Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2020-03-31
08:26:24 brinzhang_ ok, I will create it.
08:28:25 gibi thanks
08:28:38 brinzhang_ gibi: thanks too.
08:34:37 openstackgerrit Brin Zhang proposed openstack/nova master: Separate update and swap volume policies https://review.opendev.org/711194
08:34:37 openstackgerrit Brin Zhang proposed openstack/nova master: Allow PUT volume attachments API to modify delete_on_termination https://review.opendev.org/693828
08:38:04 brinzhang_ damsmith, gmann, gibi: updated the bp/destroy-instance-with-datavolume patch, and there is a question asked by gmann https://review.opendev.org/#/c/693828/21/nova/api/openstack/compute/volumes.py@475, wait dansmith to check, if need to change, that you can update that change to the latest pach
09:04:12 gibi brinzhang_: ack
09:04:19 gibi I'm reviewing the patch now
09:06:40 brinzhang_ gibi: cool, thanks~
09:21:10 openstackgerrit Luyao Zhong proposed openstack/nova master: support live migration with vpmem https://review.opendev.org/687856
09:21:10 openstackgerrit Luyao Zhong proposed openstack/nova master: partial support for live migration with specific resources https://review.opendev.org/715362
09:23:38 luyao stephenfin: Hi, stephenfin , thanks for reviewing the first patch, I hope the second patch for vpmem live migration also looks good to you. :) https://review.opendev.org/#/c/687856
09:26:53 luyao stephenfin: I haven'ted add release notes for the notification body change in patch https://review.opendev.org/#/c/715362/, if it's better to have I'll add. Thanks for gibi's comments. :)
09:27:17 gibi luyao: don't worry about that I think we never did reno for such change
09:27:45 luyao gibi: OK, get it. :)
09:53:28 openstackgerrit Stephen Finucane proposed openstack/nova master: api: Add microversion 2.85, extra spec validation https://review.opendev.org/708436
09:53:29 openstackgerrit Stephen Finucane proposed openstack/nova master: api: Add support for new cyborg extra specs https://review.opendev.org/716222
09:53:29 openstackgerrit Stephen Finucane proposed openstack/nova master: docs: Add documentation for flavor extra specs https://review.opendev.org/710037
10:07:35 openstackgerrit Stephen Finucane proposed openstack/os-vif master: trivial: Remove some rules from flake8 ignore list https://review.opendev.org/716223
10:08:31 nightmare_unreal how to deal with this error msg : This test uses methods that set internal oslo_db state, but it does not claim to use the database. This will conflict with the setup of tests that do use the database and cause failures later.
10:08:52 openstackgerrit Merged openstack/nova-specs master: Update hacking for Python3 https://review.opendev.org/715650
10:20:52 openstackgerrit Andreas Jaeger proposed openstack/python-novaclient master: Update to hacking 3.0 https://review.opendev.org/716228
10:22:00 openstackgerrit jayaditya gupta proposed openstack/nova master: Support for nova-manage placement heal_allocations --cell https://review.opendev.org/714459
11:06:50 openstackgerrit jayaditya gupta proposed openstack/nova master: Support for nova-manage placement heal_allocations --cell https://review.opendev.org/714459
11:11:05 openstackgerrit Merged openstack/nova master: Add new default roles in os-agents policies https://review.opendev.org/701648
11:27:36 bauzas gibi: stephenfin: when you're around, i have a question about a reshape
11:30:14 bauzas gibi: stephenfin: context is -1 for a reshape https://541e2403478ac154d5eb-056bfb946e355d1a1a86dc411a70c5ec.ssl.cf2.rackcdn.com/715489/1/check/nova-tox-functional-py36/481510f/testr_results.html
11:30:33 bauzas but I removed the support
11:35:45 gibi bauzas: ack. I'm on a meeting. Will ping you after
11:35:53 bauzas cool thanks
11:37:08 bauzas gibi: stephenfin tl;dr: it's about deleting some upgrade support for Rocky>Stein in https://review.opendev.org/#/c/715489/1/nova/virt/libvirt/driver.py@7008a
11:37:34 bauzas but the reshape functional test won't support it then
11:45:59 gibi brinzhang_, dansmith: the notification sample test failure shows a relevant problem in the new implementation https://review.opendev.org/#/c/693828
11:52:53 brinzhang_ gibi: the serverId is in the request path https://review.opendev.org/#/c/693828/22/nova/api/openstack/compute/schemas/volumes.py@100
11:53:09 brinzhang_ but what is 'id'?
11:53:47 gibi I think id is the attachment id
11:54:12 brinzhang_ PUT /servers/{server_id}/os-volume_attachments/{volume_id}
11:55:09 brinzhang_ gibi: I am not sure the id and serverId does need to add the request body, looks like we dont need that while we update an attachment
11:55:20 kevinz gibi: can you take a look at this https://review.opendev.org/#/c/714311/ and https://review.opendev.org/712607, those two already get one +2
11:55:24 kevinz Thanks
11:58:46 gibi brinzhang_: it is about how we define RESTFull
11:59:26 gibi brinzhang_: if we want that the GET response can be sent back as PUT request then we need that the GET rsp matches with the PUT req
12:00:11 gibi kevinz: ack, I will try
12:00:36 kevinz gibi: Thanks ~
12:02:07 gibi bauzas: so you think there is a fault in the functional test test_create_servers_with_vgpu ?
