Posted	Nick	Remark
#openstack-nova - 2019-06-17
13:43:18	stephenfin	RE: the cpu-resources spec?
13:47:11	sean-k-mooney	more or less
13:47:23	sean-k-mooney	you mean on the mailing list post right
13:47:59	sean-k-mooney	the prefilter will default to false in the config
13:48:06	sean-k-mooney	so on upgade it will be disabeld by default
13:48:41	sean-k-mooney	as you are not ment to modify config as part of upgade that means that by default we will get inveoties of both VCPU and PCPU
13:49:14	sean-k-mooney	then later we can enable the prefilter and add the dedicated_cpu_set and remove teh vcpu_pin_set
13:52:18	sean-k-mooney	personally i had assumed that defineing the dedicated_cpu_set option is what would have done the reshape fo exisitng instances
13:53:43	sean-k-mooney	reshape is a little impresise what i mean is modifying the allcoation for exsiting instance to consume PCPUs if needed
13:59:31	shilpasd	Hi, can anybody help me to test live migration? i using openstack via devstack, and facing below error
13:59:33	shilpasd	libvirtError: operation failed: Failed to connect to remote libvirt URI qemu+tcp://<vm_user>@<compute_node_name>/system: unable to connect to server at '<compute_node_name>:16509': Connection refused
14:00:08	shilpasd	FYI: referred 'https://www.mirantis.com/blog/tutorial-openstack-live-migration-with-kvm-hypervisor-and-nfs-shared-storage/' for libvirtd.conf configuration
14:01:45	shilpasd	after running telnet >>Unable to connect to remote host: Connection refused
14:02:07	shilpasd	seems configuration issue, can anybody provide me any specific link for live-migration
14:05:13	mriedem	efried: i've half a mind to revert this https://review.opendev.org/#/c/663502/
14:05:28	mriedem	he removed an unused method but it's a regression to not be using that method
14:05:32	mriedem	https://bugs.launchpad.net/nova/+bug/1830260
14:05:32	openstack	Launchpad bug 1830260 in OpenStack Compute (nova) stein "Inefficient host_status lookup when listing servers with details (regression)" [Medium,Triaged]
14:06:20	mriedem	but i guess it could be reverted when someone actually fixes the bug
14:06:57	efried	mriedem: ack. in any case, I'd like to understand the path I should have taken as a reviewer to know about that bug and that using this method was a solution
14:07:38	mriedem	it's not really your fault. i don't know, but i suspect takashi noticed it when i pointed out the regression in a recent-ish ML thread.
14:08:25	efried	okay
14:09:41	efried	mriedem: I guess a note in the bug mentioning the commit where it was removed so it can be reinstated?
14:09:58	efried	...reinstated when fixing the bug?
14:10:01	mriedem	yeah good point
14:21:06	shilpasd	mriedem: hi do you have few time, want to discuss @ live migration issue i am facing
14:21:30	openstack	Launchpad bug 1830260 in OpenStack Compute (nova) stein "Inefficient host_status lookup when listing servers with details (regression)" [High,Triaged]
14:21:30	sean-k-mooney	so if im following correctly https://bugs.launchpad.net/nova/+bug/1830260 was really intoduced by https://review.opendev.org/#/c/584590/ when that stop calling the method that https://review.opendev.org/#/c/663502/1 removed
14:23:15	sean-k-mooney	should ^ have a stein-backport-potential tag
14:33:14	sean-k-mooney	mriedem: by the way your cross cell resize series should be entering the runway in a day or so. i assume the reviewers guide you sent before is still pretty much accurate? ill try to start on the series again later this weeks but ill re read that in the mean time to reload the context
14:47:32	openstackgerrit	Stephen Finucane proposed openstack/nova master: tests: Stop starting consoleauth in functional tests https://review.opendev.org/652966
