Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2019-06-07
15:10:59 mriedem otherwise i can (1) create my server with flavor1 and numa1, (2) resize to flavor2/numa2, confirm, (3) cold migrate and now i'm scheduling with flavor1's numa1
15:11:06 mriedem sean-k-mooney: the request spec?
15:11:15 mriedem that would be an up-call...
15:11:25 mriedem that also doesn't solve your revert problem
15:11:58 sean-k-mooney ya we i think recalualate the numa toplogy in the driver and store it in the ... instance not he request spec never mind
15:12:24 sean-k-mooney so ya we need to actuly save it and then revert it
15:12:29 stephenfin sean-k-mooney: Definitely? I know we used to mess with it in the filters https://github.com/openstack/nova/blob/master/nova/scheduler/filters/numa_topology_filter.py#L61-L68
15:15:06 sean-k-mooney well see the fact the numa_fit_instance_to_host fucntion modifes the request sepc is what made me suspect this would get updated by the compute node if we booted sucessfuly
15:15:24 sean-k-mooney but honestly we need 1 trace teh code and two test it locally to confirm
15:15:27 stephenfin We probably don't want to persist that change though, in any case
15:16:32 sean-k-mooney proably. the compute node really should not need to modify the request_spec
15:16:50 sean-k-mooney well in this instance
15:16:51 mriedem the compute service should never be saving changes to the RequestSpec, ever, since it's an up-call to the API DB
15:16:57 mriedem dansmith: agree with me here? ^
15:17:24 mriedem stephenfin: as i explained above, the RequestSpec.numa_topology is used in the scheduler so yes it needs to be persisted for subsequent move operations,
15:17:36 mriedem however, on revert we have to make sure we revert the numa_topology in the request spec to the old flavor's values
15:17:37 dansmith mriedem: of course
15:17:39 sean-k-mooney we only added the request spec to the api db around rocky correct
15:17:40 mriedem which is my -1
15:17:46 mriedem jesus gd
15:17:52 stephenfin Any changes from the compute node, I mean
15:17:54 sean-k-mooney buy you are right
15:18:00 mriedem i would like to phone a friend
15:19:14 mriedem see this code on revert https://github.com/openstack/nova/blob/master/nova/compute/api.py#L3410
15:19:19 mriedem that's the thing you need to concern yourself with
15:19:29 mriedem and why i -1'ed alex_xu's change
15:19:30 mriedem and yours
15:21:42 mriedem sean-k-mooney: request spec has been around since mitaka or newton at least, which is why when i have to fix a regression bug b/c of the request spec, i have to backport it through 20 releases
15:21:48 dansmith sean-k-mooney: reqspec has been around a lot longer than rocky
15:21:50 mriedem which makes me a sad panda
15:22:39 sean-k-mooney oh i know the reqest spec has been
15:23:00 sean-k-mooney i just tought we only strated storing it in the api db more recently
15:23:03 openstackgerrit Balazs Gibizer proposed openstack/nova master: allow getting resource request of every bound ports of an instance https://review.opendev.org/655110
15:23:33 sean-k-mooney i tought melwitt added it to the api db like 2-3 release ago but i could be wrong about that
15:24:09 dansmith we've had it in the api db since we've had an api db AFAIK
15:24:29 mriedem 4th api db migration
15:24:35 dansmith yeah
15:24:51 sean-k-mooney ah ok
15:24:54 mriedem sweet laski
15:26:45 mriedem apparently we aren't keeping track of runways anymore https://etherpad.openstack.org/p/nova-runways-train
15:28:42 sean-k-mooney mriedem: i mentioned on irc that my one can be dropped form the runway
15:29:08 panda uh ?
15:29:15 panda oh, the sad me.
15:29:17 panda ok.
15:29:37 stephenfin panda: Damn it man, look sadder!
15:29:46 sean-k-mooney it still need a second +2 but its more or less done at this point
15:32:45 mriedem i've got it in a tab
15:33:07 sean-k-mooney mriedem: oh incase you missed it i set https://blueprints.launchpad.net/nova/+spec/libvirt-neutron-sriov-livemigration to implemented since the neutron docs change has merged too and updated the spec link to the renderd version instead of gerrit
15:33:46 mriedem ack thanks
15:34:35 sean-k-mooney im going to go grab lunch ill be around later for an hour or so o/
15:36:15 stephenfin sean-k-mooney: s/lunch/dinner/
15:40:51 openstackgerrit Balazs Gibizer proposed openstack/nova master: Pass network API to the conducor's MigrationTask https://review.opendev.org/655111
15:40:52 openstackgerrit Balazs Gibizer proposed openstack/nova master: Add request_spec to server move RPC calls https://review.opendev.org/655721
15:40:52 openstackgerrit Balazs Gibizer proposed openstack/nova master: re-calculate provider mapping during migration https://review.opendev.org/655112
15:40:53 openstackgerrit Balazs Gibizer proposed openstack/nova master: update allocation in binding profile during migrate https://review.opendev.org/656422
15:40:53 openstackgerrit Balazs Gibizer proposed openstack/nova master: Extend NeutronFixture to handle migrations https://review.opendev.org/655114
15:44:10 openstackgerrit Lee Yarwood proposed openstack/nova master: libvirt: Use SATA bus for cdrom devices when using Q35 machine type https://review.opendev.org/663011
15:44:10 openstackgerrit Lee Yarwood proposed openstack/nova master: DNM: Run tempest-full-py3 with q35 machine type https://review.opendev.org/662887
15:56:01 openstackgerrit Artom Lifshitz proposed openstack/nova master: Revert resize: wait for events according to hybrid plug https://review.opendev.org/644881
15:56:18 artom mriedem, bauzas ^^
16:04:52 openstackgerrit Dr. Clemens Hardewig proposed openstack/nova master: Fixes instance resizing/migration when swap volume is on lvm https://review.opendev.org/618621
16:04:52 openstackgerrit Dr. Clemens Hardewig proposed openstack/nova master: Fixes instance resizing/migration when swap volume is on lvm https://review.opendev.org/663986
16:14:16 openstackgerrit Stephen Finucane proposed openstack/nova master: Recalculate 'RequestSpec.numa_topology' on resize https://review.opendev.org/662522
16:14:36 finucannot sean-k-mooney, mriedem: ^ that should do the trick...I think
16:15:40 finucannot mriedem: I'm going to have to rebase https://review.opendev.org/#/c/651311/ otherwise Gerrit won't let it merge, so want me to add the TODO?
