions one way or the other
14:48:46 artom gibi, conceptually the new clean options seems cleaner
14:48:59 artom But maybe there are practical considerations for reviving the old one
14:49:25 stephenfin gibi: Also, in the context of https://review.opendev.org/#/c/748453/, have you seen https://review.opendev.org/#/c/746950/ ?
14:49:55 stephenfin In theory it _should_ be reproducible there, so I can put together a test for that early next week
14:52:17 gibi stephenfin: sounds cool.
14:52:32 gibi stephenfin: I wasn't eager to try to create a functional reproduce for that
14:52:49 gibi stephenfin: as that bug is a bit tangential to the sriov attach work I do righ now
14:53:40 openstackgerrit Lucas Alvares Gomes proposed openstack/nova master: DO NOT REVIEW: Test OVN devstack module https://review.opendev.org/748226
14:57:44 openstackgerrit Alexandre Arents proposed openstack/nova master: Add a lock to prevent race during detach/attach of interface https://review.opendev.org/747957
15:07:45 stephenfin aarents++ Thanks
16:23:33 sean-k-mooney stephenfin: just spoke to radolfo about a few thing one of them being the os-vif chanage for the defualt backend
16:23:55 sean-k-mooney im going to fix the patch by lazy loadign the connect to ovs via a property
16:24:05 sean-k-mooney can you take a look at the patch on monday
16:55:27 openstackgerrit Balazs Gibizer proposed openstack/nova master: Make PCI claim NUMA aware during live migration https://review.opendev.org/748453
16:55:27 openstackgerrit Balazs Gibizer proposed openstack/nova master: [WIP] Support SRIOV interface attach and detach https://review.opendev.org/740995

Earlier   Later