| Posted | Nick | Remark | |
|---|---|---|---|
| #openstack-nova - 2021-07-19 | |||
| 10:49:33 | elodilles | lyarwood: :S thanks for the heads up. and as bauzas said, too, hope that it will be sorted out soon. :X | |
| 10:58:50 | lyarwood | thanks :) | |
| 11:08:22 | opendevreview | Artom Lifshitz proposed openstack/nova master: WIP: allow service delete even if compute node record is missing https://review.opendev.org/c/openstack/nova/+/801285 | |
| 11:20:01 | opendevreview | Takashi Kajinami proposed openstack/nova master: Clean up allocations left by evacuation when deleting service https://review.opendev.org/c/openstack/nova/+/778696 | |
| 11:25:02 | opendevreview | Takashi Kajinami proposed openstack/nova master: Clean up allocations left by evacuation when deleting service https://review.opendev.org/c/openstack/nova/+/778696 | |
| 11:26:02 | opendevreview | Takashi Kajinami proposed openstack/nova master: Clean up allocations left by evacuation when deleting service https://review.opendev.org/c/openstack/nova/+/778696 | |
| 11:42:34 | kevinz | Hi lyarwood: stephenfin: could you help to review the patch https://review.opendev.org/c/openstack/nova/+/763928, to support live migration on arm64 | |
| 11:42:41 | kevinz | Thanks a lot! | |
| 11:44:18 | kashyap | kevinz: Hi, a quick question: have you tested it functionally? (As in doing a real migration w/ your patch) | |
| 11:48:04 | kevinz | kashyap: yes definitely | |
| 11:48:08 | kashyap | kevinz: I'm just taking a look; and yes, host CPU compat check indeed doesn't make sense on AArch64. Thanks for logging it | |
| 11:48:45 | kevinz | kashayap: Thanks, really appreciate | |
| 11:49:47 | kashyap | kevinz: Okay; when you've done such tests, it can help make your case if you also explicitly mention it on the change - something for next time | |
| 11:51:27 | artom | kashyap, kevinz, we have an AArch64 CI, no? Could we not run a multinode job in it to test live migration? | |
| 11:51:47 | kashyap | artom: If we do have such a proper CI, then definitely it helps | |
| 11:52:45 | kevinz | artom: The arm64 CI now is not sufficient stable now so the devstack test and nova test ci is not merged into upstream. FYI: https://review.opendev.org/c/openstack/devstack/+/708317 | |
| 11:53:24 | kevinz | kashyap: ack, will add the info | |
| 11:53:41 | artom | kashyap, aha, ack | |
| 12:03:09 | kashyap | kevinz: Oh, wait - based on that new libvir-list thread, you're selectively adding CPU comparison for AArch64, while in the past we discouraged it, because QEMU AArch64 devs themselves suggested to use 'host-passthrough' for KVM+AArch64 guests | |
| 12:03:19 | kashyap | Okay, I'll read the libvirt thread and respond back. | |
| 12:04:03 | kevinz | kashyap: OK, np, the libvirt patch has a thread discussion for this | |
| 12:04:20 | kashyap | kevinz: I have a couple of questions, I'll ask it there for the record | |
| 13:06:45 | kashyap | kevinz: I wrote some comments; please see if that makes sense to you. And, thanks gibi for Ccing me | |
| 13:17:30 | sean-k-mooney | kashyap: for what its worth i do think we should be adding the cpu comparision for aarch64 | |
| 13:20:23 | sean-k-mooney | kashyap: although i agree with you that linking to the libvirt patch is not very operator friendly | |
| 13:23:11 | kevinz | kashyap: sean-k-mooney: Thank a lot! I will address the comments :-) | |
| 13:49:32 | kashyap | sean-k-mooney: Like I wrote in the review, this discussion gets muddied by the vague term "CPU comparison" | |
| 13:49:36 | kashyap | There's two things: | |
| 13:49:54 | kashyap | (1) comparing guest CPU to host CPU -- the "traditional" use case of libvirt CPU APIs | |
| 13:50:15 | kashyap | (2) comparing source host CPU to dest host CPU -- the only thing supported for Arm/AArch64 by virtue of this new patch | |
| 13:51:07 | kashyap | Here, for AArch64/Arm, we're talking about the 2nd case, host to host CPU comparison -- that's a reasonable and valid use case, although remains to be seen how effective it is for Arm/AArch64 | |
| 13:52:15 | kashyap | In summary, for AArch64, the source host CPU to dest host CPU comparison is useful only to ensure your guest don't migrate a completely different host. | |
| 13:52:21 | kashyap | kevinz: --^ | |
| 13:53:10 | opendevreview | Balazs Gibizer proposed openstack/nova master: Reject server create with extended resource req https://review.opendev.org/c/openstack/nova/+/793619 | |
| 13:53:11 | opendevreview | Balazs Gibizer proposed openstack/nova master: Add same_subtree field to RequestLevelParams https://review.opendev.org/c/openstack/nova/+/791503 | |
| 13:53:11 | opendevreview | Balazs Gibizer proposed openstack/nova master: Reject server operations with extended resource req https://review.opendev.org/c/openstack/nova/+/793620 | |
| 13:54:07 | kashyap | kevinz: I was just thinking on a walk -- I wonder how much do you gain even with host-to-host CPU comparison. For Arm/AArch64, if you care about live migration: you're better off ensuring you have the identical host CPUs and all related virt software in your Arm/AArch64 pool + host-passthrough | |