16:18:07 mriedem sure
16:23:12 openstackgerrit Stephen Finucane proposed openstack/nova master: Remove cells v1 parameter from 'ComputeTaskAPI.resize_instance' https://review.opendev.org/651311
16:23:13 openstackgerrit Stephen Finucane proposed openstack/nova master: Stop passing 'kwargs' to 'rebuild_instance' https://review.opendev.org/651312
16:23:13 openstackgerrit Stephen Finucane proposed openstack/nova master: Stop passing 'delete_type' to 'terminate_instance' https://review.opendev.org/651313
16:23:14 openstackgerrit Stephen Finucane proposed openstack/nova master: filters: Stop handling cells v1 https://review.opendev.org/651314
16:23:14 openstackgerrit Stephen Finucane proposed openstack/nova master: Remove nova.compute.*API() shims https://review.opendev.org/660527
16:23:15 openstackgerrit Stephen Finucane proposed openstack/nova master: Add reno for removed cells v1 policies https://review.opendev.org/662031
16:23:15 openstackgerrit Stephen Finucane proposed openstack/nova master: Remove 'InstanceUnknownCell' exception https://review.opendev.org/662411
16:23:16 openstackgerrit Stephen Finucane proposed openstack/nova master: Ensure controllers all call super https://review.opendev.org/660950
16:23:19 finucannot done
16:28:30 finucannot whoops, missed the comments in the following patches. Fixed those too
16:28:32 openstackgerrit Stephen Finucane proposed openstack/nova master: Stop passing 'kwargs' to 'rebuild_instance' https://review.opendev.org/651312
16:28:33 openstackgerrit Stephen Finucane proposed openstack/nova master: Stop passing 'delete_type' to 'terminate_instance' https://review.opendev.org/651313
16:28:33 openstackgerrit Stephen Finucane proposed openstack/nova master: filters: Stop handling cells v1 https://review.opendev.org/651314
16:28:34 openstackgerrit Stephen Finucane proposed openstack/nova master: Remove nova.compute.*API() shims https://review.opendev.org/660527
16:28:34 openstackgerrit Stephen Finucane proposed openstack/nova master: Add reno for removed cells v1 policies https://review.opendev.org/662031
16:28:35 openstackgerrit Stephen Finucane proposed openstack/nova master: Remove 'InstanceUnknownCell' exception https://review.opendev.org/662411
16:28:35 openstackgerrit Stephen Finucane proposed openstack/nova master: Ensure controllers all call super https://review.opendev.org/660950
16:29:15 mriedem done
16:29:34 mriedem a couple of easy -1s to address in the next 2 after that
16:29:45 finucannot yeah, just tackled those
16:33:16 openstackgerrit Matt Riedemann proposed openstack/nova master: Refresh instance network info on deletion https://review.opendev.org/660761
16:34:09 mriedem easy fix ^
16:40:03 finucannot Indeed. +W
16:44:48 lyarwood mdbooth: replied btw, the caps stub is still required.
16:45:38 Sundar johnthetubguy: Hi, are you good with https://review.opendev.org/#/c/603955/ after the updates?
16:48:51 finucannot lyarwood: If you add TODOs for the few things mdbooth called out (I left my own comments on same), I'm +2
16:49:00 finucannot and then I'm going home
16:50:45 efried lyarwood: I gotta run, but you're hitting this code path in your tests
16:50:45 efried https://opendev.org/openstack/nova/src/branch/master/nova/virt/libvirt/driver.py#L4835
16:50:46 efried which is patently wrong - should either be ... in (fields.Architecture.AARCH64,) <== tuple -- or just ==
16:51:12 Sundar johnthetubaguy: ^
16:51:31 efried bbiab
16:54:00 lyarwood fried_rolls / finucannot: looking, thanks
16:59:14 artom Note to self: bdms != bdsm

Earlier   Later