Posted	Nick	Remark
#openstack-nova - 2019-11-13
13:43:45	bauzas	alex_xu: FWIW, I'll still continue to use WeChat
13:54:04	openstack	Launchpad bug 1852437 in OpenStack Compute (nova) "Allow ability to disable individual CPU features via `cpu_model_extra_flags`" [Undecided,New]
13:54:04	kashyap	efried: Hi, when you get a moment, is a Blueprint required for this: https://bugs.launchpad.net/nova/+bug/1852437 (Allow ability to disable individual CPU features via `cpu_model_extra_flags`)
13:55:28	efried	kashyap: at a glance, a blueprint seems entirely necessary. A spec, not as sure. Perhaps put it on tomorrow's meeting agenda for discussion?
13:56:00	kashyap	efried: Spec / BP seems like an overkill, IMHO. But sure, can discuss tomm
13:56:42	efried	we're talking about enhancing syntax of a conf option in what seems like a nontrivial way.
13:56:46	kashyap	It is essentially aims to implement what I stated at the end of this commit in paranthesis: https://opendev.org/openstack/nova/commit/cc27a2007f314
13:56:57	kashyap	efried: Hmm, phrased that way...
13:57:18	kashyap	efried: But what do you think of it? Is there a better way you can think of than the +/- notion?
13:59:08	efried	kashyap: no, that seems reasonable, just seems like a thing that ought to have a "design" somewhere written down that we can agree on. Just having it in a RFE bug might be sufficient, but really that's what blueprints are for. And it's the kind of thing I would not expect to backport.
14:01:36	kashyap	efried: Sure, can file a simple BluePrint
14:01:59	efried	thanks kashyap.
14:02:40	kashyap	I first wanted to file it, but went the bug route, thinking it is a "simple idea". :D
14:09:02	kashyap	Done: https://blueprints.launchpad.net/nova/+spec/allow-disabling-cpu-flags
14:19:26	openstack	Launchpad bug 1852437 in OpenStack Compute (nova) "Allow ability to disable individual CPU features via `cpu_model_extra_flags`" [Wishlist,New]
14:19:26	kashyap	mriedem: Maybe we can just close this, as we'll be tracking it in the Blueprint? - https://bugs.launchpad.net/nova/+bug/1852437
14:20:58	mriedem	wfm
14:43:45	efried	mriedem: what's the easiest/best way to "discover" a compute node's UUID (the one that'll match the placement root RP)?
14:44:41	efried	(I'm trying to middleman here, not actually sure if we know anything about the node beforehand)
14:48:41	mriedem	you mean to get the ComputeNode object?
14:49:00	mriedem	for non-ironic nodes the compute node uuid is randomly generated when the record is created the first time
14:49:13	mriedem	to look up the computenode record, you want to use the host/nodename
14:49:26	mriedem	if you're on compute and it's not ironic, you can just use CONF.host
14:49:42	mriedem	host == node for non-ironic
14:49:52	mriedem	otherwise you get the nodenames from the driver
14:50:05	mriedem	see ComputeManager.update_available_resource
14:50:47	mriedem	or you're trying to do something outside of nova to try to expose a vulnerability?
14:51:09	bauzas	efried: yeah, you need to know the (host, node) tuple
14:51:44	bauzas	mriedem: and no, AFAICR, for some virt drivers, node != host
14:52:04	mriedem	bauzas: yeah, ironic
14:52:05	mriedem	i said that
14:52:13	bauzas	mriedem: not only for ironic
14:52:17	mriedem	which ones?
14:52:25	mriedem	if you're thinking vcenter, you're thinking of kilo era
14:52:29	bauzas	I don't remember, lemme look
14:53:29	efried	I think I can work with this, thanks.
