Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2019-02-08
23:18:32 mriedem jroll: just a reminder that when https://review.openstack.org/#/c/635006/ is ready to go for wider review to throw it in the runways queue https://etherpad.openstack.org/p/nova-runways-stein
23:20:36 aspiers fried_rice, mriedem: was trying to understand the discussion around https://review.openstack.org/#/c/538498/ and it seems one point of debate was the corner case where a trait might need to be removed from a node, right?
23:20:51 aspiers fried_rice, mriedem: FWIW, that is not too unlikely with SEV
23:21:31 fried_rice aspiers: The basic idea is that there are some traits that are "owned" by compute, and any others are not.
23:21:36 aspiers It could be triggered by a downgrade of the kernel or QEMU or libvirt, or by disabling the sev option in the kvm_amd kernel module
23:21:50 aspiers fried_rice: yeah I got that, and I presume that the SEV trait would be owned by compute
23:22:01 fried_rice If a trait is owned by compute, you can't (un)set it as an operator or whatever. Next periodic will restore it to its "proper" (according to compute) value.
23:22:09 aspiers sure
23:22:24 fried_rice so if your trait is owned by compute, all you have to worry about is whether it needs to be set or not, and do that.
23:22:52 fried_rice If you need to be able to do that *without* compute service restarts, then I recommend doing it in update_provider_tree, not via a driver capability.
23:23:06 aspiers My point is that SEV requires support all the way up the stack, and if one link in the chain vanishes, then the trait would have to be removed. So that code under review does have to correctly handle that case, as you were saying in the discussion
23:23:28 aspiers I don't think it needs to happen without nova-compute restart
23:23:37 aspiers It's not gonna be a regular occurrence at all
23:23:51 fried_rice oh, you're worried about whether I'm right to -1 mriedem's patch?? Of *course* I am!
23:24:08 fried_rice I've *never* -1'd a mriedem patch and been *wrong* about it!
23:24:45 fried_rice aspiers: Are you looking to take over that patch and make it go? That would be appreciated, I think, by me & mriedem and others.
23:24:59 aspiers haha
23:25:38 aspiers fried_rice: I like the idea in principal, but the reality is that I'm *hopelessly* new to all of this code so I worry it would take me far too long to figure out
23:25:38 fried_rice you're referring to https://review.openstack.org/#/c/538498/7/nova/compute/resource_tracker.py@962 ?
23:26:38 aspiers yeah that was part of the discussion I was referring to
23:26:39 fried_rice aspiers: you're demonstrating an understanding that's certainly adequate to where you'd be capable of making this patch go. Matt and I would be available to advise if you got stuck.
23:26:56 aspiers fried_rice: maybe I'm just a good bluffer ;-)
23:27:19 fried_rice So yeah, I'm +2 on this patch if that condition is removed. (And also rebase, and also test fixups, I think)
23:27:34 mriedem the trick on that thing is really just functional testing the scenarios
23:27:54 mriedem 1. compute reports traits via driver capabilities, user removes them externally, run periodic and they come back
23:28:03 mriedem 2. user externally sets traits and compute periodic doesn't remove them
23:28:16 mriedem 3. driver reports traits in t0, then at t1 doesn't report the trait, so it's removedd
23:28:19 mriedem something like that
23:28:23 fried_rice ++
23:28:51 fried_rice the patch already has a start on that stuff.
23:28:53 mriedem once the functional test scenarios are in place then the code should be pretty straight-forward
23:28:58 aspiers I see
23:28:59 fried_rice https://review.openstack.org/#/c/538498/7/nova/tests/functional/test_servers.py
23:29:05 aspiers yeah, looking at that
23:29:20 mriedem with that it's 5:30 and i need to shove off
23:29:30 fried_rice o/
23:29:31 mriedem i hope y'all will do the same
23:29:42 aspiers it's 11:30pm here :-/
23:29:44 fried_rice I'm going to rip apart Sundar's PoC a little more.
23:30:04 fried_rice Where in the build flow do neutron ports get bound?
23:30:20 aspiers it looks like it doesn't yet test the behaviour of a driver-reported trait being removed, right?
23:32:36 fried_rice aspiers: I think it only really tests that capabilities are converted to traits, stuffed onto the compute rp, and new custom ones are vivified in placement.
23:32:45 fried_rice that would be scenario 0 in mriedem's list above.
23:32:51 aspiers ok
23:32:59 fried_rice So 1, 2, and 3 all still need to be implemented.
23:33:24 aspiers got it
23:33:38 fried_rice I guess the existing code is mriedem's scenario 1 before the first comma.
