Earlier  
Later
New Search
Posted Nick Remark
#openstack-nova - 2020-07-09
15:48:24 gibi stephenfin: https://review.opendev.org/#/c/739208/ looks good to me
15:51:30 gibi nova meeting starts in 9 minutes on #openstack-meeting-3
15:56:07 openstackgerrit Artom Lifshitz proposed openstack/nova master: Improve documentation of (unpin|pin)_cpus_with_siblings() https://review.opendev.org/740327
15:56:23 artom sean-k-mooney, stephenfin ^^ quick drive-by comment patch while I work on the downstream bug
16:00:59 sean-k-mooney there are docs for that :P
16:02:42 sean-k-mooney so on master if cpus & sib: should alwasy be empty/false when you have hw:cpu_thread_policy=isolate
16:02:53 sean-k-mooney we change the meaning in train
16:03:18 sean-k-mooney before it ment select host without SMT or claim both siblings if SMT is enabled on the host
16:03:32 sean-k-mooney form train on its just select a host without SMT
16:03:50 artom sean-k-mooney, *shrug* feel free to -1 :)
16:04:17 sean-k-mooney so i dont think there is a valid path to use pin_cpus_with_siblings anymore unless we are allowing the old behavior if you dont report PCPUs
16:04:26 sean-k-mooney stephenfin: are we supporting ^
16:05:04 stephenfin sean-k-mooney: https://review.opendev.org/#/c/705666/
16:05:37 sean-k-mooney hehe i even reviewd it i tought we spoke about this before
16:06:21 sean-k-mooney so we can remove those in victoria right
16:07:08 stephenfin yup
16:07:11 stephenfin just got to find the time
16:07:11 sean-k-mooney should thos be in the runway with the mixed cpus changes or just treated as seperate cleanups
16:07:16 stephenfin separate
16:07:20 stephenfin let's land the feature first
16:07:23 sean-k-mooney ok
16:24:19 openstackgerrit sean mooney proposed openstack/nova master: lookup nic feature by pci address https://review.opendev.org/739131
16:24:52 sean-k-mooney alex_xu: i need to test this on my sriov servers so ill set -w on it for now but ^ should adress your comment
16:25:07 sean-k-mooney it should be much more effiecnt then we had before
16:31:13 gmann sean-k-mooney: there are more deps need to bump lower-constraint, I am getting nova job green first and then same we can apply on os-vif- https://review.opendev.org/#/c/738126/8
16:33:04 gmann i think i need to create Focal env and test in one shot. let's see if there are more failure
16:44:17 sean-k-mooney i have folal devn envs but i have run devstack so they are not clean anymore
16:44:47 sean-k-mooney i can proabnly test the pathcs locally if you need me too i just need to boot a clean vm but not today
16:45:41 sean-k-mooney gmann: i assme infra have no plance to add the deps to the base images so we either need to fix them in bindeps in each project or bump the minor verions
16:46:35 sean-k-mooney gmann: looks like https://review.opendev.org/#/c/738130/ passed for os-vif
16:46:52 gmann sean-k-mooney: cool.
16:47:18 gmann infra side, not at lest min version as that is project side things
16:48:14 sean-k-mooney i was more thinking of the libffi-dev issue
16:48:15 sean-k-mooney but ok
16:48:49 gmann let me update cmt msg and remove depends-on as 738130 has lower-constraint update also
16:49:29 sean-k-mooney sure. the kuryr-kubernetes-tempest job is not an os-vif job by the way so we will need to fix kuryr-kubernetes before that will pass
16:49:58 gmann yeah
16:50:07 sean-k-mooney its uses os-vif form git however so we run it to make sure we dont break them
16:51:46 sean-k-mooney which reminds me with the upcoming change in devstack to use ovn by default - os-vif-ovs will be come an ovn job but we need to make sure that - os-vif-ovs-iptables keeps using ml2/ovs
16:51:58 sean-k-mooney thats unrealted but it just occured to me
16:53:16 openstackgerrit Stephen Finucane proposed openstack/nova master: libvirt: Re-enable live snapshot for paused instances https://review.opendev.org/740334
16:53:17 openstackgerrit Stephen Finucane proposed openstack/nova master: WIP: Attempt to restore some sanity to snapshot https://review.opendev.org/740336
16:53:17 openstackgerrit Stephen Finucane proposed openstack/nova master: libvirt: Remove workaround for really old QEMU https://review.opendev.org/740335
16:53:34 gmann yeah, till now we added ovn as n-v job on devstack and check its stability but yes we need to adjust few bits for ovs specific jobs
16:54:43 gmann let me try kuryr-kubernetes migration also and test all things together.