14:47:32	openstackgerrit	Stephen Finucane proposed openstack/nova master: docs: Remove references to nova-consoleauth https://review.opendev.org/652965
14:47:33	openstackgerrit	Stephen Finucane proposed openstack/nova master: xvp: Remove use of '_LI' marker https://review.opendev.org/665425
14:47:33	openstackgerrit	Stephen Finucane proposed openstack/nova master: xvp: Start using consoleauth tokens https://review.opendev.org/652967
14:47:34	openstackgerrit	Stephen Finucane proposed openstack/nova master: Remove nova-consoleauth https://review.opendev.org/652969
14:47:34	openstackgerrit	Stephen Finucane proposed openstack/nova master: nova-status: Remove consoleauth workaround check https://review.opendev.org/652968
14:47:35	openstackgerrit	Stephen Finucane proposed openstack/nova master: docs: Rework nova console diagram https://review.opendev.org/660147
14:47:35	openstackgerrit	Stephen Finucane proposed openstack/nova master: objects: Remove ConsoleAuthToken.to_dict https://review.opendev.org/652970
14:47:36	openstackgerrit	Stephen Finucane proposed openstack/nova master: Remove dead file https://review.opendev.org/665694
14:47:36	openstackgerrit	Stephen Finucane proposed openstack/nova master: docs: Integrate 'sphinx.ext.imgconverter' https://review.opendev.org/665693
14:49:46	openstackgerrit	Stephen Finucane proposed openstack/nova master: docs: Remove unused Visio diagram https://review.opendev.org/665694
14:50:16	bauzas	mriedem: thanks for having reviewed https://review.opendev.org/#/c/650963/
14:50:27	bauzas	mriedem: I just updated with comments, lemme know if you need more details
14:59:16	mriedem	sean-k-mooney: yes it's as described in the bug. and yes the cross-cell resize review guide in the ML is unchanged.
14:59:20	mriedem	shilpasd: what's up
15:13:45	openstackgerrit	Adam Spiers proposed openstack/nova master: Add extra spec parameter and image property for memory encryption https://review.opendev.org/664420
15:14:16	aspiers	efried, sean-k-mooney: ^^^ changes as discussed last week
15:15:14	openstackgerrit	Sylvain Bauza proposed openstack/nova-specs master: Proposes NUMA affinity for vGPUs https://review.opendev.org/650963
15:27:00	efried	aspiers: ack
15:27:30	aspiers	efried: I'm just adding the API-level check to it which Alex pointed out was missing
15:28:01	ralonsoh	tonyb, hi, can you take a look at https://review.opendev.org/#/c/655694/? Thank you in advance
15:28:45	openstackgerrit	Stephen Finucane proposed openstack/nova master: docs: Remove unused Visio diagram https://review.opendev.org/665694
15:28:45	openstackgerrit	Stephen Finucane proposed openstack/nova master: docs: Integrate 'sphinx.ext.imgconverter' https://review.opendev.org/665693
15:35:15	temka	sean-k-mooney, looks like you took a look at https://bugs.launchpad.net/nova/+bug/1826519 - to me that's a cloud-init problem, not a Nova one. You OK to close as Invalid?
15:35:16	openstack	Launchpad bug 1826519 in OpenStack Compute (nova) "Ephemeral disk volume was not mounted after resizing from non-ephemeral flavor" [Undecided,New]
15:35:20	artom	Whoops
15:49:13	openstackgerrit	Balazs Gibizer proposed openstack/os-vif master: Insert osprofiler trace info as external_ids to the bridge table https://review.opendev.org/665715
15:53:36	openstackgerrit	Adam Spiers proposed openstack/nova master: Add extra spec parameter and image property for memory encryption https://review.opendev.org/664420
15:57:00	openstackgerrit	Adam Spiers proposed openstack/nova master: Add extra spec parameter and image property for memory encryption https://review.opendev.org/664420
15:57:13	aspiers	efried: API-level check is now in ^^^
15:57:19	efried	ack
15:57:35	aspiers	This should be ready for review. Now it should be easy for me to fix the subsequent SEV guest config changes!
