Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2019-07-02
13:32:33 efried kashyap: note that under the latest design proposal, existing confs will continue to work, so the load on the deployer isn't as great.
13:32:37 sean-k-mooney ill quickly re reivew it too now but i was ok with the usecase in the past i think
13:33:16 kashyap efried: Will look once I address gibi's comments in my own spec
13:33:22 efried stephenfin, gibi, sean-k-mooney: ack, no big hurry I guess. Thanks kashyap.
13:33:22 kashyap stephenfin: ^
13:34:01 sean-k-mooney efried: its proably one of the spec we should be able to close out on today as part of the review day
13:34:09 kashyap gibi: On the Secure Boot thing, fixing your remarks. My eyes are glazing over from looking at my own text. Note that the "firmware descriptor files" from QEMU are nice-to-have, but doesn't block Train
13:34:15 kashyap I'll add notes there.
13:34:19 efried sean-k-mooney: yeah, that was my hope :)
13:34:28 gibi kashyap: ack
13:34:31 sean-k-mooney efried: by the way do you have the link handy to the dashboard you created
13:34:33 kashyap sean-k-mooney: I was not OK with the spec, and actually abandoned it
13:34:41 efried sean-k-mooney: https://etherpad.openstack.org/p/nova-spec-review-day
13:34:45 sean-k-mooney efried: thanks
13:34:47 kashyap But it was re-opened, and I want to see if I can convince myself again
13:34:56 efried sean-k-mooney: front matter has a couple of gerrit dashboard links.
13:35:00 stephenfin Is today a review day?
13:35:11 sean-k-mooney stephenfin: yes
13:35:11 stephenfin *spec review day
13:35:19 sean-k-mooney i think the final one for train
13:35:22 stephenfin Crap. Ugh. Too much to do
13:35:25 efried stephenfin: http://lists.openstack.org/pipermail/openstack-discuss/2019-June/007381.html
13:35:34 efried I should have sent a reminder last night, mahbad.
13:36:08 sean-k-mooney spec freeze is july 25th i think so thusday 3 weeks
13:36:47 stephenfin I'll try schedule that for this evening so. Feck the emails
13:37:07 openstackgerrit Theodoros Tsioutsias proposed openstack/nova-specs master: Add PENDING vm state https://review.opendev.org/648687
13:38:42 alex_xu sean-k-mooney: you want to use the min and max of inventory for the fixed size cache,right?
13:39:36 sean-k-mooney sorry yes min max and step size should all be set to the same
13:39:41 sean-k-mooney to prevent the fragmenation
13:39:45 sean-k-mooney not just step size
13:40:07 lyarwood sean-k-mooney: just posted some addition nits in the pqos spec if you're going over it still
13:40:37 sean-k-mooney lyarwood: i was going to leave them build up until 1600 then adress them all
13:41:02 alex_xu cool, sounds cool for me
13:41:10 sean-k-mooney but yes im checking it every so often
13:41:22 lyarwood sean-k-mooney: ack
13:42:44 sean-k-mooney alex_xu: woudl you prefer to manage the min/max/step size via the placmenet api as i was propsoeing or should i update the spec to have a config option
13:43:51 kashyap gibi: I'd need your re-ACK as I'm fixing the two other broken sentences, and add a note on the QEMU dep for nice-to-have.
13:44:26 efried Is /servers/{server_id}/action Create Image (createImage Action) aka "snapshot"?
13:44:27 gibi kashyap: just ping me
13:45:01 gibi kashyap: when you are ready
13:45:04 kashyap alex_xu: Are you convinced of that CPU models thing? WheN I explained this spec to two QEMU maintainers, they said "don't do that" (the CPU models as list). But rather allow them setting via flavor properties, if you want
13:45:10 kashyap gibi: Yep, juggling multiple things
13:45:12 alex_xu sean-k-mooney: I prefer to have config option. otherwise, I'm thinking how do we deal with the update_provider_tree and the user update
13:45:26 kashyap alex_xu: I'll comment in full, once I flush existing content to my brain's disk
13:45:32 sean-k-mooney efried: the create image action i think is snapshot yes
13:46:09 sean-k-mooney alex_xu: for update provider tree i planned to have nova only set them if it creating the inventory initailly
13:46:26 alex_xu kashyap: as my understand, that spec doesn't helpful for your flaw mitigation case, but that spec aims to live migration case
13:46:49 sean-k-mooney so update provider tree would not modify them but ill add the config option in the next version so since that is simple
13:46:58 kashyap alex_xu: Yeah, I do see the use case, though. Not denying that
13:47:12 efried looks like snapshot may also be used for shelve
13:47:40 kashyap alex_xu: Just weighing the costs, and if it'll come to bite us in a different scenario -- as it's hard to think through _all_ the cases that might be affected here.
