Tuesday, 2019-03-05

*** ttsiouts has joined #openstack-nova00:07
*** itlinux has joined #openstack-nova00:09
*** sdake has joined #openstack-nova00:09
*** sdake has quit IRC00:12
*** itlinux_ has joined #openstack-nova00:13
*** tetsuro has joined #openstack-nova00:13
*** sdake has joined #openstack-nova00:13
*** itlinux has quit IRC00:15
*** wolverineav has joined #openstack-nova00:16
*** _alastor_ has quit IRC00:17
*** lchabert has quit IRC00:19
*** lchabert has joined #openstack-nova00:21
*** _alastor_ has joined #openstack-nova00:23
*** lbragstad has quit IRC00:26
*** _alastor_ has quit IRC00:27
*** ttsiouts has quit IRC00:29
*** sdake has quit IRC00:41
*** sapd1 has quit IRC00:42
*** wolverineav has quit IRC00:47
*** markvoelker has joined #openstack-nova00:48
*** ericyoung has quit IRC00:50
*** ericyoung has joined #openstack-nova00:51
*** wxy-xiyuan has joined #openstack-nova00:53
*** rchurch has joined #openstack-nova00:54
*** openstackgerrit has joined #openstack-nova00:54
openstackgerritBrin Zhang proposed openstack/nova-specs master: Support delete_on_termination in volume attach api  https://review.openstack.org/61294900:54
*** rchurch_ has quit IRC00:55
*** Sundar has joined #openstack-nova00:56
openstackgerritBrin Zhang proposed openstack/nova-specs master: Support for changing deleted_on_termination after boot  https://review.openstack.org/58033600:58
*** wolverineav has joined #openstack-nova00:58
*** macza has joined #openstack-nova00:59
openstackgerritBrin Zhang proposed openstack/nova-specs master: Support for changing deleted_on_termination after boot  https://review.openstack.org/58033600:59
*** ileixe has joined #openstack-nova01:00
openstackgerritMerged openstack/nova stable/rocky: Avoid BadRequest error log on volume attachment  https://review.openstack.org/64011001:01
*** macza has quit IRC01:03
openstackgerritBrin Zhang proposed openstack/nova-specs master: Support delete_on_termination in volume attach api  https://review.openstack.org/61294901:05
*** sdake has joined #openstack-nova01:08
*** rchurch has quit IRC01:10
*** rchurch has joined #openstack-nova01:11
*** _alastor_ has joined #openstack-nova01:12
*** _alastor_ has quit IRC01:17
openstackgerritBrin Zhang proposed openstack/nova master: Remove the string check of the flavor attribute 'swap'  https://review.openstack.org/63901201:20
*** markvoelker has quit IRC01:22
*** brinzhang has joined #openstack-nova01:22
openstackgerritYongli He proposed openstack/nova master: Add server sub-resource topology API  https://review.openstack.org/62147601:28
*** macza has joined #openstack-nova01:35
*** dpawlik has joined #openstack-nova01:37
yongliheeveryone seems tired, take care, just update the patch. jaypipe, forgive my bad English.   ping alex_xu,  you must be energized.01:38
*** macza has quit IRC01:39
*** ileixe has quit IRC01:40
*** dpawlik has quit IRC01:42
*** dave-mccowan has joined #openstack-nova01:47
*** gouthamr has quit IRC01:48
*** ileixe has joined #openstack-nova01:51
*** ileixe has quit IRC01:52
*** sapd1 has joined #openstack-nova01:52
*** ileixe has joined #openstack-nova01:53
*** bhagyashris has joined #openstack-nova01:54
*** wolverineav has quit IRC01:57
*** whoami-rajat has joined #openstack-nova02:02
openstackgerritLuyao Zhong proposed openstack/nova master: object: Add pmem_namespaces field to the NUMACell obj  https://review.openstack.org/63454702:05
openstackgerritLuyao Zhong proposed openstack/nova master: object: Add virtual_pmems fields to the InstanceNUMACell obj  https://review.openstack.org/63454802:05
openstackgerritLuyao Zhong proposed openstack/nova master: libvirt: Enable driver configures PMEM namespace when initiating libvirt driver  https://review.openstack.org/63454902:05
openstackgerritLuyao Zhong proposed openstack/nova master: libvirt: Update PMEM namespaces info and usage  https://review.openstack.org/63455002:05
*** sdake has quit IRC02:05
openstackgerritLuyao Zhong proposed openstack/nova master: libvirt: report pmem namespaces resources by provider tree  https://review.openstack.org/63455102:05
openstackgerritLuyao Zhong proposed openstack/nova master: API: parse pmem related flavor extra spec  https://review.openstack.org/63455202:05
openstackgerritLuyao Zhong proposed openstack/nova master: scheduler: translate virtual pmems request to placement request group  https://review.openstack.org/63455302:05
openstackgerritLuyao Zhong proposed openstack/nova master: update _numa_fit_instance_cell for the support of virtual_pmems  https://review.openstack.org/63455402:05
openstackgerritLuyao Zhong proposed openstack/nova master: libvirt: boot VM with vpmems and cleanup vpmems when destroying  https://review.openstack.org/63455502:05
openstackgerritLuyao Zhong proposed openstack/nova master: support VM resizing with vpmem data migration  https://review.openstack.org/63455602:05
*** macza has joined #openstack-nova02:11
*** hongbin has joined #openstack-nova02:13
*** macza has quit IRC02:15
*** markvoelker has joined #openstack-nova02:19
*** sdake has joined #openstack-nova02:27
openstackgerritMerged openstack/nova master: Use a placement conf when testing report client  https://review.openstack.org/64085302:27
*** ircuser-1 has joined #openstack-nova02:28
*** takashin has joined #openstack-nova02:30
*** wolverineav has joined #openstack-nova02:33
*** s1061123 has quit IRC02:34
*** s1061123 has joined #openstack-nova02:37
*** sapd1 has quit IRC02:37
*** psachin has joined #openstack-nova02:39
*** gouthamr has joined #openstack-nova02:50
*** markvoelker has quit IRC02:51
*** wolverineav has quit IRC03:00
*** wolverineav has joined #openstack-nova03:01
*** wolverineav has quit IRC03:06
*** itlinux_ has quit IRC03:06
*** itlinux has joined #openstack-nova03:12
*** alanmeadows has left #openstack-nova03:21
*** itlinux has quit IRC03:21
openstackgerritFan Zhang proposed openstack/nova master: Retry after hitting libvirt error VIR_ERR_OPERATION_INVALID in live migration.  https://review.openstack.org/61227203:23
*** itlinux has joined #openstack-nova03:25
*** wolverineav has joined #openstack-nova03:31
*** tbachman has quit IRC03:33
*** dpawlik has joined #openstack-nova03:38
*** dave-mccowan has quit IRC03:42
*** dpawlik has quit IRC03:43
*** markvoelker has joined #openstack-nova03:48
*** sdake has quit IRC03:51
*** gyee has quit IRC03:54
*** wolverineav has quit IRC04:07
*** udesale has joined #openstack-nova04:18
*** markvoelker has quit IRC04:22
*** sridharg has joined #openstack-nova04:41
*** owalsh_ has joined #openstack-nova04:50
*** owalsh has quit IRC04:51
*** wolverineav has joined #openstack-nova05:00
*** wolverineav has quit IRC05:06
*** janki has joined #openstack-nova05:08
*** hongbin has quit IRC05:09
*** markvoelker has joined #openstack-nova05:19
*** udesale has quit IRC05:22
*** ratailor has joined #openstack-nova05:23
*** dpawlik has joined #openstack-nova05:39
*** dpawlik has quit IRC05:44
*** sdake has joined #openstack-nova05:47
*** jhesketh has quit IRC05:47
*** jhesketh has joined #openstack-nova05:48
*** sdake has quit IRC05:50
*** markvoelker has quit IRC05:51
*** pcaruana has joined #openstack-nova05:52
openstackgerritLajos Katona proposed openstack/python-novaclient master: Add support for microversion v2.72  https://review.openstack.org/63723405:55
*** sdake has joined #openstack-nova05:57
*** pcaruana has quit IRC06:07
*** _alastor_ has joined #openstack-nova06:08
*** _alastor_ has quit IRC06:12
*** dpawlik has joined #openstack-nova06:19
*** ccamacho has quit IRC06:21
*** dpawlik has quit IRC06:24
*** dims has quit IRC06:24
*** dims has joined #openstack-nova06:26
*** masber has joined #openstack-nova06:28
masbergood afternoon, I am trying to setup OVS offload but nova fails getting representor port for PF. The question is: How is the representor created?06:29
*** itlinux has quit IRC06:34
*** dims has quit IRC06:36
*** dims has joined #openstack-nova06:37
*** markvoelker has joined #openstack-nova06:48
*** wolverineav has joined #openstack-nova06:48
*** Luzi has joined #openstack-nova06:51
*** wolverineav has quit IRC06:53
*** sridharg has quit IRC06:57
*** sridharg has joined #openstack-nova07:02
*** mdbooth has joined #openstack-nova07:06
*** yumapath has joined #openstack-nova07:08
yumapathHi team07:08
yumapathwhile running the CI we are seeing one issue in the master devstack installation07:08
yumapath2019-03-04 23:54:21.262 | + functions:wait_for_compute:465           :   openstack --os-cloud devstack-admin --os-region RegionOne compute service list07:08
yumapaththis command is timing out07:09
*** mdbooth_ has quit IRC07:09
yumapath    :   echo 'Didn'\''t find service registered by hostname after 60 seconds' 2019-03-04 23:54:21.257 | Didn't find service registered by hostname after 60 seconds07:09
yumapathwith this error07:09
yumapaththe stable releases are all passing07:09
yumapaththe issue is seen only with master07:09
yumapathdid  anything change in the master to cause this issue07:09
*** slaweq has joined #openstack-nova07:11
*** wolverineav has joined #openstack-nova07:14
*** wolverineav has quit IRC07:14
*** markvoelker has quit IRC07:22
*** brinzhang has quit IRC07:25
*** dpawlik has joined #openstack-nova07:25
*** brinzhang has joined #openstack-nova07:25
*** stakeda has joined #openstack-nova07:30
openstackgerritYongli He proposed openstack/nova master: Add server sub-resource topology API  https://review.openstack.org/62147607:30
*** takashin has left #openstack-nova07:32
*** ralonsoh has joined #openstack-nova07:34
*** udesale has joined #openstack-nova07:51
openstackgerritmelanie witt proposed openstack/nova master: Add user_id column to the instance_mappings table  https://review.openstack.org/63334907:53
openstackgerritmelanie witt proposed openstack/nova master: Add user_id field to InstanceMapping  https://review.openstack.org/63335007:53
openstackgerritmelanie witt proposed openstack/nova master: Populate InstanceMapping.user_id during migrations and schedules  https://review.openstack.org/63857407:53
openstackgerritmelanie witt proposed openstack/nova master: Add online data migration for populating user_id  https://review.openstack.org/63335107:53
openstackgerritmelanie witt proposed openstack/nova master: Add get_counts() to InstanceMappingList  https://review.openstack.org/63807207:53
openstackgerritmelanie witt proposed openstack/nova master: WIP Count instances from mappings and cores/ram from placement  https://review.openstack.org/63807307:53
openstackgerritmelanie witt proposed openstack/nova master: Use instance mappings to count server group members  https://review.openstack.org/63832407:53
*** udesale has quit IRC08:01
*** cfriesen has quit IRC08:07
*** tesseract has joined #openstack-nova08:09
*** Sundar has quit IRC08:11
*** tssurya has joined #openstack-nova08:16
*** luksky has joined #openstack-nova08:16
*** awalende has joined #openstack-nova08:16
*** sapd1 has joined #openstack-nova08:16
*** rpittau|afk is now known as rpittau08:16
*** markvoelker has joined #openstack-nova08:18
*** pcaruana has joined #openstack-nova08:18
*** tkajinam has quit IRC08:22
*** owalsh_ is now known as owalsh08:24
*** pcaruana has quit IRC08:25
*** helenfm has joined #openstack-nova08:29
*** ccamacho has joined #openstack-nova08:29
*** yumapath has quit IRC08:31
*** sdake has quit IRC08:36
*** pcaruana has joined #openstack-nova08:37
*** priteau has joined #openstack-nova08:42
*** sajauddin has joined #openstack-nova08:43
*** pcaruana has quit IRC08:44
openstackgerritBoxiang Zhu proposed openstack/nova master: Raise error when force_raw_images and images_type are incompatible  https://review.openstack.org/64027108:47