23:36:35 aspiers I see
#openstack-nova - 2019-02-09
03:00:54 openstackgerrit Merged openstack/nova stable/queens: Null out instance.availability_zone on shelve offload https://review.openstack.org/606155
03:52:58 openstackgerrit Merged openstack/nova stable/queens: Remove stale pip-missing-reqs tox test https://review.openstack.org/628128
03:57:55 openstackgerrit Merged openstack/nova stable/queens: Removing pip-missing-reqs from default tox jobs https://review.openstack.org/628956
04:13:31 openstackgerrit Merged openstack/nova stable/queens: Don't call begin_detaching when detaching volume from shelved vm https://review.openstack.org/633962
04:13:37 openstackgerrit Merged openstack/nova stable/queens: Send context.global_id on neutron calls https://review.openstack.org/627604
04:13:44 openstackgerrit Merged openstack/nova stable/queens: Skip checking of target_dev for vhostuser https://review.openstack.org/634236
06:56:19 openstackgerrit Merged openstack/nova master: Modify InstanceMappingList.get_not_deleted_by_cell_and_project() https://review.openstack.org/635120
07:08:18 openstackgerrit Merged openstack/nova master: Plumbing for ignoring list_records_by_skipping_down_cells https://review.openstack.org/635121
07:42:49 openstackgerrit Merged openstack/nova master: Add minimum value in maximum_instance_delete_attempts https://review.openstack.org/633085
07:54:45 openstackgerrit Merged openstack/nova master: Convert CPU_TRAITS_MAPPING to use os_traits https://review.openstack.org/635919
07:54:52 openstackgerrit Merged openstack/nova master: doc: mention description field in user flavors docs https://review.openstack.org/635263
08:04:35 openstackgerrit Merged openstack/nova master: Extend RequestGroup object for mapping https://review.openstack.org/619527
13:42:43 openstackgerrit Merged openstack/nova stable/queens: Update port device_owner when unshelving https://review.openstack.org/626408
22:46:36 openstackgerrit Merged openstack/nova master: Drop nova-multiattach job https://review.openstack.org/606981
22:55:35 openstackgerrit Merged openstack/nova master: API: Remove evacuate/live-migrate 'force' parameter https://review.openstack.org/634600
#openstack-nova - 2019-02-10
08:14:13 openstackgerrit garyk proposed openstack/nova master: Better handle live migration abort https://review.openstack.org/635440
08:16:02 openstackgerrit garyk proposed openstack/nova master: Better handle live migration abort https://review.openstack.org/635440
14:13:33 openstackgerrit garyk proposed openstack/nova master: Better handle live migration abort https://review.openstack.org/635440
14:26:58 openstackgerrit Hamdy Khader proposed openstack/os-vif master: Add VIFPortProfileSmartNIC profile https://review.openstack.org/636061
14:30:44 openstackgerrit Hamdy Khader proposed openstack/os-vif master: Add VIFPortProfileSmartNIC profile https://review.openstack.org/636061
14:47:11 openstackgerrit Hamdy Khader proposed openstack/os-vif master: Add VIFPortProfileSmartNIC profile https://review.openstack.org/636061
20:36:21 openstackgerrit Matt Riedemann proposed openstack/nova master: Raise InstanceFaultRollback for UnableToMigrateToSelf from _prep_resize https://review.openstack.org/633227
20:36:21 openstackgerrit Matt Riedemann proposed openstack/nova master: Change InstanceFaultRollback handling in _error_out_instance_on_exception https://review.openstack.org/633212
20:51:29 openstackgerrit Matt Riedemann proposed openstack/nova master: Cleanup return_reservation_id in ServersController.create https://review.openstack.org/633560
20:51:29 openstackgerrit Matt Riedemann proposed openstack/nova master: Refactor bdm handling in ServersController.create method https://review.openstack.org/633535
20:51:30 openstackgerrit Matt Riedemann proposed openstack/nova master: Refactor "networks" processing in ServersController.create https://review.openstack.org/633594
#openstack-nova - 2019-02-11
01:14:46 openstackgerrit Merged openstack/nova stable/rocky: Fix InstanceMapping to always default queued_for_delete=False https://review.openstack.org/635534
01:51:21 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Don't change volume status to `in-use` when attach to shelved vm. https://review.openstack.org/627096
07:59:51 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Support volume-backed server rebuild in compute https://review.openstack.org/625893
08:54:27 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Support volume-backed server rebuild in compute https://review.openstack.org/625893
09:02:17 openstackgerrit yenai proposed openstack/nova master: Ignore VolumeAttachmentNotFound exception in compute.manager https://review.openstack.org/632657
09:16:45 openstackgerrit Zhenyu Zheng proposed openstack/nova master: Bump API microversion to support rebuild bfv instances https://review.openstack.org/627518
10:56:33 openstackgerrit Lee Yarwood proposed openstack/nova master: fixtures: Return a mocked class instead of method within fake_imagebackend https://review.openstack.org/619804
10:57:00 openstackgerrit Lee Yarwood proposed openstack/nova stable/queens: libvirt: Add workaround to cleanup instance dir when using rbd https://review.openstack.org/628726
10:57:06 openstackgerrit Lee Yarwood proposed openstack/nova stable/rocky: Don't emit warning when ironic properties are zero https://review.openstack.org/608573
11:23:30 openstackgerrit Lee Yarwood proposed openstack/nova master: compute: Reject migration requests when source is down https://review.openstack.org/623489
12:49:09 cdent I seem to remember this being talked about before: is there a way to weigh the "busy-ness" of an AZ or aggregate? Is the cpuweigher a vague proxy for that?
13:10:39 gibi bauzas: hi! could you help me pushing through two qeens/pike backports?
13:10:49 gibi bauzas: https://review.openstack.org/#/c/623355/1 and https://review.openstack.org/#/c/635897/1,
13:11:27 cdent welcome back gibi !
13:11:39 gibi cdent: thanks :)
13:30:26 bauzas gibi: hola
13:30:29 bauzas gibi: sure, lemme look
13:30:35 gibi bauzas: thanks a lot
13:31:00 bauzas gibi: FWIW, I'm also thinking about your comments like https://review.openstack.org/#/c/599208/12/nova/virt/libvirt/driver.py@6718
13:33:06 gibi bauzas: so if the TODO means that the operator today can change the config that nova does not handle then I think we have a problem
13:34:52 bauzas gibi: yeah but the problem is that we don't know yet if we need to create a new child RP
13:35:04 bauzas we only know about the root RP
13:35:15 bauzas and we don't persist the existing type
13:35:53 bauzas oh and FWIW, it doesn't work also atm
13:37:04 gibi bauzas: so, what will happen if the admin changes the enabled_vgpu_types ?
13:43:19 bauzas gibi: for the moment, we create an inventory by calling the number of left mdevs

Earlier   Later