13:47:50 sean-k-mooney efried: yes when we shelve we snapshot the root disk
13:48:06 sean-k-mooney but im not sure if we only do that for shelve offload
13:48:14 sean-k-mooney or if it happens on shelve
13:53:01 openstackgerrit sean mooney proposed openstack/nova master: [DNM] testing bug/1813789 revert resize events https://review.opendev.org/664442
13:53:11 openstackgerrit Artom Lifshitz proposed openstack/nova master: [DNM] testing bug/1813789 revert resize events https://review.opendev.org/664442
13:54:17 sean-k-mooney jinx
13:57:26 artom|gmtplus3 Dammit, 11 seconds
14:00:59 sean-k-mooney i did it via the gerrit ui
14:01:40 sean-k-mooney given the gate lag time it makes no difference :)
14:05:53 gibi sean-k-mooney: replied in https://review.opendev.org/#/c/668675
14:07:08 kashyap gibi: A small heads-up: I've also fixed two more small things related to content moving. I'll append a short summary of changes for convenience, once the docs compile locally.
14:07:21 gibi kashyap: OK, I will look at the diff
14:08:32 sean-k-mooney gibi: oh wiat thost are not test they are test helper functions because they are prefixed wtih _
14:08:51 gibi sean-k-mooney: yeah, they are steps of a biiig test case
14:08:53 sean-k-mooney so _test_shelve_and_shelve_offload_server is beeing invoked elsewhere
14:09:21 sean-k-mooney ah ok in that case your change makes sense i missed that
14:09:25 gibi sean-k-mooney: it is invoked exactly before _test_unshelve_server()
14:09:32 sean-k-mooney i was wondering why you were doing extra work when you did not need too
14:09:45 openstackgerrit Kashyap Chamarthy proposed openstack/nova-specs master: Add "Secure Boot support for KVM & QEMU guests" spec https://review.opendev.org/506720
14:09:48 gibi sean-k-mooney: this is the test case https://review.opendev.org/#/c/668675/1/nova/tests/functional/notification_sample_tests/test_instance.py@342
14:10:09 sean-k-mooney well in that case i dont think you need to do the unshelve right?
14:10:46 kashyap gibi: When you can: done. Posted the shortlog as well.
14:10:46 sean-k-mooney you are doing an extra unchelve and shelve that are not needed
14:10:54 openstackgerrit Stephen Finucane proposed openstack/nova stable/stein: Ignore hw_vif_type for direct, direct-physical vNIC types https://review.opendev.org/667354
14:11:06 openstackgerrit Stephen Finucane proposed openstack/nova stable/rocky: Ignore hw_vif_type for direct, direct-physical vNIC types https://review.opendev.org/667355
14:11:36 sean-k-mooney gibi: if _test_shelve_and_shelve_offload_server left it shelved and _test_unshelve_server required a shelved server it would speed up the test case
14:11:36 openstackgerrit Stephen Finucane proposed openstack/nova stable/rocky: Ignore hw_vif_type for direct, direct-physical vNIC types https://review.opendev.org/667355
14:12:30 gibi sean-k-mooney: I could skip that but then instance would be in SHELVED case between the steps and that would ruin the generic pre-condition of the test steps enforced in https://review.opendev.org/#/c/668675/1/nova/tests/functional/notification_sample_tests/test_instance.py@384
14:13:15 gibi sean-k-mooney: I could combine the two test step
14:13:17 sean-k-mooney i see
14:13:26 gibi kashyap: ack
14:13:42 kashyap Thanks a ton.
14:13:58 sean-k-mooney into a _test_shelve_shelve_offload_unshelve_server
14:14:10 sean-k-mooney ya that would be more efficent
14:15:13 sean-k-mooney and it would not violate the pre/post condition that the server is active between all steps
14:16:06 sean-k-mooney gibi: that said, these are fucntional tests
14:16:29 gibi kashyap: I'm +2 on the secure boot
14:16:29 sean-k-mooney so we are not actully snapshoting the server and uploading it to glance and then restoring it
14:16:52 sean-k-mooney so what you have is proably fine for the fake driver/fake libvirt whichever we are actully using
14:16:58 kashyap gibi: Thank you. And FWIW, I'm giving a "semi-strong +1" to to the CPU models thing, with a note.
14:17:08 gibi kashyap: thanks
14:17:13 kashyap As I don't want to block it any further, and we can hash out the details in the implementation.
14:18:49 gibi kashyap: ack
14:25:29 openstackgerrit Matt Riedemann proposed openstack/nova stable/queens: Fail to live migration if instance has a NUMA topology https://review.opendev.org/629597
14:25:42 openstackgerrit Matt Riedemann proposed openstack/nova stable/queens: fix up numa-topology live migration hypervisor check https://review.opendev.org/648939
14:26:21 mriedem lyarwood: can you hit https://review.opendev.org/#/c/629597/ ?
14:30:35 mriedem coreycb: can you ack this? https://review.opendev.org/#/c/668680/
14:41:45 efried kashyap: Did you forget to hit save or are you still composing?
14:42:04 kashyap efried: Forgot to hit save, bad me. A sec

Earlier   Later