kashyapgibi: When you can, please put this one through its misery :-) -- https://review.openstack.org/#/c/637578/108:48
kashyap(You've looked at it before.)08:49
*** markvoelker has quit IRC08:51
*** ttsiouts has joined #openstack-nova08:52
openstackgerritMerged openstack/nova master: Improve existing flavor and image metadata validation  https://review.openstack.org/64084108:57
*** pcaruana has joined #openstack-nova09:01
*** macza has joined #openstack-nova09:01
openstackgerritlicanwei proposed openstack/nova master: api-ref: typo service.disable_reason  https://review.openstack.org/64095609:01
openstackgerritKashyap Chamarthy proposed openstack/nova master: qemu: Make disk image conversion dramatically faster  https://review.openstack.org/64078109:03
*** jding1_ has joined #openstack-nova09:04
*** macza has quit IRC09:05
*** ttsiouts has quit IRC09:05
*** ttsiouts has joined #openstack-nova09:06
*** rha has quit IRC09:07
*** jackding has quit IRC09:07
*** rha has joined #openstack-nova09:07
*** phasespace has joined #openstack-nova09:08
*** moshele has joined #openstack-nova09:09
*** ttsiouts has quit IRC09:10
openstackgerritSilvan Kaiser proposed openstack/nova master: Exec systemd-run without --user flag in Quobyte driver  https://review.openstack.org/55419509:11
*** ttsiouts has joined #openstack-nova09:12
awalendeIs there a way to deny instances who have special characters in their hostnames like "periods" or something? It messes with internal dns09:12
*** panda|ruck|off is now known as panda|ruck|flu09:13
*** cdent has joined #openstack-nova09:15
aspierscdent: thanks for sorting that so quickly09:21
cdentaspiers: well I was the trigger for the problem, so...09:21
*** dtantsur|afk is now known as dtantsur09:22
*** IvensZambrano has joined #openstack-nova09:31
*** stakeda has quit IRC09:31
kashyapaspiers: If you want a small change in scenery: https://review.openstack.org/640781 :-)09:32
*** derekh has joined #openstack-nova09:36
*** mdbooth has quit IRC09:38
*** bhagyashris has quit IRC09:38
*** mdbooth has joined #openstack-nova09:39
openstackgerritKashyap Chamarthy proposed openstack/nova-specs master: cpu-model-selection: s/cpu_models/cpu_model_list/  https://review.openstack.org/64086609:41
openstackgerritBoxiang Zhu proposed openstack/nova master: Raise error when force_raw_images and images_type are incompatible  https://review.openstack.org/64027109:46
*** markvoelker has joined #openstack-nova09:48
openstackgerritMerged openstack/nova master: Convert driver supported capabilities to compute node provider traits  https://review.openstack.org/53849809:50
*** moshele has quit IRC09:52
*** sdake has joined #openstack-nova09:56
aspiers\o/ capabilities patch landed \o/09:58
aspierskashyap: looking09:58
kashyapaspiers: It's a one-worder, but the (long) commit message tells the story09:59
kashyapI'm off from noon; so check back in tonight / tomm.09:59
*** sajauddin has quit IRC10:02
aspierskashyap: updated https://review.openstack.org/#/c/640483/ but just realised I didn't address Artom's feedback on the commit message10:04
openstackgerritAdam Spiers proposed openstack/nova master: Parse <emulator> elements from virConnectGetCapabilities()  https://review.openstack.org/64048310:09
*** _alastor_ has joined #openstack-nova10:09
openstackgerritSylvain Bauza proposed openstack/nova master: Refactor CONTRIBUTING.rst  https://review.openstack.org/64097010:09
aspierskashyap: addressed in PS310:10
*** mdbooth has quit IRC10:10
bauzasstephenfin: gibi: I just wrote https://review.openstack.org/640970 based on some good point from ttx10:11
bauzasfeedback welcome10:12
*** _alastor_ has quit IRC10:14
gibibauzas, kashyap: ack for both patches I will try to look at both after lunch10:14
bauzasgibi: it's not a crucial thing10:15
*** mdbooth has joined #openstack-nova10:18
openstackgerritMerged openstack/nova master: Make move_allocations handle empty source allocations  https://review.openstack.org/63641210:20
*** markvoelker has quit IRC10:22
*** luksky has quit IRC10:22
*** tetsuro has quit IRC10:25
*** tetsuro has joined #openstack-nova10:29
openstackgerritBoxiang Zhu proposed openstack/nova master: Raise error when force_raw_images and images_type are incompatible  https://review.openstack.org/64027110:40
openstackgerritStephen Finucane proposed openstack/nova master: fakelibvirt: Add ability to generate fake PCI devices  https://review.openstack.org/64040910:44
openstackgerritStephen Finucane proposed openstack/nova master: tests: Create PCI tests for NUMA'y tests  https://review.openstack.org/64082610:44
*** erlon has joined #openstack-nova10:45
*** sdake has quit IRC10:51
*** macza has joined #openstack-nova10:51
*** sdake has joined #openstack-nova10:54
*** macza has quit IRC10:55
*** luksky has joined #openstack-nova11:02
*** markvoelker has joined #openstack-nova11:19
*** ttsiouts has quit IRC11:19
*** ttsiouts has joined #openstack-nova11:20
*** janki has quit IRC11:22
*** ttsiouts has quit IRC11:24
openstackgerrithuanhongda proposed openstack/nova stable/rocky: Fix resource tracker updates during instance evacuation  https://review.openstack.org/64098211:24
*** sapd1 has quit IRC11:29
openstackgerritMerged openstack/nova master: Fix wrong consumer type in logging  https://review.openstack.org/64072311:34
*** tetsuro has quit IRC11:40
openstackgerritBalazs Gibizer proposed openstack/nova master: Ensure that bandwidth and VF are from the same PF  https://review.openstack.org/62354311:49
openstackgerritBalazs Gibizer proposed openstack/nova master: Support server create with ports having resource request  https://review.openstack.org/63636011:49
*** markvoelker has quit IRC11:51
gibistephenfin, bauzas, sean-k-mooney: I've fixed / answered the concerns in https://review.openstack.org/62354311:51
sean-k-mooneygibi: thanks. just going to make tea and il review after. i was more or less happy with it before so i expect it will be fine11:54
gibisean-k-mooney: thanks11:56
stephenfinsean-k-mooney: Do you know anything about neutron's proxying of the nova metadata service?11:57
* stephenfin tried updating the config drive docs and has ended up down a rabbit hole of neutron/nova-net madness11:57
sean-k-mooneyyes11:59
sean-k-mooneyit can do it in two ways11:59
sean-k-mooneyeither via the dhcp agent or the l3 agent11:59
sean-k-mooneythe default is to do it via the l3 agent by seting up a dnat rule in the router namespace to proxy the query to the neutron api12:00
*** udesale has joined #openstack-nova12:00
*** ttsiouts has joined #openstack-nova12:01
stephenfinsean-k-mooney: and what determines which way is used? I see references to neutron-metadata-agent, but I'm not sure how that fits in12:01
sean-k-mooneyif you have an isolated network(one that is not connect to an extrenal network via routers) then you use the dhcp version which used a dhcp option field to pass a route and it does the same dnat in the dhcp namespace12:01
stephenfinsean-k-mooney: or whether nova-api-metadata or nova-api is used in this scenario12:01
sean-k-mooneystephenfin: its determind by 2 or 3 config optiosn let me get them12:05
sean-k-mooneythese in the dhcp agent https://github.com/openstack/neutron/blob/master/neutron/conf/agent/dhcp.py#L41-L6712:06
sean-k-mooneyand these in the l3 agent https://github.com/openstack/neutron/blob/master/neutron/conf/agent/l3/config.py#L49-L51 https://github.com/openstack/neutron/blob/master/neutron/conf/agent/l3/config.py#L92-L9812:07
sean-k-mooneyso by default its enabled in the l3 agent and disabled in the dhcp agent12:08
*** zhubx has quit IRC12:09
*** zhubx has joined #openstack-nova12:10
sean-k-mooneystephenfin: as far as im aware the neutron metadata agent is the thing the proxies the requrest to the nova api and the l3/dhcp agent just nat the requests to the metadata agent12:11
sean-k-mooneyso the metadata agent is what actully has the  address fo the nova metadata endpong https://github.com/openstack/neutron/blob/a388701ddfe628e9a5bd16a78422164799b11ef8/neutron/conf/agent/metadata/config.py#L47-L5312:12
*** ratailor has quit IRC12:13
*** ociuhandu has quit IRC12:14
*** mmethot has quit IRC12:15
*** awalende has quit IRC12:15
*** awalende has joined #openstack-nova12:16
*** ccamacho has quit IRC12:17
*** dave-mccowan has joined #openstack-nova12:20
*** xek has joined #openstack-nova12:27
openstackgerritChris Dent proposed openstack/nova master: Correct lower-constraints.txt and the related tox job  https://review.openstack.org/62297212:29
*** sdake has quit IRC12:39
*** mriedem has joined #openstack-nova12:41
* cdent is eagerly looking forward to when we can dump python2.x-3.512:48
*** tosky has joined #openstack-nova12:48
*** brinzhang has quit IRC12:49
*** markvoelker has joined #openstack-nova12:49
sean-k-mooneygibi: so when i was talking about macvtap i was refering to sriov in macvtap mode not the macvtap agent12:49
sean-k-mooneygibi: so if you want to say we support sriov minium bandwith we should test both modes not just direct12:50
sean-k-mooneyi guess technically we shoudl test direct_physical to but in that case we are alloccatign the pf to the guest so it gets all the bandwith12:51
sean-k-mooneyanyway i dont want to block the patch on this point but currently it looks like you have only been looking a sriov in direct mode12:51
sean-k-mooneyfor vnic_type=direct_physical we shoudl reserve the remaing bandwtih inventory for the pf as no other instances can use it in addtion to the normal allocation12:55
sean-k-mooneytechncially we should do that in some case for vnic_type=direct too as vnic_type=direct is allowed to give you a pf12:56
sean-k-mooneydirect_pysical was just intoduced to froce the alloction of a pf12:56
gibisean-k-mooney: regarding macvtap, if this is done by the sriov agent then I can create an extra functional test in nova to see if the nova side works12:58
sean-k-mooneyyep its done basically identically to direct jsut the vnic_type will be macvtap12:59
gibisean-k-mooney: yes12:59
gibisean-k-mooney: I will add a functional test for it in a separate patch12:59
sean-k-mooneyi basically jsut want to confrim that you dont have a hardcoded check for direct somewhere13:00
sean-k-mooneycool13:00
gibisean-k-mooney: regarding allocating a full pf for vnic_type=direct. When does that happen?13:00
sean-k-mooneyit happens if the pf does not have any vfs13:00
sean-k-mooneyof if the admin whitelist pf and never allocates vfs then vnic_type=direct is allowed to used them13:01
sean-k-mooneyit was how you did pf passthough before we added direct_physical13:01
*** mchlumsky has joined #openstack-nova13:01
sean-k-mooneyi can double check if we prevented that when we added direct_physical but i doubt it as it would have been a breaking  api change13:01
gibisean-k-mooney: OK. in this case the first boot will get the whole PF, and the second such boot will fail the pci claim as there is no pci device left13:02
sean-k-mooneyyep13:02
sean-k-mooneyso it wont break anything but it would be nice to tell placment that the rest of the inventory is reserved/unsuable13:03
gibisean-k-mooney: so the problem is that placement doesn't track PF inventory to signal that the PF is consumed13:03
*** phasespace has quit IRC13:03
sean-k-mooneybasically yes13:03
*** panda|ruck|flu is now known as panda|ruck|lunch13:03
sean-k-mooneyideally we would be asing for two things. 1 pf/vf resouce + x bandwith13:03
sean-k-mooneysince we dont track PFs/VFs in placment yet we only ask for the bandwith13:04
