Thursday, 2026-07-16

opendevreviewTakashi Kajinami proposed openstack/nova master: compute: Validate SEV/SEV-ES support at start up  https://review.opendev.org/c/openstack/nova/+/99434202:54
opendevreviewTakashi Kajinami proposed openstack/nova master: libvirt: Ignore SEV-ES when SEV-SNP is detected  https://review.opendev.org/c/openstack/nova/+/99434302:54
opendevreviewTakashi Kajinami proposed openstack/nova master: libvirt: Detect AMD SEV-SNP support  https://review.opendev.org/c/openstack/nova/+/99476402:55
opendevreviewTakashi Kajinami proposed openstack/nova master: libvirt: Support launchSecurity element for SEV-SNP  https://review.opendev.org/c/openstack/nova/+/99492002:55
opendevreviewTakashi Kajinami proposed openstack/nova master: libvirt: Omit optional fields for SEV/SEV-ES  https://review.opendev.org/c/openstack/nova/+/99492102:56
opendevreviewTakashi Kajinami proposed openstack/nova master: libvirt: Drop sev capability check in launch instance flow  https://review.opendev.org/c/openstack/nova/+/99492902:56
opendevreviewTakashi Kajinami proposed openstack/nova master: AMD SEV: omit iommu='on' for virtio devices  https://review.opendev.org/c/openstack/nova/+/90963502:56
opendevreviewTakashi Kajinami proposed openstack/nova master: libvirt: Launch instances with SEV-SNP memory encryption  https://review.opendev.org/c/openstack/nova/+/99493003:00
opendevreviewTakashi Kajinami proposed openstack/nova master: compute: Validate SEV/SEV-ES support at start up  https://review.opendev.org/c/openstack/nova/+/99434203:02
opendevreviewTakashi Kajinami proposed openstack/nova master: libvirt: Ignore SEV-ES when SEV-SNP is detected  https://review.opendev.org/c/openstack/nova/+/99434303:02
opendevreviewTakashi Kajinami proposed openstack/nova master: libvirt: Detect AMD SEV-SNP support  https://review.opendev.org/c/openstack/nova/+/99476403:03
opendevreviewTakashi Kajinami proposed openstack/nova master: libvirt: Support launchSecurity element for SEV-SNP  https://review.opendev.org/c/openstack/nova/+/99492003:03
opendevreviewTakashi Kajinami proposed openstack/nova master: libvirt: Omit optional fields for SEV/SEV-ES  https://review.opendev.org/c/openstack/nova/+/99492103:03
opendevreviewTakashi Kajinami proposed openstack/nova master: libvirt: Drop sev capability check in launch instance flow  https://review.opendev.org/c/openstack/nova/+/99492903:03
opendevreviewTakashi Kajinami proposed openstack/nova master: AMD SEV: omit iommu='on' for virtio devices  https://review.opendev.org/c/openstack/nova/+/90963503:04
opendevreviewTakashi Kajinami proposed openstack/nova master: libvirt: Launch instances with SEV-SNP memory encryption  https://review.opendev.org/c/openstack/nova/+/99493003:05
opendevreviewTakashi Kajinami proposed openstack/nova master: Update documentations for AMD SEV-SNP support  https://review.opendev.org/c/openstack/nova/+/99509003:05
opendevreviewTakashi Kajinami proposed openstack/nova master: Remove [libvirt] num_memory_encrypted_guests  https://review.opendev.org/c/openstack/nova/+/99512003:05
opendevreviewTakashi Kajinami proposed openstack/nova master: Use domain capabilities to detect SEV-SNP support  https://review.opendev.org/c/openstack/nova/+/99631603:05
opendevreviewTakashi Kajinami proposed openstack/nova master: Remove [DEFAULT] vcpu_pin_set  https://review.opendev.org/c/openstack/nova/+/99601603:44
*** Jeff is now known as Guest1349807:57
*** Guest13498 is now known as JeffYang08:00
opendevreviewTakashi Kajinami proposed openstack/nova master: Remove [workarounds] disable_libvirt_livesnapshot  https://review.opendev.org/c/openstack/nova/+/99749208:27
opendevreviewMerged openstack/nova stable/2025.1: Fix mutating global config in websocket proxy  https://review.opendev.org/c/openstack/nova/+/99595808:27
gibifyi tox cover is timing out a lot recently in test_schemas I filed https://bugs.launchpad.net/nova/+bug/216090109:38
sean-k-mooneygibi: i know there was some slownees with those in the past09:48
sean-k-mooneybut i dont remember exactly why09:48
sean-k-mooneyi just recall mentioning it to stephen09:48
sean-k-mooneygibi:lol we were looking at the same review09:53
tkajinam:-)10:06
opendevreviewBalazs Gibizer proposed openstack/nova master: [cover]Disable slow test_schemas test case  https://review.opendev.org/c/openstack/nova/+/99750310:15
gibiI found no better way given our current cirumstances (py314+threading mode fails)10:16
sean-k-mooneygibi: i think that is fien for now10:17