14:53:30	bauzas	I do wonder for HyperV
14:53:41	mriedem	hyperv has only 1 node per host
14:54:39	mriedem	http://cloudbase-ci.com/nova/693937/1/windows/logs/n-h2-693937-1/nova-compute.log.gz
14:54:43	mriedem	2019-11-13 05:46:51.769 5012 103574784 GreenThread-1 INFO nova.compute.resource_tracker [req-320b7c3d-8917-4c50-b48e-18dd405c7877 - - - - -] Compute node record created for n-h2-693937-1:n-h2-693937-1 with uuid: 77e9e8d8-bba6-4aba-81f9-5e9f69ca1db9
14:55:15	mriedem	ironic is the only weirdo
14:56:12	bauzas	anyway, looks you're right
14:57:08	bauzas	https://github.com/openstack/nova/blob/master/nova/compute/manager.py#L9189
14:59:33	usr2033	Hi, can i pxe boot nova/kvm instances?
14:59:43	mriedem	note that https://docs.openstack.org/nova/latest/admin/configuration/hypervisors.html doesn't have a subpage for ironic as a compute driver in nova - that's probably a decent sized gap given all of the edge cases we could describe with ironic in nova as a compute driver
14:59:47	openstackgerrit	Merged openstack/python-novaclient master: Add minor version [21] to the test_versions https://review.opendev.org/688599
15:00:08	mriedem	not to mention scaling issues when using the ironic driver to manage lots of nodes from a single compute service, re the ML threads on RT perf issues
15:02:10	mriedem	usr2033: not natively no
15:02:13	mriedem	see https://serverfault.com/questions/469479/does-nova-support-pxe-boot
15:03:38	mriedem	usr2033: you may be interested in https://openstack-virtual-baremetal.readthedocs.io/en/latest/index.html
15:06:01	efried	I may be choppy today, going to be trying to work remotely (like, more remotely than usual).
15:09:00	openstackgerrit	Stephen Finucane proposed openstack/nova master: Remove 'os-consoles' API https://review.opendev.org/687907
15:09:01	openstackgerrit	Stephen Finucane proposed openstack/nova master: Remove 'nova-xvpvncproxy' https://review.opendev.org/687909
15:09:01	openstackgerrit	Stephen Finucane proposed openstack/nova master: Remove 'nova-console' service, 'os-consoles' API https://review.opendev.org/687908
15:15:06	openstackgerrit	Merged openstack/nova master: Remove TODOs around claim_resources_on_destination https://review.opendev.org/693635
15:15:41	openstack	Launchpad bug 1852446 in OpenStack Compute (nova) "Hypervisors in nova - no subpage details for ironic" [Undecided,New]
15:15:41	mriedem	bauzas: https://bugs.launchpad.net/nova/+bug/1852446
15:16:09	bauzas	ack, good point
15:18:07	stephenfin	mriedem: It's not urgent, but if you can rebase https://review.opendev.org/#/c/693425/ today I'm happy to push it through
15:19:29	mriedem	ack, let me rebase the entire cross-cell series first quick so i can destroy the gate
15:21:44	openstackgerrit	Matt Riedemann proposed openstack/nova master: Filter duplicates from compute API get_migrations_sorted() https://review.opendev.org/636224
15:21:44	openstackgerrit	Matt Riedemann proposed openstack/nova master: Make API always RPC cast to conductor for resize/migrate https://review.opendev.org/693937
15:21:45	openstackgerrit	Matt Riedemann proposed openstack/nova master: Handle target host cross-cell cold migration in conductor https://review.opendev.org/642591
15:21:45	openstackgerrit	Matt Riedemann proposed openstack/nova master: Start functional testing for cross-cell resize https://review.opendev.org/636253
15:21:46	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add zones wrinkle to TestMultiCellMigrate https://review.opendev.org/643450
15:21:46	openstackgerrit	Matt Riedemann proposed openstack/nova master: Validate image/create during cross-cell resize functional testing https://review.opendev.org/642592
15:21:47	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add negative test for prep_snapshot_based_resize_at_source failing https://review.opendev.org/669013