gibisean-k-mooney: OK. as far as I understand the admin needs to configur such PF specially (whitelsit the PF but not the VFs) so we can also ask the admin not to configure bandwidth for that PF as that PF will be consumed in a single piece13:04
sean-k-mooneyso this will get fix in train hopefully13:04
sean-k-mooneyya they have to expclity whitelist the pf and that would be resonable intally at least13:05
gibisean-k-mooney: It will be fun moving the VF inventory to placement13:05
gibisean-k-mooney: OK then we will handle this case in the documentation for Stein13:05
sean-k-mooneythe downside is we wont be able to sue the min bandwith request to select pf with at least X bandwith13:05
artom~o~13:05
sean-k-mooneye.g. ask for a 10G pf vs a 1G or 40G or 100G pf13:06
sean-k-mooneygibi: and yes moving vfs/pfs to placement will be "fun"13:06
gibisean-k-mooney: yeah in a mixed environment it will be a limitation13:07
sean-k-mooneyits not somthing they could do before so i think its fine13:07
gibiyeah I feel the same13:08
sean-k-mooneyit would just be a nice bonus if we could enable that usecase in the future13:08
sean-k-mooneybut to do that we need to get a few other things in place13:08
mdboothlyarwood: You there? Still looking at your cinder tempest failure.13:08
mdboothIt's pretty weird. Have you had any insight?13:09
gibisean-k-mooney: totally agree13:09
sean-k-mooneyanyway i responded to your replys on patchset 39 and im now going through patchset 40 but honestly i think future change can likely be done in followup patches as part fo train.13:10
gibisean-k-mooney: and now I've responded in your responses in PS 39 :)13:13
*** janki has joined #openstack-nova13:17
*** sdake has joined #openstack-nova13:19
lyarwoodmdbooth: no, are you looking at the latest failures?13:19
lyarwoodmdbooth: I was waiting for CI to report back before taking another look13:19
mdboothlyarwood: I've been looking into the failures from yesterday or the day before?13:20
mdboothI'm currently suspecting a cinder race, tbh.13:21
mdboothRather than a tempest race, that is.13:21
*** markvoelker has quit IRC13:21
lyarwoodmdbooth: ack, I updated both changes this morning and they appear to be failing in the check queue. Once they report back I'll take a look to see if it's the same issue.13:23
mdboothlyarwood: I'll have a look at the new ones. I know what I'm looking for, now.13:26
mdboothlyarwood: In the one I've been looking into, we can see in both the tempest and c-api logs that cinder returns a 404 for the volume, then fails to delete the type because it still has a volume.13:27
mdboothI couldn't see any other volume in the logs, so my guess is that the cinder REST api is returning a 404 before updating whatever triggers the failure to delete the type.13:28
sean-k-mooneygibi: oh i didnt know rodolfo had revied the minium bandwith work for ovs. it was merged in the past and then reverted as we found edgecase that did not work. i guess he has figured out how to work aroudn them good to know13:29
mdboothWe can see in the c-vol logs that the volume deletion actually completes *after* the type deletion starts, so c-api is returning a 404 *before* the deletion completes.13:29
bauzasgibi: ack, will review it later13:29
*** _alastor_ has joined #openstack-nova13:30
*** dtantsur is now known as dtantsur|brb13:31
gibisean-k-mooney: yeah it is good that rodolfo managed to do that13:34
gibibauzas: thanks13:34
*** _alastor_ has quit IRC13:35
*** masber has quit IRC13:35
sean-k-mooneyit looks like tunneled netowrk support is still pendign but that was one of the things that did not work last time that ended up with the feature beign reverted13:35
sean-k-mooneyanyway having it work for flat and vlan networks i still a good step forward13:36
*** oanson has joined #openstack-nova13:36
openstackgerritChris Dent proposed openstack/nova master: Correct lower-constraints.txt and the related tox job  https://review.openstack.org/62297213:38
*** ivve has joined #openstack-nova13:38
*** marst has joined #openstack-nova13:38
mdboothlyarwood: It has failed again. Investigating.13:40
*** priteau has quit IRC13:41
*** sapd1 has joined #openstack-nova13:43
*** dpawlik has quit IRC13:43
openstackgerritBalazs Gibizer proposed openstack/nova master: Use Selection object to fill request group mapping  https://review.openstack.org/63871113:43
lyarwoodmdbooth: ack thanks13:43
mdboothIt seems to be the py3 version which fails13:47
*** sapd1 has quit IRC13:47
*** panda|ruck|lunch is now known as panda|ruck|flu13:50
*** mchlumsky has quit IRC13:51
*** mchlumsky has joined #openstack-nova13:53
*** priteau has joined #openstack-nova13:53
*** ratailor has joined #openstack-nova13:56
*** marst has quit IRC13:56
*** mmethot has joined #openstack-nova14:05
*** lbragstad has joined #openstack-nova14:06
cdentmriedem: the openstack-dev archives are still there, but not longer reachable by the usual paths: http://lists.openstack.org/pipermail/openstack-dev/14:08
openstackgerritMatt Riedemann proposed openstack/nova master: Handle templated cell mappings in nova-status  https://review.openstack.org/64102114:10
mriedemlyarwood: try throwing that as a dependency in your tripleo change ^; dansmith fyi ^14:10
mriedemcdent: google failed me14:10
cdentaye, it often does14:10
*** sdake has quit IRC14:11
*** janki has quit IRC14:12
*** janki has joined #openstack-nova14:12
lyarwoodmriedem: ack thanks14:13
*** dklyle has joined #openstack-nova14:13
*** dpawlik has joined #openstack-nova14:13
yonglihemriedem:  what can i do for https://review.openstack.org/#/c/621476/3914:18
*** sapd1 has joined #openstack-nova14:18
*** markvoelker has joined #openstack-nova14:18
*** sdake has joined #openstack-nova14:21
*** ttsiouts has quit IRC14:21
*** ttsiouts has joined #openstack-nova14:22
*** sdake has quit IRC14:23
*** ttsiouts has quit IRC14:25
*** ttsiouts has joined #openstack-nova14:25
dansmithmriedem: ah, dangit14:27
openstackgerritTim Rozet proposed openstack/nova stable/rocky: Fixes race condition with privsep utime  https://review.openstack.org/64102314:27
*** marst has joined #openstack-nova14:29
HD|Laptophey14:29
HD|Laptopyesterday I asked if it was possible for nova to use cinder as backing storage for images/instances... now I had an idea14:29
HD|Laptopcan I just go ahead and mount /var/lib/nova/instances as a NFS directory?14:29
HD|Laptopthat would serve the same purpose14:30
HD|Laptopeven better would be if I could share instances/_base, but I guess that this could lead to locking issues if e.g. two compute nodes try to convert the same vmdk image to qcow2?14:30
openstackgerritMerged openstack/nova master: libvirt: Omit needless check on 'CONF.serial_console'  https://review.openstack.org/63757814:30
*** sdake has joined #openstack-nova14:35
mriedemHD|Laptop: yes you can put the instances_path config on shared storage, many people do14:39
mriedemnote that has it's own perils, like https://review.openstack.org/64102314:39
mriedemand https://review.openstack.org/#/c/633042/14:39
mriedemhttps://docs.openstack.org/nova/latest/configuration/config.html#workarounds.ensure_libvirt_rbd_instance_dir_cleanup14:40
*** dtantsur|brb is now known as dtantsur14:40
efriedmriedem: Review the "stop using PlacementDirect" changes with whitespace ignores on in gerrit14:40
efriedmriedem, cdent, sean-k-mooney: but let me rebase it first...14:41
mriedemwhat if i don't want to?14:41
efriedmriedem: Then it's a bigger review that will make you sad14:41
efriedoh, what if you don't want to wait until I've rebased? Sure, then you can +2 it now, I don't mind.14:42
mriedemstrike 214:42
openstackgerritEric Fried proposed openstack/nova master: Stop using PlacementDirect  https://review.openstack.org/64088714:45
efriedcdent, mriedem, sean-k-mooney: ^14:45
cdentone moment please caller14:46
HD|Laptopmriedem: thanks. I'll just give every instance its own nfs base directory then, that should keep me safe from that shit14:46
openstackgerritEric Fried proposed openstack/nova master: Stop using PlacementDirect  https://review.openstack.org/64088714:46
efried^ fix whitespace14:47
*** hongbin has joined #openstack-nova14:47
*** sdake has quit IRC14:48
efriedcdent, sean-k-mooney: for a change, the 1..3 diff is actually empty, because the patch removes the whole class that was conflicting.14:48
openstackgerritBalazs Gibizer proposed openstack/nova master: Ensure that bandwidth and VF are from the same PF  https://review.openstack.org/62354314:49
openstackgerritBalazs Gibizer proposed openstack/nova master: Support server create with ports having resource request  https://review.openstack.org/63636014:49
gibibauzas: I've fixed up your comments ^^14:49
sean-k-mooneyefried: right cdent change modified one line in schdulerReportClientTEstBase and you are deleteing it out right14:50
*** sdake has joined #openstack-nova14:50
*** markvoelker has quit IRC14:52
cdentefried: other than me continuing to think it is keen and worth doing soon than later, you after anything in particular?14:52
efriedsean-k-mooney: three or four lines, but yeah.14:52
efriedcdent: no, just letting you know it needed a rebase and that's done now. I agree there's no reason to wait for Train. And I don't want reviewers intimidated by it because it's really like a 20LOC change.14:53
* cdent nods14:53
gibistephenfin: you were +2 before on https://review.openstack.org/623543 bauzas put +2 on the previous PS. Can you meet on PS 41 and combine your votes? :)14:53
cdentefried: I suppose if you wanted to make it really obvious you could do it in two steps: extracted the indented chunks to their own methods in 1, remove placement direct in 2, but personally I think that's overkill, the diff is readable enough as is14:54
bauzasgibi: I was about +2ing it14:55
efriedcdent: Yeah, when I was developing the change, I started off by creating a null _interceptor context manager. I didn't want to go to the effort of removing and outdenting if I wasn't going to be able to get it to work.14:55
lyarwoodmriedem: your nova-status cell_mappings change works on a local env btw14:56
lyarwoodmriedem: thanks for turning that around so quickly!14:57
mriedemanything for you lee14:57
*** pcaruana has quit IRC14:57
*** awaugama has joined #openstack-nova14:58
*** munimeha1 has joined #openstack-nova14:59
*** awalende has quit IRC14:59
*** ratailor has quit IRC15:00
*** awalende has joined #openstack-nova15:00
*** cfriesen has joined #openstack-nova15:01
*** Luzi has quit IRC15:02
*** tbachman has joined #openstack-nova15:02
gibibauzas: thanks a lot15:03
*** awaugama has quit IRC15:03
*** ccamacho has joined #openstack-nova15:03
*** awalende has quit IRC15:04
*** sdake has quit IRC15:04
*** tbachman_ has joined #openstack-nova15:05
mriedemmdbooth: fyi that https://review.openstack.org/#/c/625741/ just merged on master15:06
openstackgerritChris Dent proposed openstack/nova master: Correct lower-constraints.txt and the related tox job  https://review.openstack.org/62297215:06
*** tbachman has quit IRC15:07
*** tbachman_ is now known as tbachman15:07
mdboothmriedem: Oh, dear.15:07
mdboothmriedem: Didn't spot that was a backport.15:07
*** sdake has joined #openstack-nova15:07
sean-k-mooneyefried: i assume you rebased the patch due to a merge confilt with cdents patch?15:07
efriedsean-k-mooney: yes15:07
mdboothmriedem: Lemme read the review comments on there, perhaps my concern was already addressed.15:07
mriedemmdbooth: i doubt it, no one thinks of the image cache besides you :)15:08
sean-k-mooneyok just +1'd it again i havent used it locally but it makes sense to me15:08
mdboothmriedem: Hmm. I'd consider reverting that, tbh.15:09