16:05:41	openstackgerrit	Stephen Finucane proposed openstack/nova master: hacking: Resolve E731 (do not assign a lambda expression, use a def) https://review.opendev.org/651554
16:05:42	openstackgerrit	Stephen Finucane proposed openstack/nova master: hacking: Resolve E741 (ambiguous variable name) https://review.opendev.org/652103
16:05:42	openstackgerrit	Stephen Finucane proposed openstack/nova master: hacking: Resolve W503 (line break occurred before a binary operator) https://review.opendev.org/651555
16:05:43	openstackgerrit	Stephen Finucane proposed openstack/nova master: hacking: Resolve W605 (invalid escape sequence) https://review.opendev.org/652104
16:25:10	mriedem	sean-k-mooney: alex_xu: i went through the pmem spec https://review.opendev.org/#/c/601596/ - lots of questions and comments
16:26:21	openstackgerrit	Merged openstack/nova master: Ensure controllers all call super https://review.opendev.org/660950
16:27:26	sean-k-mooney	mriedem: ok i need to finish review that spec again too. i have not done a full review since version 12
16:28:49	mriedem	the biggest nagging feeling i get on that spec is it sounds like these are local block devices and should be modeled as BDMs
16:29:01	mriedem	they really sound like the libvirt lvm image backend with the zero'ing out and such
16:29:16	sean-k-mooney	they are presended to the guest as memory DIMs however
16:29:24	sean-k-mooney	not as block devices
16:29:32	sean-k-mooney	but form a host level they are similar
16:30:36	alex_xu	mriedem: thanks a lot, I will go through them all tomorrow
16:32:36	mriedem	sean-k-mooney: yeah i know, and i'm not saying he should change the spec, and modeling as bdms could cause confusion in a different way as well
16:32:48	sean-k-mooney	one other usecase that is out of scope for Train but i expect to see in U is the option to use PMEM inplacem of ram as similarly to how we use hugepages or filebacked memroy. for the current usecase BDMs may make sense but im not sure if it will work for those usecase too in the future
16:33:00	mriedem	i guess i just don't like new models and objects and such for new devices but i'm cranky
16:33:48	sean-k-mooney	ill go through you review comments properly this week and try to digest it.
16:34:28	sean-k-mooney	i do agree that we have to be carful with what happens if the config options chage
16:37:37	alex_xu	yea, at least from hardware view, it still memory, not the block device
16:38:29	alex_xu	sean-k-mooney: currently we have a config option which is for file backend memory
16:39:14	alex_xu	but it isn't very flexible as I remember, I didn't think about that too much also, since that isn't the usecase we are looking for now also
16:49:24	mriedem	oh yeah... https://docs.openstack.org/nova/latest/configuration/config.html#libvirt.file_backed_memory
16:51:17	mriedem	if you changed that from true to false and restarted with guests on that host we'd have problems
16:51:28	mriedem	er >0 to 0
17:24:18	openstackgerrit	Matt Riedemann proposed openstack/nova master: Remove file-backed memory live migration compat check https://review.opendev.org/665728
18:14:40	openstackgerrit	Adam Spiers proposed openstack/nova master: Use fake flavor instead of empty dict in test https://review.opendev.org/662555
18:14:41	openstackgerrit	Adam Spiers proposed openstack/nova master: Extract SEV-specific bits on host detection https://review.opendev.org/636334
18:14:41	openstackgerrit	Adam Spiers proposed openstack/nova master: Pass extra_specs to flavor in vif tests https://review.opendev.org/662556
18:14:42	openstackgerrit	Adam Spiers proposed openstack/nova master: Allow guest devices to include https://review.opendev.org/644564
18:14:42	openstackgerrit	Adam Spiers proposed openstack/nova master: Add element to guest config for AMD SEV https://review.opendev.org/636318
18:14:43	openstackgerrit	Adam Spiers proposed openstack/nova master: Use element when SEV is required https://review.opendev.org/662557
18:14:43	openstackgerrit	Adam Spiers proposed openstack/nova master: Detect that SEV is required and enable iommu for devices https://review.opendev.org/644565
18:14:44	openstackgerrit	Adam Spiers proposed openstack/nova master: Enable memory locking if SEV is requested https://review.opendev.org/662558
18:21:57	openstackgerrit	Harald Jensås proposed openstack/nova master: cell_mapping - wrap IPv6 address in format_url https://review.opendev.org/664559
18:25:41	sean-k-mooney	stephenfin: just an fyi the kolla-ansibel change for removing nova-consoleauth just merged https://review.opendev.org/#/c/661427/
18:26:13	sean-k-mooney	Jeffrey4l: thanks for completing ^
18:29:39	hjensas	dansmith: Thanks for the review on ^^, you where right and I want to apologize for the 'did you look at the bug' comment. wdyt about the new revision?
18:30:10	dansmith	hjensas: you're still changing the existing format I think, but I'm in the middle of something else.. will comment in a few

Earlier   Later