sean-k-mooneygibi: i think the answer might be to split it into multipel tests10:17
sean-k-mooneycurrently its looping over all our api contoler and then looping over all the route and then testing every schema in one masive test10:20
sean-k-mooneyhttps://github.com/openstack/nova/blob/master/nova/tests/unit/api/openstack/compute/test_schemas.py#L69-L12310:20
sean-k-mooneygibi: the fix i think is to use subtests or ddt to split up the test per contoler or route10:20
sean-k-mooneyso that each schema baseiclly get mapped to its own test10:21
sean-k-mooneythat way we wont hit the per test timeout10:21
sean-k-mooneytkajinam: speaking of masive tests that should be broken up you sev functional test have lots of repatead code that should be in shared helpers and the test shoudl be broken up in to small tests10:26
sean-k-mooneytkajinam: but im not going to block on that for now10:26
sean-k-mooneybut we shoudl clean that up in a dedicated followup10:27
gibisean-k-mooney: yeah that is also an option10:29
sean-k-mooneyim also fine with not runing that in coverage10:30
sean-k-mooneyin this particalar case im not sure that the coverage report will matirally be any better/worse by runing that one test10:31
sean-k-mooneyi just think it might be nice to do that break anyway for the unit tests10:31
sean-k-mooneyits extra slow in cover because of the profiler overhead10:31
sean-k-mooneybut its still the slowest or one of the slowest unit test today10:31
sean-k-mooneytkajinam: i didnt know that sev-snp actully needed sateless firmware?10:47
sean-k-mooneytkajinam: is that a qemu limiation or amd one10:48
sean-k-mooneyok ya that called ou tin the spec i just missed it https://specs.openstack.org/openstack/nova-specs/specs/2026.2/approved/amd-sev-snp-libvirt-support.html#:~:text=Also%20amd%2Dsev%2Dsnp%20requires%20stateless%20firmware%2C%20so%20explicit%20hw%5Ffirmware%5Fstateless%3DTrue%20image%20property%20setting%20is%20also%20required10:51
tkajinamsean-k-mooney, that's the limitation enforced by qemu, and is also needed so that libvirt can pick up the appropriate ovmf file10:59
sean-k-mooneyack11:00
tkajinamsean-k-mooney, yeah I agree the current tests are quite redundant. I'll try refactoring these later11:00
sean-k-mooneyim basiclly +2w on most of your serise now ill add +w to the bottom patch when im done11:01
tkajinamthanks !11:01
sean-k-mooneyim on the last two clean up patchs so we shoudl get this into the ci later today11:01
tkajinam:-D11:02
sean-k-mooneyi have left some comments along the way but nothing was worth holding or respinign the patches for11:02
tkajinamthanks. I'll go through these later and may address these in follow-ups11:07
sean-k-mooneyi think all that needed a recheck are now running https://zuul.openstack.org/status?change=994342 and all the approves are in place so im gong to go get coffee11:22
tkajinam\o/11:39
artpejHi, I submitted a bug (https://bugs.launchpad.net/nova/+bug/2158443) some time ago and proposed a fix (https://review.opendev.org/c/openstack/nova/+/995073). Is there anything I'm missing to get this reviewed and integrated?12:43
gibiartpej: You did the right thing12:47
gibiartpej: both by filing a bug, proposing a patch, and pinging us on IRC12:47
gibiunfortunately we are short staffed so not all patches get reviewed and the ones that gets reviewed might take a long time12:48
gibigmaan: when you are up, what does this schema stacking means? https://github.com/openstack/nova/blob/42fd9b6c8c5da09c92f5fb7efac1f5da52ca7df4/nova/api/openstack/compute/servers.py#L700-L702 Why do we have v20 top of an unversioned schema there?12:53
gibithis is related to the fix from artpej in https://review.opendev.org/c/openstack/nova/+/99507312:53
gibiit almost like that the 2.0 schema is better than the 2.0-2.18 schema. That feel backwards12:59
gibi~2.1-2.1812:59
gibiartpej: at least you found something strange :)12:59
artpejYes, my users are weird people who put spaces at the beginning of security group names :)13:01
gibithat is weird too :)13:05
gibibut the schema handling in nova for that call is weird as well13:06
gibihence my ping to gmaan 13:06
gibiwe should not have an API where 2.0 accepts "  sg  " but 2.1-2.18 does not. But 2.0 was a bit special I assume compared to 2.1. Still I feel like we might be able to say lets fix this in 2.1-2.18. 13:07
gibihm13:08
gibieven more weird13:08
gibiit might be that only 2.0 allows leading spaces there and nothing after 2.113:08