*** eharney has joined #openstack-nova15:09
mdboothmriedem: The other solution proposed was much more targetted and presumably also works around the same issue.15:10
mdboothI'd rather fail than risk leaving an empty file in the image cache.15:11
mdboothThe former you can just retry.15:11
openstackgerritBalazs Gibizer proposed openstack/nova master: Handle placement error during re-schedule  https://review.openstack.org/63960815:11
mriedemmdbooth: are you talking about this? https://review.openstack.org/#/c/633493/215:12
sean-k-mooneyany objections to me repoposing the numa and sriov migration specs for train. i think we should be ablt to wrap those up by train m1 so im hoping we can just fast appove the stien specs again?15:12
*** dpawlik has quit IRC15:13
sean-k-mooneythere is 3 days to ff so they are not going to get reivewed betwen now and then anyway and that will give us plenty fo time to work on func tests for both15:13
*** Aegil has joined #openstack-nova15:13
*** awaugama has joined #openstack-nova15:15
mriedemsean-k-mooney: no one will object to reproposing specs to train for things that will not land in stein15:16
mriedemand no one is going to review those train specs for awhile either15:16
sean-k-mooneyya i just want to have them up before the ptg15:17
sean-k-mooneyi know peopel are going to be focusing on getting RC1 ready for the next few weeks and finalising stien15:17
sean-k-mooneyon that note im hoping we can land https://review.openstack.org/#/c/602432/ in stein/rc1 its not a change i think many would want to backport so i would like to close that bug in stien if we can15:18
mdboothmriedem: Yeah15:20
mdboothmriedem: Although obviously as you say you wouldn't fix that on a stable branch.15:21
mriedemgibi: random thought on the instance_has_port_with_resource_request check in the API - could we replace the neutron GET /ports call in there now with checking the instance info cache for any VIFs whose profile has the 'allocation' key?15:21
openstackgerritMatthew Booth proposed openstack/nova master: Revert "Fixes race condition with privsep utime"  https://review.openstack.org/64103815:24
mdboothmriedem: Proposed a revert ^^^15:24
*** sapd1 has quit IRC15:25
openstackgerritChris Dent proposed openstack/nova master: Correct lower-constraints.txt and the related tox job  https://review.openstack.org/62297215:25
*** mlavalle has joined #openstack-nova15:25
mdboothEurgh.15:25
* mdbooth reformats that commit message.15:25
sean-k-mooneymriedem: your suggesting modifying https://github.com/openstack/nova/blob/6516038dea5b80af2d83906a7e23f33e76061a42/nova/api/openstack/common.py#L571-L582 to check the  instance info cache.15:26
openstackgerritMatthew Booth proposed openstack/nova master: Revert "Fixes race condition with privsep utime"  https://review.openstack.org/64103815:27
sean-k-mooneyassuming its only ever called after the info cache is populated then maybe. if not we might be able to make it condtional15:28
*** tbachman has quit IRC15:28
*** openstackgerrit has quit IRC15:28
sean-k-mooneye.g. if the instance is not in the info cache when this is called hit neutron otherwise us cache versions15:28
mriedemgibi: https://review.openstack.org/#/c/636360/15:29
gibimriedem: looking15:29
sean-k-mooneymriedem: it looks like its only called on move operations15:30
mriedemsean-k-mooney: correct15:30
mriedemso the cache should be there15:30
mriedemand ports would already be bound15:31
sean-k-mooneyyep so i guess it might be safe to always used the cache values.15:31
sean-k-mooneywe are just checkign if a port has allocation too not where they are from and since that should not change over the lifetime of a port it "should" not matter if the allcoation were stale for example because we justed moved.15:33
sean-k-mooneyim ignoring that move wont work in stien15:33
mriedembauzas: are you good with this now? https://review.openstack.org/#/c/509206/15:33
mriedemyou previously approved15:33
mriedemand then we had to fix something for tempest15:33
mriedemSylvain Bauza  Feb 18 5:34 AM  Patch Set 10: Code-Review+2 Workflow+115:34
sean-k-mooneymriedem: could we have issue with unshelve?15:34
mriedemsean-k-mooney: yes, gdi15:34
gibimriedem: is it safe to access the instance.info_cache from the API?15:34
mriedemgibi: yes15:34
mriedemgibi: the addresses portion of the server response is built from the cache15:34
gibiohh, sean-k-mooney has a good point about shelve offloaded instances15:34
mriedemyes he does15:35
mriedemgrumble15:35
sean-k-mooneygibi: in that case it would not be in the cache and we can fallback to hitting the api15:35
sean-k-mooneyso check the cache if its there use it if not hit api15:35
mriedemsean-k-mooney: but then you're going to always hit the api because not all instances have these types of ports15:35
mriedemso...15:35
mriedemyou'd have to condition based on the instance.vm_state, but anyway, let's not worry about this right now15:35
*** openstackgerrit has joined #openstack-nova15:36
openstackgerritAndrey Volkov proposed openstack/nova stable/queens: Test for plug_contrail_vif with utf8 VM name  https://review.openstack.org/64104815:36
sean-k-mooneyok there is a comment in the code to optimise this anyway so ya we can look at that in train15:36
gibiif in train we can implement the move operations then we can drop the whole check from the aPI15:37
gibibah, not ture15:37
gibitrue15:37
mriedemnot true b/c older microversions need to fail right?15:37
gibiyeah15:37
gibiwe need to support old microversions15:38
gibibut we have the microversion check in place15:38
gibiso on old microversions we can simply check the microversion not the port allocations15:38
gibiwe only need the current check as the current 2.72 is not complete15:38
gibi(or we can go back to the discussion how many microversin we need for this feature)15:39
sean-k-mooney gibi well for old microverion you still need the check as we are only failing move operation if you have a port resouce request15:39
mriedemcorrect ^15:39
mriedemjust because 2.80 supports moving those types of servers doesn't mean that all of a sudden 2.1 does also15:40
gibiassume we merge the create support in 2.72. and the in Train we fix the move with a bugfix. then if the microversion < 2.72 then we have to check if the instance has ports with allocation15:41
gibiso yeah we cannot drop that check15:41
gibithat looks into the ports15:41
gibibut we can make it smarter by checking the vm state and the info_cache15:41
sean-k-mooneyya but we shoudl definetly be about to optimise it when we have more time15:42
sean-k-mooney*able to15:42
*** pcaruana has joined #openstack-nova15:42
mriedemdid we agree to support moves without a new microversion?15:42
* gibi making notes that will end up in a new spec / bp15:42
gibimriedem: no we haven't yet.15:43
mriedemok15:43
gibimriedem: we postponed that decision to Train15:43
mriedemwe would at least need a minimum compute service version check15:43
* gibi making notes that will end up in a new spec / bp about the remaining work15:43
sean-k-mooneyi think that is dansmith's perference e.g. non new micro15:43
gibimriedem: is the compute version check due to the virt driver impact?15:43
mriedemgibi: the compute service is what unbinds and binds the ports and stuff right during a move15:44
gibimriedem: ohh so we need a new compute on both end of the migration for bind/unbind15:45
mriedemso anything you'd need to add to compute in train to support migrations would require an ABI version bump15:45
mriedemcorrect15:45
gibiunderstood15:45
*** Sundar has joined #openstack-nova15:45
sean-k-mooneythat would be a check in the conductor liek we do for the multiple port bindigns or like we will be doing for numa and sriov migration15:46
sean-k-mooneyeffectivly if both computes are not new enough reject the migration15:47
gibisean-k-mooney: ack15:48
*** markvoelker has joined #openstack-nova15:48
*** sdake has quit IRC15:49
mriedemneed another novaclient core to hit this https://review.openstack.org/#/c/640657/15:50
*** udesale has quit IRC15:50
mriedemjaypipes: are you coming back on the libvirt reshaper change today? https://review.openstack.org/#/c/599208/15:51
jaypipesmriedem: yes. it's on my screen right now.15:51
*** zhubx has quit IRC15:55
*** zhubx has joined #openstack-nova15:56
cfriesenmriedem: should I include the blueprint for the PciInvalidAlias commit, or leave it standalone?15:57
*** xek has quit IRC15:59
mriedemcfriesen: that's fixing a separate bug right?15:59
*** xek has joined #openstack-nova15:59
gibimriedem: +Ad https://review.openstack.org/#/c/640657/15:59
mriedemiow, handling PciInvalidAlias separately from the blueprint is so it could be backported right?15:59
cfriesenmriedem: yep15:59
mriedemcfriesen: if not, and it doesn't result in a 500 response today, then we don't need to backport it15:59
*** belmoreira has quit IRC16:00
mriedemcfriesen: btw are you sure that's not handled today?16:00
mriedemby this https://github.com/openstack/nova/blob/c43c1d3fb9da5dd0a13e1f15623a696212f095ff/nova/api/openstack/wsgi.py#L36716:00
*** tbachman has joined #openstack-nova16:00
mriedemPciInvalidAlias extends Invalid16:00
*** sdake has joined #openstack-nova16:01
mriedemso that should return a 40016:01
cfriesenyeah, was just trying to track that down16:01
mriedemif it is handled then i don't think we need anything separate16:02
mriedemhttps://review.openstack.org/#/c/620706/27/nova/api/openstack/compute/servers.py@a85716:02
cfriesenso Fault will end up being HTTPBadRequest?16:02
cfriesenor I guess Fault(exception.ConvertedException(16:03
mriedemwithout any testing for it it's hard to say16:04
mriedemso we get that from get_pci_requests_from_flavor which is called today from _validate_and_build_base_options in the api on server create,16:05
mriedemi don't see any existing tests that handle the PciInvalidAlias if it's raised from there16:05
mriedemi suspect it is handled and converted to a 400 response16:05
mriedembut we don't have any testing for that16:06
cfriesenI'll try a test without specially adding it to the list16:06
mriedemunlike https://review.openstack.org/#/c/640841/2/nova/tests/unit/api/openstack/compute/test_serversV21.py16:06
cfriesenmriedem: looks like it fails:  http://paste.openstack.org/show/747293/16:10
cfriesenif I add PciInvalidAlias to the list in ServersController.create() , it passes16:11
mriedemi assume those tests aren't using https://github.com/openstack/nova/blob/c43c1d3fb9da5dd0a13e1f15623a696212f095ff/nova/api/openstack/wsgi.py#L36716:11
mriedemthose are hitting the handler directly16:12
mriedemwhereas ^ wraps the handler16:12
cfriesenyeah, just realized that16:12
cfriesenokay, need to hack devstack then16:13
mriedemanywho, i think PciInvalidAlias is probably handled as a 400 and we're OK, and for completeness you could add it to that big ass list of exceptions for the validation stuff16:13
cfriesenworks16:13
mriedemyou got a 400 for an invalid alias in devstack?16:13
cfriesenhaven't done it yet16:13
mriedemoh16:14
cfriesenI get: ERROR (ClientException): Unknown Error (HTTP 500)16:16
mriedemis there a stacktrace in the api logs?16:17
cfriesenwhoops...typo.  I had "faise" istead of "raise"16:18
cfriesenokay, for real this time I get this: http://paste.openstack.org/show/747294/   checking logs16:19
cfriesenhere are the logs: http://paste.openstack.org/show/747295/16:20
cfriesenadding the exception to the list makes it a 400 as expected16:21
*** markvoelker has quit IRC16:22
cfriesenreport a bug or just put it in the git commit message?16:22
