Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2020-07-31
12:20:07 artom stephenfin, small nit on the test patch, but it'd bug me if it went in as is - just comes off as sloppy
12:21:14 stephenfin artom: Hmm, I don't understand the first comment. Can you elaborate?
12:21:25 artom stephenfin, eh?
12:21:35 stephenfin Since we only have 4 CPUs
12:21:51 stephenfin where's that coming from?
12:22:14 artom stephenfin, oh, you kept it as 8
12:22:18 sean-k-mooney we have 8
12:22:21 artom Huh, and the pinning assertion still works?
12:22:25 stephenfin I kept what as 8?
12:22:31 artom 8 CPUs
12:22:43 sean-k-mooney cpu_nodes=2, cpu_sockets=1, cpu_cores=2, cpu_threads=2,
12:22:52 stephenfin oh, right, yeah, that
12:22:53 sean-k-mooney so 8 threads
12:23:00 artom And the pinning assertion that you added will still always be true? It'll deterministic ally pin the first 4 CPUs for a 4-CPU instance?
12:23:09 artom *always pin
12:23:17 sean-k-mooney artom: stephenfin is relying on how we currenlty order the numa nodes
12:23:20 stephenfin artom: I mean, it's an implementation detail but yeah
12:23:41 sean-k-mooney which is to pass them to itertools.permutations in order
12:23:49 sean-k-mooney which will return them in order
12:24:01 sean-k-mooney so asserting 0-4 are pinned will work
12:24:06 sean-k-mooney untill we change that
12:24:34 sean-k-mooney stephenfin: i assume this test will go away once we kill vcpu_pin_set anyway or be updated right
12:24:43 sean-k-mooney since we only need to keep the ones for the new config values
12:24:45 artom OK, I'll leave it up to you then - personally I feel it'd be cleaner not to rely on the implementation, and just give the host 4 CPUs
12:24:46 stephenfin the first one, yes
12:24:48 artom Or threads
12:24:48 sean-k-mooney so i think the ordering is fine
12:25:18 stephenfin no sweat off my back to change, sec
12:25:19 sean-k-mooney so you mean just do cpu_nodes=1
12:25:26 sean-k-mooney and then 0-3
12:26:22 sean-k-mooney stephenfin: the follow up looks fine to me i guess but you still did not add a release note
12:27:19 sean-k-mooney it proably should have one given the upgrade interaction
12:27:48 stephenfin sure, forgot about that
12:31:30 openstackgerrit Stephen Finucane proposed openstack/nova master: tests: Add reproducer for bug #1889633 https://review.opendev.org/744020
12:31:30 openstack bug 1889633 in OpenStack Compute (nova) "Pinned instance with thread policy can consume VCPU" [High,In progress] https://launchpad.net/bugs/1889633 - Assigned to Stephen Finucane (stephenfinucane)
12:31:30 openstackgerrit Stephen Finucane proposed openstack/nova master: hardware: Reject requests for no hyperthreads on hosts with HT https://review.opendev.org/744021
12:31:34 stephenfin ta-da
13:01:12 artom lyarwood, can I ask you to look over https://review.opendev.org/#/c/729763/6 when you get a chance?
13:02:17 artom Err, lemme rebase that first
13:06:28 openstackgerrit Artom Lifshitz proposed openstack/nova master: Add regression test for bug 1879787 https://review.opendev.org/741230
13:06:28 openstack bug 1879787 in OpenStack Compute (nova) "post_live_migration does not handle Neutron errors" [Medium,In progress] https://launchpad.net/bugs/1879787 - Assigned to Artom Lifshitz (notartom)
13:06:28 openstackgerrit Artom Lifshitz proposed openstack/nova master: Handle Neutron errors in _post_live_migration() https://review.opendev.org/729763
13:10:04 lyarwood artom: ack can do
13:10:23 openstackgerrit Lee Yarwood proposed openstack/nova stable/ussuri: tests: Define constants in '_IntegratedTestBase' https://review.opendev.org/744158
13:10:24 openstackgerrit Lee Yarwood proposed openstack/nova stable/ussuri: func: Add _live_migrate helper to InstanceHelperMixin https://review.opendev.org/744159
13:10:24 openstackgerrit Lee Yarwood proposed openstack/nova stable/ussuri: func: Add live migration rollback volume attachment tests https://review.opendev.org/744160
13:10:26 openstackgerrit Lee Yarwood proposed openstack/nova stable/ussuri: Add regression tests for bug #1889108 https://review.opendev.org/744161
13:10:26 openstack bug 1889108 in OpenStack Compute (nova) "failures during driver.pre_live_migration remove source attachments during rollback" [High,In progress] https://launchpad.net/bugs/1889108 - Assigned to Lee Yarwood (lyarwood)
13:10:26 openstackgerrit Lee Yarwood proposed openstack/nova stable/ussuri: compute: Don't delete the original attachment during pre LM rollback https://review.opendev.org/744162
13:18:10 lyarwood artom: Ie0852a89fc9423a92baa7c29a8806c0628cae220 - can you rebase it again on that
13:18:50 artom lyarwood, ah, yeah
13:18:54 artom Damn churn
13:19:01 artom But the end state is worth it
13:20:27 lyarwood artom: yeah sorry
13:20:36 lyarwood artom: blame stephenfin for suggesting the fup ;)
13:20:50 artom Oh I blame stephenfin regardless
13:20:57 artom But in this case his OCD is useful ;)
13:30:47 artom lyarwood, btw, in a similar vein, question on https://review.opendev.org/#/c/744159/1/nova/tests/functional/integrated_helpers.py
13:32:34 lyarwood artom: I don't think so
13:32:44 lyarwood artom: we can't squash the two togther
13:32:55 lyarwood artom: as the tests used the original
13:33:05 lyarwood artom: I could backport it once they land
13:33:13 lyarwood artom: and it has landed in master
13:33:52 artom lyarwood, ah, right
13:34:00 artom Sorry, I'm a bit all over the place this morning
13:34:53 lyarwood artom: np tgif++ I'm also getting myself tied in knots trying to backport this series to stable/queens
13:35:13 lyarwood and it's 32c here that isn't helping gah!