| 13:54:29 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support same_subtree in allocation_canadidate query https://review.opendev.org/c/openstack/nova/+/791505 | |
| 13:54:29 | opendevreview | Balazs Gibizer proposed openstack/nova master: Bump min placement microversion to 1.36 https://review.opendev.org/c/openstack/nova/+/791504 | |
| 13:54:30 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] refactor assertPortMatchesAllocation https://review.opendev.org/c/openstack/nova/+/792458 | |
| 13:56:05 | kashyap | sean-k-mooney: In my two points above; point (1) -- comparing guest CPU to host CPu -- doesn't make sense for reasons noted before and in the review. If you're talking about point (2), then we're in agreement :-) | |
| 13:56:38 | kashyap | kevinz: (And yes, I've read the upstream libvirt thread. Thx for the link) | |
| 13:58:13 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] refactor asserts in qos tests https://review.opendev.org/c/openstack/nova/+/798930 | |
| 13:59:28 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] ports with both bw and pps resources https://review.opendev.org/c/openstack/nova/+/792394 | |
| 14:03:15 | opendevreview | Balazs Gibizer proposed openstack/nova master: Parse extended resource request from the port data https://review.opendev.org/c/openstack/nova/+/800085 | |
| 14:04:31 | opendevreview | Balazs Gibizer proposed openstack/nova master: Transfer RequestLevelParams from ports to scheduling https://review.opendev.org/c/openstack/nova/+/791506 | |
| 14:04:32 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support boot with extended resource request https://review.opendev.org/c/openstack/nova/+/800086 | |
| 14:07:03 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support move ops with extended resource request https://review.opendev.org/c/openstack/nova/+/800087 | |
| 14:07:04 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test]Refactor interface attach with qos https://review.opendev.org/c/openstack/nova/+/800088 | |
| 14:08:19 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support interaface attach / detach with new resource request format https://review.opendev.org/c/openstack/nova/+/800089 | |
| 14:10:51 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] move unshelve test to the proper place https://review.opendev.org/c/openstack/nova/+/793621 | |
| 14:16:39 | opendevreview | Takashi Kajinami proposed openstack/nova master: Clean up allocations left by evacuation when deleting service https://review.opendev.org/c/openstack/nova/+/778696 | |
| 16:22:44 | opendevreview | Balazs Gibizer proposed openstack/nova master: Add same_subtree field to RequestLevelParams https://review.opendev.org/c/openstack/nova/+/791503 | |
| 16:22:44 | opendevreview | Balazs Gibizer proposed openstack/nova master: Reject server operations with extended resource req https://review.opendev.org/c/openstack/nova/+/793620 | |
| 16:25:15 | opendevreview | Balazs Gibizer proposed openstack/nova master: Bump min placement microversion to 1.36 https://review.opendev.org/c/openstack/nova/+/791504 | |
| 16:26:31 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support same_subtree in allocation_canadidate query https://review.opendev.org/c/openstack/nova/+/791505 | |
| 16:27:46 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] refactor assertPortMatchesAllocation https://review.opendev.org/c/openstack/nova/+/792458 | |
| 16:29:02 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] refactor asserts in qos tests https://review.opendev.org/c/openstack/nova/+/798930 | |
| 16:30:20 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] ports with both bw and pps resources https://review.opendev.org/c/openstack/nova/+/792394 | |
| 16:31:38 | opendevreview | Balazs Gibizer proposed openstack/nova master: Parse extended resource request from the port data https://review.opendev.org/c/openstack/nova/+/800085 | |
| 16:32:55 | opendevreview | Balazs Gibizer proposed openstack/nova master: Transfer RequestLevelParams from ports to scheduling https://review.opendev.org/c/openstack/nova/+/791506 | |
| 16:34:13 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support boot with extended resource request https://review.opendev.org/c/openstack/nova/+/800086 | |
| 16:35:31 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support move ops with extended resource request https://review.opendev.org/c/openstack/nova/+/800087 | |
| 16:36:49 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test]Refactor interface attach with qos https://review.opendev.org/c/openstack/nova/+/800088 | |
| 16:38:05 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support interaface attach / detach with new resource request format https://review.opendev.org/c/openstack/nova/+/800089 | |
| 16:39:23 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] move unshelve test to the proper place https://review.opendev.org/c/openstack/nova/+/793621 | |
| 16:40:38 | opendevreview | Elod Illes proposed openstack/nova stable/stein: Make _rebase_with_qemu_img() generic https://review.opendev.org/c/openstack/nova/+/780784 | |