gibimriedem: regarding https://review.openstack.org/#/c/636360/21/releasenotes/notes/support-neutron-ports-with-resource-request-cb9ad5e9757792d0.yaml@21 I think unshelving is possible after a simple shelve without offload. And that operation is supported with bandwith as that does not need a new scheduling16:27
*** liuyulong is now known as liuyulong_zzz16:27
mriedemefried: can we do an assertEqual(1, len(mdevs)) check in here instead? https://review.openstack.org/#/c/639854/416:29
jaypipesmriedem: off it goes.16:29
efried...16:29
mriedemgibi: yeah true16:30
jaypipeswouldn't it be great if Nvidia developers actually participated in the community of people trying to enable their hardware? :(16:32
efriedmriedem: responded16:33
mriedemefried: fwiw i also commented in bauzas's change that we should just do that mdev naming conversion from dashes to underscores in _get_mediated_device_information16:35
mriedemso callers, like the libvirt driver and your test, don't need to care about that conversion16:35
mriedemgah https://review.openstack.org/#/c/639647/16:35
*** imacdonn has quit IRC16:35
mriedemi was going to -1 because there are other follow ups from this series i'd like to see16:35
mriedemi guess i'll do those16:35
*** imacdonn has joined #openstack-nova16:36
*** mrch_ has quit IRC16:36
efriedmriedem: So do you want me to fix the test patch, or are you going to fup?16:36
mriedemi'm gonna fup all night long16:37
efriedight16:37
openstackgerritSurya Seetharaman proposed openstack/nova master: Delete orphaned RS and IM records of archived instances  https://review.openstack.org/56004216:39
openstackgerritBalazs Gibizer proposed openstack/nova master: Support server create with ports having resource request  https://review.openstack.org/63636016:40
gibimriedem: fixed your comments in ^^16:40
*** janki has quit IRC16:41
*** luksky has quit IRC16:41
sean-k-mooneyjaypipes: ya nvida and opensource is not really a thing...16:41
*** janki has joined #openstack-nova16:41
sean-k-mooneyjaypipes: unless its there "opensouce" project16:41
mriedemmelwitt: there is a lot of stuff in https://blueprints.launchpad.net/nova/stein that is going to be deferred, do you want to just start moving stuff out now to make that page more clear?16:42
mriedemhttps://etherpad.openstack.org/p/nova-stein-blueprint-status is getting to be kind of a mess16:43
mriedemso would be nice if the bp pages in lp was cleaned up, or https://etherpad.openstack.org/p/nova-stein-blueprint-status was cleaned up to remove stuff that's either done or not going to make it16:43
melwittmriedem: yeah :( I can try16:43
mriedemwell if you're working on the counting stuff that's probably higher priority16:44
*** helenfm has quit IRC16:45
*** tssurya has quit IRC16:45
melwittdansmith: I updated the user_id patch last night to remove the loading stuff, fyi. from your comment it sounds like the only thing missing is to do a log message in case someone tries to load an unset user_id16:46
sean-k-mooneyam should we just add another heading for defered stuff at the bottom.16:46
melwittmriedem: oh, sorry I misunderstood "do you want to ...". yeah, I think that is fine16:46
bauzasmriedem: sorry, missed your comments on the naming thing16:46
bauzasand I was off here for an hour16:47
openstackgerritChris Friesen proposed openstack/nova master: Handle missing exception in instance creation code  https://review.openstack.org/64108216:51
openstackgerritBalazs Gibizer proposed openstack/nova master: Ensure that bandwidth and VF are from the same PF  https://review.openstack.org/62354316:53
openstackgerritBalazs Gibizer proposed openstack/nova master: Support server create with ports having resource request  https://review.openstack.org/63636016:53
*** psachin has quit IRC16:54
cfriesenmriedem: fix for PciInvalidAlias issue.  I reported a bug for it too.16:54
mriedemcfriesen: but you got a 500 for that in the api with devstack?16:54
cfriesenmriedem: yes, see above16:55
*** ccamacho has quit IRC16:55
cfriesen here are the logs: http://paste.openstack.org/show/747295/16:55
cfriesenCLI output is in the bug: https://bugs.launchpad.net/nova/+bug/181870116:56
openstackLaunchpad bug 1818701 in OpenStack Compute (nova) "invalid PCI alias in flavor results in HTTP 500 on instance create" [Medium,In progress] - Assigned to Chris Friesen (cbf123)16:56
*** mrch_ has joined #openstack-nova16:58
mriedemcfriesen: ok just robustify the test and i'm +216:58
*** ivve has quit IRC16:58
*** macza has joined #openstack-nova16:58
openstackgerritTheodoros Tsioutsias proposed openstack/nova master: Add requested_networks to RequestSpec  https://review.openstack.org/57020117:00
openstackgerritTheodoros Tsioutsias proposed openstack/nova master: Enable rebuild for instances in cell0  https://review.openstack.org/57020317:00
*** wolverineav has joined #openstack-nova17:00
mriedemgibi: can we mark this complete? https://blueprints.launchpad.net/nova/+spec/use-nested-allocation-candidates17:01
gibimriedem: there is one intentionally open patch https://review.openstack.org/#/c/608298/17:02
gibimriedem: it is an edge case that we don't need in nova right now17:02
mriedem"This cannot happen today but will in the future, when cpu and memory resource are moved to the NUMA RP and servers"17:02
gibimriedem: exactly17:02
mriedemcorrect, so i'd just say it becomes a part of those bps when they happen17:02
gibimriedem: OK. I agree. Let's mark the bp completed17:03
*** macza_ has joined #openstack-nova17:04
mriedemdone17:04
*** wolverineav has quit IRC17:04
openstackgerritChris Friesen proposed openstack/nova master: Handle missing exception in instance creation code  https://review.openstack.org/64108217:04
cfriesenmriedem: I also robustified the other PCI test17:05
mriedemoh my17:05
mriedemthat's a -517:05
mriedemi hope you ran fast817:05
mriedembut +217:05
cfriesenI can break out the other test as a separate commit if you want...17:06
mriedemi'm just joking17:06
mriedem-5 is a joke unless we're talking about forced cold migration17:06
*** macza has quit IRC17:08
*** xek_ has joined #openstack-nova17:10
-openstackstatus- NOTICE: Gerrit is being restarted for a configuration change, it will be briefly offline.17:11
sean-k-mooneyi tought forced migrations tradtional add a couple fo factorals added on like so -5!!17:11
*** xek has quit IRC17:12
mriedem-5?!?!??!!17:12
mriedemjazz it up however you want17:12
melwittreshaper + func test is finally on its way to the sarlacc pits \o/17:13
dansmithmelwitt: oh, I saw the comment wasn't on the latest PS, but I hadn't gone back to see what had changed17:14
*** jistr|sick is now known as jistr17:14
melwittdansmith: understood. just giving you the 41117:15
*** markvoelker has joined #openstack-nova17:19
*** ileixe has quit IRC17:20
mriedemmdbooth: you might want to take a look at this https://review.openstack.org/#/c/613039/17:21
mriedemif it has a chance of making stein17:21
mriedemit does have test coverage... https://review.openstack.org/#/c/623483/17:21
mdboothmriedem: ack, thanks17:21
*** luksky has joined #openstack-nova17:23
*** panda|ruck|flu is now known as panda|ruck|off17:27
*** sridharg has quit IRC17:29
openstackgerritMatthew Booth proposed openstack/nova master: Don't age swap files in the image cache  https://review.openstack.org/64109317:30
*** itlinux has joined #openstack-nova17:35
mriedemaspiers: correct me if i'm wrong, but up to https://review.openstack.org/#/c/638680/ we advertise that a compute will support sev but up to that point we don't actually configure anything in a guest that requires sev, right?17:35
*** sdake has quit IRC17:35
*** artom has quit IRC17:35
mriedemso if we were to merge that code in stein, we'd be saying "these computes support sev" but you can't actually schedule servers to them and get sev in the guest, yeah?17:36
*** ttsiouts has quit IRC17:38
mriedemi.e. something is yet to build on this https://review.openstack.org/#/c/636318/17:38
mriedemall of this being done in parallel series makes it hard to track17:38
dansmithsomeone should +W this: https://review.openstack.org/#/c/641021/17:41
*** xek_ has quit IRC17:42
*** xek_ has joined #openstack-nova17:42
* mriedem breaks17:42
melwittyou can't +W it bc it's your patch. I can look17:43
*** tbachman has quit IRC17:44
bauzasI can take a look if needed17:44
bauzasI still have 30 mins left17:44
dansmithit's easy, you all should be fighting over it17:44
*** tbachman has joined #openstack-nova17:44
bauzasI haven't followed the cross-cell resize series but looks like it's unrelated17:45
bauzasso I can chime in17:45
cfriesenphilosophical question for people.  if we resize to the same instance type, should we validate the extra-specs again in case they've been changed since we booted?  I kind of think yes.17:46
melwittbauzas: you can have it :)17:47
dansmithcfriesen: I think yes17:47
sean-k-mooneycfriesen: 1 yes 2 we proably should make exta-specs imunteable at some point once a flavor is in use17:47
cfriesen(or is that disallowed and only actually happens on migrations?)17:47
bauzascfriesen: yes please17:48
cfriesenit's confusing, there's something related to cell types in here17:48
melwittcfriesen: I think yes too17:48
bauzascfriesen: the problem is that we put too many things in extra sepcs17:49
sean-k-mooneyis there any reason we dont allow you to set extraspec when creating a flavor so we can make the whole thing readonly17:49
bauzaswe should at least validate those specs before calling the scheduler17:49
cfriesenI'm looking at https://github.com/openstack/nova/blob/master/nova/compute/api.py#L3541 which seems to indicate that you're not supposed to resize to the same flavor, but there's an exception for a "compute" cell type17:49
sean-k-mooneyi guess maybe because resize is a cold migration?17:49
openstackgerritMerged openstack/nova master: Stub out port binding create/delete in NeutronFixture  https://review.openstack.org/63641317:50
mdboothlyarwood: Would you also be able to pontificate upon https://review.openstack.org/#/c/613039/ ?17:50
*** agopi has joined #openstack-nova17:50
openstackgerritMerged openstack/python-novaclient master: Microversion 2.71 - show server group  https://review.openstack.org/64065717:51
mdboothlyarwood: FWIW I'll be +1 on that assuming the tests are good, but I don't have time to look now.17:51
*** markvoelker has quit IRC17:51
*** marst has quit IRC17:51
cfriesensean-k-mooney: yeah, but the comments say that for migration "flavor_id" is None.17:52
*** rpittau is now known as rpittau|afk17:52
sean-k-mooneycfriesen: right well resize and colde migrate are the same thing in nova except resize change the flavor and migrate does not17:53
stephenfingibi: Sorry for the delay. +2'd now :)17:53
*** gyee has joined #openstack-nova17:54
cfriesensean-k-mooney: yeah, but there's some cells complication apparently:  https://bugs.launchpad.net/nova/+bug/124381617:54
openstackLaunchpad bug 1243816 in OpenStack Compute (nova) "migrate fails with cells" [Undecided,Fix released] - Assigned to Chris Behrens (cbehrens)17:54
sean-k-mooneycfriesen: so resize to the same flavor whould be expected to do the same thing as migrate however migrate uses the embeded fravor17:54
sean-k-mooneyyes well for cells v1 the flavor may not have been the same or replicated across cell. cross cell resize for cellsv2 is something mriedem is working on currently17:55
cfriesenLooks like normally you cannot resize to the same flavor, so my concern is unfounded.18:00
openstackgerritStephen Finucane proposed openstack/nova master: docs: Rework all things metadata'y  https://review.openstack.org/64073018:01
openstackgerritStephen Finucane proposed openstack/nova master: docs: Fix nits in remote console guide  https://review.openstack.org/64109818:01