16:55:33 sean-k-mooney given that job install k8s it might be more work then other to get working
16:56:21 sean-k-mooney it will pull in docker and k8s which both shoudl be availabel on ubuntu 20.04 but there may be issue with the devstack plugin even before you get to the kuyry part
16:56:26 gmann yeah, i remember the bionic migration case. but let's try
16:56:26 sean-k-mooney or it might just work
16:57:09 sean-k-mooney ill admit its the job i know least about on the os-vif gate
16:57:26 gmann any reason it was n-v ?
16:58:13 sean-k-mooney yes it used to be quite unstable
16:58:23 sean-k-mooney well the old job did at least then the old job was deleted
16:58:58 sean-k-mooney its been fine form most of last cycle but we have had relitivly low volume of patches so it never seamed worth the effort to make it voting
16:59:29 sean-k-mooney i still check its results before mreging stuff but dont want to have it block someting it there gate is broken
17:00:07 sean-k-mooney we did plan to make it voting evnetually it has just never happened.
17:08:01 gmann ok, I created Focal env, testing locally
17:17:08 openstackgerrit Stephen Finucane proposed openstack/nova master: hardware: Tweak the 'cpu_realtime_mask' handling slightly https://review.opendev.org/461456
17:17:08 openstackgerrit Stephen Finucane proposed openstack/nova master: hardware: Allow 'hw:cpu_realtime_mask' to be omitted https://review.opendev.org/737026
17:23:29 bauzas gibi: sorry, I was off my keyboard, and I'll be on PTO from tomorrow to Wednesday morning
19:03:36 openstackgerrit Merged openstack/python-novaclient master: use stevedore to load extensions https://review.opendev.org/739331
21:05:26 openstackgerrit Merged openstack/nova stable/ussuri: hardware: Raise useful error for invalid mempage size https://review.opendev.org/736073
21:07:01 eandersson bauzas if you guys have time could you look at getting https://review.opendev.org/#/c/703116/ merged? would love to push that patch out
21:07:36 eandersson Worst case I can just cherry-pick it, but a lot nicer to just be able to rebase :D
21:11:17 eandersson https://review.opendev.org/#/q/topic:bug/1763766+(status:open)
21:39:08 openstackgerrit Merged openstack/nova stable/train: Guard against missing image cache directory https://review.opendev.org/738455
23:44:05 openstackgerrit Merged openstack/nova stable/queens: Reset the cell cache for database access in Service https://review.opendev.org/720596
#openstack-nova - 2020-07-10
00:24:22 openstackgerrit Wenping Song proposed openstack/nova-specs master: Add no user token when get Cyborg client https://review.opendev.org/740184
00:42:12 openstackgerrit Wenping Song proposed openstack/nova-specs master: Add no user token when get Cyborg client https://review.opendev.org/740184
01:34:08 openstackgerrit Keigo Noha proposed openstack/nova master: Change default num_retries for glance to 3 https://review.opendev.org/740389
03:46:01 openstackgerrit Huaqiang Wang proposed openstack/nova master: objects: Introduce 'pcpuset' field for InstanceNUMACell https://review.opendev.org/714658
03:46:01 openstackgerrit Huaqiang Wang proposed openstack/nova master: test_hardware: code formating for multiline construct https://review.opendev.org/728793
03:46:02 openstackgerrit Huaqiang Wang proposed openstack/nova master: objects: Introduce the 'CPUAllocationPolicy.MIXED' enum https://review.opendev.org/716267
03:46:02 openstackgerrit Huaqiang Wang proposed openstack/nova master: notifications: add the pcpuset info to instance numa cell payload https://review.opendev.org/728480