| 16:41:53 | opendevreview | Elod Illes proposed openstack/nova stable/stein: Use absolute path during qemu img rebase https://review.opendev.org/c/openstack/nova/+/757085 | |
| 18:32:03 | opendevreview | Lee Yarwood proposed openstack/nova master: DNM - Test tempest-integrated-compute-centos-8-stream https://review.opendev.org/c/openstack/nova/+/799996 | |
| 18:57:59 | opendevreview | Samuel proposed openstack/nova master: Add REST and compute APIs to migrate instances between projects https://review.opendev.org/c/openstack/nova/+/801377 | |
| 19:13:01 | opendevreview | Samuel proposed openstack/nova master: Add REST and compute APIs to migrate instances between projects https://review.opendev.org/c/openstack/nova/+/801377 | |
| 19:16:24 | opendevreview | Samuel proposed openstack/nova master: Add REST and compute APIs to migrate instances between projects https://review.opendev.org/c/openstack/nova/+/801377 | |
| 21:18:03 | NobodyCam | good afternoon Nova folks, off the wall question, I have a record in the (cell0) instance record that has been orphaned but is causing a phantom instance to appear in server list. my question is what is the best way to clear the phantom instance, update: "updated_at,deleted_at, deleted, vm_state, to indicate deleted or to nuke (delete) the record | |
| 21:18:24 | NobodyCam | s/instance record/instance table/ | |
| #openstack-nova - 2021-07-20 | |||
| 01:33:57 | opendevreview | Takashi Kajinami proposed openstack/nova master: Replace deprecated Table.copy() https://review.opendev.org/c/openstack/nova/+/801407 | |
| 01:36:47 | opendevreview | Takashi Kajinami proposed openstack/nova master: Replace deprecated Table.copy() https://review.opendev.org/c/openstack/nova/+/801407 | |
| 07:17:11 | opendevreview | Christian Rohmann proposed openstack/placement master: Fix SQL query counting the number of individual consumers having allocations by only selecting the aggregated consumer_id column. https://review.opendev.org/c/openstack/placement/+/801419 | |
| 09:56:11 | opendevreview | Lee Yarwood proposed openstack/nova master: WIP nova-manage: Introduce bdm show, refresh, get_connector commands https://review.opendev.org/c/openstack/nova/+/800634 | |
| 11:15:35 | opendevreview | Christian Rohmann proposed openstack/placement master: Fix SQL query counting the number of individual consumers having allocations by only selecting the aggregated consumer_id column. https://review.opendev.org/c/openstack/placement/+/801419 | |
| 12:00:03 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support boot with extended resource request https://review.opendev.org/c/openstack/nova/+/800086 | |
| 12:00:04 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support move ops with extended resource request https://review.opendev.org/c/openstack/nova/+/800087 | |
| 12:01:21 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test]Refactor interface attach with qos https://review.opendev.org/c/openstack/nova/+/800088 | |
| 12:01:22 | opendevreview | Balazs Gibizer proposed openstack/nova master: [func test] move unshelve test to the proper place https://review.opendev.org/c/openstack/nova/+/793621 | |
| 12:01:22 | opendevreview | Balazs Gibizer proposed openstack/nova master: Support interaface attach / detach with new resource request format https://review.opendev.org/c/openstack/nova/+/800089 | |
| 12:42:30 | sean-k-mooney | gibi: do you recall what the result of the last discussion on backport your event based detach code was | |
| 12:42:41 | sean-k-mooney | gibi: i see we did bring it back to wallaby which is good | |
| 12:42:43 | sean-k-mooney | https://review.opendev.org/c/openstack/nova/+/788720/2 | |
| 12:42:55 | sean-k-mooney | was there an object to trying to bring it back to train | |
| 12:43:22 | sean-k-mooney | or is it just a case that no one has fixed the merge conflcits and proposed the patch | |
| 12:43:29 | sean-k-mooney | elodilles: lyarwood ^ | |
| 12:44:15 | gibi | sean-k-mooney: I think nobody tried backporting it to older branches yet | |
| 12:44:26 | gibi | I don't know about any direct objection | |
| 12:45:53 | sean-k-mooney | ok we have a downstream bug that may be related https://bugzilla.redhat.com/show_bug.cgi?id=1981571 | |
| 12:46:25 | sean-k-mooney | they are seeing slightly odd behavior where the volume is removed form the os but not the libvirt xml | |
| 12:46:34 | sean-k-mooney | and then attachs of new volumes fail | |
| 12:46:49 | sean-k-mooney | but that could just be a timing issue with when the detach gets cancheled | |
| 12:47:29 | sean-k-mooney | so im not sure that your code would help but it may so just wanted to know how we felt about backporting in general before looking at it more closely | |
| 13:12:52 | opendevreview | Elod Illes proposed openstack/nova stable/stein: Make _rebase_with_qemu_img() generic https://review.opendev.org/c/openstack/nova/+/780784 | |
| 13:14:09 | opendevreview | Elod Illes proposed openstack/nova stable/stein: Use absolute path during qemu img rebase https://review.opendev.org/c/openstack/nova/+/757085 | |
| 15:50:19 | gibi | nova meeting starts in 10 minutes here in the channel | |