stephenfinefried: You'd probably be interested in those come Friday ^18:02
efriedack18:03
efriedFriday because... after freeze?18:04
*** agopi has quit IRC18:04
openstackgerritChris Friesen proposed openstack/nova master: Flavor extra spec and image properties validation from API  https://review.openstack.org/62070618:05
*** artom has joined #openstack-nova18:20
openstackgerritChris Friesen proposed openstack/nova master: Improve libvirt image and snapshot handling  https://review.openstack.org/61669218:22
cfriesenany chance of getting some core eyes on this?  ^  It's a host filesystem cache optimization patch, has a +1 from mdbooth.   (And if you say it's simply not going into S then I'll shut up about it till next week.)18:22
*** artom has quit IRC18:25
*** jamesdenton has quit IRC18:27
openstackgerritMerged openstack/nova master: libvirt: implement reshaper for vgpu  https://review.openstack.org/59920818:27
openstackgerritMerged openstack/nova master: Add functional test for libvirt vgpu reshape  https://review.openstack.org/63155918:28
openstackgerritMerged openstack/nova master: FUP: docs nit  https://review.openstack.org/63964718:28
dansmithcfriesen: you mean host buffer cache optimization yeah/18:28
*** jamesdenton has joined #openstack-nova18:29
dansmiththat looks like a nice win and might be why my toy all-in-one openstack at home completely destroys performance while taking snapshots18:29
*** xek_ has quit IRC18:34
*** xek has joined #openstack-nova18:35
*** mriedem has quit IRC18:37
*** artom has joined #openstack-nova18:38
*** wolverineav has joined #openstack-nova18:38
*** artom has quit IRC18:39
*** mriedem has joined #openstack-nova18:39
*** artom has joined #openstack-nova18:39
melwittdansmith: if something tries to reference unset user_id field, "NotImplementedError: Cannot load 'user_id' in the base class" happens. do you think we need to log a message in addition to that?18:41
*** itlinux has quit IRC18:42
dansmithmelwitt: I dunno, that ends up in a lot of places where we have an implementation error, so it's nice to log something about why (i.e. mappings haven't been migrated) but it's also a failing on our part, since it shouldn't be used anywhere anyway18:43
dansmithmelwitt: so you decide I guess.. but exploding is the right thing regardless of the, uh, type of explosion It hink18:43
melwittdansmith: ok, yeah, that makes sense. just log and reraise18:44
*** IvensZambrano has quit IRC18:46
*** itlinux has joined #openstack-nova18:47
*** markvoelker has joined #openstack-nova18:48
*** Sundar has quit IRC18:53
mriedemmdbooth: btw, i hope you've seen office space: https://photos.app.goo.gl/HgzGk3aR2yGHWMv6918:54
*** awaugama has quit IRC18:57
melwittlol19:04
openstackgerritsean mooney proposed openstack/nova master: libvirt: delegate ovs plug to os-vif  https://review.openstack.org/60243219:06
melwittcfriesen: from the commit message, "so that cache option can be set properly". that makes it sound like it's a bug without that change. is it a bug or no?19:06
mriedemcfriesen: approved your validation change and i'm assuming you're still working on a patch to check get_pci_requests_from_flavor  from resize?19:06
dansmithah dangit, I was just going through that19:08
dansmithmelwitt: it's a feature19:08
melwittoh k19:09
dansmithmelwitt: it's not a bug because it's not wrong, but setting the cache properly avoids trashing the host's page cache while doing it, improving performance for everything running while the snapshot is happening19:09
melwittah, thanks19:09
cfriesenmelwitt: what he said. :)19:09
dansmithaside from the consistency gain of not calling the same external binary manually in multiple places19:09
cfriesenmriedem: yes, just working on tests19:10
melwittok, kewl19:10
*** awaugama has joined #openstack-nova19:11
*** ivve has joined #openstack-nova19:12
efriedwhoops, guess I should have been paying attention here before I +A'd that sucker.19:14
melwittit's np19:16
efriedokay, can pull it if necessary.19:16
melwittnah, it's ok. looking at it further, it would take me longer to understand all the qemu-img stuff anyway19:17
*** mdbooth_ has joined #openstack-nova19:21
*** markvoelker has quit IRC19:21
*** awaugama has quit IRC19:22
*** awaugama has joined #openstack-nova19:23
*** mdbooth has quit IRC19:24
artomsean-k-mooney, around? I need your Neutron and/or os-vif expertise19:28
*** ttsiouts has joined #openstack-nova19:31
*** dtantsur is now known as dtantsur|afk19:31
mriedemefried: edmondsw: i'm assuming this is ok for powervm since the parameter is just ignored https://review.openstack.org/#/c/613039/4/nova/virt/powervm/driver.py19:32
mriedemmaybe could help optimize something with powervm volume extend though19:32
efriedmriedem: powervm oot will need to do the same, so thanks for alerting edmondsw19:32
*** sdake has joined #openstack-nova19:33
efriedgenerally I like to see a ML post when we change things like the ComputeDriver interface, since OOT drivers will need to conform.19:33
mriedemooo as will hyperv http://git.openstack.org/cgit/openstack/compute-hyperv/tree/compute_hyperv/nova/driver.py#n22419:34
efriedmriedem: but yeah, the change itself should be harmless.19:34
efriedAnd I don't see how powervm could take advantage of it to help anything.19:37
mriedemyeah me neither19:39
*** wolverineav has quit IRC19:41
artomIs the "plug" verb overloaded in Neutron and os-vif?19:41
artomLooking at https://github.com/openstack/nova/blob/master/nova/virt/libvirt/driver.py#L5653-L5656, I could be forgiven for thinking that our call to plug_vifs() is what causes the network-vif-plugged event to eventually reach19:42
*** wolverineav has joined #openstack-nova19:42
artomBut that's incorrect... right?19:42
artomWay earlier we call Neutron via setup_networks_on_host(), and *that's* how we get the event19:42
artomBecause Neutron plugs the port19:43
*** wolverineav has quit IRC19:43
*** wolverineav has joined #openstack-nova19:43
artom... is that the distinction? Neutron plugs *ports* on the host (ie, sets up host networking)19:43
artomAnd then Nova plugs the *VIF* to whatever interface Neutron has made for us...19:44
artom... except the event we get from Neutron is vif-plugged, not port-plugged19:45
*** wolverineav has quit IRC19:45
*** wolverineav has joined #openstack-nova19:46
mriedemi don't think that's accurate19:46
mriedembecause we may or may not get a network-vif-plugged event during hard reboot where the port binding host doesn't change19:46
*** wolverineav has quit IRC19:46
mriedembut only certain networking backends in neutron will send the event in that case19:46
*** wolverineav has joined #openstack-nova19:46
mriedemit led to a whole messy situation with dealing with that and reboots for things like opendaylight19:46
dansmithartom: neutron sends us vif-plugged when it sees an appropriately-named interface show up on the system,19:47
mriedemlemme dig up a ghost for you19:47
dansmithwhich is usually because we have created it as part of starting a vm19:48
mriedemartom: have fun with this https://review.openstack.org/#/c/553035/19:48
dansmithartom: there's also some plug terminology around hooking that interface up to a bridge19:48
dansmithand some around the work nova does to create an interface in the first place like during spawn19:48
artomdansmith, ok, that answers my "ulterior motive" questions, which was - can we wait for vif-plugged in the compute manager. The answer being: no. Because the VM needs to start.19:48
aspiersmriedem: Yes that's correct, there are more changes by my colleague breton which build on https://review.openstack.org/#/c/638680/ and modify the guest config to activate SEV functionality. Actually they aren't far from code complete but I think still need a bit of polish here and there. 2 or 3 are already in Gerrit and there are maybe 2 or 3 to come after that.19:48
dansmithartom: in the spawn case, that's correct19:49
dansmithartom: which is why we start the instance paused, wait for the interface to show up, neutron to notice, notify us, and then we unpause the guest19:49
dansmithartom: before, we would start the instance right away, and it might time out waiting for dhcp before all that happened19:49
openstackgerritmelanie witt proposed openstack/nova master: Add user_id field to InstanceMapping  https://review.openstack.org/63335019:49
openstackgerritmelanie witt proposed openstack/nova master: Populate InstanceMapping.user_id during migrations and schedules  https://review.openstack.org/63857419:49
openstackgerritmelanie witt proposed openstack/nova master: Add online data migration for populating user_id  https://review.openstack.org/63335119:49
openstackgerritmelanie witt proposed openstack/nova master: Add get_counts() to InstanceMappingList  https://review.openstack.org/63807219:49
openstackgerritmelanie witt proposed openstack/nova master: Count instances from mappings and cores/ram from placement  https://review.openstack.org/63807319:49
openstackgerritmelanie witt proposed openstack/nova master: Use instance mappings to count server group members  https://review.openstack.org/63832419:49
dansmithwhich was the initial point of these events19:49
artomdansmith, hrmm, so I've understood the problem wrong then19:50
aspiersmriedem: We've been mostly pushing them in a linear sequence, but there has been a bit of parallelism on the base work not specific to SEV, to avoid superfluous dependencies slowing things down.19:50
artomIt's not just that we're not listening for the event when it arrives19:50
aspiersmriedem: I'm more than happy to draw a dependency graph summarising all patches if that helps?19:51
dansmithartom: the initial problem was that some not-insignificant amount of time, we would start the instance (which created the vif) and the instance would be stillborn without networking, because it got to the OS part of the boot that waits for DHCP before neutron noticed, wired us up, setup dhcp for us etc19:51
aspiersI thought I'd seen launchpad doing that actually, but maybe that was just for dependencies between bps19:51
artomIn case of spawn, we can't ever hit that situation because we start listening before we start the VM19:51
dansmithartom: in the boot case, we're definitely waiting at the right time, and that's why we don't have 10% DOA instances in the gate since like 201319:52
dansmithartom: exactlyu19:52
artomThe problem arises in other cases, where the interface is already present on the host19:52
dansmithartom: right, which is because we've possibly blindly applied this to other cases like moves19:52
artomTook me 219:52
dansmithartom: which have similar issues, of course, but potentially different timing19:52
artomTook me 2 days, but I got there19:52
artomSo it's not a blanket "let's start waiting before we poke Neutron" kinda of thing19:53
artomI need to take it case by case19:53
artomSee when we poke Neutron, whether the interface is present when we do, and when do we start waiting19:54
artomFuuun19:54
dansmithartom: that's why in the previous discussion, I was careful to say "before we do the thing that will trigger the event"19:54
dansmithwhich may or may not be talking to neutron (in fact, it usually isn't I think)19:54
dansmithand also why I said bringing the trigger under the umbrella context is the right thing, for whichever operations that isn't already the case19:54
artomdansmith, a very carefully worded non-specificity ;)19:55
*** dave-mccowan has quit IRC19:55
*** itlinux has quit IRC19:56
artomAnd apparently what Neutron does also depend on the VIF type?19:56
dansmithAFAIK, all of this depends on vif type19:57