03:46:03 openstackgerrit Huaqiang Wang proposed openstack/nova master: Validate CPU pinning configuration for mixed instance https://review.opendev.org/716268
03:46:03 openstackgerrit Huaqiang Wang proposed openstack/nova master: libvirt: apply mixed instance CPU policy https://review.opendev.org/728481
03:46:04 openstackgerrit Huaqiang Wang proposed openstack/nova master: scheduler: mixed instance CPU resource translating https://review.opendev.org/728482
03:46:04 openstackgerrit Huaqiang Wang proposed openstack/nova master: Calculate the CPU usage for mixed instance https://review.opendev.org/716269
03:46:05 openstackgerrit Huaqiang Wang proposed openstack/nova master: compute: bump nova-compute service version for mixed instance https://review.opendev.org/716271
03:46:05 openstackgerrit Huaqiang Wang proposed openstack/nova master: flavor: Enable 'hw:cpu_dedicated_mask' for creating a mixed instance https://review.opendev.org/714706
03:46:06 openstackgerrit Huaqiang Wang proposed openstack/nova master: hardware: refactor numa_get_constraints https://review.opendev.org/718294
03:46:06 openstackgerrit Huaqiang Wang proposed openstack/nova master: hardware: create 'mixed' instance for realtime and shared CPUs https://review.opendev.org/728485
03:46:07 openstackgerrit Huaqiang Wang proposed openstack/nova master: release note: add new features introduces by use-pcpu-and-vcpu-in-one-instance https://review.opendev.org/736997
03:46:07 openstackgerrit Huaqiang Wang proposed openstack/nova master: metadata: export the vCPU IDs that are pinning on the host CPUs https://review.opendev.org/688936
05:17:13 logan- hello, is it possible to set cpu_mode in flavor metadata or something similar? ideally i want to use host-model for most things, and set aside some special flavors that use host-passthrough.
05:50:19 huaqiang login-: I think answer is no.
05:50:53 huaqiang we don't have per-instance cpu_model
05:52:39 huaqiang logan-: also curious for what scenario need this kind of per-instance cpu model?
05:57:34 logan- thanks. reason is: some instances need passthrough for nested virt stability, but generally we'd like to avoid running passthrough on all instances because of the live migration limitations etc. I was thinking it'd be ideal to segment by flavor in this situation.
06:41:07 openstackgerrit Kevin Zhao proposed openstack/nova master: [WIP] CI: add tempest-integrated-compute-aarch64 job https://review.opendev.org/714439
07:38:37 openstackgerrit Balazs Gibizer proposed openstack/nova stable/stein: Guard against missing image cache directory https://review.opendev.org/740409
07:39:26 openstackgerrit Balazs Gibizer proposed openstack/nova stable/stein: Guard against missing image cache directory https://review.opendev.org/740409
07:56:27 gibi stephenfin, lyarwood: hi! could you hit this again https://review.opendev.org/#/q/topic:bug/1886418 ? It is now connected to a proper bugreport
08:10:25 openstackgerrit Balazs Gibizer proposed openstack/nova master: Document nova in tree virt drivers https://review.opendev.org/740061
08:11:41 openstackgerrit Elod Illes proposed openstack/nova stable/pike: Fix os_CODENAME detection and repo refresh during ceph tests https://review.opendev.org/740419
08:15:54 openstackgerrit Alexandre Arents proposed openstack/nova master: Rebase qcow2 images when unshelving an instance https://review.opendev.org/696084
08:40:04 openstackgerrit Lee Yarwood proposed openstack/nova stable/queens: Clean up allocation if unshelve fails due to neutron https://review.opendev.org/729540
08:40:35 lyarwood gibi: ack
08:50:57 gibi thanks
09:28:10 stephenfin gibi: comments on https://review.opendev.org/#/c/704865/

Earlier   Later