gibistill we have a strong case, if neutron allows leading spaces then we should not stop it13:09
opendevreviewMerged openstack/nova master: Do not try to return empty attachment_id from API  https://review.opendev.org/c/openstack/nova/+/99675713:35
opendevreviewMerged openstack/nova master: compute: Validate SEV/SEV-ES support at start up  https://review.opendev.org/c/openstack/nova/+/99434213:41
opendevreviewMerged openstack/nova master: libvirt: Ignore SEV-ES when SEV-SNP is detected  https://review.opendev.org/c/openstack/nova/+/99434313:41
tkajinamsean-k-mooney, could you also check https://review.opendev.org/c/openstack/nova-specs/+/996394 and https://review.opendev.org/c/openstack/nova-specs/+/996395 when you have time ?14:17
tkajinamthese two update the sev-snp spec according to the latest implementation14:18
opendevreviewTakashi Kajinami proposed openstack/nova master: Do not report SEV capability for non kvm virt_type  https://review.opendev.org/c/openstack/nova/+/99649814:28
sean-k-mooneytkajinam: bothg of those look good to me as well14:55
tkajinamsean-k-mooney, thanks !14:58
opendevreviewKamil Sambor proposed openstack/nova master: Replace StaticallyDelayingWrapper with futurist DelayingExecutor  https://review.opendev.org/c/openstack/nova/+/99741014:59
opendevreviewMerged openstack/nova-specs master: Fix SEV-SNP feature detection in start up  https://review.opendev.org/c/openstack/nova-specs/+/99639415:11
opendevreviewMerged openstack/nova-specs master: Use domain capability to detect SEV-SNP support  https://review.opendev.org/c/openstack/nova-specs/+/99639515:13
gmaangibi: artpej: that seems strange as v2.0 is less strict than >=v2.1 and not other way. anyways I will check it today  16:00
gibigmaan: thanks16:13
opendevreviewBalazs Gibizer proposed openstack/placement master: Check overall capacity before generating candidates  https://review.opendev.org/c/openstack/placement/+/99761716:35
gibisean-k-mooney: it is not terrible to check the flattened capacity in python ^^16:36
sean-k-mooneycool16:36
sean-k-mooneyi mean if we get it workign we can make it faster later16:36
gibiit works and it is not super ugly as I had all the data available in the right place, it is just a bit of data juggling to summarize and compare16:39
sean-k-mooneyack, on a call but ill look after16:39
gibithanks16:40
sean-k-mooneyif its "good enough" then we can start there and defer doign it in sql till later16:40
sean-k-mooneyoh your system how long these it take i assume you updated the coments in the tests?16:41
gibiit is so fast that I bumped N to 64 and removed the comments as it runs in sub seconds16:42
gibiit fully eliminates the hot path16:43
gibiI also added a test that shows the the extra check is not too costly for the non edge case16:43
gibi{3} placement.tests.functional.test_allocation_candidates.TestAlmostFittingRoot.test_one_less_device_than_requested_no_wide_tree_optimization [0.231623s] ... ok16:44
gibithis is with 64 RP 63 request16:44
gibiI had enough for today :) See you folks tomorrow16:52
sean-k-mooneygibi: cool enjoy your evenigng17:19
sean-k-mooneyjust finished the calls so ill take a look shortly17:20
opendevreviewTakashi Kajinami proposed openstack/nova master: Refactor CPU traits functional tests for AMD SEV support  https://review.opendev.org/c/openstack/nova/+/99763518:40
opendevreviewTakashi Kajinami proposed openstack/nova master: Refactor CPU traits functional tests for AMD SEV support  https://review.opendev.org/c/openstack/nova/+/99763519:10
gmaanartpej: gibi RE SG name with leading/trailing whitespace, yes it is changed/disallowed with v2.1 strict validation on resource name. But agree that it does not make sense for Nova to block it when neuston allow. I added comment in bug as well as gerrit.19:12
gmaanonly things is we need to fix this for higher microversion and should not change older microvesion behavior even this is regression. 19:13
opendevreviewTakashi Kajinami proposed openstack/nova master: Refactor CPU traits functional tests for AMD SEV support  https://review.opendev.org/c/openstack/nova/+/99763519:14
opendevreviewTakashi Kajinami proposed openstack/nova master: Refactor CPU traits functional tests for AMD SEV support  https://review.opendev.org/c/openstack/nova/+/99763519:33
tkajinam-889 +234 ...19:34
opendevreviewSylvain Desgrais proposed openstack/nova master: Allow security group names with leading/trailing spaces  https://review.opendev.org/c/openstack/nova/+/99507319:57

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!