dansmithwhich is why we don't wait for events for certain vif types19:57
dansmithand for certain vif types during certain operations, IIRC (i.e. move vs. spawn)19:57
artomWhere's the code that decides that?19:58
dansmiththere's _get_neutron_events() in libvirt driver, but that doesn't have the vif type filter I was thinking of19:59
dansmithoh,19:59
dansmithsahid wrote a patch that introduced one vif type filter that we reverted20:00
artomYeah, I was looking at that20:00
dansmithso that's no longer in there20:00
dansmiththere's one other place I remember, ahgn on20:00
artomUnless it's implicit in the 'ACTIVE' check20:00
dansmithhmm, I thought there was also a vif type filter in mriedem's more generic check for live migration, but I don't see it there either20:01
*** dave-mccowan has joined #openstack-nova20:01
dansmithso maybe I'm just thinking about that patch from sahid that we reverted20:01
artomAnd actually in libvirt _create_domain_and_network is the only time we call wait_for_instance_event20:01
dansmith...in the libvirt driver, yes20:01
artomAh, there's _get_neutron_events_for_live_migration in the compute manager20:03
artomNo vif type filter though20:03
*** xek has quit IRC20:03
*** xek has joined #openstack-nova20:03
dansmithyeah, I really thought that one had a filter on it, maybe mriedem can comment when he sees this20:03
dansmithlemme dig up the sahid patch to prove there was one at least20:03
mriedemvif type == linuxbridge?20:04
mriedemvif type goes out the window because we can't rely on it20:04
mriedemala https://review.openstack.org/#/c/553035/20:04
mriedemopendaylight is ovs vif type but doesn't send network-vif-plugged on vif plugging, only port binding20:05
mriedemaspiers: no i don't need a graph of the code patches,20:05
dansmithartom: https://review.openstack.org/#/c/497457/30/nova/virt/libvirt/driver.py@553620:05
mriedemaspiers: the reason i was asking is because there doesn't seem to be a reason, to me, to land the sev discovery / exposing the capability from the host change (which jaypipes is +2 on) until we actually have code that can plumb the guest20:05
dansmithmriedem: ah, did we have a check for that before we had to remove because of ODL?20:05
artomdansmith, hrmm, why'd that get reverted out?20:06
dansmithartom: for so many reasons20:06
dansmithartom: amazingly broken :)20:06
mriedemartom: for one thing it was setting up the waiter *after* we'd plugged vifs on the dest20:06
mriedemand the waiter code was waiting on the source20:06
mriedemso a race and all that20:06
artomI was thinking more narrowly in terms of the VIF type logic20:06
dansmithartom: https://review.openstack.org/#/q/Ibf2b5eeafd962e93ae4ab6290015d58c3302413220:07
artomIOW, is "Currently we only use the mechanism of waiting for neutron events during live-migration for linux-bridge" false?20:07
dansmithartom: it turned out that neutron was sending those events, this just excluded waiting for them because of other bugs.. papering over the real problem20:08
dansmithit's been a while, but there was a LOT of analysis and discussion20:08
mriedemartom: true that is false :)20:08
aspiersmriedem: well one reason is that we (SUSE) will need to carry our own downstream backport to Rocky, so the more which lands in Stein the easier it will get for us ... but if that is in conflict with upstream goals then of course I would entirely understand20:08
mriedemartom: https://docs.openstack.org/nova/latest/configuration/config.html#compute.live_migration_wait_for_vif_plug20:08
*** itlinux has joined #openstack-nova20:08
mriedem^ we wait for networking events during live migration based on config20:08
mriedemsince rocky20:09
mriedemand part of that was backported20:09
mriedemby mr smith20:09
mriedemaspiers: heh, sure - i just don't think we want the sev capability showing up automatically in stein on the compute node resource provider when there is really nothing you can do with it20:09
*** dpawlik has joined #openstack-nova20:09
dansmithah right, artom, the Note:: in that doc explains why the vif_type thing isn't usable and why you have to config your way out of this for certain backends20:10
dansmithwhich sucsk20:10
aspiersmriedem: yeah, totally understood ... we were really hoping to get all the patches ready in time for stein but it looks like we just weren't quite quick enough20:10
*** ttsiouts has quit IRC20:11
mriedemaspiers: no problem, you got that capabilities patch landed which i'd abandoned essentially, which is pretty good for your first big contribution to nova20:11
*** ttsiouts has joined #openstack-nova20:11
artomdansmith, I get all of it except for "without an accompanying port binding:host_id"20:11
mriedemplus you had that acid trip of a venn diagram20:11
mriedemartom: that's what i was saying earlier,20:12
dansmithartom: ever get the feeling you stepped into a hornet's nest?20:12
mriedemODL will only send network-vif-plugged for port binding changes to the host20:12
*** eharney has quit IRC20:12
mriedemif the host doesn't change on the port binding, but we plug/unplug vifs or whatever, ODL doesn't send the event20:12
aspiersmriedem: thanks :) feels good to be giving to upstream not just "taking" (i.e. asking for our own pet feature to be reviewed and merged)20:12
mriedemartom: hence why it had to be reverted from that hard reboot change20:12
mriedemthis: https://review.openstack.org/#/c/553035/20:13
artomdansmith, yes20:13
aspiersmriedem: haha, well my first version of that venn diagram was quite different, you can blame^Wthank efried for a lot of that second version20:13
mriedemlinuxbridge also had some problems with that in the neutron lb job because something about polling for changes and it would miss them and not send the event20:13
melwittI concur, thanks aspiers++20:13
artommriedem, ah, so the note is just worded poorly to say "ODL will only send us vif-plugged if the host has changed"20:13
dansmithmriedem: ah, yeah that's the thing I was expecting to find in there, forgot that had to be reverted too20:13
aspiersthanks melwitt :)20:13
artom... which obviously it doesn't on hard reboot20:14
mriedemartom: it is worded perfectly20:14
artomNuh uh, communication is only perfect if the thickest of idiots understand it. I didn't understand this. QED.20:14
mriedemof course the port binding host *does* change during live migration,20:14
*** itlinux has quit IRC20:15
mriedembut that happens after / during the guest transfer, not as a result of plug_vifs during pre_live_migration20:15
mriedemwhich is what that config is used for20:15
*** ttsiouts has quit IRC20:15
mriedemalso, let me say, that i'm afraid sean-k-mooney might be dead or something because this entire conversation is the kind of thing he stays up until 3am for20:15
artomSeriously.20:16
aspiersLOL20:16
artomRIP sean I guess. We'll do a minute of silence i Denver.20:16
*** dpawlik has quit IRC20:16
*** sdake has quit IRC20:17
artommriedem, wait, so why is that note even there? It's talking about not sending events when the host binding doesn't change.20:18
artomThat doesn't apply to live migration...20:18
jaypipesmriedem: was I not supposed to +2 something?20:18
*** markvoelker has joined #openstack-nova20:19
*** sdake has joined #openstack-nova20:19
aspiersjaypipes: hopefully +2 is OK, W+1 might not have been though :)20:20
artomSo, I think I should ignore the whole Neutron backend thing for now. It's out of scope, and I shouldn't introduce new bugs by ignoring it because it's currently ignored anyways. We can continue to wait for events in the same situations as currently, we just need to change the timing of the waits.20:21
*** whoami-rajat has quit IRC20:22
melwittpoor wording, that reminds me, I was supposed to update the consoleauth reno (or was it nova-status?) yet again to make more sense. I need to find IRC log from that convo with imacdonn20:23
*** igordc has joined #openstack-nova20:28
imacdonnmelwitt: http://eavesdrop.openstack.org/irclogs/%23openstack-nova/%23openstack-nova.2019-02-20.log.html#t2019-02-20T18:46:5220:29
*** ralonsoh has quit IRC20:29
edmondswmriedem yeah, I agree with you and efried. Should be fine, except that we do need to go update the OOT driver as well, so the heads up is definitely appreciated.20:29
melwittimacdonn: thanks, you're awesome20:30
imacdonn:)20:30
*** awaugama has quit IRC20:30
*** wolverineav has quit IRC20:31
mriedemartom: ...20:32
mriedemartom: it's saying where this config is used, might not make sense for all networking backends (ODL),20:32
*** ttsiouts has joined #openstack-nova20:32
mriedembecause the config is only for vif plugging, not port binding host changes,20:32
*** wolverineav has joined #openstack-nova20:32
mriedemODL doesn't send events for the former, but does for the latter,20:32
mriedembut the latter is not where that config is used20:32
mriedemport binding happens during/after the guest transfer,20:32
mriedemwe're waiting for the event before starting the guest transfer20:32
mriedembecause if the dest plug fails or something, we don't want to start transferring the guest20:33
mriedemdansmith: check my comments on takashin's fix here https://review.openstack.org/#/c/636271/ - should i just not care about a functional test for live migration or stack it up on top myself if i care?20:33
mriedemi'm probably reaching20:34
mriedemjaypipes: see my comment on https://review.openstack.org/#/c/638680/20:34
dansmithmriedem: I dunno, the real change is in that object loader, which seems pretty well covered by two operations, so .. seemed like enough to me20:35
artommriedem, ah, got it, I think. Sorry, am slow.20:35
dansmithI pulled it down and poked at it too, both fail in the expected way...20:35
mriedemdansmith: yeah ok, i just don't like being told it's not an issue for live migration :)20:35
dansmithwell, I don't think stacking one on top is a problem if it makes you feel better20:36
mriedemit won't fill the hole inside20:36
dansmithhah20:36
artommriedem, oh, ODL *doesn't* send events for binding host changes20:36
*** wolverineav has quit IRC20:37
mriedemWRONG20:37
artomDAMMIT20:37
mriedemit *only* sends events in that case20:37
dansmithARGH20:37
artomThat note is damn confusing, I'm not backing down from that >_<20:37
mriedemis it time for a truth table in that config option note?20:38
mriedemif x and y but not z = False?!20:38
artom...20:38
artomOK, sorry, it's fine.20:39
artomI misread mriedem's line here.20:39
artomI mean, mostly fine, it could be worded better.20:39
dansmithbut, the note is for an operator,20:40
dansmithand all they need to know is ODL is not okay20:41
dansmithit's not meant to explain the inner workings of both services, like you want it to, right?20:41
artomYeah20:41
artomI just think explicitly saying what "OK" (or "will not currently work") would have been better20:41
artomI need to be off anyways20:42
openstackgerritMerged openstack/nova master: Fix an error when generating a host ID  https://review.openstack.org/63993620:42
mriedemfor those that *need* to know the inner details, the bug report is linked20:43
openstackgerritAdam Spiers proposed openstack/nova master: Add detection of SEV support from QEMU/AMD-SP/libvirt on AMD hosts  https://review.openstack.org/63385520:44
openstackgerritAdam Spiers proposed openstack/nova master: Add detection of SEV support from QEMU/AMD-SP/libvirt on AMD hosts  https://review.openstack.org/63385520:46
*** cdent has quit IRC20:51
*** markvoelker has quit IRC20:52
sean-k-mooneymriedem: :) i was having dinner but i see ye were dicussing how odl breaks all the things when it comes to vif-plugged events20:55