12:02:14 bauzas gibi: not really
12:02:32 bauzas gibi: tbc we supported a reshape for Stein
12:02:40 bauzas for Rocky>Stein
12:02:48 bauzas now, we're in Ussuri
12:02:53 brinzhang_ gibi: yeah, as you think it's should keep the same, but we add the tag, device, because swap volume need these parameters, so we should add them in the reqeust body, if we are just update the delete flag for the attachemt, these are invalid, do we need to add these check, such as attachment_id and serverId cannot be changed
12:03:05 bauzas gibi: so I can remove the upgrade support that we created
12:03:21 bauzas gibi: but then, of course the reshape method won't longer work
12:03:37 bauzas gibi: so I should probably remove it too, right? (and the tests)
12:03:50 bauzas but then I think about FFU
12:04:01 brinzhang_ gibi: so form this side, I think they are redundant to add the PUT request body, maybe dansmith have some idea of this.
12:04:02 gibi brinzhang_: let's ask dansmith how serious he want the two json body to match
12:04:31 gibi bauzas: yeah FFU support is a question
12:06:34 gibi bauzas: if we can say that we dont support FFU between Rocky - Ussuri then we can remove the reshape code
12:07:04 bauzas gibi: ... or I would just leave the upgrade support until we agree on that
12:07:24 bauzas gibi: but do we have already some consensus about those kind of questions ?
12:07:30 gibi bauzas: yeah, I think we had the agreement in the past that we keep reshape support for a while
12:07:40 bauzas am I the first folk asking about it ?
12:07:41 gibi bauzas: I think we don't have
12:07:55 gibi or at least I dont rememer
12:07:55 openstackgerrit Huaqiang Wang proposed openstack/nova master: tox: Integrate mypy https://review.opendev.org/676208
12:07:56 bauzas hum
12:07:56 openstackgerrit Huaqiang Wang proposed openstack/nova master: libvirt: Add typing information https://review.opendev.org/714695
12:07:56 openstackgerrit Huaqiang Wang proposed openstack/nova master: hardware: Update and correct typing information https://review.opendev.org/714694
12:07:57 openstackgerrit Huaqiang Wang proposed openstack/nova master: objects: Replace 'cpu_pinning_requested' helper https://review.opendev.org/714697
12:07:57 openstackgerrit Huaqiang Wang proposed openstack/nova master: tests: Split instance NUMA object tests https://review.opendev.org/714696
12:07:58 openstackgerrit Huaqiang Wang proposed openstack/nova master: hardware: Remove handling of pre-Train compute nodes https://review.opendev.org/714699
12:07:58 openstackgerrit Huaqiang Wang proposed openstack/nova master: hardware: Don't consider overhead CPUs for unpinned instances https://review.opendev.org/714698
12:07:59 openstackgerrit Huaqiang Wang proposed openstack/nova master: hardware: Tweak the 'cpu_realtime_mask' handling slightly https://review.opendev.org/461456
12:07:59 openstackgerrit Huaqiang Wang proposed openstack/nova master: hardware: Add validation for 'cpu_realtime_mask' https://review.opendev.org/468203
12:08:00 openstackgerrit Huaqiang Wang proposed openstack/nova master: hardware: Invert order of NUMA topology generation https://review.opendev.org/714701
12:08:00 openstackgerrit Huaqiang Wang proposed openstack/nova master: hardware: Rework 'get_realtime_constraint' https://review.opendev.org/714700
12:08:01 openstackgerrit Huaqiang Wang proposed openstack/nova master: Introduce 'pcpuset' field for InstanceNUMACell https://review.opendev.org/714658
12:08:01 openstackgerrit Huaqiang Wang proposed openstack/nova master: hardware: Remove '_numa_fit_instance_cell_with_pinning' https://review.opendev.org/714703
12:08:02 openstackgerrit Huaqiang Wang proposed openstack/nova master: libvirt: set host CPUs for the mixed instance https://review.opendev.org/714655
12:08:02 openstackgerrit Huaqiang Wang proposed openstack/nova master: hardware: Add support for 'hw:cpu_dedicated_mask' extra spec https://review.opendev.org/714706
12:08:03 openstackgerrit Huaqiang Wang proposed openstack/nova master: metadata: export the vCPU IDs that are pinning on the host CPUs https://review.opendev.org/688936
12:08:03 openstackgerrit Huaqiang Wang proposed openstack/nova master: Setup 'mixed' instance through 'PCPU' and 'VCPU' resource https://review.opendev.org/713355
12:08:04 openstackgerrit Huaqiang Wang proposed openstack/nova master: Introduce the 'CPUAllocationPolicy.MIXED' enum https://review.opendev.org/716267
12:08:36 gibi bauzas: I think it is worth to ask the others how they think about it
12:08:39 bauzas gibi: I think we would need to have some policy about it
12:08:43 gibi yeah
12:08:52 bauzas gibi: because before FFU, it was simple
12:08:58 gibi I can imagine that the default policy is not to remove reshape, eer
12:09:00 gibi ever
12:09:12 bauzas gibi: we were just removing upgrade support after one release
12:09:24 bauzas gibi: but now, I no longer know when we should do it
12:09:53 bauzas gibi: and like you say, some operators could tell us to just support reshapes for a whole
12:10:01 bauzas if so, uhu
12:13:04 brinzhang_ gibi: http://paste.openstack.org/show/791394/ https://review.opendev.org/#/c/693828/22/nova/tests/functional/api_sample_tests/test_volumes.py@305
12:13:34 brinzhang_ I debuged in http://paste.openstack.org/show/791394/
12:16:49 gibi brinzhang_: what do you mean?
12:17:05 gibi I simply removed the sub and the test still passed.
12:17:15 gibi so the sub is unused
12:17:48 brinzhang_ gibi: yes, it unused
12:18:22 brinzhang_ gibi: gibi: when the data volume attached to the server, the delete_on_termination will be set to False by default https://opendev.org/openstack/nova/src/branch/master/nova/compute/api.py:_attach_volume()

Earlier   Later