15:21:47	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add negative test for cross-cell finish_resize failing https://review.opendev.org/643451
15:21:48	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add ConfirmResizeTask https://review.opendev.org/637070
15:21:48	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add confirm_snapshot_based_resize_at_source compute method https://review.opendev.org/637058
15:21:49	openstackgerrit	Matt Riedemann proposed openstack/nova master: Confirm cross-cell resize from the API https://review.opendev.org/637316
15:21:49	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add confirm_snapshot_based_resize conductor RPC method https://review.opendev.org/637075
15:21:50	openstackgerrit	Matt Riedemann proposed openstack/nova master: Deal with cross-cell resize in _remove_deleted_instances_allocations https://review.opendev.org/639453
15:21:50	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add revert_snapshot_based_resize_at_dest compute method https://review.opendev.org/637630
15:21:51	openstackgerrit	Matt Riedemann proposed openstack/nova master: WIP: Add RevertResizeTask https://review.opendev.org/638046
15:21:51	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add finish_revert_snapshot_based_resize_at_source compute method https://review.opendev.org/637647
15:21:52	openstackgerrit	Matt Riedemann proposed openstack/nova master: Revert cross-cell resize from the API https://review.opendev.org/638048
15:21:52	openstackgerrit	Matt Riedemann proposed openstack/nova master: Add revert_snapshot_based_resize conductor RPC method https://review.opendev.org/638047
15:22:11	mriedem	dansmith: gibi_off: https://review.opendev.org/#/c/693937/ is fixed now
15:22:15	mriedem	py2 vs py3 weirdness
15:27:09	openstackgerrit	Merged openstack/nova master: Remove now invalid TODO from ComputeManager._confirm_resize https://review.opendev.org/693427
15:27:21	openstackgerrit	Merged openstack/nova master: Use ListOfUUIDField from oslo.versionedobjects https://review.opendev.org/693258
15:27:37	openstackgerrit	Merged openstack/nova master: Add known limitation about resize not resizing ephemeral disks https://review.opendev.org/691915
15:27:46	openstackgerrit	Merged openstack/nova master: api-ref: re-work resize action post-conditions https://review.opendev.org/691918
15:27:56	openstackgerrit	Merged openstack/nova master: Provide a better error when _verify_response hits a TypeError https://review.opendev.org/693042
15:30:24	openstackgerrit	Matt Riedemann proposed openstack/nova master: Use named kwargs in compute.API.resize https://review.opendev.org/693425
15:30:26	mriedem	stephenfin: done ^
15:36:05	openstackgerrit	Elod Illes proposed openstack/nova stable/pike: cleanup evacuated instances not on hypervisor https://review.opendev.org/687912
15:36:27	mdbooth	lyarwood: Re https://review.opendev.org/#/c/694033/ What did you think about combining these 2 traits?
15:37:00	mdbooth	IIUC their window of usefulness is limited to the period between stable disks and BFV rescue landing, right?
15:37:24	mdbooth	COMPUTE_RESCUE_BFV implies COMPUTE_RESCUE_STABLE_DEVICES
15:39:02	lyarwood	mdbooth: for the libvirt driver implementation yeah
15:40:48	sean-k-mooney	i think having two traits is fine
15:41:01	mdbooth	lyarwood: As I said in the spec, though, if this discussion gets in the way of getting this done, it's not worth it.
15:41:01	sean-k-mooney	glad to see you put them under compute :)
15:41:32	mdbooth	But I think 1 trait would be better.
15:42:07	sean-k-mooney	its only better if we support rescue for BFV in the same step
15:42:21	sean-k-mooney	or release
15:42:44	mdbooth	sean-k-mooney: Right, but the change from stable disks to rescue BFV is trivial.
15:42:58	mdbooth	IIRC it's just removing a check in the api which prevents it.

Earlier   Later