sean-k-mooneyartom: by the way i straed on https://review.openstack.org/#/c/635083/ to recored which ml2 driver bound a port so we can use it in nova for thing and i intend to add another feature that report when each ml2 driver sends events. e.g. on portbinding vs when the backend wires it up20:57
*** wolverineav has joined #openstack-nova21:02
*** eharney has joined #openstack-nova21:03
mriedemeasy bug fix here https://review.openstack.org/#/c/638821/21:04
sean-k-mooneyartom: what im hpoing we can do in train is take all the guess work out of when the event will be sent and have neutron tell us but i need to bring this up with the neutron folks and work with them to create a spec.21:04
*** artom has quit IRC21:08
sean-k-mooneyi know we made the DriverBlockDevice inherit from dict to make tanstioning to it simpler but at some point it would be better to bit the bullet and just make it a reall nova object/ovo21:08
*** pcaruana has quit IRC21:10
*** wolverineav has quit IRC21:11
*** wolverineav has joined #openstack-nova21:29
*** janki has quit IRC21:29
*** ivve has quit IRC21:44
openstackgerritChris Friesen proposed openstack/nova master: Validate PCI aliases early in resize  https://review.openstack.org/64113121:45
cfriesenmriedem: ^21:45
*** rchurch_ has joined #openstack-nova21:45
openstackgerritAdam Spiers proposed openstack/nova master: Add new "supports_amd_sev" capability to libvirt driver  https://review.openstack.org/63868021:46
*** xek has quit IRC21:46
*** rchurch has quit IRC21:47
*** wolverineav has quit IRC21:47
*** wolverineav has joined #openstack-nova21:48
*** markvoelker has joined #openstack-nova21:49
mriedemcfriesen: comments inline21:49
*** mchlumsky has quit IRC21:50
cfriesenmriedem: wasn't sure whether to copy the other docstring or refer to it. :)21:50
cfriesenwill update based on other comments...do you think I should test that pci is not validated by default?21:51
mriedemwell, if you changed validate_pci=True, i'm guessing some existing tests for create/rebuild would fail b/c they haven't mocked something out?21:51
mriedemif so, then you probably don't need an explicit test for that21:52
cfriesenworks21:52
*** wolverineav has quit IRC21:53
openstackgerritMatt Riedemann proposed openstack/nova master: De-cruftify the finish_resize methods  https://review.openstack.org/63534921:59
openstackgerritMatt Riedemann proposed openstack/nova master: Delete the obj_as_admin context manager  https://review.openstack.org/63535121:59
openstackgerritMatt Riedemann proposed openstack/nova master: neutron: make migrate_instance_finish call setup_instance_network_on_host  https://review.openstack.org/63534321:59
openstackgerritMatt Riedemann proposed openstack/nova master: Temporarily mutate migration object in finish_revert_resize  https://review.openstack.org/64113721:59
openstackgerritMatt Riedemann proposed openstack/nova master: Use errors_out_migration decorator on finish_resize  https://review.openstack.org/64113821:59
*** priteau has quit IRC22:01
mriedemdansmith: you might want to gander at https://bugs.launchpad.net/nova/+bug/181823922:05
openstackLaunchpad bug 1818239 in nova (Ubuntu) "scheduler: build failure high negative weighting" [High,Triaged]22:05
*** wolverineav has joined #openstack-nova22:12
*** eharney has quit IRC22:14
openstackgerritMatt Riedemann proposed openstack/nova master: Fix WeighedHost logging regression  https://review.openstack.org/64114322:18
mriedemsuper easy regression fix that i've now seen in 2 reported bugs ^22:18
*** markvoelker has quit IRC22:22
*** mvkr has quit IRC22:25
*** erlon has quit IRC22:25
*** sdake has quit IRC22:28
mriedemsomeone want to send the bw provider microversion change to it's watery grave? https://review.openstack.org/#/c/636360/22:29
*** sdake has joined #openstack-nova22:30
dansmithmriedem: I think I give up22:30
mriedemon that build failure weigher? heh i know22:30
mriedem"this seems bad if you have build failures...."22:30
mriedemcoreycb: maybe you and/or jamespage want to talk with us about bug 181823922:31
openstackbug 1818239 in nova (Ubuntu) "scheduler: build failure high negative weighting" [High,Triaged] https://launchpad.net/bugs/181823922:31
dansmithit also seems to assume that you can deploy with the completely standard config and expect everything to be fine, regardless of size or the bits of config you want to tweak22:31
dansmithtotally cool with charms defaulting it off22:31
*** sdake has joined #openstack-nova22:32
*** sdake has quit IRC22:32
dansmithI tried to make the rationale for the default weight completely sane based on other default weighers and such22:32
dansmithbut I dunno what else to do, other than engineer the periodic de-de-prioritizing in the compute manager we discussed22:32
*** mmethot has quit IRC22:34
dansmiththe latter of course will demand some config option to allow tweaking the bleed rate22:35
openstackgerritChris Friesen proposed openstack/nova master: Validate PCI aliases early in resize  https://review.openstack.org/64113122:35
mriedemi tried to summarize https://bugs.launchpad.net/ubuntu/+source/nova/+bug/1818239/comments/1622:35
openstackLaunchpad bug 1818239 in nova (Ubuntu) "scheduler: build failure high negative weighting" [High,Triaged]22:35
openstackgerritMichael Still proposed openstack/nova master: Move set_vf_interface_vlan to be with its only caller.  https://review.openstack.org/62422922:35
openstackgerritMichael Still proposed openstack/nova master: Move DHCP releasing to privsep.  https://review.openstack.org/62423022:35
openstackgerritMichael Still proposed openstack/nova master: Convert additional IP management calls to privsep.  https://review.openstack.org/62423122:35
openstackgerritMichael Still proposed openstack/nova master: Move route management to privsep.  https://review.openstack.org/62423222:35
openstackgerritMichael Still proposed openstack/nova master: Move additional IP address management to privsep.  https://review.openstack.org/62459122:35
openstackgerritMichael Still proposed openstack/nova master: Move slight bonkers IP management to privsep.  https://review.openstack.org/62459222:35
openstackgerritMichael Still proposed openstack/nova master: Move arping to privsep.  https://review.openstack.org/62459322:36
openstackgerritMichael Still proposed openstack/nova master: Move cleaning conntrack to privsep.  https://review.openstack.org/62459422:36
openstackgerritMichael Still proposed openstack/nova master: Privsepify ipv4 forwarding enablement.  https://review.openstack.org/63543122:36
openstackgerritMichael Still proposed openstack/nova master: Move killing processes to privsep.  https://review.openstack.org/63543222:36
openstackgerritMichael Still proposed openstack/nova master: Move FP device creation and deletion to privsep.  https://review.openstack.org/63543322:36
openstackgerritMichael Still proposed openstack/nova master: Cleanup the exec_ebtables code a little.  https://review.openstack.org/63543422:36
openstackgerritMichael Still proposed openstack/nova master: Privsep the ebtables modification code.  https://review.openstack.org/63543522:36
openstackgerritMichael Still proposed openstack/nova master: Move adding vlans to interfaces to privsep.  https://review.openstack.org/63543622:36
openstackgerritMichael Still proposed openstack/nova master: Move iptables rule fetching and setting to privsep.  https://review.openstack.org/63650822:36
openstackgerritMichael Still proposed openstack/nova master: Move dnsmasq restarts to privsep.  https://review.openstack.org/63928022:36
openstackgerritMichael Still proposed openstack/nova master: Move router advertisement daemon restarts to privsep.  https://review.openstack.org/63928122:36
openstackgerritMichael Still proposed openstack/nova master: Move calls to ovs-vsctl to privsep.  https://review.openstack.org/63928222:36
dansmithmriedem: ack thanks22:36
*** sapd1 has joined #openstack-nova22:36
mriedemno one puts mikal in a corner22:37
*** sdake has joined #openstack-nova22:37
openstackgerritMatt Riedemann proposed openstack/nova master: Validate PCI aliases early in resize  https://review.openstack.org/64113122:39
cfriesenmriedem: gah...thanks for the fixup22:40
*** sdake has quit IRC22:45
*** wolverineav has quit IRC22:47
*** sdake has joined #openstack-nova22:51
*** munimeha1 has quit IRC22:52
*** tkajinam has joined #openstack-nova22:54
efriedDo we allow backports of fixes to client libs, and corresponding requirements bumps in nova??22:58
mriedemyes on the former, no on the latter22:59
efriedswhat I thought, thanks.23:00
efriedmriedem: fyi https://review.openstack.org/#/c/640879/2/nova/virt/ironic/client_wrapper.py@10623:00
efriedmriedem: I guess deployers wanting the bug fix can pull in the updated client lib version, even though it exceeds upper constraints for that release?23:01
efriedor do we bump upper constraints for the release? /me confused.23:01
mriedemwe can bump upper-constraints in stalbe23:02
mriedem*stalbe23:02
mriedemgah23:02
mriedemwe don't bump lower-constraints23:02
mriedembecause those could be frozen for people at GA23:02
*** sdake has quit IRC23:02
efriedokay, got it.23:03
mriedemso yeah they can backport the ironicclient fix and release it with a patch version on stable23:03
mriedembut we can't depend on it with a requirements/lower-constraints update on stable23:03
mriedemif we can hack around it in a backportable way and then fix on master-only with a lower-constraints bump on ironicclient, then sure23:03
efriedack23:05
*** igordc has quit IRC23:10
*** markvoelker has joined #openstack-nova23:19
*** wolverineav has joined #openstack-nova23:24
*** wolverineav has quit IRC23:30
*** sdake has joined #openstack-nova23:33
*** takashin has joined #openstack-nova23:35
*** derekh has quit IRC23:36
*** wolverineav has joined #openstack-nova23:38
*** tesseract has quit IRC23:42
*** markvoelker has quit IRC23:52
mriedemWOO F TO THE HOO23:53
mriedemtime to fix a multi-cell bug in our functional tests23:55
openstackgerritMerged openstack/nova master: Parse <emulator> elements from virConnectGetCapabilities()  https://review.openstack.org/64048323:56
eanderssonAnyone else having issues with soft-anti-affinity etc in Rocky?23:56
eandersson> Invalid input for field/attribute 0. Value: soft-anti-affinity. u'soft-anti-affinity' is not one of ['anti-affinity', 'affinity']23:57
eandersson> openstack --os-compute-api-version 2.15 server group create anti-affinity-test --policy soft-anti-affinity23:57
openstackgerritMerged openstack/nova master: Correct instance port binding for rebuilds  https://review.openstack.org/60384423:57
eanderssonTested both with openstack client and terraform23:57
openstackgerritMerged openstack/nova master: Fix resetting non-persistent fields when saving obj  https://review.openstack.org/63627123:57
mriedemeandersson: that looks correct https://docs.openstack.org/python-openstackclient/latest/cli/command-objects/server-group.html#server-group-create23:58
mriedemwhich version of osc?23:58
eandersson3.16.323:58
eanderssonIt was working before the upgrade with the same clients23:58
mriedemhmm, this is the last change to the server groups module in osc https://github.com/openstack/python-openstackclient/commit/d0917cd14b94ce13fd92237490efb5722761eda4#diff-f8f1dd5d07b16e47671071cb4fede85d23:59
*** wolverineav has quit IRC23:59
eanderssonWe havne't changed osc version23:59
eanderssonjust the nova version23:59
eanderssonwent to nova rocky23:59
*** wolverineav has joined #openstack-nova23:59
eanderssonbrb23:59
mriedemthat error isn't from nova23:59

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!