13:38:25 sean-k-mooney its raining here
13:38:30 sean-k-mooney becaue ireland
13:38:35 artom lyarwood, heat pump yo
13:38:41 artom I have no idea why in Europe you use them for floors
13:38:50 artom Blast that cold air right your face
13:39:08 artom Motorboat that wall unit
13:39:18 lyarwood this house is definitely built to keep heat in
13:39:25 lyarwood and it's doing a great job
13:39:30 lyarwood shame it's 33c now
13:39:32 lyarwood :|
13:39:39 sean-k-mooney same timber frame makes a big difference
13:39:54 openstackgerrit Artom Lifshitz proposed openstack/nova master: Add regression test for bug 1879787 https://review.opendev.org/741230
13:39:54 openstack bug 1879787 in OpenStack Compute (nova) "post_live_migration does not handle Neutron errors" [Medium,In progress] https://launchpad.net/bugs/1879787 - Assigned to Artom Lifshitz (notartom)
13:39:54 openstackgerrit Artom Lifshitz proposed openstack/nova master: Handle Neutron errors in _post_live_migration() https://review.opendev.org/729763
13:40:07 sean-k-mooney you should install vlux windows if you can
13:40:28 sean-k-mooney they are great for getting rid of excess heat in a controled way
13:46:19 stephenfin artom, lyarwood: fwiw, I want to eventually drop the 'ADMIN_API' parameter too
13:46:44 stephenfin rationale being that you have 'self.admin_api' to hand, and it seems to be a good thing to have to think about the kind of user that can do a given op
13:55:55 melwitt lyarwood: ack, will do
14:01:13 openstackgerrit Merged openstack/nova master: Make _rebase_with_qemu_img() generic https://review.opendev.org/743537
14:31:25 openstackgerrit Stephen Finucane proposed openstack/nova master: tests: Add helpers for suspend, resume and reboot of server https://review.opendev.org/741285
14:31:26 openstackgerrit Stephen Finucane proposed openstack/nova master: libvirt: Pass context, instance to '_create_domain' https://review.opendev.org/741286
14:31:27 openstackgerrit Stephen Finucane proposed openstack/nova master: api: Reject non-spawn operations for vTPM https://review.opendev.org/741500
14:31:27 openstackgerrit Stephen Finucane proposed openstack/nova master: libvirt: Add emulated TPM support to Nova https://review.opendev.org/631363
14:31:27 openstackgerrit Stephen Finucane proposed openstack/nova master: docs: Add docs for vTPM support https://review.opendev.org/739213
14:31:28 openstackgerrit Stephen Finucane proposed openstack/nova master: Add type hints to 'nova.compute.manager' https://review.opendev.org/742863
14:31:28 openstackgerrit Stephen Finucane proposed openstack/nova master: Don't unset Instance.old_flavor, new_flavor until necessary https://review.opendev.org/741995
14:31:29 openstackgerrit Stephen Finucane proposed openstack/nova master: privsep: Add support for recursive chown, move_tree operations https://review.opendev.org/742864
14:31:29 openstackgerrit Stephen Finucane proposed openstack/nova master: Add type hints to 'nova.virt.libvirt.utils' https://review.opendev.org/742865
14:31:30 openstackgerrit Stephen Finucane proposed openstack/nova master: Add support for resize and cold migration of emulated TPM files https://review.opendev.org/639934
14:33:53 openstackgerrit Stephen Finucane proposed openstack/nova master: tests: Further usage of new server helpers https://review.opendev.org/743204
14:35:24 openstackgerrit Andrew Bonney proposed openstack/nova master: Fix misleading documentation for live_migration_inbound_addr https://review.opendev.org/741473
14:39:56 stephenfin kashyap: Want to give this an eye over once more https://review.opendev.org/#/c/741473/ ?

Earlier   Later